From b1ec5525bcf2fb101c57d93beeba26c8c4420638 Mon Sep 17 00:00:00 2001 From: 3gg <3gg@shellblade.net> Date: Sun, 4 May 2025 12:06:00 -0700 Subject: Textures WIP --- Spear.cabal | 42 +- Spear/Asset/Image.hsc | 86 + Spear/Assets/Image.hsc | 126 - Spear/Assets/Image/BMP/BMP_load.c | 257 - Spear/Assets/Image/BMP/BMP_load.h | 23 - Spear/Assets/Image/Image.c | 8 - Spear/Assets/Image/Image.h | 32 - Spear/Assets/Image/Image_error_code.h | 15 - Spear/Assets/Image/sys_types.h | 16 - Spear/Assets/Model.hsc | 440 -- Spear/Assets/Model/MD2/MD2_load.c | 480 -- Spear/Assets/Model/MD2/MD2_load.h | 22 - Spear/Assets/Model/Model.c | 112 - Spear/Assets/Model/Model.h | 100 - Spear/Assets/Model/Model_error_code.h | 16 - Spear/Assets/Model/OBJ/Makefile | 15 - Spear/Assets/Model/OBJ/OBJ_load.c | 274 - Spear/Assets/Model/OBJ/OBJ_load.h | 25 - Spear/Assets/Model/OBJ/cvector.c | 90 - Spear/Assets/Model/OBJ/cvector.h | 36 - Spear/Assets/Model/sys_types.h | 16 - Spear/GL.hs | 800 --- Spear/Math/Collision.hs | 239 - Spear/Math/Octree.hs | 228 - Spear/Render/AnimatedModel.hs | 262 - Spear/Render/Asset.hs | 65 + Spear/Render/Box.hs | 193 - Spear/Render/Core.hs | 2 + Spear/Render/Core/Shader.hs | 58 +- Spear/Render/Core/State.hs | 58 +- Spear/Render/Core/Texture.hs | 235 + Spear/Render/Immediate.hs | 16 +- Spear/Render/Light.hs | 25 - Spear/Render/Material.hs | 16 - Spear/Render/Model.hsc | 54 - Spear/Render/Program.hs | 102 - Spear/Render/RenderModel.c | 232 - Spear/Render/RenderModel.h | 49 - Spear/Render/Shaders/immediate_mode.frag | 23 +- Spear/Render/Sphere.hs | 45 - Spear/Render/StaticModel.hs | 152 - Spear/Render/Triangle.hs | 8 - Spear/Scene/Graph.hs | 143 - Spear/Scene/Loader.hs | 364 -- Spear/Scene/SceneResources.hs | 70 - Spear/cbits/stb/LICENSE | 16 + Spear/cbits/stb/stb_image.c | 2 + Spear/cbits/stb/stb_image.h | 7988 +++++++++++++++++++++++ Spear/old/Spear/Assets/Image.hsc | 126 + Spear/old/Spear/Assets/Image/BMP/BMP_load.c | 257 + Spear/old/Spear/Assets/Image/BMP/BMP_load.h | 23 + Spear/old/Spear/Assets/Image/Image.c | 8 + Spear/old/Spear/Assets/Image/Image.h | 32 + Spear/old/Spear/Assets/Image/Image_error_code.h | 15 + Spear/old/Spear/Assets/Image/sys_types.h | 16 + Spear/old/Spear/Assets/Model.hsc | 440 ++ Spear/old/Spear/Assets/Model/MD2/MD2_load.c | 480 ++ Spear/old/Spear/Assets/Model/MD2/MD2_load.h | 22 + Spear/old/Spear/Assets/Model/Model.c | 112 + Spear/old/Spear/Assets/Model/Model.h | 100 + Spear/old/Spear/Assets/Model/Model_error_code.h | 16 + Spear/old/Spear/Assets/Model/OBJ/Makefile | 15 + Spear/old/Spear/Assets/Model/OBJ/OBJ_load.c | 274 + Spear/old/Spear/Assets/Model/OBJ/OBJ_load.h | 25 + Spear/old/Spear/Assets/Model/OBJ/cvector.c | 90 + Spear/old/Spear/Assets/Model/OBJ/cvector.h | 36 + Spear/old/Spear/Assets/Model/sys_types.h | 16 + Spear/old/Spear/GL.hs | 800 +++ Spear/old/Spear/Math/Collision.hs | 239 + Spear/old/Spear/Math/Octree.hs | 228 + Spear/old/Spear/Render/AnimatedModel.hs | 262 + Spear/old/Spear/Render/Box.hs | 193 + Spear/old/Spear/Render/Light.hs | 25 + Spear/old/Spear/Render/Material.hs | 16 + Spear/old/Spear/Render/Model.hsc | 54 + Spear/old/Spear/Render/Program.hs | 102 + Spear/old/Spear/Render/RenderModel.c | 232 + Spear/old/Spear/Render/RenderModel.h | 49 + Spear/old/Spear/Render/Sphere.hs | 45 + Spear/old/Spear/Render/StaticModel.hs | 152 + Spear/old/Spear/Scene/Graph.hs | 143 + Spear/old/Spear/Scene/Loader.hs | 364 ++ Spear/old/Spear/Scene/SceneResources.hs | 70 + 83 files changed, 13609 insertions(+), 5144 deletions(-) create mode 100644 Spear/Asset/Image.hsc delete mode 100644 Spear/Assets/Image.hsc delete mode 100644 Spear/Assets/Image/BMP/BMP_load.c delete mode 100644 Spear/Assets/Image/BMP/BMP_load.h delete mode 100644 Spear/Assets/Image/Image.c delete mode 100644 Spear/Assets/Image/Image.h delete mode 100644 Spear/Assets/Image/Image_error_code.h delete mode 100644 Spear/Assets/Image/sys_types.h delete mode 100644 Spear/Assets/Model.hsc delete mode 100644 Spear/Assets/Model/MD2/MD2_load.c delete mode 100644 Spear/Assets/Model/MD2/MD2_load.h delete mode 100644 Spear/Assets/Model/Model.c delete mode 100644 Spear/Assets/Model/Model.h delete mode 100644 Spear/Assets/Model/Model_error_code.h delete mode 100644 Spear/Assets/Model/OBJ/Makefile delete mode 100644 Spear/Assets/Model/OBJ/OBJ_load.c delete mode 100644 Spear/Assets/Model/OBJ/OBJ_load.h delete mode 100644 Spear/Assets/Model/OBJ/cvector.c delete mode 100644 Spear/Assets/Model/OBJ/cvector.h delete mode 100644 Spear/Assets/Model/sys_types.h delete mode 100644 Spear/GL.hs delete mode 100644 Spear/Math/Collision.hs delete mode 100644 Spear/Math/Octree.hs delete mode 100644 Spear/Render/AnimatedModel.hs create mode 100644 Spear/Render/Asset.hs delete mode 100644 Spear/Render/Box.hs create mode 100644 Spear/Render/Core/Texture.hs delete mode 100644 Spear/Render/Light.hs delete mode 100644 Spear/Render/Material.hs delete mode 100644 Spear/Render/Model.hsc delete mode 100644 Spear/Render/Program.hs delete mode 100644 Spear/Render/RenderModel.c delete mode 100644 Spear/Render/RenderModel.h delete mode 100644 Spear/Render/Sphere.hs delete mode 100644 Spear/Render/StaticModel.hs delete mode 100644 Spear/Render/Triangle.hs delete mode 100644 Spear/Scene/Graph.hs delete mode 100644 Spear/Scene/Loader.hs delete mode 100644 Spear/Scene/SceneResources.hs create mode 100644 Spear/cbits/stb/LICENSE create mode 100644 Spear/cbits/stb/stb_image.c create mode 100644 Spear/cbits/stb/stb_image.h create mode 100644 Spear/old/Spear/Assets/Image.hsc create mode 100644 Spear/old/Spear/Assets/Image/BMP/BMP_load.c create mode 100644 Spear/old/Spear/Assets/Image/BMP/BMP_load.h create mode 100644 Spear/old/Spear/Assets/Image/Image.c create mode 100644 Spear/old/Spear/Assets/Image/Image.h create mode 100644 Spear/old/Spear/Assets/Image/Image_error_code.h create mode 100644 Spear/old/Spear/Assets/Image/sys_types.h create mode 100644 Spear/old/Spear/Assets/Model.hsc create mode 100644 Spear/old/Spear/Assets/Model/MD2/MD2_load.c create mode 100644 Spear/old/Spear/Assets/Model/MD2/MD2_load.h create mode 100644 Spear/old/Spear/Assets/Model/Model.c create mode 100644 Spear/old/Spear/Assets/Model/Model.h create mode 100644 Spear/old/Spear/Assets/Model/Model_error_code.h create mode 100644 Spear/old/Spear/Assets/Model/OBJ/Makefile create mode 100644 Spear/old/Spear/Assets/Model/OBJ/OBJ_load.c create mode 100644 Spear/old/Spear/Assets/Model/OBJ/OBJ_load.h create mode 100644 Spear/old/Spear/Assets/Model/OBJ/cvector.c create mode 100644 Spear/old/Spear/Assets/Model/OBJ/cvector.h create mode 100644 Spear/old/Spear/Assets/Model/sys_types.h create mode 100644 Spear/old/Spear/GL.hs create mode 100644 Spear/old/Spear/Math/Collision.hs create mode 100644 Spear/old/Spear/Math/Octree.hs create mode 100644 Spear/old/Spear/Render/AnimatedModel.hs create mode 100644 Spear/old/Spear/Render/Box.hs create mode 100644 Spear/old/Spear/Render/Light.hs create mode 100644 Spear/old/Spear/Render/Material.hs create mode 100644 Spear/old/Spear/Render/Model.hsc create mode 100644 Spear/old/Spear/Render/Program.hs create mode 100644 Spear/old/Spear/Render/RenderModel.c create mode 100644 Spear/old/Spear/Render/RenderModel.h create mode 100644 Spear/old/Spear/Render/Sphere.hs create mode 100644 Spear/old/Spear/Render/StaticModel.hs create mode 100644 Spear/old/Spear/Scene/Graph.hs create mode 100644 Spear/old/Spear/Scene/Loader.hs create mode 100644 Spear/old/Spear/Scene/SceneResources.hs diff --git a/Spear.cabal b/Spear.cabal index 56eb302..0be03fe 100644 --- a/Spear.cabal +++ b/Spear.cabal @@ -19,6 +19,7 @@ library OpenGL >= 3, OpenGLRaw -any, StateVar -any, + --stb-image >= 0.3.0, TODO: Remove base -any, bytestring -any, directory -any, @@ -37,20 +38,16 @@ library exposed-modules: Spear.App - Spear.Assets.Image - Spear.Assets.Model + Spear.Asset.Image Spear.Game - Spear.GL Spear.Math.AABB Spear.Math.Algebra Spear.Math.Camera Spear.Math.Circle - Spear.Math.Collision Spear.Math.Frustum Spear.Math.Matrix3 Spear.Math.Matrix4 Spear.Math.MatrixUtils - Spear.Math.Octree Spear.Math.Plane Spear.Math.Quaternion Spear.Math.Ray @@ -69,7 +66,7 @@ library Spear.Physics.Collision Spear.Physics.RigidBody Spear.Prelude - Spear.Render.AnimatedModel + Spear.Render.Asset Spear.Render.Core Spear.Render.Core.Buffer Spear.Render.Core.Constants @@ -77,15 +74,9 @@ library Spear.Render.Core.Pipeline Spear.Render.Core.Shader Spear.Render.Core.State + Spear.Render.Core.Texture Spear.Render.Immediate - Spear.Render.Material - Spear.Render.Model - Spear.Render.Program Spear.Render.Shaders - Spear.Render.StaticModel - Spear.Scene.Graph - Spear.Scene.Loader - Spear.Scene.SceneResources Spear.Sound.Sound Spear.Sound.State Spear.Step @@ -101,36 +92,17 @@ library build-tools: hsc2hs -any c-sources: - Spear/Assets/Image/Image.c - Spear/Assets/Image/BMP/BMP_load.c - Spear/Assets/Model/Model.c - Spear/Assets/Model/MD2/MD2_load.c - Spear/Assets/Model/OBJ/cvector.c - Spear/Assets/Model/OBJ/OBJ_load.c - Spear/Render/RenderModel.c + Spear/cbits/stb/stb_image.c Spear/Sys/Timer/timer.c includes: - Spear/Assets/Image/BMP/BMP_load.h - Spear/Assets/Image/Image.h - Spear/Assets/Image/Image_error_code.h - Spear/Assets/Image/sys_types.h - Spear/Assets/Model/MD2/MD2_load.h - Spear/Assets/Model/OBJ/OBJ_load.h - Spear/Assets/Model/OBJ/cvector.h - Spear/Assets/Model/Model.h - Spear/Assets/Model/Model_error_code.h - Spear/Assets/Model/sys_types.h - Spear/Render/RenderModel.h + cbits/stb/stb_image.h Spear/Sys/Timer/timer.h include-dirs: . Spear - Spear/Assets/Image - Spear/Assets/Image/BMP - Spear/Assets/Model - Spear/Render + Spear/cbits Spear/Sys hs-source-dirs: . diff --git a/Spear/Asset/Image.hsc b/Spear/Asset/Image.hsc new file mode 100644 index 0000000..f907f39 --- /dev/null +++ b/Spear/Asset/Image.hsc @@ -0,0 +1,86 @@ +{-# LANGUAGE CPP, ForeignFunctionInterface #-} + +module Spear.Asset.Image +( + Pixels +, Image(..) +, loadImage +, freeImage +, withImage +, setFlipVerticallyOnLoad +) +where + + +import Control.Monad.IO.Class +import Foreign.C.String +import Foreign.C.Types +import Foreign.Marshal.Alloc +import Foreign.Marshal.Utils +import Foreign.Ptr +import Foreign.Storable + + +type Pixels = Ptr CUChar + +data Image = Image + { width :: Int + , height :: Int + , components :: Int + , pixels :: Pixels + } + + +#include "stb/stb_image.h" + +foreign import ccall safe "stb_image.h stbi_load" + stbi_load :: CString -> Ptr Int -> Ptr Int -> Ptr Int -> Int -> IO Pixels + +foreign import ccall safe "stb_image.h stbi_set_flip_vertically_on_load" + stbi_set_flip_vertically_on_load :: Int -> IO () + +foreign import ccall safe "stb_image.h stbi_free_image" + stbi_free_image :: Pixels -> IO () + +foreign import ccall safe "stbi_image.h stbi_failure_reason" + stbi_failure_reason :: IO CString + + +-- | Load an image. +loadImage :: MonadIO m => FilePath -> m (Either String Image) +loadImage filepath = liftIO $ + withCString filepath $ \cFilePath -> + alloca $ \pWidth -> + alloca $ \pHeight -> + alloca $ \pComponents -> do + pixels <- stbi_load cFilePath pWidth pHeight pComponents 0 + width <- peek pWidth + height <- peek pHeight + components <- peek pComponents + if pixels == nullPtr + then do + err <- stbi_failure_reason + Left <$> peekCString err + else do + return . Right $ Image width height components pixels + +-- | Free the image's pixels from memory. +freeImage :: MonadIO m => Image -> m () +freeImage = liftIO . stbi_free_image . pixels + +-- | Load an image, perform an operation on it, then free the image. +-- On error, return an error message. +withImage :: MonadIO m => FilePath -> (Image -> m a) -> m (Either String a) +withImage filepath f = do + result <- loadImage filepath + case result of + Left err -> return . Left $ err + Right image -> do + a <- f image + freeImage image + return . Right $ a + +-- | Flip subsequently loaded images vertically. +setFlipVerticallyOnLoad :: MonadIO m => Bool -> m () +setFlipVerticallyOnLoad = + liftIO . stbi_set_flip_vertically_on_load . fromIntegral . fromEnum diff --git a/Spear/Assets/Image.hsc b/Spear/Assets/Image.hsc deleted file mode 100644 index db90afe..0000000 --- a/Spear/Assets/Image.hsc +++ /dev/null @@ -1,126 +0,0 @@ -{-# LANGUAGE CPP, ForeignFunctionInterface #-} - -module Spear.Assets.Image -( - -- * Data types - Image - -- * Loading and unloading -, loadImage - -- * Accessors -, width -, height -, bpp -, pixels -) -where - -import Spear.Game -import Foreign.Ptr -import Foreign.Storable -import Foreign.C.Types -import Foreign.C.String -import Foreign.Marshal.Utils as Foreign (with) -import Foreign.Marshal.Alloc (alloca) -import Data.List (splitAt, elemIndex) -import Data.Char (toLower) - -#include "Image.h" -#include "BMP/BMP_load.h" - -data ImageErrorCode - = ImageSuccess - | ImageReadError - | ImageMemoryAllocationError - | ImageFileNotFound - | ImageInvalidFormat - | ImageNoSuitableLoader - deriving (Eq, Enum, Show) - -data CImage = CImage - { cwidth :: CInt - , cheight :: CInt - , cbpp :: CInt - , cpixels :: Ptr CUChar - } - -instance Storable CImage where - sizeOf _ = #{size Image} - alignment _ = alignment (undefined :: CInt) - - peek ptr = do - width <- #{peek Image, width} ptr - height <- #{peek Image, height} ptr - bpp <- #{peek Image, bpp} ptr - pixels <- #{peek Image, pixels} ptr - return $ CImage width height bpp pixels - - poke ptr (CImage width height bpp pixels) = do - #{poke Image, width} ptr width - #{poke Image, height} ptr height - #{poke Image, bpp} ptr bpp - #{poke Image, pixels} ptr pixels - --- | An image resource. -data Image = Image - { imageData :: CImage - , rkey :: ReleaseKey - } - -instance ResourceClass Image where - getResource = rkey - -foreign import ccall "Image.h image_free" - image_free :: Ptr CImage -> IO () - -foreign import ccall "BMP_load.h BMP_load" - bmp_load' :: Ptr CChar -> Ptr CImage -> IO Int - -bmp_load :: Ptr CChar -> Ptr CImage -> IO ImageErrorCode -bmp_load file image = bmp_load' file image >>= \code -> return . toEnum $ code - --- | Load the image specified by the given file. -loadImage :: FilePath -> Game s Image -loadImage file = do - dotPos <- case elemIndex '.' file of - Nothing -> gameError $ "file name has no extension: " ++ file - Just p -> return p - - let ext = map toLower . tail . snd $ splitAt dotPos file - - result <- liftIO . alloca $ \ptr -> do - status <- withCString file $ \fileCstr -> do - case ext of - "bmp" -> bmp_load fileCstr ptr - _ -> return ImageNoSuitableLoader - - case status of - ImageSuccess -> peek ptr >>= return . Right - ImageReadError -> return . Left $ "read error" - ImageMemoryAllocationError -> return . Left $ "memory allocation error" - ImageFileNotFound -> return . Left $ "file not found" - ImageInvalidFormat -> return . Left $ "invalid format" - ImageNoSuitableLoader -> return . Left $ "no suitable loader for extension " ++ ext - - case result of - Right image -> register (freeImage image) >>= return . Image image - Left err -> gameError $ "loadImage: " ++ err - --- | Free the given 'CImage'. -freeImage :: CImage -> IO () -freeImage image = Foreign.with image image_free - --- | Return the given image's width. -width :: Image -> Int -width = fromIntegral . cwidth . imageData - --- | Return the given image's height. -height :: Image -> Int -height = fromIntegral . cheight . imageData - --- | Return the given image's bits per pixel. -bpp :: Image -> Int -bpp = fromIntegral . cbpp . imageData - --- | Return the given image's pixels. -pixels :: Image -> Ptr CUChar -pixels = cpixels . imageData diff --git a/Spear/Assets/Image/BMP/BMP_load.c b/Spear/Assets/Image/BMP/BMP_load.c deleted file mode 100644 index 5c1b195..0000000 --- a/Spear/Assets/Image/BMP/BMP_load.c +++ /dev/null @@ -1,257 +0,0 @@ -#include "BMP_load.h" -#include -#include -#include -#include - -#define BITMAP_ID 0x4D42 - - -/// Bitmap file header structure. -typedef struct -{ - U16 type; // Specifies the image type; must be BM (0x4D42). - U32 size; // Specifies the size in bytes of the bitmap file. - U32 reserved; // Reserved; must be zero. - U32 offBits; // Specifies the offset, in bytes, from the BitmapFileHeader structure to the bitmap bits. -} -BitmapFileHeader; - - -/// Bitmap info header structure. -typedef struct -{ - U32 size; // Specifies the number of bytes required by the structure. - U32 width; // Specifies the width of the bitmap, in pixels. - U32 height; // Specifies the height of the bitmap, in pixels. - U16 planes; // Specifies the number of color planes; must be 1. - U16 bitCount; // Specifies the number of bits per pixel; must be 1, 4, 16, 24, or 32. - U32 compression; // Specifies the type of compression. - U32 imageSize; // Specifies the size of the image in bytes. - U32 xPelsPerMeter; // Specifies the number of pixels per meter on the x axis. - U32 yPelsPerMeter; // Specifies the number of pixels per meter on the y axis. - U32 clrUsed; // Specifies the number of colours used by the bitmap. - U32 clrImportant; // Specifies the number of colours that are important. -} -BitmapInfoHeader; - - -static void safe_free (void* ptr) -{ - if (ptr) free (ptr); -} - - -static Image_error_code read_raw_data( - FILE* filePtr, const BitmapFileHeader* bitmapFileHeader, - const BitmapInfoHeader* bitmapInfoHeader, U8** data) -{ - U8* bitmapImage; - U8* auxrow; - size_t row_size = bitmapInfoHeader->width * 3; - size_t numBytes = bitmapInfoHeader->height * row_size; - size_t bytes_read; - - // Allocate memory for the bitmap data and the auxiliary row. - bitmapImage = (U8*) malloc (numBytes); - auxrow = (U8*) malloc (row_size); - if (!bitmapImage || !auxrow) - { - safe_free (bitmapImage); - safe_free (auxrow); - return Image_Memory_Allocation_Error; - } - - // Move the file pointer to the beginning of bitmap data and read the data. - fseek(filePtr, bitmapFileHeader->offBits, SEEK_SET); - bytes_read = fread(bitmapImage, 1, numBytes, filePtr); - if (bytes_read != numBytes) - { - free(bitmapImage); - return Image_Read_Error; - } - - size_t i; - - // Reverse rows. - /*size_t h = bitmapInfoHeader->height / 2; - for (i = 0; i < h; ++i) - { - U8* row1 = bitmapImage + i * row_size; - U8* row2 = bitmapImage + (bitmapInfoHeader->height - i - 1) * row_size; - - memcpy (auxrow, row1, row_size); - memcpy (row1, row2, row_size); - memcpy (row2, auxrow, row_size); - }*/ - - // Swap B and R channels. BGR -> RGB. - for (i = 0; i < bytes_read; i+= 3) - { - U8 tmp = bitmapImage[i]; - bitmapImage[i] = bitmapImage[i+2]; - bitmapImage[i+2] = tmp; - } - - *data = bitmapImage; - - return Image_Success; -} - - -static Image_error_code read_paletised_data8 - (FILE* filePtr, const BitmapInfoHeader *bitmapInfoHeader, U8** data) -{ - U8* bitmapImage; - U8* palette; - U8* colourIndices; - size_t bytes_read; - - size_t paletteSize = pow(2, bitmapInfoHeader->bitCount) * 4; - size_t colourIndicesSize = bitmapInfoHeader->width * bitmapInfoHeader->height; - int bitmapImageSize = colourIndicesSize * 3; - - // Save memory for the palette, colour indices and bitmap image. - palette = (U8*) malloc (paletteSize); - colourIndices = (U8*) malloc(colourIndicesSize); - bitmapImage = (U8*) malloc(bitmapImageSize); - if (!palette | !colourIndices || !bitmapImage) - { - safe_free (palette); - safe_free (colourIndices); - safe_free (bitmapImage); - return Image_Memory_Allocation_Error; - } - - // Read the colour palette. - bytes_read = fread(palette, 1, paletteSize, filePtr); - if (bytes_read != paletteSize) return Image_Read_Error; - - // Read the colour indices. - bytes_read = fread(colourIndices, 1, colourIndicesSize, filePtr); - if (bytes_read != colourIndicesSize) return Image_Read_Error; - - // Decode the image data. - U8* imgptr = &bitmapImage[bitmapImageSize - (bitmapInfoHeader->width * 4)]; - - size_t i; - for (i = 0; i < colourIndicesSize; i++) - { - int index = colourIndices[i]; - - memcpy(imgptr, (const void*) &palette[index * 4], 3); - imgptr += 3; - - if (!((i+1) % bitmapInfoHeader->width)) - { - imgptr -= (bitmapInfoHeader->width * 4 * 2); - } - } - - free(palette); - free(colourIndices); - - *data = bitmapImage; - - return Image_Success; -} - - -Image_error_code BMP_load (const char* filename, Image* image) -{ - FILE* filePtr; - BitmapFileHeader bitmapFileHeader; - BitmapInfoHeader bitmapInfoHeader; - U8 buf[40]; - U8* data; - - // Open the file in binary read-only mode. - filePtr = fopen(filename, "rb"); - if (!filePtr) return Image_File_Not_Found; - - if ((fread(buf, 14, 1, filePtr)) != 1) - { - fclose(filePtr); - return Image_Read_Error; - } - - bitmapFileHeader.type = *((U16*)buf); - bitmapFileHeader.size = *((U32*)(buf+2)); - bitmapFileHeader.reserved = *((U32*)(buf+6)); - bitmapFileHeader.offBits = *((U32*)(buf+10)); - - // Check that this is in fact a BMP file. - if (bitmapFileHeader.type != BITMAP_ID) - { - fprintf(stderr, "Not a valid BMP file\n"); - fclose(filePtr); - return Image_Invalid_Format; - } - - if ((fread(&buf, 40, 1, filePtr)) != 1) - { - fclose(filePtr); - return Image_Read_Error; - } - - bitmapInfoHeader.size = *((U32*)(buf)); - bitmapInfoHeader.width = *((U32*)(buf+4)); - bitmapInfoHeader.height = *((U32*)(buf+8)); - bitmapInfoHeader.planes = *((U16*)(buf+12)); - bitmapInfoHeader.bitCount = *((U16*)(buf+14)); - bitmapInfoHeader.compression = *((U32*)(buf+16)); - bitmapInfoHeader.imageSize = *((U32*)(buf+20)); - bitmapInfoHeader.xPelsPerMeter = *((U32*)(buf+24)); - bitmapInfoHeader.yPelsPerMeter = *((U32*)(buf+28)); - bitmapInfoHeader.clrUsed = *((U32*)(buf+32)); - bitmapInfoHeader.clrImportant = *((U32*)(buf+36)); - - // Check that no compression is used. - // Compression is not supported at the moment. - if (bitmapInfoHeader.compression != 0) - { - fprintf(stderr, "Compression not supported\n"); - fclose(filePtr); - return Image_Invalid_Format; - } - - // Check that this is a Windows BMP file. - // Other formats are not supported. - if (bitmapInfoHeader.size != 40) - { - fprintf(stderr, "Only Windows BMP files supported\n"); - fclose(filePtr); - return Image_Invalid_Format; - } - - Image_error_code status; - - if (bitmapInfoHeader.bitCount == 8) - { - // The BMP file uses a colour palette. - // We are already positioned at the colour palette. - status = read_paletised_data8 (filePtr, &bitmapInfoHeader, &data); - } - else if (bitmapInfoHeader.bitCount >= 16) - { - // The BMP file uses no colour palette. - status = read_raw_data (filePtr, &bitmapFileHeader, &bitmapInfoHeader, &data); - } - else - { - fprintf(stderr, "Only 24-bit and 16-bit palette images supported\n"); - fclose(filePtr); - return Image_Invalid_Format; - } - - fclose(filePtr); - - if (data == 0) return status; - - image->width = bitmapInfoHeader.width; - image->height = bitmapInfoHeader.height; - image->bpp = 3; - image->pixels = data; - - return Image_Success; -} diff --git a/Spear/Assets/Image/BMP/BMP_load.h b/Spear/Assets/Image/BMP/BMP_load.h deleted file mode 100644 index f4ad32f..0000000 --- a/Spear/Assets/Image/BMP/BMP_load.h +++ /dev/null @@ -1,23 +0,0 @@ -#ifndef _BMP_LOAD_H -#define _BMP_LOAD_H - - -#include "../Image.h" -#include "../Image_error_code.h" - - -#ifdef __cplusplus -extern "C" { -#endif - - -/// Loads the BMP file specified by the given string. -/// (0,0) corresponds to the top left corner of the image. -Image_error_code BMP_load (const char* filename, Image* image); - - -#ifdef __cplusplus -} -#endif - -#endif // _BMP_LOAD_H diff --git a/Spear/Assets/Image/Image.c b/Spear/Assets/Image/Image.c deleted file mode 100644 index f4150e1..0000000 --- a/Spear/Assets/Image/Image.c +++ /dev/null @@ -1,8 +0,0 @@ -#include "Image.h" -#include - - -void image_free (Image* image) -{ - free (image->pixels); -} diff --git a/Spear/Assets/Image/Image.h b/Spear/Assets/Image/Image.h deleted file mode 100644 index aaca5e9..0000000 --- a/Spear/Assets/Image/Image.h +++ /dev/null @@ -1,32 +0,0 @@ -#ifndef _SPEAR_IMAGE_H -#define _SPEAR_IMAGE_H - -#include "sys_types.h" - - -typedef struct -{ - int width; - int height; - int bpp; // Bits per pixel. - // If bpp = 3 then format = RGB. - // If bpp = 4 then format = RGBA. - U8* pixels; -} -Image; - - -#ifdef __cplusplus -extern "C" { -#endif - -/// Frees the given Image from memory. -/// The 'image' pointer itself is not freed. -void image_free (Image* image); - -#ifdef __cplusplus -} -#endif - - -#endif // _SPEAR_IMAGE_H diff --git a/Spear/Assets/Image/Image_error_code.h b/Spear/Assets/Image/Image_error_code.h deleted file mode 100644 index dc54fc2..0000000 --- a/Spear/Assets/Image/Image_error_code.h +++ /dev/null @@ -1,15 +0,0 @@ -#ifndef _SPEAR_IMAGE_ERROR_CODE_H -#define _SPEAR_IMAGE_ERROR_CODE_H - -typedef enum -{ - Image_Success, - Image_Read_Error, - Image_Memory_Allocation_Error, - Image_File_Not_Found, - Image_Invalid_Format, - Image_No_Suitable_Loader, -} -Image_error_code; - -#endif // _SPEAR_IMAGE_ERROR_CODE_H diff --git a/Spear/Assets/Image/sys_types.h b/Spear/Assets/Image/sys_types.h deleted file mode 100644 index 6aca9e9..0000000 --- a/Spear/Assets/Image/sys_types.h +++ /dev/null @@ -1,16 +0,0 @@ -#ifndef _SPEAR_SYS_TYPES_H -#define _SPEAR_SYS_TYPES_H - -#include - -typedef int8_t I8; -typedef int16_t I16; -typedef int32_t I32; -typedef int64_t I64; -typedef uint8_t U8; -typedef uint16_t U16; -typedef uint32_t U32; -typedef uint64_t U64; - -#endif // _SPEAR_SYS_TYPES_H - diff --git a/Spear/Assets/Model.hsc b/Spear/Assets/Model.hsc deleted file mode 100644 index 02e1edf..0000000 --- a/Spear/Assets/Model.hsc +++ /dev/null @@ -1,440 +0,0 @@ -{-# LANGUAGE CPP, ForeignFunctionInterface #-} - -module Spear.Assets.Model -( - -- * Data types - Vec2(..) -, Vec3(..) -, TexCoord(..) -, CTriangle(..) -, Box(..) -, Skin(..) -, Animation(..) -, Triangle(..) -, Model(..) - -- * Loading -, loadModel - -- * Accessors -, animated -, animation -, animationByName -, triangles' - -- * Manipulation -, transformVerts -, transformNormals -, toGround -, modelBoxes -) -where - -import Spear.Game - -import qualified Data.ByteString.Char8 as B -import Data.Char (toLower) -import Data.List (splitAt, elemIndex) -import qualified Data.Vector as V -import qualified Data.Vector.Storable as S -import Foreign.Ptr -import Foreign.Storable -import Foreign.C.Types -import Foreign.C.String -import Foreign.Marshal.Utils as Foreign (with) -import Foreign.Marshal.Alloc (alloca, allocaBytes) -import Foreign.Marshal.Array (allocaArray, copyArray, peekArray) -import Unsafe.Coerce (unsafeCoerce) - -#include "Model.h" -#include "MD2/MD2_load.h" -#include "OBJ/OBJ_load.h" - -data ModelErrorCode - = ModelSuccess - | ModelReadError - | ModelMemoryAllocationError - | ModelFileNotFound - | ModelFileMismatch - | ModelNoSuitableLoader - deriving (Eq, Enum, Show) - -sizeFloat = #{size float} -sizePtr = #{size int*} - --- | A 2D vector. -data Vec2 = Vec2 {-# UNPACK #-} !Float {-# UNPACK #-} !Float - -instance Storable Vec2 where - sizeOf _ = 2*sizeFloat - alignment _ = alignment (undefined :: CFloat) - - peek ptr = do - f0 <- peekByteOff ptr 0 - f1 <- peekByteOff ptr sizeFloat - return $ Vec2 f0 f1 - - poke ptr (Vec2 f0 f1) = do - pokeByteOff ptr 0 f0 - pokeByteOff ptr sizeFloat f1 - --- | A 3D vector. -data Vec3 = Vec3 {-# UNPACK #-} !Float {-# UNPACK #-} !Float {-# UNPACK #-} !Float - -instance Storable Vec3 where - sizeOf _ = 3*sizeFloat - alignment _ = alignment (undefined :: CFloat) - - peek ptr = do - f0 <- peekByteOff ptr 0 - f1 <- peekByteOff ptr sizeFloat - f2 <- peekByteOff ptr (2*sizeFloat) - return $ Vec3 f0 f1 f2 - - poke ptr (Vec3 f0 f1 f2) = do - pokeByteOff ptr 0 f0 - pokeByteOff ptr sizeFloat f1 - pokeByteOff ptr (2*sizeFloat) f2 - --- | A 2D texture coordinate. -data TexCoord = TexCoord {-# UNPACK #-} !Float {-# UNPACK #-} !Float - -instance Storable TexCoord where - sizeOf _ = 2*sizeFloat - alignment _ = alignment (undefined :: CFloat) - - peek ptr = do - f0 <- peekByteOff ptr 0 - f1 <- peekByteOff ptr sizeFloat - return $ TexCoord f0 f1 - - poke ptr (TexCoord f0 f1) = do - pokeByteOff ptr 0 f0 - pokeByteOff ptr sizeFloat f1 - --- | A raw triangle holding vertex/normal and texture indices. -data CTriangle = CTriangle - { vertexIndex0 :: {-# UNPACK #-} !CUShort - , vertexIndex1 :: {-# UNPACK #-} !CUShort - , vertexIndex2 :: {-# UNPACK #-} !CUShort - , textureIndex1 :: {-# UNPACK #-} !CUShort - , textureIndex2 :: {-# UNPACK #-} !CUShort - , textureIndex3 :: {-# UNPACK #-} !CUShort - } - -instance Storable CTriangle where - sizeOf _ = #{size triangle} - alignment _ = alignment (undefined :: CUShort) - - peek ptr = do - v0 <- #{peek triangle, vertexIndices[0]} ptr - v1 <- #{peek triangle, vertexIndices[1]} ptr - v2 <- #{peek triangle, vertexIndices[2]} ptr - - t0 <- #{peek triangle, textureIndices[0]} ptr - t1 <- #{peek triangle, textureIndices[1]} ptr - t2 <- #{peek triangle, textureIndices[2]} ptr - - return $ CTriangle v0 v1 v2 t0 t1 t2 - - poke ptr (CTriangle v0 v1 v2 t0 t1 t2) = do - #{poke triangle, vertexIndices[0]} ptr v0 - #{poke triangle, vertexIndices[1]} ptr v1 - #{poke triangle, vertexIndices[2]} ptr v2 - - #{poke triangle, textureIndices[0]} ptr t0 - #{poke triangle, textureIndices[1]} ptr t1 - #{poke triangle, textureIndices[2]} ptr t2 - --- | A 3D axis-aligned bounding box. -data Box = Box {-# UNPACK #-} !Vec3 {-# UNPACK #-} !Vec3 - -instance Storable Box where - sizeOf _ = 6 * sizeFloat - alignment _ = alignment (undefined :: CFloat) - - peek ptr = do - xmin <- peekByteOff ptr 0 - ymin <- peekByteOff ptr sizeFloat - zmin <- peekByteOff ptr $ 2*sizeFloat - xmax <- peekByteOff ptr $ 3*sizeFloat - ymax <- peekByteOff ptr $ 4*sizeFloat - zmax <- peekByteOff ptr $ 5*sizeFloat - return $ Box (Vec3 xmin ymin zmin) (Vec3 xmax ymax zmax) - - poke ptr (Box (Vec3 xmin ymin zmin) (Vec3 xmax ymax zmax)) = do - pokeByteOff ptr 0 xmin - pokeByteOff ptr sizeFloat ymin - pokeByteOff ptr (2*sizeFloat) zmin - pokeByteOff ptr (3*sizeFloat) xmax - pokeByteOff ptr (4*sizeFloat) ymax - pokeByteOff ptr (5*sizeFloat) zmax - --- | A model skin. -newtype Skin = Skin { skinName :: B.ByteString } - -instance Storable Skin where - sizeOf (Skin s) = 64 - alignment _ = 1 - - peek ptr = do - s <- B.packCString $ unsafeCoerce ptr - return $ Skin s - - poke ptr (Skin s) = do - B.useAsCStringLen s $ \(sptr, len) -> copyArray (unsafeCoerce ptr) sptr len - --- | A model animation. --- --- See also: 'animation', 'animationByName', 'numAnimations'. -data Animation = Animation - { name :: B.ByteString - , start :: Int - , end :: Int - } - -instance Storable Animation where - sizeOf _ = #{size animation} - alignment _ = alignment (undefined :: CUInt) - - peek ptr = do - name <- B.packCString (unsafeCoerce ptr) - start <- #{peek animation, start} ptr - end <- #{peek animation, end} ptr - return $ Animation name start end - - poke ptr (Animation name start end) = do - B.useAsCStringLen name $ \(sptr, len) -> copyArray (unsafeCoerce ptr) sptr len - #{poke animation, start} ptr start - #{poke animation, end} ptr end - --- | A 3D model. -data Model = Model - { vertices :: S.Vector Vec3 -- ^ Array of 'numFrames' * 'numVerts' vertices. - , normals :: S.Vector Vec3 -- ^ Array of 'numFrames' * 'numVerts' normals. - , texCoords :: S.Vector TexCoord -- ^ Array of 'numTexCoords' texture coordinates. - , triangles :: S.Vector CTriangle -- ^ Array of 'numTriangles' triangles. - , skins :: S.Vector Skin -- ^ Array of 'numSkins' skins. - , animations :: S.Vector Animation -- ^ Array of 'numAnimations' animations. - , numFrames :: Int -- ^ Number of frames. - , numVerts :: Int -- ^ Number of vertices (and normals) per frame. - , numTriangles :: Int -- ^ Number of triangles in one frame. - , numTexCoords :: Int -- ^ Number of texture coordinates in one frame. - , numSkins :: Int -- ^ Number of skins. - , numAnimations :: Int -- ^ Number of animations. - } - -instance Storable Model where - sizeOf _ = #{size Model} - alignment _ = alignment (undefined :: CUInt) - - peek ptr = do - numFrames <- #{peek Model, numFrames} ptr - numVertices <- #{peek Model, numVertices} ptr - numTriangles <- #{peek Model, numTriangles} ptr - numTexCoords <- #{peek Model, numTexCoords} ptr - numSkins <- #{peek Model, numSkins} ptr - numAnimations <- #{peek Model, numAnimations} ptr - pVerts <- peek (unsafeCoerce ptr) - pNormals <- peekByteOff ptr sizePtr - pTexCoords <- peekByteOff ptr (2*sizePtr) - pTriangles <- peekByteOff ptr (3*sizePtr) - pSkins <- peekByteOff ptr (4*sizePtr) - pAnimations <- peekByteOff ptr (5*sizePtr) - vertices <- fmap S.fromList $ peekArray (numVertices*numFrames) pVerts - normals <- fmap S.fromList $ peekArray (numVertices*numFrames) pNormals - texCoords <- fmap S.fromList $ peekArray numTexCoords pTexCoords - triangles <- fmap S.fromList $ peekArray numTriangles pTriangles - skins <- fmap S.fromList $ peekArray numSkins pSkins - animations <- fmap S.fromList $ peekArray numAnimations pAnimations - return $ - Model vertices normals texCoords triangles skins animations - numFrames numVertices numTriangles numTexCoords numSkins numAnimations - - poke ptr - (Model verts normals texCoords tris skins animations - numFrames numVerts numTris numTex numSkins numAnimations) = - S.unsafeWith verts $ \pVerts -> - S.unsafeWith normals $ \pNormals -> - S.unsafeWith texCoords $ \pTexCoords -> - S.unsafeWith tris $ \pTris -> - S.unsafeWith skins $ \pSkins -> - S.unsafeWith animations $ \pAnimations -> do - #{poke Model, vertices} ptr pVerts - #{poke Model, normals} ptr pNormals - #{poke Model, texCoords} ptr pTexCoords - #{poke Model, triangles} ptr pTris - #{poke Model, skins} ptr pSkins - #{poke Model, animations} ptr pAnimations - #{poke Model, numFrames} ptr numFrames - #{poke Model, numVertices} ptr numVerts - #{poke Model, numTriangles} ptr numTris - #{poke Model, numTexCoords} ptr numTex - #{poke Model, numSkins} ptr numSkins - #{poke Model, numAnimations} ptr numAnimations - --- | A model triangle. --- --- See also: 'triangles''. -data Triangle = Triangle - { v0 :: Vec3 - , v1 :: Vec3 - , v2 :: Vec3 - , n0 :: Vec3 - , n1 :: Vec3 - , n2 :: Vec3 - , t0 :: TexCoord - , t1 :: TexCoord - , t2 :: TexCoord - } - -instance Storable Triangle where - sizeOf _ = #{size model_triangle} - alignment _ = alignment (undefined :: Float) - - peek ptr = do - v0 <- #{peek model_triangle, v0} ptr - v1 <- #{peek model_triangle, v1} ptr - v2 <- #{peek model_triangle, v2} ptr - n0 <- #{peek model_triangle, n0} ptr - n1 <- #{peek model_triangle, n1} ptr - n2 <- #{peek model_triangle, n2} ptr - t0 <- #{peek model_triangle, t0} ptr - t1 <- #{peek model_triangle, t1} ptr - t2 <- #{peek model_triangle, t2} ptr - return $ Triangle v0 v1 v2 n0 n1 n2 t0 t1 t2 - - poke ptr (Triangle v0 v1 v2 n0 n1 n2 t0 t1 t2) = do - #{poke model_triangle, v0} ptr v0 - #{poke model_triangle, v1} ptr v1 - #{poke model_triangle, v2} ptr v2 - #{poke model_triangle, n0} ptr n0 - #{poke model_triangle, n1} ptr n1 - #{poke model_triangle, n2} ptr n2 - #{poke model_triangle, t0} ptr t0 - #{poke model_triangle, t1} ptr t1 - #{poke model_triangle, t2} ptr t2 - -foreign import ccall "Model.h model_free" - model_free :: Ptr Model -> IO () - -foreign import ccall "MD2_load.h MD2_load" - md2_load' :: Ptr CChar -> CChar -> CChar -> Ptr Model -> IO Int - -foreign import ccall "OBJ_load.h OBJ_load" - obj_load' :: Ptr CChar -> CChar -> CChar -> Ptr Model -> IO Int - -md2_load :: Ptr CChar -> CChar -> CChar -> Ptr Model -> IO ModelErrorCode -md2_load file clockwise leftHanded model = - md2_load' file clockwise leftHanded model >>= \code -> return . toEnum $ code - -obj_load :: Ptr CChar -> CChar -> CChar -> Ptr Model -> IO ModelErrorCode -obj_load file clockwise leftHanded model = - obj_load' file clockwise leftHanded model >>= \code -> return . toEnum $ code - --- | Load the model specified by the given file. -loadModel :: FilePath -> Game s Model -loadModel file = do - dotPos <- case elemIndex '.' file of - Nothing -> gameError $ "file name has no extension: " ++ file - Just p -> return p - - let ext = map toLower . tail . snd $ splitAt dotPos file - - result <- liftIO . alloca $ \ptr -> do - status <- withCString file $ \fileCstr -> do - case ext of - "md2" -> md2_load fileCstr 0 0 ptr - "obj" -> obj_load fileCstr 0 0 ptr - _ -> return ModelNoSuitableLoader - - case status of - ModelSuccess -> do - model <- peek ptr - model_free ptr - return . Right $ model - ModelReadError -> return . Left $ "read error" - ModelMemoryAllocationError -> return . Left $ "memory allocation error" - ModelFileNotFound -> return . Left $ "file not found" - ModelFileMismatch -> return . Left $ "file mismatch" - ModelNoSuitableLoader -> return . Left $ "no suitable loader for extension " ++ ext - - case result of - Right model -> return model - Left err -> gameError $ "loadModel: " ++ err - --- | Return 'True' if the model is animated, 'False' otherwise. -animated :: Model -> Bool -animated = (>1) . numFrames - --- | Return the model's ith animation. -animation :: Model -> Int -> Animation -animation model i = animations model S.! i - --- | Return the animation specified by the given string. -animationByName :: Model -> String -> Maybe Animation -animationByName model anim = - let anim' = B.pack anim in S.find ((==) anim' . name) $ animations model - --- | Return a copy of the model's triangles. -triangles' :: Model -> IO [Triangle] -triangles' model = - let n = numVerts model * numFrames model - in with model $ \modelPtr -> - allocaArray n $ \arrayPtr -> do - model_copy_triangles modelPtr arrayPtr - tris <- peekArray n arrayPtr - return tris - -foreign import ccall "Model.h model_copy_triangles" - model_copy_triangles :: Ptr Model -> Ptr Triangle -> IO () - --- | Transform the model's vertices. -transformVerts :: Model -> (Vec3 -> Vec3) -> Model -transformVerts model f = model { vertices = vertices' } - where - n = numVerts model * numFrames model - vertices' = S.generate n f' - f' i = f $ vertices model S.! i - --- | Transform the model's normals. -transformNormals :: Model -> (Vec3 -> Vec3) -> Model -transformNormals model f = model { normals = normals' } - where - n = numVerts model * numFrames model - normals' = S.generate n f' - f' i = f $ normals model S.! i - --- | Translate the model such that its lowest point has y = 0. -toGround :: Model -> IO Model -toGround model = - let model' = model { vertices = S.generate n $ \i -> vertices model S.! i } - n = numVerts model * numFrames model - in - with model' model_to_ground >> return model' - -foreign import ccall "Model.h model_to_ground" - model_to_ground :: Ptr Model -> IO () - --- | Get the model's 3D bounding boxes. -modelBoxes :: Model -> IO (V.Vector Box) -modelBoxes model = - with model $ \modelPtr -> - allocaArray (numVerts model * numFrames model * 6) $ \pointsPtr -> do - model_compute_boxes modelPtr pointsPtr - let n = numFrames model - getBoxes = peekBoxes pointsPtr n 0 0 $ return [] - peekBoxes ptr n cur off l - | cur == n = l - | otherwise = do - xmin <- peekByteOff ptr off - ymin <- peekByteOff ptr $ off + sizeFloat - zmin <- peekByteOff ptr $ off + 2*sizeFloat - xmax <- peekByteOff ptr $ off + 3*sizeFloat - ymax <- peekByteOff ptr $ off + 4*sizeFloat - zmax <- peekByteOff ptr $ off + 5*sizeFloat - let pmin = Vec3 xmin ymin zmin - pmax = Vec3 xmax ymax zmax - box = Box pmin pmax - peekBoxes ptr n (cur+1) (off + 6*sizeFloat) $ fmap (box:) l - fmap (V.fromList . reverse) getBoxes - -foreign import ccall "Model.h model_compute_boxes" - model_compute_boxes :: Ptr Model -> Ptr Vec2 -> IO () diff --git a/Spear/Assets/Model/MD2/MD2_load.c b/Spear/Assets/Model/MD2/MD2_load.c deleted file mode 100644 index 92b1ac2..0000000 --- a/Spear/Assets/Model/MD2/MD2_load.c +++ /dev/null @@ -1,480 +0,0 @@ -#include "MD2_load.h" -#include -#include -#include // malloc -#include // sqrt - -//! The MD2 magic number used to identify MD2 files. -#define MD2_ID 0x32504449 - -//! Limit values for the MD2 file format. -#define MD2_MAX_TRIANGLES 4096 -#define MD2_MAX_VERTICES 2048 -#define MD2_MAX_TEXCOORDS 2048 -#define MD2_MAX_FRAMES 512 -#define MD2_MAX_SKINS 32 - - -/// MD2 file header. -typedef struct -{ - I32 magic; /// The magic number "IDP2"; 844121161 in decimal; 0x32504449 - I32 version; /// Version number, always 8. - I32 skinWidth; /// Width of the skin(s) in pixels. - I32 skinHeight; /// Height of the skin(s) in pixels. - I32 frameSize; /// Size of a single frame in bytes. - I32 numSkins; /// Number of skins. - I32 numVertices; /// Number of vertices in a single frame. - I32 numTexCoords; /// Number of texture coordinates. - I32 numTriangles; /// Number of triangles. - I32 numGlCommands; /// Number of dwords in the Gl command list. - I32 numFrames; /// Number of frames. - I32 offsetSkins; /// Offset from the start of the file to the array of skins. - I32 offsetTexCoords; /// Offset from the start of the file to the array of texture coordinates. - I32 offsetTriangles; /// Offset from the start of the file to the array of triangles. - I32 offsetFrames; /// Offset from the start of the file to the array of frames. - I32 offsetGlCommands; /// Offset from the start of the file to the array of Gl commands. - I32 offsetEnd; /// Offset from the start of the file to the end of the file (the file size). -} -md2Header_t; - - -/// Represents a texture coordinate index. -typedef struct -{ - I16 s; - I16 t; -} -texCoord_t; - - -/// Represents a frame point. -typedef struct -{ - U8 x, y, z; - U8 lightNormalIndex; -} -vertex_t; - - -/// Represents a single frame. -typedef struct -{ - vec3 scale; - vec3 translate; - I8 name[16]; - vertex_t vertices[1]; -} -frame_t; - - -static void normalise (vec3* v) -{ - float x = v->x; - float y = v->y; - float z = v->z; - float mag = sqrt (x*x + y*y + z*z); - mag = mag == 0 ? 1 : mag; - v->x = x / mag; - v->y = y / mag; - v->z = z / mag; -} - - -static void cross (const vec3* a, const vec3* b, vec3* c) -{ - c->x = a->y * b->z - a->z * b->y; - c->y = a->z * b->x - a->x * b->z; - c->z = a->x * b->y - a->y * b->x; -} - - -static void vec3_sub (const vec3* a, const vec3* b, vec3* out) -{ - out->x = a->x - b->x; - out->y = a->y - b->y; - out->z = a->z - b->z; -} - - -static void normal (char clockwise, const vec3* p1, const vec3* p2, const vec3* p3, vec3* n) -{ - vec3 v1, v2; - if (clockwise) - { - vec3_sub (p3, p2, &v1); - vec3_sub (p1, p2, &v2); - } - else - { - vec3_sub (p1, p2, &v1); - vec3_sub (p3, p2, &v2); - } - cross (&v1, &v2, n); - normalise (n); -} - - -typedef struct -{ - vec3* normals; - vec3* base; - unsigned int N; -} -normal_map; - - -static void normal_map_initialise (normal_map* m, unsigned int N) -{ - m->N = N; -} - - -static void normal_map_clear (normal_map* m, vec3* normals, vec3* base) -{ - memset (normals, 0, m->N * sizeof(vec3)); - m->normals = normals; - m->base = base; -} - - -static void normal_map_insert (normal_map* m, vec3* vec, vec3 normal) -{ - unsigned int i = vec - m->base; - vec3* n = m->normals + i; - n->x += normal.x; - n->y += normal.y; - n->z += normal.z; -} - - -static void compute_normals (normal_map* m, char left_handed) -{ - vec3* n = m->normals; - unsigned int i; - for (i = 0; i < m->N; ++i) - { - if (!left_handed) - { - n->x = -n->x; - n->y = -n->y; - n->z = -n->z; - } - normalise (n); - n++; - } -} - - -static void safe_free (void* ptr) -{ - if (ptr) free (ptr); -} - - -static char frame_equal (const char* name1, const char* name2) -{ - char equal = 1; - int i; - - if (((name1 == 0) && (name2 != 0)) || ((name1 != 0) && (name2 == 0))) - { - return 0; - } - - for (i = 0; i < 16; ++i) - { - char c1 = *name1; - char c2 = *name2; - if ((c1 >= '0' && c1 <= '9') || (c2 >= '0' && c2 <= '9')) break; - if (c1 != c2) - { - equal = 0; - break; - } - if (c1 == '_' || c2 == '_') break; - name1++; - name2++; - } - return equal; -} - - -static void animation_remove_numbers (char* name) -{ - int i; - for (i = 0; i < 16; ++i) - { - char c = *name; - if (c == 0) break; - if (c >= '0' && c <= '9') *name = 0; - name++; - } -} - - -Model_error_code MD2_load (const char* filename, char clockwise, char left_handed, Model* model) -{ - FILE* filePtr; - vec3* vertices; - vec3* normals; - texCoord* texCoords; - triangle* triangles; - skin* skins; - animation* animations; - int i; - - // Open the file for reading. - filePtr = fopen(filename, "rb"); - if (!filePtr) return Model_File_Not_Found; - - // Make sure it is an MD2 file. - int magic; - if ((fread(&magic, 4, 1, filePtr)) != 1) - { - fclose(filePtr); - return Model_Read_Error; - } - - if (magic != MD2_ID) return Model_File_Mismatch; - - // Find out the file size. - long int fileSize; - fseek(filePtr, 0, SEEK_END); - fileSize = ftell(filePtr); - fseek(filePtr, 0, SEEK_SET); - - // Allocate a chunk of data to store the file in. - char *buffer = (char*) malloc(fileSize); - if (!buffer) - { - fclose(filePtr); - return Model_Memory_Allocation_Error; - } - - // Read the entire file into memory. - if ((fread(buffer, 1, fileSize, filePtr)) != (unsigned int)fileSize) - { - fclose(filePtr); - free(buffer); - return Model_Read_Error; - } - - // File stream is no longer needed. - fclose(filePtr); - - // Set a pointer to the header for parsing. - md2Header_t* header = (md2Header_t*) buffer; - - // Compute the number of animations. - unsigned numAnimations = 1; - int currentFrame; - const char* name = 0; - for (currentFrame = 0; currentFrame < header->numFrames; ++currentFrame) - { - frame_t* frame = (frame_t*) &buffer[header->offsetFrames + currentFrame * header->frameSize]; - if (name == 0) - { - name = frame->name; - } - else if (!frame_equal(name, frame->name)) - { - numAnimations++; - name = frame->name; - } - } - - // Allocate memory for arrays. - vertices = (vec3*) malloc(sizeof(vec3) * header->numVertices * header->numFrames); - normals = (vec3*) malloc(sizeof(vec3) * header->numVertices * header->numFrames); - texCoords = (texCoord*) malloc(sizeof(texCoord) * header->numTexCoords); - triangles = (triangle*) malloc(sizeof(triangle) * header->numTriangles); - skins = (skin*) malloc(sizeof(skin) * header->numSkins); - animations = (animation*) malloc (numAnimations * sizeof(animation)); - - if (!vertices || !normals || !texCoords || !triangles || !skins || !animations) - { - safe_free (animations); - safe_free (skins); - safe_free (triangles); - safe_free (texCoords); - safe_free (normals); - safe_free (vertices); - free (buffer); - return Model_Memory_Allocation_Error; - } - - // Load the model's vertices. - // Loop through each frame, grab the vertices that make it up, transform them back - // to their real coordinates and store them in the model's vertex array. - for (currentFrame = 0; currentFrame < header->numFrames; ++currentFrame) - { - // Set a frame pointer to the current frame. - frame_t* frame = (frame_t*) &buffer[header->offsetFrames + currentFrame * header->frameSize]; - - // Set a vertex pointer to the model's vertex array, at the appropiate position. - vec3* vert = &vertices[currentFrame * header->numVertices]; - - // Now parse those vertices and transform them back. - int currentVertex; - for (currentVertex = 0; currentVertex != header->numVertices; ++currentVertex) - { - vert[currentVertex].x = frame->vertices[currentVertex].x * frame->scale.x + frame->translate.x; - vert[currentVertex].y = frame->vertices[currentVertex].y * frame->scale.y + frame->translate.y; - vert[currentVertex].z = frame->vertices[currentVertex].z * frame->scale.z + frame->translate.z; - } - } - - // Load the model's triangles. - - // Set a pointer to the triangles array in the buffer. - triangle* t = (triangle*) &buffer[header->offsetTriangles]; - - if (clockwise) - { - for (i = 0; i < header->numTriangles; ++i) - { - triangles[i].vertexIndices[0] = t[i].vertexIndices[0]; - triangles[i].vertexIndices[1] = t[i].vertexIndices[1]; - triangles[i].vertexIndices[2] = t[i].vertexIndices[2]; - - triangles[i].textureIndices[0] = t[i].textureIndices[0]; - triangles[i].textureIndices[1] = t[i].textureIndices[1]; - triangles[i].textureIndices[2] = t[i].textureIndices[2]; - } - } - else - { - for (i = 0; i < header->numTriangles; ++i) - { - triangles[i].vertexIndices[0] = t[i].vertexIndices[0]; - triangles[i].vertexIndices[1] = t[i].vertexIndices[2]; - triangles[i].vertexIndices[2] = t[i].vertexIndices[1]; - - triangles[i].textureIndices[0] = t[i].textureIndices[0]; - triangles[i].textureIndices[1] = t[i].textureIndices[2]; - triangles[i].textureIndices[2] = t[i].textureIndices[1]; - } - } - - // Load the texture coordinates. - float sw = (float) header->skinWidth; - float sh = (float) header->skinHeight; - texCoord_t* texc = (texCoord_t*) &buffer[header->offsetTexCoords]; - for (i = 0; i < header->numTexCoords; ++i) - { - texCoords[i].s = (float)texc->s / sw; - texCoords[i].t = 1.0f - (float)texc->t / sh; - texc++; - } - - // Iterate over every frame and compute normals for every triangle. - vec3 n; - - normal_map map; - normal_map_initialise (&map, header->numVertices); - - for (currentFrame = 0; currentFrame < header->numFrames; ++currentFrame) - { - // Set a pointer to the triangle array. - triangle* t = triangles; - - // Set a pointer to the vertex array at the appropiate position. - vec3* vertex_array = vertices + header->numVertices * currentFrame; - - // Set a pointer to the normals array at the appropiate position. - vec3* normals_ptr = normals + header->numVertices * currentFrame; - - normal_map_clear (&map, normals_ptr, vertex_array); - - for (i = 0; i < header->numTriangles; ++i) - { - // Compute face normal. - vec3* v0 = &vertex_array[t->vertexIndices[0]]; - vec3* v1 = &vertex_array[t->vertexIndices[1]]; - vec3* v2 = &vertex_array[t->vertexIndices[2]]; - normal (clockwise, v0, v1, v2, &n); - - // Add face normal to each of the face's vertices. - normal_map_insert (&map, v0, n); - normal_map_insert (&map, v1, n); - normal_map_insert (&map, v2, n); - - t++; - } - - compute_normals (&map, left_handed); - } - - // Load the model's skins. - const skin* s = (const skin*) &buffer[header->offsetSkins]; - for (i = 0; i < header->numSkins; ++i) - { - memcpy (skins[i].name, s->name, 64); - s++; - } - - // Load the model's animations. - unsigned start = 0; - name = 0; - animation* currentAnimation = animations; - for (currentFrame = 0; currentFrame < header->numFrames; ++currentFrame) - { - frame_t* frame = (frame_t*) &buffer[header->offsetFrames + currentFrame * header->frameSize]; - if (name == 0) - { - name = frame->name; - } - else if (!frame_equal(name, frame->name)) - { - memcpy (currentAnimation->name, name, 16); - animation_remove_numbers (currentAnimation->name); - currentAnimation->start = start; - currentAnimation->end = currentFrame-1; - if (currentAnimation != animations) - { - animation* prev = currentAnimation; - prev--; - prev->end = start-1; - } - name = frame->name; - currentAnimation++; - start = currentFrame; - } - } - currentAnimation = animations + numAnimations - 1; - memcpy (currentAnimation->name, name, 16); - animation_remove_numbers (currentAnimation->name); - currentAnimation->start = start; - currentAnimation->end = header->numFrames-1; - - /*printf ("finished loading model %s\n", filename); - printf ("numAnimations: %u\n", numAnimations); - printf ("animations: %p\n", animations); - - currentAnimation = animations; - for (i = 0; i < numAnimations; ++i) - { - printf ("Animation %d, name: %s, start: %d, end %d\n", - i, currentAnimation->name, currentAnimation->start, currentAnimation->end); - currentAnimation++; - }*/ - - model->vertices = vertices; - model->normals = normals; - model->texCoords = texCoords; - model->triangles = triangles; - model->skins = skins; - model->animations = animations; - - model->numFrames = header->numFrames; - model->numVertices = header->numVertices; - model->numTriangles = header->numTriangles; - model->numTexCoords = header->numTexCoords; - model->numSkins = header->numSkins; - model->numAnimations = numAnimations; - - free(buffer); - - return Model_Success; -} diff --git a/Spear/Assets/Model/MD2/MD2_load.h b/Spear/Assets/Model/MD2/MD2_load.h deleted file mode 100644 index 689f9d1..0000000 --- a/Spear/Assets/Model/MD2/MD2_load.h +++ /dev/null @@ -1,22 +0,0 @@ -#ifndef _MD2_LOAD_H -#define _MD2_LOAD_H - -#include "../Model.h" -#include "../Model_error_code.h" - - -#ifdef __cplusplus -extern "C" { -#endif - -/// Loads the MD2 file specified by the given string. -/// 'clockwise' should be 1 if you plan to render the model in a clockwise environment, 0 otherwise. -/// 'smooth_normals' should be 1 if you want the loader to compute smooth normals, 0 otherwise. -Model_error_code MD2_load (const char* filename, char clockwise, char left_handed, Model* model); - -#ifdef __cplusplus -} -#endif - - -#endif // _MD2_LOAD_H diff --git a/Spear/Assets/Model/Model.c b/Spear/Assets/Model/Model.c deleted file mode 100644 index fd588ec..0000000 --- a/Spear/Assets/Model/Model.c +++ /dev/null @@ -1,112 +0,0 @@ -#include "Model.h" -#include // free -#include - - -#define TO_RAD M_PI / 180.0 - - -static void safe_free (void* ptr) -{ - if (ptr) - { - free (ptr); - ptr = 0; - } -} - - -void model_free (Model* model) -{ - safe_free (model->vertices); - safe_free (model->normals); - safe_free (model->texCoords); - safe_free (model->triangles); - safe_free (model->skins); - safe_free (model->animations); -} - - -void model_to_ground (Model* model) -{ - unsigned i, f; - vec3* v = model->vertices; - - // Compute the minimum y coordinate for each frame and translate - // the model appropriately. - for (f = 0; f < model->numFrames; ++f) - { - vec3* w = v; - float y = v->y; - - for (i = 0; i < model->numVertices; ++i, ++v) - { - y = fmin (y, v->y); - } - - v = w; - for (i = 0; i < model->numVertices; ++i, ++v) - { - v->y -= y; - } - } -} - - -void model_copy_triangles (Model* model, unsigned frame, model_triangle* tris) -{ - int i; - int j = model->numVertices; - - vec3* v = model->vertices + j * frame; - vec3* n = model->normals + j * frame; - texCoord* t = model->texCoords; - triangle* tri = model->triangles; - - - for (i = 0; i < j; ++i, ++tri, ++tris) - { - tris->v0 = v[tri->vertexIndices[0]]; - tris->v1 = v[tri->vertexIndices[1]]; - tris->v2 = v[tri->vertexIndices[2]]; - - tris->n0 = n[tri->vertexIndices[0]]; - tris->n1 = n[tri->vertexIndices[1]]; - tris->n2 = n[tri->vertexIndices[2]]; - - tris->t0 = t[tri->textureIndices[0]]; - tris->t1 = t[tri->textureIndices[1]]; - tris->t2 = t[tri->textureIndices[2]]; - } -} - - -void model_compute_boxes (Model* model, float* points) -{ - vec3* v = model->vertices; - - unsigned f; - for (f = 0; f < model->numFrames; ++f) - { - float xmin = v->x; - float xmax = v->x; - float ymin = v->y; - float ymax = v->y; - float zmin = v->z; - float zmax = v->z; - - unsigned i; - for (i = 0; i < model->numVertices; ++i, ++v) - { - xmin = fmin (xmin, v->x); - ymin = fmin (ymin, v->y); - zmin = fmin (zmin, v->z); - xmax = fmax (xmax, v->x); - ymax = fmax (ymax, v->y); - zmax = fmax (zmax, v->z); - } - - *points++ = xmin; *points++ = ymin; *points++ = zmin; - *points++ = xmax; *points++ = ymax; *points++ = zmax; - } -} diff --git a/Spear/Assets/Model/Model.h b/Spear/Assets/Model/Model.h deleted file mode 100644 index 0f2aece..0000000 --- a/Spear/Assets/Model/Model.h +++ /dev/null @@ -1,100 +0,0 @@ -#ifndef _SPEAR_MODEL_H -#define _SPEAR_MODEL_H - -#include "sys_types.h" - - -typedef struct -{ - char name[64]; -} -skin; - - -typedef struct -{ - float x, y, z; -} -vec3; - - -typedef struct -{ - float s, t; -} -texCoord; - - -typedef struct -{ - U16 vertexIndices[3]; - U16 textureIndices[3]; -} -triangle; - - -typedef struct -{ - char name[16]; - unsigned int start; - unsigned int end; -} -animation; - - -typedef struct -{ - vec3* vertices; // One array per frame. - vec3* normals; // One array per frame. One normal per vertex per frame. - texCoord* texCoords; // One array for all frames. - triangle* triangles; // One array for all frames. - skin* skins; // Holds the model's texture files. - animation* animations; // Holds the model's animations. - - unsigned int numFrames; - unsigned int numVertices; // Number of vertices per frame. - unsigned int numTriangles; // Number of triangles in one frame. - unsigned int numTexCoords; // Number of texture coordinates in one frame. - unsigned int numSkins; - unsigned int numAnimations; -} -Model; - - -typedef struct -{ - vec3 v0; - vec3 v1; - vec3 v2; - vec3 n0; - vec3 n1; - vec3 n2; - texCoord t0; - texCoord t1; - texCoord t2; -} -model_triangle; - - -#ifdef __cplusplus -extern "C" { -#endif - -/// Frees the given Model from memory. -/// The 'model' pointer itself is not freed. -void model_free (Model* model); - -/// Translate the Model such that its lowest point has y = 0. -void model_to_ground (Model* model); - -/// Copy the triangles of the given frame from the Model into the given array. -void model_copy_triangles (Model* model, unsigned frame, model_triangle* tris); - -/// Compute the model's 2d AABBs. -void model_compute_boxes (Model* model, float* points); - -#ifdef __cplusplus -} -#endif - -#endif // _SPEAR_MODEL_H diff --git a/Spear/Assets/Model/Model_error_code.h b/Spear/Assets/Model/Model_error_code.h deleted file mode 100644 index d306052..0000000 --- a/Spear/Assets/Model/Model_error_code.h +++ /dev/null @@ -1,16 +0,0 @@ -#ifndef _SPEAR_MODEL_ERROR_CODE_H -#define _SPEAR_MODEL_ERROR_CODE_H - -typedef enum -{ - Model_Success, - Model_Read_Error, - Model_Memory_Allocation_Error, - Model_File_Not_Found, - Model_File_Mismatch, - Model_No_Suitable_Loader, -} -Model_error_code; - -#endif // _SPEAR_MODEL_ERROR_CODE_H - diff --git a/Spear/Assets/Model/OBJ/Makefile b/Spear/Assets/Model/OBJ/Makefile deleted file mode 100644 index 9630c9d..0000000 --- a/Spear/Assets/Model/OBJ/Makefile +++ /dev/null @@ -1,15 +0,0 @@ -test: ../Model.o OBJ_load.o cvector.o test.o - $(CC) Model.o OBJ_load.o cvector.o test.o -o $@ -lm - -vector: cvector.o vector-test.o - $(CC) cvector.o vector-test.o -o vector - - -%.o: %.c %.h - $(CC) -g -c $< - - -clean: - @rm -f test vector - @rm -f *.o - diff --git a/Spear/Assets/Model/OBJ/OBJ_load.c b/Spear/Assets/Model/OBJ/OBJ_load.c deleted file mode 100644 index cdd39c9..0000000 --- a/Spear/Assets/Model/OBJ/OBJ_load.c +++ /dev/null @@ -1,274 +0,0 @@ -#include "OBJ_load.h" -#include "cvector.h" -#include -#include // free -#include // memcpy -#include // sqrt - - -char lastError [128]; - - -static void safe_free (void* ptr) -{ - if (ptr) - { - free (ptr); - ptr = 0; - } -} - - -static void cross (vec3 a, vec3 b, vec3* c) -{ - c->x = a.y * b.z - a.z * b.y; - c->y = a.z * b.x - a.x * b.z; - c->z = a.x * b.y - a.y * b.x; -} - - -static void vec3_sub (vec3 a, vec3 b, vec3* out) -{ - out->x = a.x - b.x; - out->y = a.y - b.y; - out->z = a.z - b.z; -} - - -static void compute_normal (char clockwise, vec3 p1, vec3 p2, vec3 p3, vec3* n) -{ - vec3 v1, v2; - if (!clockwise) - { - vec3_sub (p3, p2, &v1); - vec3_sub (p1, p2, &v2); - } - else - { - vec3_sub (p1, p2, &v1); - vec3_sub (p3, p2, &v2); - } - cross (v1, v2, n); -} - - -static void normalise (vec3* v) -{ - float x = v->x; - float y = v->y; - float z = v->z; - float mag = sqrt (x*x + y*y + z*z); - mag = mag == 0.0f ? 1.0f : mag; - v->x /= mag; - v->y /= mag; - v->z /= mag; -} - - -static void vec3_add (vec3 a, vec3* b) -{ - b->x += a.x; - b->y += a.y; - b->z += a.z; -} - - -static void read_vertex (FILE* file, vec3* vert) -{ - fscanf (file, "%f %f", &vert->x, &vert->y); - if (fscanf(file, "%f", &vert->z) == 0) vert->z = 0.0f; -} - - -static void read_normal (FILE* file, vec3* normal) -{ - fscanf (file, "%f %f %f", &normal->x, &normal->y, &normal->z); -} - - -static void read_tex_coord (FILE* file, texCoord* texc) -{ - fscanf (file, "%f %f", &texc->s, &texc->t); -} - - -static void read_face (FILE* file, - char clockwise, - vector* vertices, - vector* normals, - vector* triangles) -{ - vector idxs; - vector texCoords; - - vector_new (&idxs, sizeof(int), 4); - vector_new (&texCoords, sizeof(int), 4); - - unsigned int index; - unsigned int normal; - unsigned int texc; - - fscanf (file, "f"); - - while (!feof(file) && fscanf(file, "%d", &index) > 0) - { - vector_append (&idxs, &index); - - if (fgetc (file) == '/') - { - fscanf (file, "%d", &texc); - vector_append (&texCoords, &texc); - } - else fseek (file, -1, SEEK_CUR); - - if (fgetc (file) == '/') - { - fscanf (file, "%d", &normal); - } - else fseek (file, -1, SEEK_CUR); - } - - // Triangulate the face and add its triangles to the triangles vector. - triangle tri; - tri.vertexIndices[0] = *((int*) vector_ith (&idxs, 0)) - 1; - tri.textureIndices[0] = *((int*) vector_ith (&texCoords, 0)) - 1; - - int i; - for (i = 1; i < vector_size(&idxs)-1; i++) - { - tri.vertexIndices[1] = *((int*) vector_ith (&idxs, i)) - 1; - tri.textureIndices[1] = *((int*) vector_ith (&texCoords, i)) - 1; - tri.vertexIndices[2] = *((int*) vector_ith (&idxs, i+1)) - 1; - tri.textureIndices[2] = *((int*) vector_ith (&texCoords, i+1)) - 1; - vector_append (triangles, &tri); - } - - // Compute face normal and add contribution to each of the face's vertices. - unsigned int i0 = tri.vertexIndices[0]; - unsigned int i1 = tri.vertexIndices[1]; - unsigned int i2 = tri.vertexIndices[2]; - - vec3 n; - vec3 v0 = *((vec3*) vector_ith (vertices, i0)); - vec3 v1 = *((vec3*) vector_ith (vertices, i1)); - vec3 v2 = *((vec3*) vector_ith (vertices, i2)); - compute_normal (clockwise, v0, v1, v2, &n); - - for (i = 0; i < vector_size (&idxs); i++) - { - int j = *((int*) vector_ith (&idxs, i)) - 1; - vec3* normal = (vec3*) vector_ith (normals, j); - vec3_add (n, normal); - } - - vector_free (&idxs); - vector_free (&texCoords); -} - - -Model_error_code OBJ_load (const char* filename, char clockwise, char left_handed, Model* model) -{ - vec3* norms = 0; - vec3* verts = 0; - texCoord* texcs = 0; - triangle* tris = 0; - - FILE* file = fopen (filename, "r"); - if (file == NULL) return Model_File_Not_Found; - - vec3 vert; - vec3 normal; - texCoord texc; - - vector vertices; - vector normals; - vector texCoords; - vector triangles; - - int result = vector_new (&vertices, sizeof(vec3), 0) - | vector_new (&normals, sizeof(vec3), 0) - | vector_new (&texCoords, sizeof(texCoord), 0) - | vector_new (&triangles, sizeof(triangle), 0); - - if (result != 0) - { - safe_free (vertices.data); - safe_free (normals.data); - safe_free (texCoords.data); - safe_free (triangles.data); - return Model_Memory_Allocation_Error; - } - - while (!feof(file)) - { - switch (fgetc(file)) - { - case 'v': - switch (fgetc(file)) - { - case 't': - read_tex_coord (file, &texc); - vector_append (&texCoords, &texc); - break; - - case 'n': - read_normal (file, &normal); - vector_append (&normals, &normal); - break; - - default: - read_vertex (file, &vert); - vector_append (&vertices, &vert); - break; - } - break; - - case 'f': - // Initialise the normals vector if it is empty. - if (vector_size(&normals) == 0) - { - vec3 zero; - zero.x = 0.0f; zero.y = 0.0f; zero.z = 0.0f; - vector_new (&normals, sizeof(vec3), vector_size(&vertices)); - vector_initialise (&normals, &zero); - } - read_face (file, clockwise, &vertices, &normals, &triangles); - break; - - case '#': - { - int x = 17; - while (x != '\n' && x != EOF) x = fgetc(file); - break; - } - - default: break; - } - } - - fclose (file); - - unsigned numVertices = vector_size (&vertices); - - // Normalise normals. - unsigned i; - for (i = 0; i < numVertices; ++i) - { - normalise (vector_ith (&normals, i)); - } - - model->vertices = (vec3*) vertices.data; - model->normals = (vec3*) normals.data; - model->texCoords = (texCoord*) texCoords.data; - model->triangles = (triangle*) triangles.data; - model->skins = 0; - model->animations = 0; - model->numFrames = 1; - model->numVertices = numVertices; - model->numTriangles = vector_size (&triangles); - model->numTexCoords = vector_size (&texCoords); - model->numSkins = 0; - model->numAnimations = 0; - - return Model_Success; -} diff --git a/Spear/Assets/Model/OBJ/OBJ_load.h b/Spear/Assets/Model/OBJ/OBJ_load.h deleted file mode 100644 index 485d3cc..0000000 --- a/Spear/Assets/Model/OBJ/OBJ_load.h +++ /dev/null @@ -1,25 +0,0 @@ -#ifndef _OBJ_LOAD_H -#define _OBJ_LOAD_H - -#include "../Model.h" -#include "../Model_error_code.h" - - -#ifdef __cplusplus -extern "C" { -#endif - -/// Loads the OBJ file specified by the given string. -/// 'clockwise' should be 1 if you plan to render the model in a clockwise environment, 0 otherwise. -/// 'smooth_normals' should be 1 if you want the loader to compute smooth normals, 0 otherwise. -Model_error_code OBJ_load (const char* filename, char clockwise, char left_handed, Model* model); - -/// Gets the last error generated by the OBJ loader. -char* get_last_error (); - -#ifdef __cplusplus -} -#endif - - -#endif // _OBJ_LOAD_H diff --git a/Spear/Assets/Model/OBJ/cvector.c b/Spear/Assets/Model/OBJ/cvector.c deleted file mode 100644 index 9213d8d..0000000 --- a/Spear/Assets/Model/OBJ/cvector.c +++ /dev/null @@ -1,90 +0,0 @@ -#include "cvector.h" -#include // malloc, realloc, free -#include // memcpy - - -int max (int a, int b) -{ - if (a > b) return a; - return b; -} - - -int vector_new (vector* v, int elem_size, int num_elems) -{ - int n = num_elems * elem_size; - - char* data = 0; - if (num_elems > 0) - { - data = (char*) malloc (n); - if (data == NULL) return 1; - } - - v->data = data; - v->next = data; - v->chunk_size = n; - v->elem_size = elem_size; - - return 0; -} - - -void vector_free (vector* v) -{ - if (v->data != 0) free (v->data); -} - - -void vector_initialise (vector* v, void* value) -{ - char* ptr = v->data; - int esize = v->elem_size; - int n = vector_size (v); - - int i; - for (i = 0; i < n; ++i) - { - memcpy (ptr, value, esize); - ptr += esize; - } -} - - -int vector_append (vector* v, void* elem) -{ - // Realloc a bigger chunk when the vector runs out of space. - if (v->next == v->data + v->chunk_size) - { - int old_chunk_size = v->chunk_size; - int n = max (v->elem_size, 2 * old_chunk_size); - - char* data = (char*) realloc (v->data, n); - if (data == NULL) return 1; - - v->data = data; - v->next = data + old_chunk_size; - v->chunk_size = n; - } - - memcpy ((void*)v->next, elem, v->elem_size); - v->next += v->elem_size; -} - - -void* vector_ith (vector* v, int i) -{ - return (void*) (v->data + i*v->elem_size); -} - - -int vector_size (vector* v) -{ - return (v->next - v->data) / v->elem_size; -} - - -int vector_capacity (vector* v) -{ - return v->chunk_size / v->elem_size; -} diff --git a/Spear/Assets/Model/OBJ/cvector.h b/Spear/Assets/Model/OBJ/cvector.h deleted file mode 100644 index 2c269e4..0000000 --- a/Spear/Assets/Model/OBJ/cvector.h +++ /dev/null @@ -1,36 +0,0 @@ -#ifndef _C_SPEAR_VECTOR_H -#define _C_SPEAR_VECTOR_H - -typedef struct -{ - char* data; - char* next; - int chunk_size; - int elem_size; -} -vector; - -/// Construct a new vector. -/// Returns non-zero on error. -int vector_new (vector* v, int elem_size, int num_elems); - -/// Free the vector. -void vector_free (vector* v); - -/// Initialise every position to the given value. -void vector_initialise (vector* v, void* value); - -/// Append an element. -/// Returns non-zero on error. -int vector_append (vector* v, void* elem); - -/// Access the ith element. -void* vector_ith (vector* v, int i); - -/// Return the number of elements in the vector. -int vector_size (vector* v); - -/// Return the vector's capacity. -int vector_capacity (vector* v); - -#endif // _C_SPEAR_VECTOR_H diff --git a/Spear/Assets/Model/sys_types.h b/Spear/Assets/Model/sys_types.h deleted file mode 100644 index 6aca9e9..0000000 --- a/Spear/Assets/Model/sys_types.h +++ /dev/null @@ -1,16 +0,0 @@ -#ifndef _SPEAR_SYS_TYPES_H -#define _SPEAR_SYS_TYPES_H - -#include - -typedef int8_t I8; -typedef int16_t I16; -typedef int32_t I32; -typedef int64_t I64; -typedef uint8_t U8; -typedef uint16_t U16; -typedef uint32_t U32; -typedef uint64_t U64; - -#endif // _SPEAR_SYS_TYPES_H - diff --git a/Spear/GL.hs b/Spear/GL.hs deleted file mode 100644 index 3c1734b..0000000 --- a/Spear/GL.hs +++ /dev/null @@ -1,800 +0,0 @@ -{-# LANGUAGE FlexibleInstances #-} - -module Spear.GL - ( -- * Programs - GLSLProgram, - newProgram, - linkProgram, - useProgram, - unuseProgram, - withGLSLProgram, - - -- ** Locations - attribLocation, - fragLocation, - uniformLocation, - - -- ** Uniforms - Uniform (..), - - -- * Shaders - GLSLShader, - ShaderType (..), - attachShader, - detachShader, - loadShader, - newShader, - - -- ** Source loading - loadSource, - shaderSource, - readSource, - compile, - - -- * Helper functions - ($=), - Data.StateVar.get, - - -- * VAOs - VAO(..), - newVAO, - bindVAO, - unbindVAO, - enableVAOAttrib, - attribVAOPointer, - - -- ** Rendering - drawArrays, - drawElements, - - -- * Buffers - GLBuffer(..), - TargetBuffer (..), - BufferUsage (..), - newBuffer, - bindBuffer, - unbindBuffer, - BufferData (..), - bufferData', - withGLBuffer, - - -- * Textures - Texture, - SettableStateVar, - ($), - - -- ** Creation and destruction - newTexture, - loadTextureImage, - - -- ** Manipulation - bindTexture, - unbindTexture, - loadTextureData, - texParami, - texParamf, - activeTexture, - - -- * Error Handling - getGLError, - printGLError, - assertGL, - - -- * OpenGL - module Graphics.GL.Core46, - Ptr, - nullPtr, - ) -where - -import Spear.Assets.Image -import Spear.Game -import Spear.Math.Algebra -import Spear.Math.Matrix3 (Matrix3) -import Spear.Math.Matrix4 (Matrix4) -import Spear.Math.Vector - -import Control.Monad -import Control.Monad.Trans.Class -import Control.Monad.Trans.State as State -import qualified Data.ByteString.Char8 as B -import Data.StateVar -import Data.Word -import Foreign.C.String -import Foreign.C.Types -import Foreign.Marshal.Alloc (alloca) -import Foreign.Marshal.Array (withArray) -import Foreign.Marshal.Utils as Foreign (with) -import Foreign.Ptr -import Foreign.Storable -import Foreign.Storable (peek) -import Graphics.GL.Core46 -import Prelude hiding ((*)) -import System.Directory (doesFileExist, getCurrentDirectory, - setCurrentDirectory) -import System.IO (hPutStrLn, stderr) -import Unsafe.Coerce - --- --- MANAGEMENT --- - --- | A GLSL shader handle. -data GLSLShader = GLSLShader - { getShader :: GLuint, - getShaderKey :: ReleaseKey - } - -instance ResourceClass GLSLShader where - getResource = getShaderKey - --- | A GLSL program handle. -data GLSLProgram = GLSLProgram - { getProgram :: GLuint, - getProgramKey :: ReleaseKey - } - -instance ResourceClass GLSLProgram where - getResource = getProgramKey - --- | Supported shader types. -data ShaderType = VertexShader | FragmentShader | GeometryShader deriving (Eq, Show) - -toGLShader :: ShaderType -> GLenum -toGLShader VertexShader = GL_VERTEX_SHADER -toGLShader FragmentShader = GL_FRAGMENT_SHADER -toGLShader GeometryShader = GL_GEOMETRY_SHADER - --- | Apply the given function to the program's id. -withGLSLProgram :: GLSLProgram -> (GLuint -> a) -> a -withGLSLProgram prog f = f $ getProgram prog - --- | Get the location of the given uniform variable within the given program. -uniformLocation :: GLSLProgram -> String -> GettableStateVar GLint -uniformLocation prog var = makeGettableStateVar $ - withCString var $ \str -> glGetUniformLocation (getProgram prog) (unsafeCoerce str) - --- | Get or set the location of the given variable to a fragment shader colour number. -fragLocation :: GLSLProgram -> String -> StateVar GLint -fragLocation prog var = makeStateVar get set - where - get = withCString var $ \str -> glGetFragDataLocation (getProgram prog) (unsafeCoerce str) - set idx = withCString var $ \str -> - glBindFragDataLocation (getProgram prog) (unsafeCoerce idx) (unsafeCoerce str) - --- | Get or set the location of the given attribute within the given program. -attribLocation :: GLSLProgram -> String -> StateVar GLint -attribLocation prog var = makeStateVar get set - where - get = withCString var $ \str -> glGetAttribLocation (getProgram prog) (unsafeCoerce str) - set idx = withCString var $ \str -> - glBindAttribLocation (getProgram prog) (unsafeCoerce idx) (unsafeCoerce str) - --- | Create a new program. -newProgram :: [GLSLShader] -> Game s GLSLProgram -newProgram shaders = do - h <- liftIO glCreateProgram - when (h == 0) $ gameError "glCreateProgram failed" - rkey <- register $ deleteProgram h - let program = GLSLProgram h rkey - mapM_ (liftIO . attachShader program) shaders - linkProgram program - return program - --- Delete the program. -deleteProgram :: GLuint -> IO () ---deleteProgram = glDeleteProgram -deleteProgram prog = do - putStrLn $ "Deleting shader program " ++ show prog - glDeleteProgram prog - --- | Link the program. -linkProgram :: GLSLProgram -> Game s () -linkProgram prog = do - let h = getProgram prog - err <- liftIO $ do - glLinkProgram h - alloca $ \statptr -> do - glGetProgramiv h GL_LINK_STATUS statptr - status <- peek statptr - case status of - 0 -> getStatus glGetProgramiv glGetProgramInfoLog h - _ -> return "" - - case length err of - 0 -> return () - _ -> gameError err - --- | Use the program. -useProgram :: GLSLProgram -> IO () -useProgram prog = glUseProgram $ getProgram prog - --- | Deactivate the active program. -unuseProgram :: IO () -unuseProgram = glUseProgram 0 - --- | Attach the given shader to the given program. -attachShader :: GLSLProgram -> GLSLShader -> IO () -attachShader prog shader = glAttachShader (getProgram prog) (getShader shader) - --- | Detach the given GLSL from the given program. -detachShader :: GLSLProgram -> GLSLShader -> IO () -detachShader prog shader = glDetachShader (getProgram prog) (getShader shader) - --- | Load a shader from the file specified by the given string. --- --- This function creates a new shader. To load source code into an existing shader, --- see 'loadSource', 'shaderSource' and 'readSource'. -loadShader :: ShaderType -> FilePath -> Game s GLSLShader -loadShader shaderType file = do - shader <- newShader shaderType - loadSource file shader - compile file shader - return shader - --- | Create a new shader. -newShader :: ShaderType -> Game s GLSLShader -newShader shaderType = do - h <- liftIO $ glCreateShader (toGLShader shaderType) - case h of - 0 -> gameError "glCreateShader failed" - _ -> do - rkey <- register $ deleteShader h - return $ GLSLShader h rkey - --- | Free the shader. -deleteShader :: GLuint -> IO () ---deleteShader = glDeleteShader -deleteShader shader = do - putStrLn $ "Deleting shader " ++ show shader - glDeleteShader shader - --- | Load a shader source from the file specified by the given string --- into the shader. -loadSource :: FilePath -> GLSLShader -> Game s () -loadSource file h = do - exists <- liftIO $ doesFileExist file - case exists of - False -> gameError "the specified shader file does not exist" - True -> liftIO $ do - code <- readSource file - withCString code $ shaderSource h - --- | Load the given shader source into the shader. -shaderSource :: GLSLShader -> CString -> IO () -shaderSource shader str = - let ptr = unsafeCoerce str - in withArray [ptr] $ flip (glShaderSource (getShader shader) 1) nullPtr - --- | Compile the shader. -compile :: FilePath -> GLSLShader -> Game s () -compile file shader = do - let h = getShader shader - - -- Compile - liftIO $ glCompileShader h - - -- Verify status - err <- liftIO $ - alloca $ \statusPtr -> do - glGetShaderiv h GL_COMPILE_STATUS statusPtr - result <- peek statusPtr - case result of - 0 -> getStatus glGetShaderiv glGetShaderInfoLog h - _ -> return "" - - case length err of - 0 -> return () - _ -> gameError $ "Unable to compile shader " ++ file ++ ":\n" ++ err - -type StatusCall = GLuint -> GLenum -> Ptr GLint -> IO () - -type LogCall = GLuint -> GLsizei -> Ptr GLsizei -> Ptr GLchar -> IO () - -getStatus :: StatusCall -> LogCall -> GLuint -> IO String -getStatus getStatus getLog h = do - alloca $ \lenPtr -> do - getStatus h GL_INFO_LOG_LENGTH lenPtr - len <- peek lenPtr - case len of - 0 -> return "" - _ -> withCString (replicate (unsafeCoerce len) '\0') $ getErrorString getLog h (unsafeCoerce len) - -getErrorString :: LogCall -> GLuint -> GLsizei -> CString -> IO String -getErrorString getLog h len str = do - let ptr = unsafeCoerce str - getLog h len nullPtr ptr - peekCString str - --- | Load the shader source specified by the given file. --- --- This function implements an #include mechanism, so the given file can --- refer to other files. -readSource :: FilePath -> IO String -readSource = fmap B.unpack . readSource' - -readSource' :: FilePath -> IO B.ByteString -readSource' file = do - let includeB = B.pack "#include" - newLineB = B.pack "\n" - isInclude = ((==) includeB) . B.take 8 - clean = B.dropWhile (\c -> c == ' ') - cleanInclude = B.filter (\c -> c /= '"') . B.dropWhile (\c -> c /= ' ') - toLines = B.splitWith (\c -> c == '\n' || c == '\r') - addNewLine s = if (B.last s /= '\n') then B.append s newLineB else s - parse = - fmap B.concat . (fmap . fmap $ flip B.append newLineB) . sequence - . fmap (processLine . clean) - . toLines - processLine l = - if isInclude l - then readSource' $ B.unpack . clean . cleanInclude $ l - else return l - - contents <- B.readFile file - - dir <- getCurrentDirectory - let dir' = dir ++ "/" ++ (reverse . dropWhile (\c -> c /= '/') . reverse) file - - setCurrentDirectory dir' - code <- parse contents - setCurrentDirectory dir - - return code - -class Uniform a where - -- | Load a list of uniform values. - uniform :: GLint -> a -> IO () - -instance Uniform Int where uniform loc a = glUniform1i loc (fromIntegral a) - -instance Uniform Float where uniform loc a = glUniform1f loc a - -instance Uniform CFloat where uniform loc a = glUniform1f loc (unsafeCoerce a) - -instance Uniform (Int, Int) where - uniform loc (x, y) = glUniform2i loc (fromIntegral x) (fromIntegral y) - -instance Uniform (Float, Float) where - uniform loc (x, y) = glUniform2f loc x y - -instance Uniform (Int, Int, Int) where - uniform loc (x, y, z) = glUniform3i loc (fromIntegral x) (fromIntegral y) (fromIntegral z) - -instance Uniform (Float, Float, Float) where - uniform loc (x, y, z) = glUniform3f loc x y z - -instance Uniform (Int, Int, Int, Int) where - uniform loc (x, y, z, w) = - glUniform4i - loc - (fromIntegral x) - (fromIntegral y) - (fromIntegral z) - (fromIntegral w) - -instance Uniform (Float, Float, Float, Float) where - uniform loc (x, y, z, w) = glUniform4f loc x y z w - -instance Uniform Vector2 where - uniform loc v = glUniform2f loc x' y' - where - x' = unsafeCoerce $ x v - y' = unsafeCoerce $ y v - -instance Uniform Vector3 where - uniform loc v = glUniform3f loc x' y' z' - where - x' = unsafeCoerce $ x v - y' = unsafeCoerce $ y v - z' = unsafeCoerce $ z v - -instance Uniform Vector4 where - uniform loc v = glUniform4f loc x' y' z' w' - where - x' = unsafeCoerce $ x v - y' = unsafeCoerce $ y v - z' = unsafeCoerce $ z v - w' = unsafeCoerce $ w v - -instance Uniform Matrix3 where - uniform loc mat = - with mat $ \ptrMat -> - glUniformMatrix3fv loc 1 (toEnum 0) (unsafeCoerce ptrMat) - -instance Uniform Matrix4 where - uniform loc mat = - with mat $ \ptrMat -> - glUniformMatrix4fv loc 1 (toEnum 0) (unsafeCoerce ptrMat) - -instance Uniform [Float] where - uniform loc vals = withArray (map unsafeCoerce vals) $ \ptr -> - case length vals of - 1 -> glUniform1fv loc 1 ptr - 2 -> glUniform2fv loc 1 ptr - 3 -> glUniform3fv loc 1 ptr - 4 -> glUniform4fv loc 1 ptr - -instance Uniform [CFloat] where - uniform loc vals = withArray vals $ \ptr -> - case length vals of - 1 -> glUniform1fv loc 1 $ castPtr ptr - 2 -> glUniform2fv loc 1 $ castPtr ptr - 3 -> glUniform3fv loc 1 $ castPtr ptr - 4 -> glUniform4fv loc 1 $ castPtr ptr - -instance Uniform [Int] where - uniform loc vals = withArray (map fromIntegral vals) $ \ptr -> - case length vals of - 1 -> glUniform1iv loc 1 ptr - 2 -> glUniform2iv loc 1 ptr - 3 -> glUniform3iv loc 1 ptr - 4 -> glUniform4iv loc 1 ptr - --- --- VAOs --- - --- | A vertex array object. -data VAO = VAO - { getVAO :: GLuint, - vaoKey :: ReleaseKey - } - -instance ResourceClass VAO where - getResource = vaoKey - -instance Eq VAO where - vao1 == vao2 = getVAO vao1 == getVAO vao2 - -instance Ord VAO where - vao1 < vao2 = getVAO vao1 < getVAO vao2 - vao1 <= vao2 = getVAO vao1 <= getVAO vao2 - --- | Create a new vao. -newVAO :: Game s VAO -newVAO = do - h <- liftIO . alloca $ \ptr -> do - glGenVertexArrays 1 ptr - peek ptr - - rkey <- register $ deleteVAO h - return $ VAO h rkey - --- | Delete the vao. -deleteVAO :: GLuint -> IO () -deleteVAO vao = Foreign.with vao $ glDeleteVertexArrays 1 - --- | Bind the vao. -bindVAO :: VAO -> IO () -bindVAO = glBindVertexArray . getVAO - --- | Unbind the bound vao. -unbindVAO :: IO () -unbindVAO = glBindVertexArray 0 - --- | Enable the given vertex attribute of the bound vao. --- --- See also 'bindVAO'. -enableVAOAttrib :: - -- | Attribute index. - GLuint -> - IO () -enableVAOAttrib = glEnableVertexAttribArray - --- | Bind the bound buffer to the given point. -attribVAOPointer :: - -- | The index of the generic vertex attribute to be modified. - GLuint -> - -- | The number of components per generic vertex attribute. Must be 1,2,3,4. - GLint -> - -- | The data type of each component in the array. - GLenum -> - -- | Whether fixed-point data values should be normalized. - Bool -> - -- | Stride. Byte offset between consecutive generic vertex attributes. - GLsizei -> - -- | Offset to the first component in the array. - Int -> - IO () -attribVAOPointer idx ncomp dattype normalise stride off = - glVertexAttribPointer idx ncomp dattype normalise' stride (unsafeCoerce off) - where - normalise' = if normalise then 1 else 0 - --- | Draw the bound vao. -drawArrays :: - -- | The kind of primitives to render. - GLenum -> - -- | Starting index in the enabled arrays. - Int -> - -- | The number of indices to be rendered. - Int -> - IO () -drawArrays mode first count = glDrawArrays mode (unsafeCoerce first) (unsafeCoerce count) - --- | Draw the bound vao, indexed mode. -drawElements :: - -- | The kind of primitives to render. - GLenum -> - -- | The number of elements to be rendered. - Int -> - -- | The type of the index values. Must be one of gl_UNSIGNED_BYTE, gl_UNSIGNED_SHORT, or gl_UNSIGNED_INT. - GLenum -> - -- | Pointer to the location where indices are stored, or offset into the index array when there is a bound ElementArrayBuffer. - Ptr a -> - IO () -drawElements mode count t idxs = glDrawElements mode (unsafeCoerce count) t idxs - --- --- BUFFER --- - --- | An OpenGL buffer. -data GLBuffer = GLBuffer - { getBuffer :: GLuint, - bufferKey :: ReleaseKey - } - -instance ResourceClass GLBuffer where - getResource = bufferKey - --- | The type of target buffer. -data TargetBuffer - = ArrayBuffer - | ElementArrayBuffer - | PixelPackBuffer - | PixelUnpackBuffer - deriving (Eq, Show) - -fromTarget :: TargetBuffer -> GLenum -fromTarget ArrayBuffer = GL_ARRAY_BUFFER -fromTarget ElementArrayBuffer = GL_ELEMENT_ARRAY_BUFFER -fromTarget PixelPackBuffer = GL_PIXEL_PACK_BUFFER -fromTarget PixelUnpackBuffer = GL_PIXEL_UNPACK_BUFFER - --- | A buffer usage. -data BufferUsage - = StreamDraw - | StreamRead - | StreamCopy - | StaticDraw - | StaticRead - | StaticCopy - | DynamicDraw - | DynamicRead - | DynamicCopy - deriving (Eq, Show) - -fromUsage :: BufferUsage -> GLenum -fromUsage StreamDraw = GL_STREAM_DRAW -fromUsage StreamRead = GL_STREAM_READ -fromUsage StreamCopy = GL_STREAM_COPY -fromUsage StaticDraw = GL_STATIC_DRAW -fromUsage StaticRead = GL_STATIC_READ -fromUsage StaticCopy = GL_STATIC_COPY -fromUsage DynamicDraw = GL_DYNAMIC_DRAW -fromUsage DynamicRead = GL_DYNAMIC_READ -fromUsage DynamicCopy = GL_DYNAMIC_COPY - --- | Create a new buffer. -newBuffer :: Game s GLBuffer -newBuffer = do - h <- liftIO . alloca $ \ptr -> do - glGenBuffers 1 ptr - peek ptr - - rkey <- register $ deleteBuffer h - return $ GLBuffer h rkey - --- | Delete the buffer. -deleteBuffer :: GLuint -> IO () -deleteBuffer buf = Foreign.with buf $ glDeleteBuffers 1 - --- | Bind the buffer. -bindBuffer :: TargetBuffer -> GLBuffer -> IO () -bindBuffer target buf = glBindBuffer (fromTarget target) $ getBuffer buf - --- | Unbind the bound buffer. -unbindBuffer :: TargetBuffer -> IO () -unbindBuffer target = glBindBuffer (fromTarget target) 0 - -class Storable a => BufferData a where - -- | Set the buffer's data. - bufferData :: TargetBuffer -> [a] -> BufferUsage -> IO () - bufferData tgt vals usage = - let n = sizeOf (head vals) * length vals - in withArray vals $ \ptr -> bufferData' tgt n ptr usage - -instance BufferData Word8 - -instance BufferData Word16 - -instance BufferData Word32 - -instance BufferData CChar - -instance BufferData CInt - -instance BufferData CFloat - -instance BufferData CDouble - -instance BufferData Int - -instance BufferData Float - -instance BufferData Double - -{-bufferData :: Storable a - => TargetBuffer - -> Int -- ^ The size in bytes of an element in the data list. - -> [a] -- ^ The data list. - -> BufferUsage - -> IO () -bufferData target n bufData usage = withArray bufData $ - \ptr -> bufferData target (n * length bufData) ptr usage-} - --- | Set the buffer's data. -bufferData' :: - TargetBuffer -> - -- | Buffer size in bytes. - Int -> - Ptr a -> - BufferUsage -> - IO () -bufferData' target n bufData usage = glBufferData (fromTarget target) (unsafeCoerce n) bufData (fromUsage usage) - --- | Apply the given function the buffer's id. -withGLBuffer :: GLBuffer -> (GLuint -> a) -> a -withGLBuffer buf f = f $ getBuffer buf - --- --- TEXTURE --- - --- | Represents a texture resource. -data Texture = Texture - { getTex :: GLuint, - texKey :: ReleaseKey - } - -instance Eq Texture where - t1 == t2 = getTex t1 == getTex t2 - -instance Ord Texture where - t1 < t2 = getTex t1 < getTex t2 - t1 <= t2 = getTex t1 <= getTex t2 - -instance ResourceClass Texture where - getResource = texKey - --- | Create a new texture. -newTexture :: Game s Texture -newTexture = do - tex <- liftIO . alloca $ \ptr -> do - glGenTextures 1 ptr - peek ptr - - rkey <- register $ deleteTexture tex - return $ Texture tex rkey - --- | Delete the texture. -deleteTexture :: GLuint -> IO () ---deleteTexture tex = with tex $ glDeleteTextures 1 -deleteTexture tex = do - putStrLn $ "Releasing texture " ++ show tex - with tex $ glDeleteTextures 1 - --- | Load the 'Texture' specified by the given file. -loadTextureImage :: - FilePath -> - -- | Texture's min filter. - GLenum -> - -- | Texture's mag filter. - GLenum -> - Game s Texture -loadTextureImage file minFilter magFilter = do - image <- loadImage file - tex <- newTexture - liftIO $ do - let w = width image - h = height image - pix = pixels image - rgb = fromIntegral . fromEnum $ GL_RGB - - bindTexture tex - loadTextureData GL_TEXTURE_2D 0 rgb w h 0 GL_RGB GL_UNSIGNED_BYTE pix - texParami GL_TEXTURE_2D GL_TEXTURE_MIN_FILTER $= minFilter - texParami GL_TEXTURE_2D GL_TEXTURE_MAG_FILTER $= magFilter - - return tex - --- | Bind the texture. -bindTexture :: Texture -> IO () -bindTexture = glBindTexture GL_TEXTURE_2D . getTex - --- | Unbind the bound texture. -unbindTexture :: IO () -unbindTexture = glBindTexture GL_TEXTURE_2D 0 - --- | Load data onto the bound texture. --- --- See also 'bindTexture'. -loadTextureData :: - GLenum -> - -- | Target - Int -> - -- | Level - Int -> - -- | Internal format - Int -> - -- | Width - Int -> - -- | Height - Int -> - -- | Border - GLenum -> - -- | Texture type - GLenum -> - -- | Texture data - Ptr a -> - IO () -loadTextureData target level internalFormat width height border format texType texData = do - glTexImage2D - target - (fromIntegral level) - (fromIntegral internalFormat) - (fromIntegral width) - (fromIntegral height) - (fromIntegral border) - (fromIntegral format) - texType - texData - --- | Set the bound texture's parameter to the given value. -texParami :: GLenum -> GLenum -> SettableStateVar GLenum -texParami target param = makeSettableStateVar $ \val -> glTexParameteri target param $ fromIntegral . fromEnum $ val - --- | Set the bound texture's parameter to the given value. -texParamf :: GLenum -> GLenum -> SettableStateVar Float -texParamf target param = makeSettableStateVar $ \val -> glTexParameterf target param (unsafeCoerce val) - --- | Set the active texture unit. -activeTexture :: SettableStateVar GLenum -activeTexture = makeSettableStateVar glActiveTexture - --- --- ERROR --- - --- | Get the last OpenGL error. -getGLError :: IO (Maybe String) -getGLError = fmap translate glGetError - where - translate err - | err == GL_NO_ERROR = Nothing - | err == GL_INVALID_ENUM = Just "Invalid enum" - | err == GL_INVALID_VALUE = Just "Invalid value" - | err == GL_INVALID_OPERATION = Just "Invalid operation" - | err == GL_OUT_OF_MEMORY = Just "Out of memory" - | otherwise = Just "Unknown error" - --- | Print the last OpenGL error. -printGLError :: IO () -printGLError = - getGLError >>= \err -> case err of - Nothing -> return () - Just str -> hPutStrLn stderr str - --- | Run the given setup action and check for OpenGL errors. --- --- If an OpenGL error is produced, an exception is thrown containing --- the given string appended to the string describing the error. -assertGL :: Game s a -> String -> Game s a -assertGL action err = do - result <- action - status <- liftIO getGLError - case status of - Just str -> gameError $ "OpenGL error raised: " ++ err ++ "; " ++ str - Nothing -> return result diff --git a/Spear/Math/Collision.hs b/Spear/Math/Collision.hs deleted file mode 100644 index 4412b10..0000000 --- a/Spear/Math/Collision.hs +++ /dev/null @@ -1,239 +0,0 @@ -{-# LANGUAGE NoImplicitPrelude #-} - -module Spear.Math.Collision -( - CollisionType(..) - -- * 2D Collision -, Collisionable2(..) -, Collisioner2(..) - -- ** Construction -, aabb2Collisioner -, circleCollisioner -, mkCols - -- ** Collision test -, collide - -- ** Manipulation -, move - -- ** Helpers -, buildAABB2 -, aabb2FromCircle -, circleFromAABB2 - -- * 3D Collision -, Collisionable3(..) - -- ** Helpers -, aabb3FromSphere -) -where - -import Spear.Assets.Model -import Spear.Math.AABB -import Spear.Math.Algebra -import Spear.Math.Circle -import qualified Spear.Math.Matrix4 as M4 -import Spear.Math.Plane -import Spear.Math.Sphere -import Spear.Math.Vector -import Spear.Prelude - -import Data.List (foldl') - -data CollisionType = NoCollision | Collision | FullyContains | FullyContainedBy - deriving (Eq, Show) - --- 2D collision - -class Collisionable2 a where - -- | Collide the object with an AABB. - collideAABB2 :: AABB2 -> a -> CollisionType - - -- | Collide the object with a circle. - collideCircle :: Circle -> a -> CollisionType - -instance Collisionable2 AABB2 where - collideAABB2 box1@(AABB2 min1 max1) box2@(AABB2 min2 max2) - | (x max1) < (x min2) = NoCollision - | (x min1) > (x max2) = NoCollision - | (y max1) < (y min2) = NoCollision - | (y min1) > (y max2) = NoCollision - | box1 `aabb2pt` min2 && box1 `aabb2pt` max2 = FullyContains - | box2 `aabb2pt` min1 && box2 `aabb2pt` max1 = FullyContainedBy - | otherwise = Collision - - collideCircle circle@(Circle c r) aabb@(AABB2 min max) - | test == FullyContains || test == FullyContainedBy = test - | normSq (c - boxC) > (l + r)^2 = NoCollision - | otherwise = Collision - where - test = collideAABB2 aabb $ aabb2FromCircle circle - boxC = min + (max-min) / (2::Float) - l = norm $ min + (vec2 (x boxC) (y min)) - min - -instance Collisionable2 Circle where - collideAABB2 box circle = case collideCircle circle box of - FullyContains -> FullyContainedBy - FullyContainedBy -> FullyContains - x -> x - - collideCircle s1@(Circle c1 r1) s2@(Circle c2 r2) - | distance_centers <= sub_radii = if (r1 > r2) then FullyContains else FullyContainedBy - | distance_centers <= sum_radii = Collision - | otherwise = NoCollision - where - distance_centers = normSq $ c1 - c2 - sum_radii = (r1 + r2)^2 - sub_radii = (r1 - r2)^2 - -instance Collisionable2 Collisioner2 where - collideAABB2 box (AABB2Col self) = collideAABB2 box self - collideAABB2 box (CircleCol self) = collideAABB2 box self - - collideCircle circle (AABB2Col self) = collideCircle circle self - collideCircle circle (CircleCol self) = collideCircle circle self - - -aabbPoints :: AABB2 -> [Vector2] -aabbPoints (AABB2 min max) = [p1,p2,p3,p4,p5,p6,p7,p8] - where - p1 = vec2 (x min) (y min) - p2 = vec2 (x min) (y min) - p3 = vec2 (x min) (y max) - p4 = vec2 (x min) (y max) - p5 = vec2 (x max) (y min) - p6 = vec2 (x max) (y min) - p7 = vec2 (x max) (y max) - p8 = vec2 (x max) (y max) - - --- | A collisioner component. -data Collisioner2 - -- | An axis-aligned bounding box. - = AABB2Col {-# UNPACK #-} !AABB2 - -- | A bounding circle. - | CircleCol {-# UNPACK #-} !Circle - - --- | Create a collisioner from the specified box. -aabb2Collisioner :: AABB2 -> Collisioner2 -aabb2Collisioner = AABB2Col - --- | Create a collisioner from the specified circle. -circleCollisioner :: Circle -> Collisioner2 -circleCollisioner = CircleCol - --- | Compute AABB collisioners in view space from the given AABB. -mkCols :: M4.Matrix4 -- ^ Modelview matrix - -> Box - -> [Collisioner2] -mkCols modelview (Box (Vec3 xmin ymin zmin) (Vec3 xmax ymax zmax)) = - let - toVec2 v = vec2 (x v) (y v) - p1 = toVec2 $ modelview `M4.mulp` vec3 xmin ymin zmax - p2 = toVec2 $ modelview `M4.mulp` vec3 xmax ymin zmin - p3 = toVec2 $ modelview `M4.mulp` vec3 xmax ymax zmin - col1 = AABB2Col $ AABB2 p1 p2 - col2 = AABB2Col $ AABB2 p1 p3 - in - [col1, col2] - --- | Create the minimal AABB fully containing the specified collisioners. -buildAABB2 :: [Collisioner2] -> AABB2 -buildAABB2 cols = aabb2 $ generatePoints cols - --- | Create the minimal box fully containing the specified circle. -aabb2FromCircle :: Circle -> AABB2 -aabb2FromCircle (Circle c r) = AABB2 bot top - where - bot = c - vec2 r r - top = c + vec2 r r - --- | Create the minimal circle fully containing the specified box. -circleFromAABB2 :: AABB2 -> Circle -circleFromAABB2 (AABB2 min max) = Circle c r - where - c = (0.5::Float) * (min + max) - r = norm . (*(0.5::Float)) $ max - min - -generatePoints :: [Collisioner2] -> [Vector2] -generatePoints = foldl' generate [] - where - generate acc (AABB2Col (AABB2 pmin pmax)) = p1:p2:p3:p4:p5:p6:p7:p8:acc - where - p1 = vec2 (x pmin) (y pmin) - p2 = vec2 (x pmin) (y pmin) - p3 = vec2 (x pmin) (y pmax) - p4 = vec2 (x pmin) (y pmax) - p5 = vec2 (x pmax) (y pmin) - p6 = vec2 (x pmax) (y pmin) - p7 = vec2 (x pmax) (y pmax) - p8 = vec2 (x pmax) (y pmax) - - generate acc (CircleCol (Circle c r)) = p1:p2:p3:p4:acc - where - p1 = c + unitx2 * vec2 r r - p2 = c - unitx2 * vec2 r r - p3 = c + unity2 * vec2 r r - p4 = c - unity2 * vec2 r r - --- | Collide the given collisioners. -collide :: Collisioner2 -> Collisioner2 -> CollisionType -collide (AABB2Col box1) (AABB2Col box2) = collideAABB2 box1 box2 -collide (AABB2Col box) (CircleCol circle) = collideAABB2 box circle -collide (CircleCol s1) (CircleCol s2) = collideCircle s1 s2 -collide (CircleCol circle) (AABB2Col box) = collideCircle circle box - --- | Move the collisioner. -move :: Vector2 -> Collisioner2 -> Collisioner2 -move v (AABB2Col (AABB2 min max)) = AABB2Col (AABB2 (min+v) (max+v)) -move v (CircleCol (Circle c r)) = CircleCol (Circle (c+v) r) - - --- | 3D collision -class Collisionable3 a where - -- | Collide the object with an AABB. - collideAABB3 :: AABB3 -> a -> CollisionType - - -- | Collide the object with a sphere. - collideSphere :: Sphere -> a -> CollisionType - -instance Collisionable3 AABB3 where - collideAABB3 box1@(AABB3 min1 max1) box2@(AABB3 min2 max2) - | x max1 < x min2 = NoCollision - | x min1 > x max2 = NoCollision - | y max1 < y min2 = NoCollision - | y min1 > y max2 = NoCollision - | box1 `aabb3pt` min2 && box1 `aabb3pt` max2 = FullyContains - | box2 `aabb3pt` min1 && box2 `aabb3pt` max1 = FullyContainedBy - | otherwise = Collision - - collideSphere sphere@(Sphere c r) aabb@(AABB3 min max) - | test == FullyContains || test == FullyContainedBy = test - | normSq (c - boxC) > (l + r)^2 = NoCollision - | otherwise = Collision - where - test = collideAABB3 aabb $ aabb3FromSphere sphere - boxC = min + v - l = norm v - v = (max-min) / (2::Float) - -instance Collisionable3 Sphere where - collideAABB3 box sphere = case collideSphere sphere box of - FullyContains -> FullyContainedBy - FullyContainedBy -> FullyContains - x -> x - - collideSphere s1@(Sphere c1 r1) s2@(Sphere c2 r2) - | distance_centers <= sub_radii = - if r1 > r2 then FullyContains else FullyContainedBy - | distance_centers <= sum_radii = Collision - | otherwise = NoCollision - where - distance_centers = normSq $ c1 - c2 - sum_radii = (r1 + r2)^2 - sub_radii = (r1 - r2)^2 - --- | Create the minimal box fully containing the specified sphere. -aabb3FromSphere :: Sphere -> AABB3 -aabb3FromSphere (Sphere c r) = AABB3 bot top - where - bot = c - vec3 r r r - top = c + vec3 r r r diff --git a/Spear/Math/Octree.hs b/Spear/Math/Octree.hs deleted file mode 100644 index 6c22468..0000000 --- a/Spear/Math/Octree.hs +++ /dev/null @@ -1,228 +0,0 @@ -module Spear.Math.Octree -( - Octree -, makeOctree -, clone -, Spear.Math.Octree.insert -, Spear.Math.Octree.map -, gmap -) -where - -import Spear.Math.AABB -import Spear.Math.Collision -import Spear.Math.Vector - -import Control.Applicative ((<*>)) -import Data.List -import Data.Functor -import Data.Monoid -import qualified Data.Foldable as F - --- | An octree. -data Octree e - = Octree - { root :: !AABB2 - , ents :: ![e] - , c1 :: !(Octree e) - , c2 :: !(Octree e) - , c3 :: !(Octree e) - , c4 :: !(Octree e) - , c5 :: !(Octree e) - , c6 :: !(Octree e) - , c7 :: !(Octree e) - , c8 :: !(Octree e) - } - | - Leaf - { root :: !AABB2 - , ents :: ![e] - } - --- | Construct an octree using the specified AABB as the root and having the specified depth. -makeOctree :: Int -> AABB2 -> Octree e -makeOctree d root@(AABB2 min max) - | d == 0 = Leaf root [] - | otherwise = Octree root [] c1 c2 c3 c4 c5 c6 c7 c8 - where - boxes = subdivide root - c1 = makeOctree (d-1) $ boxes !! 0 - c2 = makeOctree (d-1) $ boxes !! 1 - c3 = makeOctree (d-1) $ boxes !! 2 - c4 = makeOctree (d-1) $ boxes !! 3 - c5 = makeOctree (d-1) $ boxes !! 4 - c6 = makeOctree (d-1) $ boxes !! 5 - c7 = makeOctree (d-1) $ boxes !! 6 - c8 = makeOctree (d-1) $ boxes !! 7 - -subdivide :: AABB2 -> [AABB2] -subdivide (AABB2 min max) = [a1, a2, a3, a4, a5, a6, a7, a8] - where - v = (max-min) / 2 - c = vec2 (x min + x v) (y min + y v) - a1 = AABB2 min c - a2 = AABB2 ( vec2 (x min) (y min)) ( vec2 (x c) (y c) ) - a3 = AABB2 ( vec2 (x min) (y c) ) ( vec2 (x c) (y max)) - a4 = AABB2 ( vec2 (x min) (y c) ) ( vec2 (x c) (y max)) - a5 = AABB2 ( vec2 (x c) (y min)) ( vec2 (x max) (y c) ) - a6 = AABB2 ( vec2 (x c) (y min)) ( vec2 (x max) (y c) ) - a7 = AABB2 ( vec2 (x c) (y c) ) ( vec2 (x max) (y max)) - a8 = AABB2 c max - --- | Clone the structure of the octree. The new octree has no entities. -clone :: Octree e -> Octree e -clone (Leaf root ents) = Leaf root [] -clone (Octree root ents c1 c2 c3 c4 c5 c6 c7 c8) = Octree root [] c1' c2' c3' c4' c5' c6' c7' c8' - where - c1' = clone c1 - c2' = clone c2 - c3' = clone c3 - c4' = clone c4 - c5' = clone c5 - c6' = clone c6 - c7' = clone c7 - c8' = clone c8 - -keep :: (e -> AABB2 -> CollisionType) -> AABB2 -> e -> Bool -keep testAABB2 aabb e = test == FullyContainedBy - where test = e `testAABB2` aabb - --- | Insert a list of entities into the octree. -insert :: (e -> AABB2 -> CollisionType) -> Octree e -> [e] -> Octree e -insert testAABB2 octree es = octree' where (octree', _) = insert' testAABB2 es octree - -insert' :: (e -> AABB2 -> CollisionType) -> [e] -> Octree e -> (Octree e, [e]) - -insert' testAABB2 es (Leaf root ents) = (Leaf root ents', outliers) - where - ents' = ents ++ ents_kept - ents_kept = filter (keep testAABB2 root) es - outliers = filter (not . keep testAABB2 root) es - -insert' testAABB2 es (Octree root ents c1 c2 c3 c4 c5 c6 c7 c8) = - (Octree root ents' c1' c2' c3' c4' c5' c6' c7' c8', outliers) - where - ents' = ents ++ ents_kept - new_ents = es ++ ents1 ++ ents2 ++ ents3 ++ ents4 ++ ents5 ++ ents6 ++ ents7 ++ ents8 - ents_kept = filter (keep testAABB2 root) new_ents - outliers = filter (not . keep testAABB2 root) new_ents - (c1', ents1) = insert' testAABB2 es c1 - (c2', ents2) = insert' testAABB2 es c2 - (c3', ents3) = insert' testAABB2 es c3 - (c4', ents4) = insert' testAABB2 es c4 - (c5', ents5) = insert' testAABB2 es c5 - (c6', ents6) = insert' testAABB2 es c6 - (c7', ents7) = insert' testAABB2 es c7 - (c8', ents8) = insert' testAABB2 es c8 - --- | Extract all entities from the octree. The resulting octree has no entities. -extract :: Octree e -> (Octree e, [e]) -extract (Leaf root ents) = (Leaf root [], ents) -extract (Octree root ents c1 c2 c3 c4 c5 c6 c7 c8) = (Octree root [] c1' c2' c3' c4' c5' c6' c7' c8', ents') - where - (c1', ents1) = extract c1 - (c2', ents2) = extract c2 - (c3', ents3) = extract c3 - (c4', ents4) = extract c4 - (c5', ents5) = extract c5 - (c6', ents6) = extract c6 - (c7', ents7) = extract c7 - (c8', ents8) = extract c8 - ents' = ents ++ ents1 ++ ents2 ++ ents3 ++ ents4 ++ ents5 ++ ents6 ++ ents7 ++ ents8 - --- | Apply the given function to the entities in the octree. --- --- Entities that break out of their cell are reallocated appropriately. -map :: (e -> AABB2 -> CollisionType) -> (e -> e) -> Octree e -> Octree e -map testAABB2 f o = - let (o', outliers) = map' testAABB2 f o - in Spear.Math.Octree.insert testAABB2 o' outliers - -map' :: (e -> AABB2 -> CollisionType) -> (e -> e) -> Octree e -> (Octree e, [e]) - -map' testAABB2 f (Leaf root ents) = (Leaf root ents_kept, outliers) - where - ents' = fmap f ents - ents_kept = filter (keep testAABB2 root) ents' - outliers = filter (not . keep testAABB2 root) ents' - -map' testAABB2 f (Octree root ents c1 c2 c3 c4 c5 c6 c7 c8) = - (Octree root ents_kept c1' c2' c3' c4' c5' c6' c7' c8', outliers) - where - ents' = (fmap f ents) ++ out1 ++ out2 ++ out3 ++ out4 ++ out5 ++ out6 ++ out7 ++ out8 - ents_kept = filter (keep testAABB2 root) ents' - outliers = filter (not . keep testAABB2 root) ents' - (c1', out1) = map' testAABB2 f c1 - (c2', out2) = map' testAABB2 f c2 - (c3', out3) = map' testAABB2 f c3 - (c4', out4) = map' testAABB2 f c4 - (c5', out5) = map' testAABB2 f c5 - (c6', out6) = map' testAABB2 f c6 - (c7', out7) = map' testAABB2 f c7 - (c8', out8) = map' testAABB2 f c8 - - --- | Apply a function to the entity groups in the octree. --- --- Entities that break out of their cell are reallocated appropriately. -gmap :: (e -> AABB2 -> CollisionType) -> (e -> e -> e) -> Octree e -> Octree e -gmap testAABB2 f o = - let (o', outliers) = gmap' testAABB2 f o - in Spear.Math.Octree.insert testAABB2 o' outliers - -gmap' :: (e -> AABB2 -> CollisionType) -> (e -> e -> e) -> Octree e -> (Octree e, [e]) - -gmap' testAABB2 f (Leaf root ents) = (Leaf root ents_kept, outliers) - where - ents' = f <$> ents <*> ents - ents_kept = filter (keep testAABB2 root) ents' - outliers = filter (not . keep testAABB2 root) ents' - -gmap' testAABB2 f (Octree root ents c1 c2 c3 c4 c5 c6 c7 c8) = - (Octree root ents_kept c1' c2' c3' c4' c5' c6' c7' c8', outliers) - where - ents'= (f <$> ents <*> ents) ++ out1 ++ out2 ++ out3 ++ out4 ++ out5 ++ out6 ++ out7 ++ out8 - ents_kept = filter (keep testAABB2 root) ents' - outliers = filter (not . keep testAABB2 root) ents' - (c1', out1) = gmap' testAABB2 f c1 - (c2', out2) = gmap' testAABB2 f c2 - (c3', out3) = gmap' testAABB2 f c3 - (c4', out4) = gmap' testAABB2 f c4 - (c5', out5) = gmap' testAABB2 f c5 - (c6', out6) = gmap' testAABB2 f c6 - (c7', out7) = gmap' testAABB2 f c7 - (c8', out8) = gmap' testAABB2 f c8 - -instance Functor Octree where - - fmap f (Leaf root ents) = Leaf root $ fmap f ents - - fmap f (Octree root ents c1 c2 c3 c4 c5 c6 c7 c8) = - Octree root (fmap f ents) c1' c2' c3' c4' c5' c6' c7' c8' - where - c1' = fmap f c1 - c2' = fmap f c2 - c3' = fmap f c3 - c4' = fmap f c4 - c5' = fmap f c5 - c6' = fmap f c6 - c7' = fmap f c7 - c8' = fmap f c8 - -instance F.Foldable Octree where - - foldMap f (Leaf root ents) = mconcat . fmap f $ ents - - foldMap f (Octree root ents c1 c2 c3 c4 c5 c6 c7 c8) = - mconcat (fmap f ents) `mappend` - c1' `mappend` c2' `mappend` c3' `mappend` c4' `mappend` - c5' `mappend` c6' `mappend` c7' `mappend` c8' - where - c1' = F.foldMap f c1 - c2' = F.foldMap f c2 - c3' = F.foldMap f c3 - c4' = F.foldMap f c4 - c5' = F.foldMap f c5 - c6' = F.foldMap f c6 - c7' = F.foldMap f c7 - c8' = F.foldMap f c8 diff --git a/Spear/Render/AnimatedModel.hs b/Spear/Render/AnimatedModel.hs deleted file mode 100644 index e5b29ec..0000000 --- a/Spear/Render/AnimatedModel.hs +++ /dev/null @@ -1,262 +0,0 @@ - -{-# LANGUAGE NoImplicitPrelude #-} - -module Spear.Render.AnimatedModel - ( -- * Data types - AnimatedModelResource, - AnimatedModelRenderer, - AnimationSpeed, - - -- * Construction and destruction - animatedModelResource, - animatedModelRenderer, - - -- * Accessors - animationSpeed, - box, - currentAnimation, - currentFrame, - frameProgress, - modelRes, - nextFrame, - - -- * Manipulation - update, - setAnimation, - setAnimationSpeed, - - -- * Rendering - bind, - render, - - -- * Collision - mkColsFromAnimated, - ) -where - -import Spear.Assets.Model -import Spear.Game -import Spear.GL -import Spear.Math.AABB -import Spear.Math.Algebra -import Spear.Math.Collision -import Spear.Math.Matrix4 (Matrix4) -import Spear.Math.Vector -import Spear.Prelude -import Spear.Render.Material -import Spear.Render.Model -import Spear.Render.Program - -import Control.Applicative ((<$>), (<*>)) -import qualified Data.Vector as V -import Foreign.C.Types -import Unsafe.Coerce (unsafeCoerce) - - -type AnimationSpeed = Float - --- | An animated model resource. --- --- Contains model data necessary to render an animated model. -data AnimatedModelResource = AnimatedModelResource - { model :: Model, - vao :: VAO, - nFrames :: Int, - nVertices :: Int, - material :: Material, - texture :: Texture, - boxes :: V.Vector Box, - rkey :: ReleaseKey - } - -instance Eq AnimatedModelResource where - m1 == m2 = vao m1 == vao m2 - -instance Ord AnimatedModelResource where - m1 < m2 = vao m1 < vao m2 - m1 <= m2 = vao m1 <= vao m2 - -instance ResourceClass AnimatedModelResource where - getResource = rkey - --- | An animated model renderer. --- --- Holds animation data necessary to render an animated model and a reference --- to an 'AnimatedModelResource'. --- --- Model data is kept separate from animation data. This allows instances --- of 'AnimatedModelRenderer' to share the underlying 'AnimatedModelResource', --- minimising the amount of data in memory and allowing one to minimise OpenGL --- state changes by sorting 'AnimatedModelRenderer's by their underlying --- 'AnimatedModelResource' when rendering the scene. -data AnimatedModelRenderer = AnimatedModelRenderer - { modelResource :: AnimatedModelResource, - currentAnim :: Int, - frameStart :: Int, - frameEnd :: Int, - -- | Get the renderer's current frame. - currentFrame :: Int, - -- | Get the renderer's frame progress. - frameProgress :: Float, - -- | Get the renderer's animation speed. - animationSpeed :: Float - } - -instance Eq AnimatedModelRenderer where - m1 == m2 = modelResource m1 == modelResource m2 - -instance Ord AnimatedModelRenderer where - m1 < m2 = modelResource m1 < modelResource m2 - m1 <= m2 = modelResource m1 <= modelResource m2 - --- | Create an model resource from the given model. -animatedModelResource :: - AnimatedProgramChannels -> - Material -> - Texture -> - Model -> - Game s AnimatedModelResource -animatedModelResource - (AnimatedProgramChannels vertChan1 vertChan2 normChan1 normChan2 texChan) - material - texture - model = do - RenderModel elements numFrames numVertices <- liftIO . renderModelFromModel $ model - elementBuf <- newBuffer - vao <- newVAO - boxes <- liftIO $ modelBoxes model - - liftIO $ do - let elemSize = 56::CUInt - elemSize' = fromIntegral elemSize - n = numVertices * numFrames - - bindVAO vao - - bindBuffer ArrayBuffer elementBuf - bufferData' ArrayBuffer (unsafeCoerce $ elemSize * n) elements StaticDraw - - attribVAOPointer vertChan1 3 GL_FLOAT False elemSize' 0 - attribVAOPointer vertChan2 3 GL_FLOAT False elemSize' 12 - attribVAOPointer normChan1 3 GL_FLOAT False elemSize' 24 - attribVAOPointer normChan2 3 GL_FLOAT False elemSize' 36 - attribVAOPointer texChan 2 GL_FLOAT False elemSize' 48 - - enableVAOAttrib vertChan1 - enableVAOAttrib vertChan2 - enableVAOAttrib normChan1 - enableVAOAttrib normChan2 - enableVAOAttrib texChan - - rkey <- register $ do - release $ vaoKey vao - release $ bufferKey elementBuf - - return $ - AnimatedModelResource - model - vao - (unsafeCoerce numFrames) - (unsafeCoerce numVertices) - material - texture - boxes - rkey - --- | Create a renderer from the given model resource. -animatedModelRenderer :: AnimationSpeed -> AnimatedModelResource -> AnimatedModelRenderer -animatedModelRenderer animSpeed modelResource = - AnimatedModelRenderer modelResource 0 0 0 0 0 animSpeed - --- | Update the renderer. -update :: Float -> AnimatedModelRenderer -> AnimatedModelRenderer -update dt (AnimatedModelRenderer model curAnim startFrame endFrame curFrame fp s) = - AnimatedModelRenderer model curAnim startFrame endFrame curFrame' fp' s - where - f = fp + dt * s - nextFrame = f >= 1.0 - fp' = if nextFrame then f - (1::Float) else f - curFrame' = - if nextFrame - then - let x = curFrame + (1::Int) - in if x > endFrame then startFrame else x - else curFrame - --- | Get the model's ith bounding box. -box :: Int -> AnimatedModelResource -> Box -box i model = boxes model V.! i - --- | Get the renderer's current animation. -currentAnimation :: Enum a => AnimatedModelRenderer -> a -currentAnimation = toEnum . currentAnim - --- | Get the renderer's model resource. -modelRes :: AnimatedModelRenderer -> AnimatedModelResource -modelRes = modelResource - --- | Get the renderer's next frame. -nextFrame :: AnimatedModelRenderer -> Int -nextFrame rend = - let curFrame = currentFrame rend - in if curFrame == frameEnd rend - then frameStart rend - else curFrame + (1::Int) - --- | Set the active animation to the given one. -setAnimation :: Enum a => a -> AnimatedModelRenderer -> AnimatedModelRenderer -setAnimation anim modelRend = - let (Animation _ f1 f2) = animation (model . modelResource $ modelRend) anim' - anim' = fromEnum anim - in modelRend {currentAnim = anim', frameStart = f1, frameEnd = f2, currentFrame = f1} - --- | Set the renderer's animation speed. -setAnimationSpeed :: AnimationSpeed -> AnimatedModelRenderer -> AnimatedModelRenderer -setAnimationSpeed s r = r {animationSpeed = s} - --- | Bind the given renderer to prepare it for rendering. -bind :: AnimatedProgramUniforms -> AnimatedModelRenderer -> IO () -bind (AnimatedProgramUniforms kaLoc kdLoc ksLoc shiLoc texLoc _ _ _ _) modelRend = - let model' = modelResource modelRend - in do - bindVAO . vao $ model' - bindTexture $ texture model' - activeTexture $= GL_TEXTURE0 - glUniform1i texLoc 0 - --- | Render the model described by the given renderer. -render :: AnimatedProgramUniforms -> AnimatedModelRenderer -> IO () -render uniforms (AnimatedModelRenderer model _ _ _ curFrame fp _) = - let n = nVertices model - (Material _ ka kd ks shi) = material model - in do - uniform (kaLoc uniforms) ka - uniform (kdLoc uniforms) kd - uniform (ksLoc uniforms) ks - glUniform1f (shiLoc uniforms) $ unsafeCoerce shi - glUniform1f (fpLoc uniforms) (unsafeCoerce fp) - drawArrays GL_TRIANGLES (n * curFrame) n - --- | Compute AABB collisioners in view space from the given model. -mkColsFromAnimated :: - -- | Source frame - Int -> - -- | Dest frame - Int -> - -- | Frame progress - Float -> - -- | Modelview matrix - Matrix4 -> - AnimatedModelResource -> - [Collisioner2] -mkColsFromAnimated f1 f2 fp modelview modelRes = - let (Box (Vec3 xmin1 ymin1 zmin1) (Vec3 xmax1 ymax1 zmax1)) = box f1 modelRes - (Box (Vec3 xmin2 ymin2 zmin2) (Vec3 xmax2 ymax2 zmax2)) = box f2 modelRes - min1 = vec3 xmin1 ymin1 zmin1 - max1 = vec3 xmax1 ymax1 zmax1 - min2 = vec3 xmin2 ymin2 zmin2 - max2 = vec3 xmax2 ymax2 zmax2 - min = min1 + fp * (min2 - min1) - max = max1 + fp * (max2 - max1) - in mkCols modelview $ - Box (Vec3 (x min) (y min) (z min)) (Vec3 (x max) (y max) (z max)) diff --git a/Spear/Render/Asset.hs b/Spear/Render/Asset.hs new file mode 100644 index 0000000..626fff5 --- /dev/null +++ b/Spear/Render/Asset.hs @@ -0,0 +1,65 @@ +module Spear.Render.Asset +( + TextureColourSpace(..) +, TextureOrigin(..) +, TextureLoadDesc(..) +, loadTexture +) +where + + +import Spear.Asset.Image as Image +import Spear.Game +import Spear.Render.Core.State +import Spear.Render.Core.Texture + +import Foreign.Ptr + + +data TextureColourSpace + = SRGBColourSpace + | LinearColourSpace + deriving (Eq, Show) + +newtype TextureOrigin + = TextureFromFile FilePath + deriving Show + +data TextureLoadDesc = TextureLoadDesc + { textureLoadDescOrigin :: TextureOrigin + , textureLoadDescColourSpace :: TextureColourSpace + , textureLoadDescFilter :: TextureFiltering + , textureLoadDescWrap :: TextureWrapping + , textureLoadDescMips :: Bool + } + deriving Show + + +loadTexture :: TextureLoadDesc -> Game RenderCoreState Texture +loadTexture desc = case textureLoadDescOrigin desc of + TextureFromFile filepath -> loadTextureFromFile desc + +-- Private + +loadTextureFromFile :: TextureLoadDesc -> Game RenderCoreState Texture +loadTextureFromFile (TextureLoadDesc (TextureFromFile filepath) space filter wrap mips) = do + result <- withImage filepath $ \image -> + makeTexture $ TextureDesc + (Image.width image) + (Image.height image) + 0 -- depth + Texture2D + (getTextureFormat (Image.components image) space) + filter + wrap + mips + (TextureData . castPtr $ Image.pixels image) + case result of + Left err -> gameError err + Right _ -> return undefined + +getTextureFormat :: Int -> TextureColourSpace -> TextureFormat +getTextureFormat 3 SRGBColourSpace = TextureSRGB8 +getTextureFormat 3 LinearColourSpace = TextureRGB8 +getTextureFormat 4 SRGBColourSpace = TextureSRGBA8 +getTextureFormat 4 LinearColourSpace = TextureRGBA8 diff --git a/Spear/Render/Box.hs b/Spear/Render/Box.hs deleted file mode 100644 index 305ef32..0000000 --- a/Spear/Render/Box.hs +++ /dev/null @@ -1,193 +0,0 @@ -module Spear.Render.Box -( - render -, renderOutwards -, renderInwards -, renderEdges -) -where - - -import Spear.Math.Vector3 -import Spear.Math.Matrix -import Graphics.Rendering.OpenGL.Raw -import Unsafe.Coerce -import Control.Monad.Instances - -type Center = Vector3 -type Colour = Vector4 -type Length = Float -type Normals = [Vector3] -type GenerateTexCoords = Bool - - -applyColour :: Colour -> IO () ---applyColour col = glColor4f (unsafeCoerce $ x col) (unsafeCoerce $ y col) (unsafeCoerce $ z col) (unsafeCoerce $ w col) -applyColour = do - ax <- unsafeCoerce . x - ay <- unsafeCoerce . y - az <- unsafeCoerce . z - aw <- unsafeCoerce . w - glColor4f ax ay az aw - - -applyNormal :: Vector3 -> IO () ---applyNormal v = glNormal3f (unsafeCoerce $ x v) (unsafeCoerce $ y v) (unsafeCoerce $ z v) -applyNormal = do - nx <- unsafeCoerce . x - ny <- unsafeCoerce . y - nz <- unsafeCoerce . z - glNormal3f nx ny nz - - --- | Renders a box. -render :: Center -- ^ The box's center. - -> Length -- ^ The perpendicular distance from the box's center to any of its sides. - -> Colour -- ^ The box's colour. - -> Normals -- ^ The box's normals, of the form [front, back, right, left, top, bottom]. - -> IO () -render c l col normals = do - glPushMatrix - glTranslatef (unsafeCoerce $ x c) (unsafeCoerce $ y c) (unsafeCoerce $ z c) - applyColour col - - let d = unsafeCoerce l - glBegin gl_QUADS - - --Front - --glNormal3f 0 0 (-1) - applyNormal $ normals !! 0 - glVertex3f d (-d) (-d) - glVertex3f d d (-d) - glVertex3f (-d) d (-d) - glVertex3f (-d) (-d) (-d) - - --Back - --glNormal3f 0 0 1 - applyNormal $ normals !! 1 - glVertex3f (-d) (-d) d - glVertex3f (-d) d d - glVertex3f d d d - glVertex3f d (-d) d - - --Right - --glNormal3f 1 0 0 - applyNormal $ normals !! 2 - glVertex3f d (-d) (-d) - glVertex3f d (-d) d - glVertex3f d d d - glVertex3f d d (-d) - - --Left - --glNormal3f (-1) 0 0 - applyNormal $ normals !! 3 - glVertex3f (-d) (-d) (-d) - glVertex3f (-d) d (-d) - glVertex3f (-d) d d - glVertex3f (-d) (-d) d - - --Top - --glNormal3f 0 1 0 - applyNormal $ normals !! 4 - glVertex3f (-d) d (-d) - glVertex3f d d (-d) - glVertex3f d d d - glVertex3f (-d) d d - - --Bottom - --glNormal3f 0 (-1) 0 - applyNormal $ normals !! 5 - glVertex3f d (-d) d - glVertex3f d (-d) (-d) - glVertex3f (-d) (-d) (-d) - glVertex3f (-d) (-d) d - - glEnd - - glPopMatrix - - -normals = [vec3 0 0 (-1), vec3 0 0 1, vec3 1 0 0, vec3 (-1) 0 0, vec3 0 1 0, vec3 0 (-1) 0] - - --- | Renders a box with normals facing outwards. -renderOutwards :: Center -- ^ The box's center. - -> Length -- ^ The perpendicular distance from the box's center to any of its sides. - -> Colour -- ^ The box's colour. - -> IO () -renderOutwards c l col = render c l col normals - - --- | Renders a box with normals facing inwards. -renderInwards :: Center -- ^ The box's center. - -> Length -- ^ The perpendicular distance from the box's center to any of its sides. - -> Colour -- ^ The box's colour. - -> IO () -renderInwards c l col = do - glFrontFace gl_CW - render c l col $ Prelude.map neg normals - glFrontFace gl_CCW - - -renderEdges :: Center -- ^ The box's center. - -> Length -- ^ The perpendicular distance from the box's center to any of its sides. - -> Colour -- ^ The box's colour. - -> IO () -renderEdges c l col = do - glPushMatrix - glTranslatef (unsafeCoerce $ x c) (unsafeCoerce $ y c) (unsafeCoerce $ z c) - applyColour col - - let d = unsafeCoerce l - - --Front - glBegin gl_LINE_STRIP - glVertex3f d (-d) (-d) - glVertex3f d d (-d) - glVertex3f (-d) d (-d) - glVertex3f (-d) (-d) (-d) - glEnd - - --Back - glBegin gl_LINE_STRIP - glVertex3f (-d) (-d) d - glVertex3f (-d) d d - glVertex3f d d d - glVertex3f d (-d) d - glVertex3f (-d) (-d) d - glEnd - - --Right - glBegin gl_LINE_STRIP - glVertex3f d (-d) (-d) - glVertex3f d (-d) d - glVertex3f d d d - glVertex3f d d (-d) - glEnd - - --Left - glBegin gl_LINE_STRIP - glVertex3f (-d) (-d) (-d) - glVertex3f (-d) d (-d) - glVertex3f (-d) d d - glVertex3f (-d) (-d) d - glEnd - - --Top - glBegin gl_LINE_STRIP - glVertex3f (-d) d (-d) - glVertex3f d d (-d) - glVertex3f d d d - glVertex3f (-d) d d - glEnd - - --Bottom - glBegin gl_LINE_STRIP - glVertex3f d (-d) d - glVertex3f d (-d) (-d) - glVertex3f (-d) (-d) (-d) - glVertex3f (-d) (-d) d - glEnd - - glPopMatrix - \ No newline at end of file diff --git a/Spear/Render/Core.hs b/Spear/Render/Core.hs index b5308ce..fc49ce7 100644 --- a/Spear/Render/Core.hs +++ b/Spear/Render/Core.hs @@ -6,6 +6,7 @@ module Spear.Render.Core , module Spear.Render.Core.Pipeline , module Spear.Render.Core.Shader , module Spear.Render.Core.State +, module Spear.Render.Core.Texture ) where @@ -15,3 +16,4 @@ import Spear.Render.Core.Geometry import Spear.Render.Core.Pipeline import Spear.Render.Core.Shader import Spear.Render.Core.State +import Spear.Render.Core.Texture diff --git a/Spear/Render/Core/Shader.hs b/Spear/Render/Core/Shader.hs index ce29d4b..bff5e5d 100644 --- a/Spear/Render/Core/Shader.hs +++ b/Spear/Render/Core/Shader.hs @@ -21,7 +21,7 @@ import Spear.Math.Matrix4 import Spear.Math.Vector import Spear.Render.Core.State -import Control.Monad (mapM_) +import Control.Monad import Control.Monad.IO.Class import Data.Bits import Data.ByteString as B @@ -174,7 +174,10 @@ setUniform uniform program = liftIO $ applyUniforms :: MonadIO io => ShaderProgram -> io () applyUniforms program = - let update (FloatUniform name value) = + let handle = shaderProgramHandle program + isSubroutine (SubroutineUniform {}) = True + isSubroutine _ = False + update (FloatUniform name value) = glGetUniformLocation' handle name >>= \location -> glUniform1f (fromIntegral location) value update (Vec3Uniform name (Vector3 x y z)) = @@ -191,18 +194,55 @@ applyUniforms program = glGetUniformLocation' handle name >>= \location -> withArray mat4s $ \ptrMat4s -> glUniformMatrix4fv location (fromIntegral $ Prelude.length mat4s) GL_FALSE (unsafeCoerce ptrMat4s) - handle = shaderProgramHandle program + updateSubroutines shaderType subroutines = do + let glShaderType = toGLShaderType shaderType + let uniformNames = uniformName <$> subroutines + let subroutineNames = subroutineName <$> subroutines + -- Uniform locations are actually indices into the uniform locations array. + uniformLocations <- mapM (glGetSubroutineUniformLocation' handle glShaderType) uniformNames + -- These are the indices, for each subroutine uniform, of the selected subroutine. + functionIndices <- mapM (glGetSubroutineIndex' handle glShaderType) subroutineNames + -- This maps uniform locations (positions of the array) with the selected subroutine. + let subroutineMappings = Prelude.zip uniformLocations functionIndices + glUniformSubroutinesuiv' handle glShaderType subroutineMappings in liftIO $ do uniforms <- readIORef (shaderProgramUniforms program) - mapM_ update uniforms + --print uniforms writeIORef (shaderProgramUniforms program) [] + let subroutines = Prelude.filter isSubroutine uniforms + let vsSubroutines = Prelude.filter ((==VertexShader) . subroutineShaderType) subroutines + let fsSubroutines = Prelude.filter ((==FragmentShader) . subroutineShaderType) subroutines + let otherUniforms = Prelude.filter (not . isSubroutine) uniforms + mapM_ update otherUniforms + unless (Prelude.null vsSubroutines) $ updateSubroutines VertexShader vsSubroutines + unless (Prelude.null fsSubroutines) $ updateSubroutines FragmentShader fsSubroutines -- Private glGetUniformLocation' :: GLuint -> String -> IO GLint -glGetUniformLocation' handle name = liftIO $ - withCString name $ \nameCStr -> - glGetUniformLocation (fromIntegral handle) (unsafeCoerce nameCStr) +glGetUniformLocation' programHandle name = + withCString name $ \pNameCstr -> + glGetUniformLocation (fromIntegral programHandle) pNameCstr + +glGetSubroutineUniformLocation' :: GLuint -> GLuint -> String -> IO GLint +glGetSubroutineUniformLocation' programHandle glShaderType name = + withCString name $ \pNameCstr -> + glGetSubroutineUniformLocation (fromIntegral programHandle) glShaderType pNameCstr + +glGetSubroutineIndex' :: GLuint -> GLuint -> String -> IO GLuint +glGetSubroutineIndex' programHandle glShaderType subroutineName = + withCString subroutineName $ \pNameCstr -> + glGetSubroutineIndex programHandle glShaderType pNameCstr + +glUniformSubroutinesuiv' :: GLuint -> GLuint -> [(GLint,GLuint)] -> IO () +glUniformSubroutinesuiv' programHandle glShaderType subroutineMappings = do + numActiveLocations <- alloca $ \ptr -> do + glGetProgramStageiv programHandle glShaderType GL_ACTIVE_SUBROUTINE_UNIFORM_LOCATIONS ptr + peek ptr + allocaArray (fromIntegral numActiveLocations) $ \pLocations -> do + forM_ subroutineMappings $ \(uniformLocation, functionIndex) -> + poke (advancePtr pLocations $ fromIntegral uniformLocation) functionIndex + glUniformSubroutinesuiv glShaderType numActiveLocations pLocations deleteShader' :: GLuint -> IO () deleteShader' = glDeleteShader @@ -220,5 +260,5 @@ toGLShaderType FragmentShader = GL_FRAGMENT_SHADER toGLShaderType ComputeShader = GL_COMPUTE_SHADER makeDefinesString :: [Define] -> ByteString -makeDefinesString defines = B.concat[B.intercalate "\n" body, "\n"] - where body = (\(name, value) -> B.concat["#define ", name, " ", value]) <$> defines +makeDefinesString defines = B.concat [B.intercalate "\n" body, "\n"] + where body = (\(name, value) -> B.concat ["#define ", name, " ", value]) <$> defines diff --git a/Spear/Render/Core/State.hs b/Spear/Render/Core/State.hs index aa42635..b316e3e 100644 --- a/Spear/Render/Core/State.hs +++ b/Spear/Render/Core/State.hs @@ -93,30 +93,68 @@ data ShaderType -- | A shader uniform. data ShaderUniform - = FloatUniform { uniformName :: String, uniformFloat :: Float } - | Vec3Uniform { uniformName :: String, uniformVec3 :: Vector3 } - | Vec4Uniform { uniformName :: String, uniformVec4 :: Vector4 } - | Mat4Uniform { uniformName :: String, uniformMat4 :: Matrix4 } - | Mat4ArrayUniform { uniformName :: String, uniformMat4s :: [Matrix4] } + = FloatUniform { uniformName :: String, uniformFloat :: Float } + | Vec3Uniform { uniformName :: String, uniformVec3 :: Vector3 } + | Vec4Uniform { uniformName :: String, uniformVec4 :: Vector4 } + | Mat4Uniform { uniformName :: String, uniformMat4 :: Matrix4 } + | Mat4ArrayUniform { uniformName :: String, uniformMat4s :: [Matrix4] } + | SubroutineUniform { uniformName :: String, subroutineName :: String, + subroutineShaderType :: ShaderType } + deriving (Show) -- | A shader program. data ShaderProgram = ShaderProgram { shaderProgramHandle :: GLuint , shaderProgramResource :: ReleaseKey , shaderProgramHash :: Int - -- Dirty set of uniforms that have been set since the last time uniforms were - -- applied. OpenGL retains the values of uniforms for a program until the - -- program is linked again, so we only need to store the updates here. + -- https://www.khronos.org/opengl/wiki/Shader_Subroutine#Runtime_selection + -- OpenGL distinguishes between two types of uniforms: + -- + -- Program State: preserved across shader program activations and + -- deactivations. These uniforms only need to be reset if the program is + -- relinked. + -- + -- Context State: not preserved across activations. If the program is + -- deactivated and re-activated, these uniforms must be reset. + -- + -- Context State uniforms include: texture bindings, uniform buffer bindings, + -- and subroutine uniforms. These are typically properties of a material, so + -- they are not stored here in ShaderProgram. + -- + -- Program State uniforms include: floats, vectors and matrices. These may + -- also be properties of a material, or could be more general properties that + -- configure the shader's behaviour. + -- + -- Below is the set of Program State uniforms that have been set since the + -- last time uniforms were applied (dirty set). State Uniform changes are + -- stored in this list and only flushed on a call to `applyUniforms` (see + -- Shader.hs). + -- Since the state of this type of uniforms is preserved across activations, + -- we only need to reset the values of the dirty ones. , shaderProgramUniforms :: IORef [ShaderUniform] } +-- | A texture. +data Texture = Texture + { textureHandle :: GLuint + , textureResource :: ReleaseKey + , textureTarget :: GLenum + , textureFormat :: GLenum + , textureType :: GLenum + , textureWidth :: Int + , textureHeight :: Int + , textureDepth :: Int + } + + -- | Core render state. data RenderCoreState = RenderCoreState { buffers :: Map GLuint Buffer , geometries :: Map GLuint Geometry , shaders :: Map ShaderHash Shader , shaderPrograms :: Map ShaderProgramHash ShaderProgram + , textures :: Map GLuint Texture } type ShaderHash = Int @@ -135,6 +173,9 @@ instance ResourceClass Shader where instance ResourceClass ShaderProgram where getResource = shaderProgramResource +instance ResourceClass Texture where + getResource = textureResource + newRenderCoreState :: RenderCoreState newRenderCoreState = RenderCoreState @@ -142,4 +183,5 @@ newRenderCoreState = RenderCoreState , geometries = HashMap.empty , shaders = HashMap.empty , shaderPrograms = HashMap.empty + , textures = HashMap.empty } diff --git a/Spear/Render/Core/Texture.hs b/Spear/Render/Core/Texture.hs new file mode 100644 index 0000000..5934235 --- /dev/null +++ b/Spear/Render/Core/Texture.hs @@ -0,0 +1,235 @@ +module Spear.Render.Core.Texture +( + TextureDimension(..) +, TextureFormat(..) +, TextureFiltering(..) +, TextureWrapping(..) +, TextureDataDesc(..) +, TextureDesc(..) +, makeTexture +, deleteTexture +, updateTexture +) +where + +import Spear.Game +import Spear.Render.Core.State + +import Control.Monad +import Data.HashMap as HashMap +import Foreign.Marshal.Alloc +import Foreign.Ptr +import Foreign.Storable +import Graphics.GL.Core46 + + +data CubemapFace + = CubemapFacePosX + | CubemapFaceNegX + | CubemapFacePosY + | CubemapFaceNegY + | CubemapFacePosZ + | CubemapFaceNegZ + deriving (Eq, Show) + +data TextureDimension + = Texture1D + | Texture2D + | Texture3D + deriving (Eq, Show) + +data TextureFormat + = TextureDepth + | TextureRG16 + | TextureRG16F + | TextureRGB8 + | TextureR11G11B10F + | TextureRGBA8 + | TextureSRGB8 + | TextureSRGBA8 + deriving (Eq, Show) + +data TextureFiltering + = NearestFiltering + | LinearFiltering + deriving (Eq, Show) + +data TextureWrapping + = TextureRepeat + | TextureClampToEdge + deriving (Eq, Show) + +data TextureDataDesc + = TextureData { pixels :: Ptr () } + | CubemapData + { pixelsPosX :: Ptr () + , pixelsNegX :: Ptr () + , pixelsPosY :: Ptr () + , pixelsNegY :: Ptr () + , pixelsPosZ :: Ptr () + , pixelsNegZ :: Ptr () + } deriving (Show) + +data TextureDesc = TextureDesc + { textureDescWidth :: Int + , textureDescHeight :: Int + , textureDescDepth :: Int + , textureDescDimension :: TextureDimension + , textureDescFormat :: TextureFormat + , textureDescFiltering :: TextureFiltering + , textureDescWrapping :: TextureWrapping + , textureDescMipmaps :: Bool + , textureDescData :: TextureDataDesc + } deriving (Show) + + +-- | Create a texture. +makeTexture :: TextureDesc -> Game RenderCoreState Texture +makeTexture desc@(TextureDesc width height depth dim format filter wrap mips texData) = do + handle <- liftIO . alloca $ \ptr -> glGenTextures 1 ptr >> peek ptr + resourceKey <- register $ deleteTexture' handle + let + glTarget = toGLDimension dim + glFormat = toGLFormat format + glType = toGLType format + glInternalFormat = toGLInternalFormat format + texture = Texture handle resourceKey glTarget glFormat glType width height depth + levels = if mips then floor $ logBase 2 (fromIntegral $ maximum [width, height, depth]) + 1 else 1 + glWrap = fromIntegral . fromEnum $ toGLWrap wrap + glMinFilter = fromIntegral $ toGLMinFilter mips filter + glMagFilter = fromIntegral $ toGLMagFilter filter + glWidth = fromIntegral width + glHeight = fromIntegral height + glDepth = fromIntegral depth + glBindTexture glTarget handle + case glTarget of + GL_TEXTURE_1D -> glTexStorage1D glTarget levels glInternalFormat glWidth + GL_TEXTURE_2D -> glTexStorage2D glTarget levels glInternalFormat glWidth glHeight + GL_TEXTURE_3D -> glTexStorage3D glTarget levels glInternalFormat glWidth glHeight glDepth + GL_TEXTURE_CUBE_MAP -> glTexStorage2D glTarget levels glInternalFormat glWidth glHeight + forM_ [GL_TEXTURE_WRAP_R, GL_TEXTURE_WRAP_S, GL_TEXTURE_WRAP_T] $ \wrapDim -> + glTexParameteri glTarget wrapDim glWrap + glTexParameteri glTarget GL_TEXTURE_MIN_FILTER glMinFilter + glTexParameteri glTarget GL_TEXTURE_MAG_FILTER glMagFilter + updateTexture' texture texData + when mips $ glGenerateMipmap glTarget + glBindTexture glTarget 0 + modify $ \state -> state { + textures = HashMap.insert handle texture (textures state) + } + return texture + +-- | Delete the texture. +deleteTexture :: Texture -> Game RenderCoreState () +deleteTexture texture = do + modify $ \state -> state { + textures = HashMap.delete (textureHandle texture) (textures state) + } + release' texture + +-- | Update the texture's data. +updateTexture :: Texture -> TextureDataDesc -> Game RenderCoreState () +updateTexture texture desc = do + let glTarget = textureTarget texture + glBindTexture glTarget (textureHandle texture) + updateTexture' texture desc + glBindTexture glTarget 0 + +-- Private + +deleteTexture' :: GLuint -> IO () +deleteTexture' handle = undefined + +-- Assumes texture is already bound. +updateTexture' :: Texture -> TextureDataDesc -> Game RenderCoreState () +updateTexture' texture desc = do + let + glTarget = textureTarget texture + glFormat = textureFormat texture + glType = textureType texture + width = fromIntegral $ textureWidth texture + height = fromIntegral $ textureHeight texture + depth = fromIntegral $ textureDepth texture + level = 0 + xoffset = 0 + yoffset = 0 + zoffset = 0 + case glTarget of + GL_TEXTURE_1D -> glTexSubImage1D glTarget level xoffset width glFormat glType (pixels desc) + GL_TEXTURE_2D -> glTexSubImage2D glTarget level xoffset yoffset width height glFormat glType (pixels desc) + GL_TEXTURE_3D -> glTexSubImage3D glTarget level xoffset yoffset zoffset width height depth glFormat glType (pixels desc) + GL_TEXTURE_CUBE_MAP -> forM_ [0..5] $ \i -> + let i' = fromIntegral i + in glTexSubImage2D (GL_TEXTURE_CUBE_MAP_POSITIVE_X + i) level xoffset yoffset width height glFormat glType (cubemapFacePixels desc i') + +cubemapFacePixels :: TextureDataDesc -> Int -> Ptr () +cubemapFacePixels desc@(CubemapData posX negX posY negY posZ negZ) i = case i of + 0 -> posX + 1 -> negX + 2 -> posY + 3 -> negY + 4 -> posZ + 5 -> negZ + +toGLDimension :: TextureDimension -> GLenum +toGLDimension dim + | dim == Texture1D = GL_TEXTURE_1D + | dim == Texture2D = GL_TEXTURE_2D + | dim == Texture3D = GL_TEXTURE_3D + +toGLInternalFormat :: TextureFormat -> GLenum +toGLInternalFormat format + | format == TextureDepth = GL_DEPTH_COMPONENT + | format == TextureRG16 = GL_RG16 + | format == TextureRG16F = GL_RG16F + | format == TextureRGB8 = GL_RGB8 + | format == TextureR11G11B10F = GL_R11F_G11F_B10F + | format == TextureRGBA8 = GL_RGBA8 + | format == TextureSRGB8 = GL_SRGB8 + | format == TextureSRGBA8 = GL_SRGB8_ALPHA8 + +toGLFormat :: TextureFormat -> GLenum +toGLFormat format + | format == TextureDepth = GL_DEPTH_COMPONENT + | format == TextureRG16 = GL_RG + | format == TextureRG16F = GL_RG + | format == TextureRGB8 = GL_RGB + | format == TextureR11G11B10F = GL_RGB + | format == TextureRGBA8 = GL_RGBA + | format == TextureSRGB8 = GL_RGB + | format == TextureSRGBA8 = GL_RGBA + +toGLType :: TextureFormat -> GLenum +toGLType format + | format == TextureDepth = GL_FLOAT + | format == TextureRG16 = GL_UNSIGNED_BYTE + | format == TextureRG16F = GL_FLOAT + | format == TextureRGB8 = GL_UNSIGNED_BYTE + | format == TextureR11G11B10F = GL_FLOAT + | format == TextureRGBA8 = GL_UNSIGNED_BYTE + | format == TextureSRGB8 = GL_UNSIGNED_BYTE + | format == TextureSRGBA8 = GL_UNSIGNED_BYTE + +toGLCubemapFace :: CubemapFace -> GLenum +toGLCubemapFace face + | face == CubemapFacePosX = GL_TEXTURE_CUBE_MAP_POSITIVE_X + | face == CubemapFaceNegX = GL_TEXTURE_CUBE_MAP_NEGATIVE_X + | face == CubemapFacePosY = GL_TEXTURE_CUBE_MAP_POSITIVE_Y + | face == CubemapFaceNegY = GL_TEXTURE_CUBE_MAP_NEGATIVE_Y + | face == CubemapFacePosZ = GL_TEXTURE_CUBE_MAP_POSITIVE_Z + | face == CubemapFaceNegZ = GL_TEXTURE_CUBE_MAP_NEGATIVE_Z + +toGLWrap :: TextureWrapping -> GLenum +toGLWrap wrap + | wrap == TextureRepeat = GL_REPEAT + | wrap == TextureClampToEdge = GL_CLAMP_TO_EDGE + +toGLMinFilter :: Bool -> TextureFiltering -> GLenum +toGLMinFilter False NearestFiltering = GL_NEAREST +toGLMinFilter False LinearFiltering = GL_LINEAR +toGLMinFilter True NearestFiltering = GL_NEAREST_MIPMAP_NEAREST +toGLMinFilter True LinearFiltering = GL_LINEAR_MIPMAP_LINEAR + +toGLMagFilter :: TextureFiltering -> GLenum +toGLMagFilter NearestFiltering = GL_NEAREST +toGLMagFilter LinearFiltering = GL_LINEAR diff --git a/Spear/Render/Immediate.hs b/Spear/Render/Immediate.hs index 94a7abf..95f533d 100644 --- a/Spear/Render/Immediate.hs +++ b/Spear/Render/Immediate.hs @@ -17,6 +17,7 @@ module Spear.Render.Immediate , immPopMatrix , immPreservingMatrix , immSetColour +, immSetTexture , immSetModelMatrix , immSetViewProjectionMatrix ) @@ -135,18 +136,21 @@ immPreservingMatrix f = do immSetColour :: Vector4 -> Game ImmRenderState () immSetColour colour = do - state <- get - setUniform (Vec4Uniform "Colour" colour) (shader state) + shaderProgram <- shader <$> get + setUniform (Vec4Uniform "Colour" colour) shaderProgram + setUniform (SubroutineUniform "GetColour" "GetConstantColour" FragmentShader) shaderProgram + +immSetTexture = undefined -- TODO immSetModelMatrix :: Matrix4 -> Game ImmRenderState () immSetModelMatrix model = do - state <- get - setUniform (Mat4Uniform "Model" model) (shader state) + shaderProgram <- shader <$> get + setUniform (Mat4Uniform "Model" model) shaderProgram immSetViewProjectionMatrix :: Matrix4 -> Game ImmRenderState () immSetViewProjectionMatrix viewProjection = do - state <- get - setUniform (Mat4Uniform "ViewProjection" viewProjection) (shader state) + shaderProgram <- shader <$> get + setUniform (Mat4Uniform "ViewProjection" viewProjection) shaderProgram -- Private diff --git a/Spear/Render/Light.hs b/Spear/Render/Light.hs deleted file mode 100644 index cc1de1f..0000000 --- a/Spear/Render/Light.hs +++ /dev/null @@ -1,25 +0,0 @@ -module Spear.Render.Light -where - - -import Spear.Vector -import Graphics.Rendering.OpenGL.Raw - - -data LightData = LightData { - ambient :: Vector Float, - diffuse :: Vector Float, - spec :: Vector Float -} - - -data Light = - ParallelLight { - ld :: LightData - } -| PointLight { - ld :: LightData - } -| SpotLight { - ld :: LightData - } diff --git a/Spear/Render/Material.hs b/Spear/Render/Material.hs deleted file mode 100644 index d9c60ea..0000000 --- a/Spear/Render/Material.hs +++ /dev/null @@ -1,16 +0,0 @@ -module Spear.Render.Material -( Material(..) -) -where - - -import Spear.Math.Vector - - -data Material = Material - { ke :: Vector4 - , ka :: Vector4 - , kd :: Vector4 - , ks :: Vector4 - , shininess :: Float - } diff --git a/Spear/Render/Model.hsc b/Spear/Render/Model.hsc deleted file mode 100644 index ba6bf39..0000000 --- a/Spear/Render/Model.hsc +++ /dev/null @@ -1,54 +0,0 @@ -{-# LANGUAGE CPP, ForeignFunctionInterface #-} - -module Spear.Render.Model -( - RenderModel(..) -, renderModelFromModel -) -where - -import qualified Spear.Assets.Model as Assets -import Spear.Game - -import Foreign.Ptr -import Foreign.C.Types -import Foreign.Marshal.Alloc -import Foreign.Marshal.Array -import Foreign.Marshal.Utils (with) -import Foreign.Storable - -#include "RenderModel.h" - -data Vec3 = Vec3 !CFloat !CFloat !CFloat - -data TexCoord = TexCoord !CFloat !CFloat - -data RenderModel = RenderModel - { elements :: Ptr CChar - , numFrames :: CUInt - , numVertices :: CUInt -- ^ Number of vertices per frame. - } - -instance Storable RenderModel where - sizeOf _ = #{size RenderModel} - alignment _ = alignment (undefined :: CUInt) - - peek ptr = do - elements <- #{peek RenderModel, elements} ptr - numFrames <- #{peek RenderModel, numFrames} ptr - numVertices <- #{peek RenderModel, numVertices} ptr - return $ RenderModel elements numFrames numVertices - - poke ptr (RenderModel elements numFrames numVertices) = do - #{poke RenderModel, elements} ptr elements - #{poke RenderModel, numFrames} ptr numFrames - #{poke RenderModel, numVertices} ptr numVertices - -foreign import ccall "RenderModel.h render_model_from_model_asset" - render_model_from_model_asset :: Ptr Assets.Model -> Ptr RenderModel -> IO Int - --- | Convert the given 'Model' to a 'ModelData' instance. -renderModelFromModel :: Assets.Model -> IO RenderModel -renderModelFromModel m = with m $ \mPtr -> alloca $ \mdPtr -> do - render_model_from_model_asset mPtr mdPtr - peek mdPtr diff --git a/Spear/Render/Program.hs b/Spear/Render/Program.hs deleted file mode 100644 index b5a8658..0000000 --- a/Spear/Render/Program.hs +++ /dev/null @@ -1,102 +0,0 @@ -module Spear.Render.Program -( - StaticProgram(..) -, AnimatedProgram(..) -, Program(..) -, ProgramUniforms(..) -, StaticProgramChannels(..) -, StaticProgramUniforms(..) -, AnimatedProgramChannels(..) -, AnimatedProgramUniforms(..) -) -where - -import Spear.GL - -data StaticProgram = StaticProgram - { staticProgram :: GLSLProgram - , staticProgramChannels :: StaticProgramChannels - , staticProgramUniforms :: StaticProgramUniforms - } - -data AnimatedProgram = AnimatedProgram - { animatedProgram :: GLSLProgram - , animatedProgramChannels :: AnimatedProgramChannels - , animatedProgramUniforms :: AnimatedProgramUniforms - } - -data StaticProgramChannels = StaticProgramChannels - { vertexChannel :: GLuint -- ^ Vertex channel. - , normalChannel :: GLuint -- ^ Normal channel. - , stexChannel :: GLuint -- ^ Texture channel. - } - -data AnimatedProgramChannels = AnimatedProgramChannels - { vertexChannel1 :: GLuint -- ^ Vertex channel 1. - , vertexChannel2 :: GLuint -- ^ Vertex channel 2. - , normalChannel1 :: GLuint -- ^ Normal channel 1. - , normalChannel2 :: GLuint -- ^ Normal channel 2. - , atexChannel :: GLuint -- ^ Texture channel. - } - -data StaticProgramUniforms = StaticProgramUniforms - { skaLoc :: GLint -- ^ Material ambient uniform location. - , skdLoc :: GLint -- ^ Material diffuse uniform location. - , sksLoc :: GLint -- ^ Material specular uniform location. - , sshiLoc :: GLint -- ^ Material shininess uniform location. - , stexLoc :: GLint -- ^ Texture sampler location. - , smodelviewLoc :: GLint -- ^ Modelview matrix location. - , snormalmatLoc :: GLint -- ^ Normal matrix location. - , sprojLoc :: GLint -- ^ Projection matrix location. - } - -data AnimatedProgramUniforms = AnimatedProgramUniforms - { akaLoc :: GLint -- ^ Material ambient uniform location. - , akdLoc :: GLint -- ^ Material diffuse uniform location. - , aksLoc :: GLint -- ^ Material specular uniform location. - , ashiLoc :: GLint -- ^ Material shininess uniform location. - , atexLoc :: GLint -- ^ Texture sampler location. - , fpLoc :: GLint -- ^ Frame progress uniform location. - , amodelviewLoc :: GLint -- ^ Modelview matrix location. - , anormalmatLoc :: GLint -- ^ Normal matrix location. - , aprojLoc :: GLint -- ^ Projection matrix location. - } - -class Program a where - program :: a -> GLSLProgram - -instance Program StaticProgram where - program = staticProgram - -instance Program AnimatedProgram where - program = animatedProgram - -class ProgramUniforms a where - kaLoc :: a -> GLint - kdLoc :: a -> GLint - ksLoc :: a -> GLint - shiLoc :: a -> GLint - texLoc :: a -> GLint - modelviewLoc :: a -> GLint - normalmatLoc :: a -> GLint - projLoc :: a -> GLint - -instance ProgramUniforms StaticProgramUniforms where - kaLoc = skaLoc - kdLoc = skdLoc - ksLoc = sksLoc - shiLoc = sshiLoc - texLoc = stexLoc - modelviewLoc = smodelviewLoc - normalmatLoc = snormalmatLoc - projLoc = sprojLoc - -instance ProgramUniforms AnimatedProgramUniforms where - kaLoc = akaLoc - kdLoc = akdLoc - ksLoc = aksLoc - shiLoc = ashiLoc - texLoc = atexLoc - modelviewLoc = amodelviewLoc - normalmatLoc = anormalmatLoc - projLoc = aprojLoc diff --git a/Spear/Render/RenderModel.c b/Spear/Render/RenderModel.c deleted file mode 100644 index 1543052..0000000 --- a/Spear/Render/RenderModel.c +++ /dev/null @@ -1,232 +0,0 @@ -#include "RenderModel.h" -#include // free -#include // memcpy -#include - - -static void safe_free (void* ptr) -{ - if (ptr) - { - free (ptr); - ptr = 0; - } -} - - -/// Populate elements of an animated model to be rendered from -/// start to end in a loop. -/*int populate_elements_animated (Model* model_asset, RenderModel* model) -{ - size_t nverts = model_asset->numVertices; - size_t ntriangles = model_asset->numTriangles; - size_t nframes = model_asset->numFrames; - size_t n = nframes * ntriangles * 3; - - model->elements = malloc (56 * n); - if (!model->elements) return -1; - - // Populate elements. - - size_t f, i; - - char* elem = (char*) model->elements; - vec3* v1 = model_asset->vertices; - vec3* v2 = v1 + nverts; - vec3* n1 = model_asset->normals; - vec3* n2 = n1 + nverts; - texCoord* tex = model_asset->texCoords; - - for (f = 0; f < nframes; ++f) - { - triangle* t = model_asset->triangles; - - for (i = 0; i < ntriangles; ++i) - { - *((vec3*) elem) = v1[t->vertexIndices[0]]; - *((vec3*) (elem + 12)) = v2[t->vertexIndices[0]]; - *((vec3*) (elem + 24)) = n1[t->vertexIndices[0]]; - *((vec3*) (elem + 36)) = n2[t->vertexIndices[0]]; - *((texCoord*) (elem + 48)) = tex[t->textureIndices[0]]; - elem += 56; - - *((vec3*) elem) = v1[t->vertexIndices[1]]; - *((vec3*) (elem + 12)) = v2[t->vertexIndices[1]]; - *((vec3*) (elem + 24)) = n1[t->vertexIndices[1]]; - *((vec3*) (elem + 36)) = n2[t->vertexIndices[1]]; - *((texCoord*) (elem + 48)) = tex[t->textureIndices[1]]; - elem += 56; - - *((vec3*) elem) = v1[t->vertexIndices[2]]; - *((vec3*) (elem + 12)) = v2[t->vertexIndices[2]]; - *((vec3*) (elem + 24)) = n1[t->vertexIndices[2]]; - *((vec3*) (elem + 36)) = n2[t->vertexIndices[2]]; - *((texCoord*) (elem + 48)) = tex[t->textureIndices[2]]; - elem += 56; - - t++; - } - - v1 += nverts; - v2 += nverts; - n1 += nverts; - n2 += nverts; - - if (f == nframes-2) - { - v2 = model_asset->vertices; - n2 = model_asset->normals; - } - } - - return 0; -}*/ - - -/// Populate elements of an animated model according to its frames -/// of animation. -int populate_elements_animated (Model* model_asset, RenderModel* model) -{ - size_t nverts = model_asset->numVertices; - size_t ntriangles = model_asset->numTriangles; - size_t nframes = model_asset->numFrames; - size_t n = nframes * ntriangles * 3; - - model->elements = malloc (56 * n); - if (!model->elements) return -1; - - // Populate elements. - - unsigned f, i, j, u; - - char* elem = (char*) model->elements; - animation* anim = model_asset->animations; - - for (i = 0; i < model_asset->numAnimations; ++i, anim++) - { - unsigned start = anim->start; - unsigned end = anim->end; - - char singleFrameAnim = start == end; - - vec3* v1 = model_asset->vertices + start*nverts; - vec3* v2 = singleFrameAnim ? v1 : v1 + nverts; - vec3* n1 = model_asset->normals + start*nverts; - vec3* n2 = singleFrameAnim ? n1 : n1 + nverts; - texCoord* tex = model_asset->texCoords; - - for (u = start; u <= end; ++u) - { - triangle* t = model_asset->triangles; - - for (j = 0; j < ntriangles; ++j, t++) - { - *((vec3*) elem) = v1[t->vertexIndices[0]]; - *((vec3*) (elem + 12)) = v2[t->vertexIndices[0]]; - *((vec3*) (elem + 24)) = n1[t->vertexIndices[0]]; - *((vec3*) (elem + 36)) = n2[t->vertexIndices[0]]; - *((texCoord*) (elem + 48)) = tex[t->textureIndices[0]]; - elem += 56; - - *((vec3*) elem) = v1[t->vertexIndices[1]]; - *((vec3*) (elem + 12)) = v2[t->vertexIndices[1]]; - *((vec3*) (elem + 24)) = n1[t->vertexIndices[1]]; - *((vec3*) (elem + 36)) = n2[t->vertexIndices[1]]; - *((texCoord*) (elem + 48)) = tex[t->textureIndices[1]]; - elem += 56; - - *((vec3*) elem) = v1[t->vertexIndices[2]]; - *((vec3*) (elem + 12)) = v2[t->vertexIndices[2]]; - *((vec3*) (elem + 24)) = n1[t->vertexIndices[2]]; - *((vec3*) (elem + 36)) = n2[t->vertexIndices[2]]; - *((texCoord*) (elem + 48)) = tex[t->textureIndices[2]]; - elem += 56; - } - - // Advance to the next frame of animation of the current - // animation. - v1 += nverts; - v2 += nverts; - n1 += nverts; - n2 += nverts; - - // Reset the secondary pointers to the beginning of the - // animation when we are about to reach the last frame. - if (u == end-1) - { - v2 = model_asset->vertices + start*nverts; - n2 = model_asset->normals + start*nverts; - } - } - } - - return 0; -} - - -int populate_elements_static (Model* model_asset, RenderModel* model) -{ - size_t nverts = model_asset->numVertices; - size_t ntriangles = model_asset->numTriangles; - size_t n = ntriangles * 3; - - model->elements = malloc (32 * n); - if (!model->elements) return -1; - - // Populate elements. - - size_t f, i; - - char* elem = (char*) model->elements; - vec3* vert = model_asset->vertices; - vec3* norm = model_asset->normals; - texCoord* tex = model_asset->texCoords; - - triangle* t = model_asset->triangles; - - for (i = 0; i < ntriangles; ++i) - { - *((vec3*) elem) = vert[t->vertexIndices[0]]; - *((vec3*) (elem + 12)) = norm[t->vertexIndices[0]]; - *((texCoord*) (elem + 24)) = tex[t->textureIndices[0]]; - elem += 32; - - *((vec3*) elem) = vert[t->vertexIndices[1]]; - *((vec3*) (elem + 12)) = norm[t->vertexIndices[1]]; - *((texCoord*) (elem + 24)) = tex[t->textureIndices[1]]; - elem += 32; - - *((vec3*) elem) = vert[t->vertexIndices[2]]; - *((vec3*) (elem + 12)) = norm[t->vertexIndices[2]]; - *((texCoord*) (elem + 24)) = tex[t->textureIndices[2]]; - elem += 32; - - t++; - } - - return 0; -} - - -int render_model_from_model_asset (Model* model_asset, RenderModel* model) -{ - U32 ntriangles = model_asset->numTriangles; - U32 nframes = model_asset->numFrames; - - int result; - if (nframes > 1) result = populate_elements_animated (model_asset, model); - else result = populate_elements_static (model_asset, model); - - if (result != 0) return result; - - model->numFrames = nframes; - model->numVertices = ntriangles * 3; // Number of vertices per frame. - - return 0; -} - - -void render_model_free (RenderModel* model) -{ - safe_free (model->elements); -} diff --git a/Spear/Render/RenderModel.h b/Spear/Render/RenderModel.h deleted file mode 100644 index 6a5fb5e..0000000 --- a/Spear/Render/RenderModel.h +++ /dev/null @@ -1,49 +0,0 @@ -#ifndef _SPEAR_RENDER_MODEL_H -#define _SPEAR_RENDER_MODEL_H - -#include "Model.h" - - -/// Represents a renderable model. -/** - * If the model is animated: - * - * Buffer layout: - * vert1 vert2 norm1 norm2 texc - * - * element size = (3 + 3 + 3 + 3 + 2)*4 = 56 B - * buffer size = element size * num vertices = 56n - * - * If the model is static: - * - * Buffer layout: - * vert norm texc - * - * element size = (3 + 3 + 2)*4 = 32 B - * buffer size = element size * num vertices = 32n - * - **/ -typedef struct -{ - void* elements; - U32 numFrames; - U32 numVertices; // Number of vertices per frame. -} -RenderModel; - - -#ifdef __cplusplus -extern "C" { -#endif - -int render_model_from_model_asset (Model* model_asset, RenderModel* render_model); - -void render_model_free (RenderModel* model); - -#ifdef __cplusplus -} -#endif - - -#endif // _SPEAR_RENDER_MODEL_H - diff --git a/Spear/Render/Shaders/immediate_mode.frag b/Spear/Render/Shaders/immediate_mode.frag index ac23b5c..8884a1e 100644 --- a/Spear/Render/Shaders/immediate_mode.frag +++ b/Spear/Render/Shaders/immediate_mode.frag @@ -1,10 +1,31 @@ precision highp float; +subroutine vec4 GetColour_f(); +subroutine uniform GetColour_f GetColour; + uniform vec4 Colour; +uniform sampler2D ColourTexture; +in vec2 Texcoord; + out vec4 FragColour; +subroutine(GetColour_f) +vec4 GetConstantColour() +{ + return Colour; +} + +subroutine(GetColour_f) +vec4 GetTextureColour() +{ + return texture(ColourTexture, Texcoord).rgba; +} + +vec3 Gamma(vec3 colour) { return pow(colour, vec3(1.0 / 2.2)); } + void main() { - FragColour = vec4(pow(Colour.rgb, vec3(1.0/2.2)), Colour.a); + vec4 colour = GetColour(); + FragColour = vec4(Gamma(colour.rgb), Colour.a); } diff --git a/Spear/Render/Sphere.hs b/Spear/Render/Sphere.hs deleted file mode 100644 index 4e74375..0000000 --- a/Spear/Render/Sphere.hs +++ /dev/null @@ -1,45 +0,0 @@ -module Spear.Render.Sphere -( - render -) -where - - -import Spear.Math.Vector as Vector -import Spear.Math.Matrix -import Graphics.Rendering.OpenGL.Raw -import Graphics.Rendering.OpenGL.GL.Colors -import qualified Graphics.Rendering.OpenGL.GLU as GLU -import Unsafe.Coerce - - -type Center = Vector R -type Radius = R -type Colour = Vector R - - -applyColour :: Colour -> IO () -applyColour col = - if Vector.length col == 4 then - glColor4f (unsafeCoerce $ x col) (unsafeCoerce $ y col) (unsafeCoerce $ z col) - (unsafeCoerce $ w col) - else - glColor3f (unsafeCoerce $ x col) (unsafeCoerce $ y col) (unsafeCoerce $ z col) - - --- | Renders a sphere. --- Center is the sphere's center. --- Radius is the sphere's radius. --- Colour is a Vector representing the sphere's colour. Colour may hold an alpha channel. -render :: Center -> Radius -> Colour -> IO () -render c radius col = do - glPushMatrix - glTranslatef (unsafeCoerce $ x c) (unsafeCoerce $ y c) (unsafeCoerce $ z c) - applyColour col - - let r = unsafeCoerce $ (realToFrac radius :: Double) - let style = GLU.QuadricStyle (Just Smooth) GLU.NoTextureCoordinates GLU.Outside GLU.FillStyle - GLU.renderQuadric style $ GLU.Sphere r 16 16 - - glPopMatrix - \ No newline at end of file diff --git a/Spear/Render/StaticModel.hs b/Spear/Render/StaticModel.hs deleted file mode 100644 index 5168cf2..0000000 --- a/Spear/Render/StaticModel.hs +++ /dev/null @@ -1,152 +0,0 @@ -module Spear.Render.StaticModel - ( -- * Data types - StaticModelResource, - StaticModelRenderer, - - -- * Construction and destruction - staticModelResource, - staticModelRenderer, - - -- * Manipulation - box, - modelRes, - - -- * Rendering - bind, - render, - - -- * Collision - mkColsFromStatic, - ) -where - -import Spear.Assets.Model -import Spear.Game -import Spear.GL -import Spear.Math.AABB -import Spear.Math.Algebra -import Spear.Math.Collision -import Spear.Math.Matrix4 (Matrix4) -import Spear.Math.Vector -import Spear.Render.Material -import Spear.Render.Model -import Spear.Render.Program - -import qualified Data.Vector as V -import Foreign.C.Types -import Prelude hiding ((*)) -import Unsafe.Coerce (unsafeCoerce) - - -data StaticModelResource = StaticModelResource - { vao :: VAO, - nVertices :: Int, - material :: Material, - texture :: Texture, - boxes :: V.Vector Box, - rkey :: ReleaseKey - } - -instance Eq StaticModelResource where - m1 == m2 = vao m1 == vao m2 - -instance Ord StaticModelResource where - m1 < m2 = vao m1 < vao m2 - m1 <= m2 = vao m1 <= vao m2 - -instance ResourceClass StaticModelResource where - getResource = rkey - -data StaticModelRenderer = StaticModelRenderer {model :: StaticModelResource} - -instance Eq StaticModelRenderer where - m1 == m2 = model m1 == model m2 - -instance Ord StaticModelRenderer where - m1 < m2 = model m1 < model m2 - m1 <= m2 = model m1 <= model m2 - --- | Create a model resource from the given model. -staticModelResource :: - StaticProgramChannels -> - Material -> - Texture -> - Model -> - Game s StaticModelResource -staticModelResource (StaticProgramChannels vertChan normChan texChan) material texture model = do - RenderModel elements _ numVertices <- liftIO . renderModelFromModel $ model - elementBuf <- newBuffer - vao <- newVAO - boxes <- liftIO $ modelBoxes model - - liftIO $ do - let elemSize = 32::CUInt - elemSize' = fromIntegral elemSize - n = numVertices - - bindVAO vao - - bindBuffer ArrayBuffer elementBuf - bufferData' ArrayBuffer (fromIntegral $ elemSize * n) elements StaticDraw - - attribVAOPointer vertChan 3 GL_FLOAT False elemSize' 0 - attribVAOPointer normChan 3 GL_FLOAT False elemSize' 12 - attribVAOPointer texChan 2 GL_FLOAT False elemSize' 24 - - enableVAOAttrib vertChan - enableVAOAttrib normChan - enableVAOAttrib texChan - - rkey <- register $ do - release $ vaoKey vao - release $ bufferKey elementBuf - - return $ - StaticModelResource - vao - (unsafeCoerce numVertices) - material - texture - boxes - rkey - --- | Create a renderer from the given model resource. -staticModelRenderer :: StaticModelResource -> StaticModelRenderer -staticModelRenderer = StaticModelRenderer - --- | Get the model's ith bounding box. -box :: Int -> StaticModelResource -> Box -box i model = boxes model V.! i - --- | Get the renderer's model resource. -modelRes :: StaticModelRenderer -> StaticModelResource -modelRes = model - --- | Bind the given renderer to prepare it for rendering. -bind :: StaticProgramUniforms -> StaticModelRenderer -> IO () -bind (StaticProgramUniforms kaLoc kdLoc ksLoc shiLoc texLoc _ _ _) (StaticModelRenderer model) = - let (Material _ ka kd ks shi) = material model - in do - bindVAO . vao $ model - bindTexture $ texture model - activeTexture $= GL_TEXTURE0 - glUniform1i texLoc 0 - --- | Render the given renderer. -render :: StaticProgramUniforms -> StaticModelRenderer -> IO () -render uniforms (StaticModelRenderer model) = - let (Material _ ka kd ks shi) = material model - in do - uniform (kaLoc uniforms) ka - uniform (kdLoc uniforms) kd - uniform (ksLoc uniforms) ks - glUniform1f (shiLoc uniforms) $ unsafeCoerce shi - drawArrays GL_TRIANGLES 0 $ nVertices model - --- | Compute AABB collisioners in view space from the given model. -mkColsFromStatic :: - -- | Modelview matrix - Matrix4 -> - StaticModelResource -> - [Collisioner2] -mkColsFromStatic modelview modelRes = mkCols modelview (box 0 modelRes) diff --git a/Spear/Render/Triangle.hs b/Spear/Render/Triangle.hs deleted file mode 100644 index 49f4418..0000000 --- a/Spear/Render/Triangle.hs +++ /dev/null @@ -1,8 +0,0 @@ -module Spear.Render.Triangle -( -) -where - - -import Spear.GL - diff --git a/Spear/Scene/Graph.hs b/Spear/Scene/Graph.hs deleted file mode 100644 index 8f8b5f9..0000000 --- a/Spear/Scene/Graph.hs +++ /dev/null @@ -1,143 +0,0 @@ -module Spear.Scene.Graph -( - Property -, SceneGraph(..) -, ParseError -, loadSceneGraph -, loadSceneGraphFromFile -, node -) -where - - -import qualified Data.ByteString.Char8 as B -import Data.List (find, intersperse) -import Data.Maybe (isJust) -import Text.Parsec.Char -import Text.Parsec.Combinator -import Text.Parsec.Error -import Text.Parsec.Prim -import qualified Text.Parsec.ByteString as P -import qualified Text.Parsec.Token as PT - - -type Property = (String, [String]) - - -data SceneGraph - = SceneNode - { nodeID :: String - , properties :: [Property] - , children :: [SceneGraph] - } - | SceneLeaf - { nodeID :: String - , properties :: [Property] - } - - -instance Show SceneGraph where - show sceneGraph = show' "" sceneGraph - where - show' tab (SceneNode nid props children) = - tab ++ nid ++ "\n" ++ tab ++ "{\n" ++ (printProps tab props) ++ - (concat . fmap (show' $ " " ++ tab) $ children) ++ '\n':tab ++ "}\n" - - show' tab (SceneLeaf nid props) = - tab ++ nid ++ '\n':tab ++ "{\n" ++ tab ++ (printProps tab props) ++ '\n':tab ++ "}\n" - - -printProp :: Property -> String -printProp (key, vals) = key ++ " = " ++ (concat $ intersperse ", " vals) - - -printProps :: String -> [Property] -> String -printProps tab props = - let - tab' = '\n':(tab ++ tab) - longestKeyLen = maximum . fmap (length . fst) $ props - - align :: Int -> String -> String - align len str = - let (key, vals) = break ((==) '=') str - thisLen = length key - padLen = len - thisLen + 1 - pad = replicate padLen ' ' - in - key ++ pad ++ vals - in - case concat . intersperse tab' . fmap (align longestKeyLen . printProp) $ props of - [] -> [] - xs -> tab ++ xs - - --- | Load the scene graph from the given string. -loadSceneGraph :: String -> Either ParseError SceneGraph -loadSceneGraph str = parse sceneGraph "(unknown)" $ B.pack str - - --- | Load the scene graph specified by the given file. -loadSceneGraphFromFile :: FilePath -> IO (Either ParseError SceneGraph) -loadSceneGraphFromFile = P.parseFromFile sceneGraph - - --- | Get the node identified by the given string from the given scene graph. -node :: String -> SceneGraph -> Maybe SceneGraph -node str SceneLeaf {} = Nothing -node str n@(SceneNode nid _ children) - | str == nid = Just n - | otherwise = case find isJust $ fmap (node str) children of - Nothing -> Nothing - Just x -> x - - -sceneGraph :: P.Parser SceneGraph -sceneGraph = do - g <- graph - whitespace - eof - return g - - -graph :: P.Parser SceneGraph -graph = do - nid <- name - whitespace - char '{' - props <- many . try $ whitespace >> property - children <- many . try $ whitespace >> graph - whitespace - char '}' - - return $ case null children of - True -> SceneLeaf nid props - False -> SceneNode nid props children - - -property :: P.Parser Property -property = do - key <- name - spaces >> char '=' >> spaces - vals <- cells name - return (key, vals) - - -cells :: P.Parser String -> P.Parser [String] -cells p = do - val <- p - vals <- remainingCells p - return $ val:vals - - -remainingCells :: P.Parser String -> P.Parser [String] -remainingCells p = - try (whitespace >> char ',' >> whitespace >> cells p) - <|> (return []) - - -name :: P.Parser String -name = many1 $ choice [oneOf "-/.()?_", alphaNum] - - -whitespace :: P.Parser () -whitespace = skipMany $ choice [space, newline] diff --git a/Spear/Scene/Loader.hs b/Spear/Scene/Loader.hs deleted file mode 100644 index 0593c77..0000000 --- a/Spear/Scene/Loader.hs +++ /dev/null @@ -1,364 +0,0 @@ -{-# LANGUAGE FlexibleContexts #-} - -module Spear.Scene.Loader - ( SceneResources (..), - loadScene, - validate, - resourceMap, - value, - unspecified, - mandatory, - asString, - asFloat, - asVec3, - asVec4, - ) -where - -import Control.Monad.State.Strict -import Control.Monad.Trans (lift) -import Data.List as L (find) -import Data.Map as M -import qualified Data.StateVar as SV (get) -import Prelude hiding ((*)) -import Spear.Assets.Model as Model -import Spear.Game -import qualified Spear.GL as GL -import Spear.Math.Algebra -import Spear.Math.Collision -import Spear.Math.Matrix3 as M3 -import Spear.Math.Matrix4 as M4 -import Spear.Math.MatrixUtils (fastNormalMatrix) -import Spear.Math.Vector -import Spear.Render.AnimatedModel as AM -import Spear.Render.Material -import Spear.Render.Program -import Spear.Render.StaticModel as SM -import Spear.Scene.Graph -import Spear.Scene.SceneResources -import Text.Printf (printf) - -type Loader = Game SceneResources - --- | Load the scene specified by the given file. -loadScene :: FilePath -> Game s (SceneResources, SceneGraph) -loadScene file = do - result <- liftIO $ loadSceneGraphFromFile file - case result of - Left err -> gameError $ show err - Right g -> case validate g of - Nothing -> do - sceneRes <- resourceMap g - return (sceneRes, g) - Just err -> gameError err - --- | Validate the given SceneGraph. -validate :: SceneGraph -> Maybe String -validate _ = Nothing - --- | Load the scene described by the given 'SceneGraph'. -resourceMap :: SceneGraph -> Game s SceneResources -resourceMap g = exec runChildGame emptySceneResources (resourceMap' g) - -resourceMap' :: SceneGraph -> Loader () -resourceMap' node@(SceneLeaf nid props) = do - case nid of - "shader-program" -> newShaderProgram node - "model" -> newModel node - x -> return () -resourceMap' node@(SceneNode nid props children) = do - mapM_ resourceMap' children - --- Lookup the given resource in the data pool. Load it if it is not present, otherwise return it. -loadResource :: - -- | Resource name. - String -> - -- | Map getter. - (SceneResources -> Map String a) -> - -- | Function to modify resources. - (String -> a -> Loader ()) -> - -- | Resource loader. - Loader a -> - Loader a -loadResource key field modifyResources load = do - sceneData <- get - case M.lookup key $ field sceneData of - Just val -> return val - Nothing -> do - liftIO $ printf "Loading %s..." key - resource <- load - liftIO $ printf "done\n" - modifyResources key resource - return resource - -addShader name shader = modify $ \sceneData -> - sceneData {shaders = M.insert name shader $ shaders sceneData} - -addCustomProgram name prog = modify $ \sceneData -> - sceneData {customPrograms = M.insert name prog $ customPrograms sceneData} - -addStaticProgram name prog = modify $ \sceneData -> - sceneData {staticPrograms = M.insert name prog $ staticPrograms sceneData} - -addAnimatedProgram name prog = modify $ \sceneData -> - sceneData {animatedPrograms = M.insert name prog $ animatedPrograms sceneData} - -addTexture name tex = modify $ \sceneData -> - sceneData {textures = M.insert name tex $ textures sceneData} - -addStaticModel name model = modify $ - \sceneData -> sceneData {staticModels = M.insert name model $ staticModels sceneData} - -addAnimatedModel name model = modify $ - \sceneData -> sceneData {animatedModels = M.insert name model $ animatedModels sceneData} - --- Get the given resource from the data pool. -getResource :: (SceneResources -> Map String a) -> String -> Loader a -getResource field key = do - sceneData <- get - case M.lookup key $ field sceneData of - Just val -> return val - Nothing -> gameError $ "Oops, the given resource has not been loaded: " ++ key - ----------------------- --- Resource Loading -- ----------------------- - -newModel :: SceneGraph -> Loader () -newModel (SceneLeaf _ props) = do - name <- asString $ mandatory' "name" props - file <- asString $ mandatory' "file" props - tex <- asString $ mandatory' "texture" props - prog <- asString $ mandatory' "shader-program" props - ke <- asVec4 $ mandatory' "ke" props - ka <- asVec4 $ mandatory' "ka" props - kd <- asVec4 $ mandatory' "kd" props - ks <- asVec4 $ mandatory' "ks" props - shi <- asFloat $ mandatory' "shi" props - - let rotation = asRotation $ value "rotation" props - scale = asVec3 $ value "scale" props - - liftIO $ printf "Loading model %s..." name - model <- loadModel' file rotation scale - liftIO . putStrLn $ "done" - texture <- loadTexture tex - sceneRes <- get - - let material = Material ke ka kd ks shi - - case animated model of - False -> - case M.lookup prog $ staticPrograms sceneRes of - Nothing -> (gameError $ "Static shader program " ++ prog ++ " does not exist") >> return () - Just p -> - let StaticProgram _ channels _ = p - in do - model' <- staticModelResource channels material texture model - loadResource name staticModels addStaticModel (return model') - return () - True -> - case M.lookup prog $ animatedPrograms sceneRes of - Nothing -> (gameError $ "Animated shader program " ++ prog ++ " does not exist") >> return () - Just p -> - let AnimatedProgram _ channels _ = p - in do - model' <- animatedModelResource channels material texture model - loadResource name animatedModels addAnimatedModel (return model') - return () - -loadModel' :: FilePath -> Maybe Rotation -> Maybe Vector3 -> Game s Model -loadModel' file rotation scale = do - let transform = - ( case rotation of - Nothing -> Prelude.id - Just rot -> rotateModel rot - ) - . ( case scale of - Nothing -> Prelude.id - Just s -> flip Model.transformVerts $ - \(Vec3 x' y' z') -> Vec3 (x s * x') (y s * y') (z s * z') - ) - - (fmap transform $ Model.loadModel file) >>= liftIO . toGround - -rotateModel :: Rotation -> Model -> Model -rotateModel (Rotation ax ay az order) model = - let mat = case order of - XYZ -> rotateZ az * rotateY ay * rotateX ax - XZY -> rotateY ay * rotateZ az * rotateX ax - YXZ -> rotateZ az * rotateX ax * rotateY ay - YZX -> rotateX ax * rotateZ az * rotateY ay - ZXY -> rotateY ay * rotateX ax * rotateZ az - ZYX -> rotateX ax * rotateY ay * rotateZ az - normalMat = fastNormalMatrix mat - - vTransform (Vec3 x' y' z') = - let v = mat `M4.mulp` (vec3 x' y' z') in Vec3 (x v) (y v) (z v) - - nTransform (Vec3 x' y' z') = - let v = normalMat `M3.mul` (vec3 x' y' z') in Vec3 (x v) (y v) (z v) - in flip Model.transformVerts vTransform . flip Model.transformNormals nTransform $ model - -loadTexture :: FilePath -> Loader GL.Texture -loadTexture file = - loadResource file textures addTexture $ - GL.loadTextureImage file GL.GL_LINEAR GL.GL_LINEAR - -newShaderProgram :: SceneGraph -> Loader () -newShaderProgram (SceneLeaf _ props) = do - (vsName, vertShader) <- Spear.Scene.Loader.loadShader GL.VertexShader props - (fsName, fragShader) <- Spear.Scene.Loader.loadShader GL.FragmentShader props - name <- asString $ mandatory' "name" props - stype <- asString $ mandatory' "type" props - prog <- GL.newProgram [vertShader, fragShader] - - let getUniformLoc name = (liftIO . SV.get $ GL.uniformLocation prog name) `GL.assertGL` name - - case stype of - "static" -> do - ambient <- asString $ mandatory' "ambient" props - diffuse <- asString $ mandatory' "diffuse" props - specular <- asString $ mandatory' "specular" props - shininess <- asString $ mandatory' "shininess" props - texture <- asString $ mandatory' "texture" props - modelview <- asString $ mandatory' "modelview" props - normalmat <- asString $ mandatory' "normalmat" props - projection <- asString $ mandatory' "projection" props - - ka <- getUniformLoc ambient - kd <- getUniformLoc diffuse - ks <- getUniformLoc specular - shi <- getUniformLoc shininess - tex <- getUniformLoc texture - mview <- getUniformLoc modelview - nmat <- getUniformLoc normalmat - proj <- getUniformLoc projection - - vertChan <- fmap read $ asString $ mandatory' "vertex-channel" props - normChan <- fmap read $ asString $ mandatory' "normal-channel" props - texChan <- fmap read $ asString $ mandatory' "texture-channel" props - - let channels = StaticProgramChannels vertChan normChan texChan - uniforms = StaticProgramUniforms ka kd ks shi tex mview nmat proj - - loadResource name staticPrograms addStaticProgram $ - return $ StaticProgram prog channels uniforms - return () - "animated" -> do - ambient <- asString $ mandatory' "ambient" props - diffuse <- asString $ mandatory' "diffuse" props - specular <- asString $ mandatory' "specular" props - shininess <- asString $ mandatory' "shininess" props - texture <- asString $ mandatory' "texture" props - modelview <- asString $ mandatory' "modelview" props - normalmat <- asString $ mandatory' "normalmat" props - projection <- asString $ mandatory' "projection" props - - ka <- getUniformLoc ambient - kd <- getUniformLoc diffuse - ks <- getUniformLoc specular - shi <- getUniformLoc shininess - tex <- getUniformLoc texture - mview <- getUniformLoc modelview - nmat <- getUniformLoc normalmat - proj <- getUniformLoc projection - - vertChan1 <- fmap read $ asString $ mandatory' "vertex-channel1" props - vertChan2 <- fmap read $ asString $ mandatory' "vertex-channel2" props - normChan1 <- fmap read $ asString $ mandatory' "normal-channel1" props - normChan2 <- fmap read $ asString $ mandatory' "normal-channel2" props - texChan <- fmap read $ asString $ mandatory' "texture-channel" props - fp <- asString $ mandatory' "fp" props - p <- getUniformLoc fp - - let channels = AnimatedProgramChannels vertChan1 vertChan2 normChan1 normChan2 texChan - uniforms = AnimatedProgramUniforms ka kd ks shi tex p mview nmat proj - - loadResource name animatedPrograms addAnimatedProgram $ - return $ AnimatedProgram prog channels uniforms - return () - _ -> do - loadResource name customPrograms addCustomProgram $ return prog - return () - -loadShader :: GL.ShaderType -> [Property] -> Loader (String, GL.GLSLShader) -loadShader _ [] = gameError $ "Loader::vertexShader: empty list" -loadShader shaderType ((stype, file) : xs) = - if shaderType == GL.VertexShader && stype == "vertex-shader" - || shaderType == GL.FragmentShader && stype == "fragment-shader" - then - let f = concat file - in loadShader' f shaderType >>= \shader -> return (f, shader) - else Spear.Scene.Loader.loadShader shaderType xs - -loadShader' :: String -> GL.ShaderType -> Loader GL.GLSLShader -loadShader' file shaderType = loadResource file shaders addShader $ GL.loadShader shaderType file - ----------------------- --- Helper functions -- ----------------------- - --- Get the value of the given key. -value :: String -> [Property] -> Maybe [String] -value name props = case L.find ((==) name . fst) props of - Nothing -> Nothing - Just prop -> Just . snd $ prop - -unspecified :: Maybe a -> a -> a -unspecified (Just x) _ = x -unspecified Nothing x = x - -mandatory :: String -> [Property] -> Game s [String] -mandatory name props = case value name props of - Nothing -> gameError $ "Loader::mandatory: key not found: " ++ name - Just x -> return x - -mandatory' :: String -> [Property] -> Loader [String] -mandatory' name props = mandatory name props - -asString :: Functor f => f [String] -> f String -asString = fmap concat - -asFloat :: Functor f => f [String] -> f Float -asFloat = fmap (read . concat) - -asVec2 :: Functor f => f [String] -> f Vector2 -asVec2 val = fmap toVec2 val - where - toVec2 (x : y : _) = vec2 (read x) (read y) - toVec2 (x : []) = let x' = read x in vec2 x' x' - -asVec3 :: Functor f => f [String] -> f Vector3 -asVec3 val = fmap toVec3 val - where - toVec3 (x : y : z : _) = vec3 (read x) (read y) (read z) - toVec3 (x : []) = let x' = read x in vec3 x' x' x' - -asVec4 :: Functor f => f [String] -> f Vector4 -asVec4 val = fmap toVec4 val - where - toVec4 (x : y : z : w : _) = vec4 (read x) (read y) (read z) (read w) - toVec4 (x : []) = let x' = read x in vec4 x' x' x' x' - -asRotation :: Functor f => f [String] -> f Rotation -asRotation val = fmap parseRotation val - where - parseRotation (ax : ay : az : order : _) = Rotation (read ax) (read ay) (read az) (readOrder order) - -data Rotation = Rotation - { ax :: Float, - ay :: Float, - az :: Float, - order :: RotationOrder - } - -data RotationOrder = XYZ | XZY | YXZ | YZX | ZXY | ZYX deriving (Eq) - -readOrder :: String -> RotationOrder -readOrder "xyz" = XYZ -readOrder "xzy" = XZY -readOrder "yxz" = YXZ -readOrder "yzx" = YZX -readOrder "zxy" = ZXY -readOrder "zyx" = ZYX diff --git a/Spear/Scene/SceneResources.hs b/Spear/Scene/SceneResources.hs deleted file mode 100644 index de2fc80..0000000 --- a/Spear/Scene/SceneResources.hs +++ /dev/null @@ -1,70 +0,0 @@ -module Spear.Scene.SceneResources -( - -- * Data types - SceneResources(..) -, StaticProgram(..) -, AnimatedProgram(..) - -- * Construction -, emptySceneResources - -- * Accessors -, getShader -, getCustomProgram -, getStaticProgram -, getAnimatedProgram -, getTexture -, getStaticModel -, getAnimatedModel -) -where - -import Spear.Assets.Model as Model -import Spear.GL as GL -import Spear.Math.Vector -import Spear.Render.AnimatedModel -import Spear.Render.Material -import Spear.Render.Program -import Spear.Render.StaticModel - -import Data.Map as M - -data SceneResources = SceneResources - { shaders :: Map String GLSLShader - , customPrograms :: Map String GLSLProgram - , staticPrograms :: Map String StaticProgram - , animatedPrograms :: Map String AnimatedProgram - , textures :: Map String Texture - , staticModels :: Map String StaticModelResource - , animatedModels :: Map String AnimatedModelResource - } - --- | Build an empty instance of 'SceneResources'. -emptySceneResources = - SceneResources M.empty M.empty M.empty M.empty M.empty M.empty M.empty - --- | Get the shader specified by the given string. -getShader :: SceneResources -> String -> Maybe GLSLShader -getShader res key = M.lookup key $ shaders res - --- | Get the custom program specified by the given string. -getCustomProgram :: SceneResources -> String -> Maybe GLSLProgram -getCustomProgram res key = M.lookup key $ customPrograms res - --- | Get the static program specified by the given string. -getStaticProgram :: SceneResources -> String -> Maybe StaticProgram -getStaticProgram res key = M.lookup key $ staticPrograms res - --- | Get the animated program specified by the given string. -getAnimatedProgram :: SceneResources -> String -> Maybe AnimatedProgram -getAnimatedProgram res key = M.lookup key $ animatedPrograms res - --- | Get the texture specified by the given string. -getTexture :: SceneResources -> String -> Maybe Texture -getTexture res key = M.lookup key $ textures res - --- | Get the static model resource specified by the given string. -getStaticModel :: SceneResources -> String -> Maybe StaticModelResource -getStaticModel res key = M.lookup key $ staticModels res - --- | Get the animated model resource specified by the given string. -getAnimatedModel :: SceneResources -> String -> Maybe AnimatedModelResource -getAnimatedModel res key = M.lookup key $ animatedModels res diff --git a/Spear/cbits/stb/LICENSE b/Spear/cbits/stb/LICENSE new file mode 100644 index 0000000..8681adb --- /dev/null +++ b/Spear/cbits/stb/LICENSE @@ -0,0 +1,16 @@ +Copyright (c) 2017 Sean Barrett +Permission is hereby granted, free of charge, to any person obtaining a copy of +this software and associated documentation files (the "Software"), to deal in +the Software without restriction, including without limitation the rights to +use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies +of the Software, and to permit persons to whom the Software is furnished to do +so, subject to the following conditions: +The above copyright notice and this permission notice shall be included in all +copies or substantial portions of the Software. +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +SOFTWARE. diff --git a/Spear/cbits/stb/stb_image.c b/Spear/cbits/stb/stb_image.c new file mode 100644 index 0000000..8ddfd1f --- /dev/null +++ b/Spear/cbits/stb/stb_image.c @@ -0,0 +1,2 @@ +#define STB_IMAGE_IMPLEMENTATION +#include "stb_image.h" diff --git a/Spear/cbits/stb/stb_image.h b/Spear/cbits/stb/stb_image.h new file mode 100644 index 0000000..9eedabe --- /dev/null +++ b/Spear/cbits/stb/stb_image.h @@ -0,0 +1,7988 @@ +/* stb_image - v2.30 - public domain image loader - http://nothings.org/stb + no warranty implied; use at your own risk + + Do this: + #define STB_IMAGE_IMPLEMENTATION + before you include this file in *one* C or C++ file to create the implementation. + + // i.e. it should look like this: + #include ... + #include ... + #include ... + #define STB_IMAGE_IMPLEMENTATION + #include "stb_image.h" + + You can #define STBI_ASSERT(x) before the #include to avoid using assert.h. + And #define STBI_MALLOC, STBI_REALLOC, and STBI_FREE to avoid using malloc,realloc,free + + + QUICK NOTES: + Primarily of interest to game developers and other people who can + avoid problematic images and only need the trivial interface + + JPEG baseline & progressive (12 bpc/arithmetic not supported, same as stock IJG lib) + PNG 1/2/4/8/16-bit-per-channel + + TGA (not sure what subset, if a subset) + BMP non-1bpp, non-RLE + PSD (composited view only, no extra channels, 8/16 bit-per-channel) + + GIF (*comp always reports as 4-channel) + HDR (radiance rgbE format) + PIC (Softimage PIC) + PNM (PPM and PGM binary only) + + Animated GIF still needs a proper API, but here's one way to do it: + http://gist.github.com/urraka/685d9a6340b26b830d49 + + - decode from memory or through FILE (define STBI_NO_STDIO to remove code) + - decode from arbitrary I/O callbacks + - SIMD acceleration on x86/x64 (SSE2) and ARM (NEON) + + Full documentation under "DOCUMENTATION" below. + + +LICENSE + + See end of file for license information. + +RECENT REVISION HISTORY: + + 2.30 (2024-05-31) avoid erroneous gcc warning + 2.29 (2023-05-xx) optimizations + 2.28 (2023-01-29) many error fixes, security errors, just tons of stuff + 2.27 (2021-07-11) document stbi_info better, 16-bit PNM support, bug fixes + 2.26 (2020-07-13) many minor fixes + 2.25 (2020-02-02) fix warnings + 2.24 (2020-02-02) fix warnings; thread-local failure_reason and flip_vertically + 2.23 (2019-08-11) fix clang static analysis warning + 2.22 (2019-03-04) gif fixes, fix warnings + 2.21 (2019-02-25) fix typo in comment + 2.20 (2019-02-07) support utf8 filenames in Windows; fix warnings and platform ifdefs + 2.19 (2018-02-11) fix warning + 2.18 (2018-01-30) fix warnings + 2.17 (2018-01-29) bugfix, 1-bit BMP, 16-bitness query, fix warnings + 2.16 (2017-07-23) all functions have 16-bit variants; optimizations; bugfixes + 2.15 (2017-03-18) fix png-1,2,4; all Imagenet JPGs; no runtime SSE detection on GCC + 2.14 (2017-03-03) remove deprecated STBI_JPEG_OLD; fixes for Imagenet JPGs + 2.13 (2016-12-04) experimental 16-bit API, only for PNG so far; fixes + 2.12 (2016-04-02) fix typo in 2.11 PSD fix that caused crashes + 2.11 (2016-04-02) 16-bit PNGS; enable SSE2 in non-gcc x64 + RGB-format JPEG; remove white matting in PSD; + allocate large structures on the stack; + correct channel count for PNG & BMP + 2.10 (2016-01-22) avoid warning introduced in 2.09 + 2.09 (2016-01-16) 16-bit TGA; comments in PNM files; STBI_REALLOC_SIZED + + See end of file for full revision history. + + + ============================ Contributors ========================= + + Image formats Extensions, features + Sean Barrett (jpeg, png, bmp) Jetro Lauha (stbi_info) + Nicolas Schulz (hdr, psd) Martin "SpartanJ" Golini (stbi_info) + Jonathan Dummer (tga) James "moose2000" Brown (iPhone PNG) + Jean-Marc Lienher (gif) Ben "Disch" Wenger (io callbacks) + Tom Seddon (pic) Omar Cornut (1/2/4-bit PNG) + Thatcher Ulrich (psd) Nicolas Guillemot (vertical flip) + Ken Miller (pgm, ppm) Richard Mitton (16-bit PSD) + github:urraka (animated gif) Junggon Kim (PNM comments) + Christopher Forseth (animated gif) Daniel Gibson (16-bit TGA) + socks-the-fox (16-bit PNG) + Jeremy Sawicki (handle all ImageNet JPGs) + Optimizations & bugfixes Mikhail Morozov (1-bit BMP) + Fabian "ryg" Giesen Anael Seghezzi (is-16-bit query) + Arseny Kapoulkine Simon Breuss (16-bit PNM) + John-Mark Allen + Carmelo J Fdez-Aguera + + Bug & warning fixes + Marc LeBlanc David Woo Guillaume George Martins Mozeiko + Christpher Lloyd Jerry Jansson Joseph Thomson Blazej Dariusz Roszkowski + Phil Jordan Dave Moore Roy Eltham + Hayaki Saito Nathan Reed Won Chun + Luke Graham Johan Duparc Nick Verigakis the Horde3D community + Thomas Ruf Ronny Chevalier github:rlyeh + Janez Zemva John Bartholomew Michal Cichon github:romigrou + Jonathan Blow Ken Hamada Tero Hanninen github:svdijk + Eugene Golushkov Laurent Gomila Cort Stratton github:snagar + Aruelien Pocheville Sergio Gonzalez Thibault Reuille github:Zelex + Cass Everitt Ryamond Barbiero github:grim210 + Paul Du Bois Engin Manap Aldo Culquicondor github:sammyhw + Philipp Wiesemann Dale Weiler Oriol Ferrer Mesia github:phprus + Josh Tobin Neil Bickford Matthew Gregan github:poppolopoppo + Julian Raschke Gregory Mullen Christian Floisand github:darealshinji + Baldur Karlsson Kevin Schmidt JR Smith github:Michaelangel007 + Brad Weinberger Matvey Cherevko github:mosra + Luca Sas Alexander Veselov Zack Middleton [reserved] + Ryan C. Gordon [reserved] [reserved] + DO NOT ADD YOUR NAME HERE + + Jacko Dirks + + To add your name to the credits, pick a random blank space in the middle and fill it. + 80% of merge conflicts on stb PRs are due to people adding their name at the end + of the credits. +*/ + +#ifndef STBI_INCLUDE_STB_IMAGE_H +#define STBI_INCLUDE_STB_IMAGE_H + +// DOCUMENTATION +// +// Limitations: +// - no 12-bit-per-channel JPEG +// - no JPEGs with arithmetic coding +// - GIF always returns *comp=4 +// +// Basic usage (see HDR discussion below for HDR usage): +// int x,y,n; +// unsigned char *data = stbi_load(filename, &x, &y, &n, 0); +// // ... process data if not NULL ... +// // ... x = width, y = height, n = # 8-bit components per pixel ... +// // ... replace '0' with '1'..'4' to force that many components per pixel +// // ... but 'n' will always be the number that it would have been if you said 0 +// stbi_image_free(data); +// +// Standard parameters: +// int *x -- outputs image width in pixels +// int *y -- outputs image height in pixels +// int *channels_in_file -- outputs # of image components in image file +// int desired_channels -- if non-zero, # of image components requested in result +// +// The return value from an image loader is an 'unsigned char *' which points +// to the pixel data, or NULL on an allocation failure or if the image is +// corrupt or invalid. The pixel data consists of *y scanlines of *x pixels, +// with each pixel consisting of N interleaved 8-bit components; the first +// pixel pointed to is top-left-most in the image. There is no padding between +// image scanlines or between pixels, regardless of format. The number of +// components N is 'desired_channels' if desired_channels is non-zero, or +// *channels_in_file otherwise. If desired_channels is non-zero, +// *channels_in_file has the number of components that _would_ have been +// output otherwise. E.g. if you set desired_channels to 4, you will always +// get RGBA output, but you can check *channels_in_file to see if it's trivially +// opaque because e.g. there were only 3 channels in the source image. +// +// An output image with N components has the following components interleaved +// in this order in each pixel: +// +// N=#comp components +// 1 grey +// 2 grey, alpha +// 3 red, green, blue +// 4 red, green, blue, alpha +// +// If image loading fails for any reason, the return value will be NULL, +// and *x, *y, *channels_in_file will be unchanged. The function +// stbi_failure_reason() can be queried for an extremely brief, end-user +// unfriendly explanation of why the load failed. Define STBI_NO_FAILURE_STRINGS +// to avoid compiling these strings at all, and STBI_FAILURE_USERMSG to get slightly +// more user-friendly ones. +// +// Paletted PNG, BMP, GIF, and PIC images are automatically depalettized. +// +// To query the width, height and component count of an image without having to +// decode the full file, you can use the stbi_info family of functions: +// +// int x,y,n,ok; +// ok = stbi_info(filename, &x, &y, &n); +// // returns ok=1 and sets x, y, n if image is a supported format, +// // 0 otherwise. +// +// Note that stb_image pervasively uses ints in its public API for sizes, +// including sizes of memory buffers. This is now part of the API and thus +// hard to change without causing breakage. As a result, the various image +// loaders all have certain limits on image size; these differ somewhat +// by format but generally boil down to either just under 2GB or just under +// 1GB. When the decoded image would be larger than this, stb_image decoding +// will fail. +// +// Additionally, stb_image will reject image files that have any of their +// dimensions set to a larger value than the configurable STBI_MAX_DIMENSIONS, +// which defaults to 2**24 = 16777216 pixels. Due to the above memory limit, +// the only way to have an image with such dimensions load correctly +// is for it to have a rather extreme aspect ratio. Either way, the +// assumption here is that such larger images are likely to be malformed +// or malicious. If you do need to load an image with individual dimensions +// larger than that, and it still fits in the overall size limit, you can +// #define STBI_MAX_DIMENSIONS on your own to be something larger. +// +// =========================================================================== +// +// UNICODE: +// +// If compiling for Windows and you wish to use Unicode filenames, compile +// with +// #define STBI_WINDOWS_UTF8 +// and pass utf8-encoded filenames. Call stbi_convert_wchar_to_utf8 to convert +// Windows wchar_t filenames to utf8. +// +// =========================================================================== +// +// Philosophy +// +// stb libraries are designed with the following priorities: +// +// 1. easy to use +// 2. easy to maintain +// 3. good performance +// +// Sometimes I let "good performance" creep up in priority over "easy to maintain", +// and for best performance I may provide less-easy-to-use APIs that give higher +// performance, in addition to the easy-to-use ones. Nevertheless, it's important +// to keep in mind that from the standpoint of you, a client of this library, +// all you care about is #1 and #3, and stb libraries DO NOT emphasize #3 above all. +// +// Some secondary priorities arise directly from the first two, some of which +// provide more explicit reasons why performance can't be emphasized. +// +// - Portable ("ease of use") +// - Small source code footprint ("easy to maintain") +// - No dependencies ("ease of use") +// +// =========================================================================== +// +// I/O callbacks +// +// I/O callbacks allow you to read from arbitrary sources, like packaged +// files or some other source. Data read from callbacks are processed +// through a small internal buffer (currently 128 bytes) to try to reduce +// overhead. +// +// The three functions you must define are "read" (reads some bytes of data), +// "skip" (skips some bytes of data), "eof" (reports if the stream is at the end). +// +// =========================================================================== +// +// SIMD support +// +// The JPEG decoder will try to automatically use SIMD kernels on x86 when +// supported by the compiler. For ARM Neon support, you must explicitly +// request it. +// +// (The old do-it-yourself SIMD API is no longer supported in the current +// code.) +// +// On x86, SSE2 will automatically be used when available based on a run-time +// test; if not, the generic C versions are used as a fall-back. On ARM targets, +// the typical path is to have separate builds for NEON and non-NEON devices +// (at least this is true for iOS and Android). Therefore, the NEON support is +// toggled by a build flag: define STBI_NEON to get NEON loops. +// +// If for some reason you do not want to use any of SIMD code, or if +// you have issues compiling it, you can disable it entirely by +// defining STBI_NO_SIMD. +// +// =========================================================================== +// +// HDR image support (disable by defining STBI_NO_HDR) +// +// stb_image supports loading HDR images in general, and currently the Radiance +// .HDR file format specifically. You can still load any file through the existing +// interface; if you attempt to load an HDR file, it will be automatically remapped +// to LDR, assuming gamma 2.2 and an arbitrary scale factor defaulting to 1; +// both of these constants can be reconfigured through this interface: +// +// stbi_hdr_to_ldr_gamma(2.2f); +// stbi_hdr_to_ldr_scale(1.0f); +// +// (note, do not use _inverse_ constants; stbi_image will invert them +// appropriately). +// +// Additionally, there is a new, parallel interface for loading files as +// (linear) floats to preserve the full dynamic range: +// +// float *data = stbi_loadf(filename, &x, &y, &n, 0); +// +// If you load LDR images through this interface, those images will +// be promoted to floating point values, run through the inverse of +// constants corresponding to the above: +// +// stbi_ldr_to_hdr_scale(1.0f); +// stbi_ldr_to_hdr_gamma(2.2f); +// +// Finally, given a filename (or an open file or memory block--see header +// file for details) containing image data, you can query for the "most +// appropriate" interface to use (that is, whether the image is HDR or +// not), using: +// +// stbi_is_hdr(char *filename); +// +// =========================================================================== +// +// iPhone PNG support: +// +// We optionally support converting iPhone-formatted PNGs (which store +// premultiplied BGRA) back to RGB, even though they're internally encoded +// differently. To enable this conversion, call +// stbi_convert_iphone_png_to_rgb(1). +// +// Call stbi_set_unpremultiply_on_load(1) as well to force a divide per +// pixel to remove any premultiplied alpha *only* if the image file explicitly +// says there's premultiplied data (currently only happens in iPhone images, +// and only if iPhone convert-to-rgb processing is on). +// +// =========================================================================== +// +// ADDITIONAL CONFIGURATION +// +// - You can suppress implementation of any of the decoders to reduce +// your code footprint by #defining one or more of the following +// symbols before creating the implementation. +// +// STBI_NO_JPEG +// STBI_NO_PNG +// STBI_NO_BMP +// STBI_NO_PSD +// STBI_NO_TGA +// STBI_NO_GIF +// STBI_NO_HDR +// STBI_NO_PIC +// STBI_NO_PNM (.ppm and .pgm) +// +// - You can request *only* certain decoders and suppress all other ones +// (this will be more forward-compatible, as addition of new decoders +// doesn't require you to disable them explicitly): +// +// STBI_ONLY_JPEG +// STBI_ONLY_PNG +// STBI_ONLY_BMP +// STBI_ONLY_PSD +// STBI_ONLY_TGA +// STBI_ONLY_GIF +// STBI_ONLY_HDR +// STBI_ONLY_PIC +// STBI_ONLY_PNM (.ppm and .pgm) +// +// - If you use STBI_NO_PNG (or _ONLY_ without PNG), and you still +// want the zlib decoder to be available, #define STBI_SUPPORT_ZLIB +// +// - If you define STBI_MAX_DIMENSIONS, stb_image will reject images greater +// than that size (in either width or height) without further processing. +// This is to let programs in the wild set an upper bound to prevent +// denial-of-service attacks on untrusted data, as one could generate a +// valid image of gigantic dimensions and force stb_image to allocate a +// huge block of memory and spend disproportionate time decoding it. By +// default this is set to (1 << 24), which is 16777216, but that's still +// very big. + +#ifndef STBI_NO_STDIO +#include +#endif // STBI_NO_STDIO + +#define STBI_VERSION 1 + +enum +{ + STBI_default = 0, // only used for desired_channels + + STBI_grey = 1, + STBI_grey_alpha = 2, + STBI_rgb = 3, + STBI_rgb_alpha = 4 +}; + +#include +typedef unsigned char stbi_uc; +typedef unsigned short stbi_us; + +#ifdef __cplusplus +extern "C" { +#endif + +#ifndef STBIDEF +#ifdef STB_IMAGE_STATIC +#define STBIDEF static +#else +#define STBIDEF extern +#endif +#endif + +////////////////////////////////////////////////////////////////////////////// +// +// PRIMARY API - works on images of any type +// + +// +// load image by filename, open file, or memory buffer +// + +typedef struct +{ + int (*read) (void *user,char *data,int size); // fill 'data' with 'size' bytes. return number of bytes actually read + void (*skip) (void *user,int n); // skip the next 'n' bytes, or 'unget' the last -n bytes if negative + int (*eof) (void *user); // returns nonzero if we are at end of file/data +} stbi_io_callbacks; + +//////////////////////////////////// +// +// 8-bits-per-channel interface +// + +STBIDEF stbi_uc *stbi_load_from_memory (stbi_uc const *buffer, int len , int *x, int *y, int *channels_in_file, int desired_channels); +STBIDEF stbi_uc *stbi_load_from_callbacks(stbi_io_callbacks const *clbk , void *user, int *x, int *y, int *channels_in_file, int desired_channels); + +#ifndef STBI_NO_STDIO +STBIDEF stbi_uc *stbi_load (char const *filename, int *x, int *y, int *channels_in_file, int desired_channels); +STBIDEF stbi_uc *stbi_load_from_file (FILE *f, int *x, int *y, int *channels_in_file, int desired_channels); +// for stbi_load_from_file, file pointer is left pointing immediately after image +#endif + +#ifndef STBI_NO_GIF +STBIDEF stbi_uc *stbi_load_gif_from_memory(stbi_uc const *buffer, int len, int **delays, int *x, int *y, int *z, int *comp, int req_comp); +#endif + +#ifdef STBI_WINDOWS_UTF8 +STBIDEF int stbi_convert_wchar_to_utf8(char *buffer, size_t bufferlen, const wchar_t* input); +#endif + +//////////////////////////////////// +// +// 16-bits-per-channel interface +// + +STBIDEF stbi_us *stbi_load_16_from_memory (stbi_uc const *buffer, int len, int *x, int *y, int *channels_in_file, int desired_channels); +STBIDEF stbi_us *stbi_load_16_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *channels_in_file, int desired_channels); + +#ifndef STBI_NO_STDIO +STBIDEF stbi_us *stbi_load_16 (char const *filename, int *x, int *y, int *channels_in_file, int desired_channels); +STBIDEF stbi_us *stbi_load_from_file_16(FILE *f, int *x, int *y, int *channels_in_file, int desired_channels); +#endif + +//////////////////////////////////// +// +// float-per-channel interface +// +#ifndef STBI_NO_LINEAR + STBIDEF float *stbi_loadf_from_memory (stbi_uc const *buffer, int len, int *x, int *y, int *channels_in_file, int desired_channels); + STBIDEF float *stbi_loadf_from_callbacks (stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *channels_in_file, int desired_channels); + + #ifndef STBI_NO_STDIO + STBIDEF float *stbi_loadf (char const *filename, int *x, int *y, int *channels_in_file, int desired_channels); + STBIDEF float *stbi_loadf_from_file (FILE *f, int *x, int *y, int *channels_in_file, int desired_channels); + #endif +#endif + +#ifndef STBI_NO_HDR + STBIDEF void stbi_hdr_to_ldr_gamma(float gamma); + STBIDEF void stbi_hdr_to_ldr_scale(float scale); +#endif // STBI_NO_HDR + +#ifndef STBI_NO_LINEAR + STBIDEF void stbi_ldr_to_hdr_gamma(float gamma); + STBIDEF void stbi_ldr_to_hdr_scale(float scale); +#endif // STBI_NO_LINEAR + +// stbi_is_hdr is always defined, but always returns false if STBI_NO_HDR +STBIDEF int stbi_is_hdr_from_callbacks(stbi_io_callbacks const *clbk, void *user); +STBIDEF int stbi_is_hdr_from_memory(stbi_uc const *buffer, int len); +#ifndef STBI_NO_STDIO +STBIDEF int stbi_is_hdr (char const *filename); +STBIDEF int stbi_is_hdr_from_file(FILE *f); +#endif // STBI_NO_STDIO + + +// get a VERY brief reason for failure +// on most compilers (and ALL modern mainstream compilers) this is threadsafe +STBIDEF const char *stbi_failure_reason (void); + +// free the loaded image -- this is just free() +STBIDEF void stbi_image_free (void *retval_from_stbi_load); + +// get image dimensions & components without fully decoding +STBIDEF int stbi_info_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp); +STBIDEF int stbi_info_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp); +STBIDEF int stbi_is_16_bit_from_memory(stbi_uc const *buffer, int len); +STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *clbk, void *user); + +#ifndef STBI_NO_STDIO +STBIDEF int stbi_info (char const *filename, int *x, int *y, int *comp); +STBIDEF int stbi_info_from_file (FILE *f, int *x, int *y, int *comp); +STBIDEF int stbi_is_16_bit (char const *filename); +STBIDEF int stbi_is_16_bit_from_file(FILE *f); +#endif + + + +// for image formats that explicitly notate that they have premultiplied alpha, +// we just return the colors as stored in the file. set this flag to force +// unpremultiplication. results are undefined if the unpremultiply overflow. +STBIDEF void stbi_set_unpremultiply_on_load(int flag_true_if_should_unpremultiply); + +// indicate whether we should process iphone images back to canonical format, +// or just pass them through "as-is" +STBIDEF void stbi_convert_iphone_png_to_rgb(int flag_true_if_should_convert); + +// flip the image vertically, so the first pixel in the output array is the bottom left +STBIDEF void stbi_set_flip_vertically_on_load(int flag_true_if_should_flip); + +// as above, but only applies to images loaded on the thread that calls the function +// this function is only available if your compiler supports thread-local variables; +// calling it will fail to link if your compiler doesn't +STBIDEF void stbi_set_unpremultiply_on_load_thread(int flag_true_if_should_unpremultiply); +STBIDEF void stbi_convert_iphone_png_to_rgb_thread(int flag_true_if_should_convert); +STBIDEF void stbi_set_flip_vertically_on_load_thread(int flag_true_if_should_flip); + +// ZLIB client - used by PNG, available for other purposes + +STBIDEF char *stbi_zlib_decode_malloc_guesssize(const char *buffer, int len, int initial_size, int *outlen); +STBIDEF char *stbi_zlib_decode_malloc_guesssize_headerflag(const char *buffer, int len, int initial_size, int *outlen, int parse_header); +STBIDEF char *stbi_zlib_decode_malloc(const char *buffer, int len, int *outlen); +STBIDEF int stbi_zlib_decode_buffer(char *obuffer, int olen, const char *ibuffer, int ilen); + +STBIDEF char *stbi_zlib_decode_noheader_malloc(const char *buffer, int len, int *outlen); +STBIDEF int stbi_zlib_decode_noheader_buffer(char *obuffer, int olen, const char *ibuffer, int ilen); + + +#ifdef __cplusplus +} +#endif + +// +// +//// end header file ///////////////////////////////////////////////////// +#endif // STBI_INCLUDE_STB_IMAGE_H + +#ifdef STB_IMAGE_IMPLEMENTATION + +#if defined(STBI_ONLY_JPEG) || defined(STBI_ONLY_PNG) || defined(STBI_ONLY_BMP) \ + || defined(STBI_ONLY_TGA) || defined(STBI_ONLY_GIF) || defined(STBI_ONLY_PSD) \ + || defined(STBI_ONLY_HDR) || defined(STBI_ONLY_PIC) || defined(STBI_ONLY_PNM) \ + || defined(STBI_ONLY_ZLIB) + #ifndef STBI_ONLY_JPEG + #define STBI_NO_JPEG + #endif + #ifndef STBI_ONLY_PNG + #define STBI_NO_PNG + #endif + #ifndef STBI_ONLY_BMP + #define STBI_NO_BMP + #endif + #ifndef STBI_ONLY_PSD + #define STBI_NO_PSD + #endif + #ifndef STBI_ONLY_TGA + #define STBI_NO_TGA + #endif + #ifndef STBI_ONLY_GIF + #define STBI_NO_GIF + #endif + #ifndef STBI_ONLY_HDR + #define STBI_NO_HDR + #endif + #ifndef STBI_ONLY_PIC + #define STBI_NO_PIC + #endif + #ifndef STBI_ONLY_PNM + #define STBI_NO_PNM + #endif +#endif + +#if defined(STBI_NO_PNG) && !defined(STBI_SUPPORT_ZLIB) && !defined(STBI_NO_ZLIB) +#define STBI_NO_ZLIB +#endif + + +#include +#include // ptrdiff_t on osx +#include +#include +#include + +#if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) +#include // ldexp, pow +#endif + +#ifndef STBI_NO_STDIO +#include +#endif + +#ifndef STBI_ASSERT +#include +#define STBI_ASSERT(x) assert(x) +#endif + +#ifdef __cplusplus +#define STBI_EXTERN extern "C" +#else +#define STBI_EXTERN extern +#endif + + +#ifndef _MSC_VER + #ifdef __cplusplus + #define stbi_inline inline + #else + #define stbi_inline + #endif +#else + #define stbi_inline __forceinline +#endif + +#ifndef STBI_NO_THREAD_LOCALS + #if defined(__cplusplus) && __cplusplus >= 201103L + #define STBI_THREAD_LOCAL thread_local + #elif defined(__GNUC__) && __GNUC__ < 5 + #define STBI_THREAD_LOCAL __thread + #elif defined(_MSC_VER) + #define STBI_THREAD_LOCAL __declspec(thread) + #elif defined (__STDC_VERSION__) && __STDC_VERSION__ >= 201112L && !defined(__STDC_NO_THREADS__) + #define STBI_THREAD_LOCAL _Thread_local + #endif + + #ifndef STBI_THREAD_LOCAL + #if defined(__GNUC__) + #define STBI_THREAD_LOCAL __thread + #endif + #endif +#endif + +#if defined(_MSC_VER) || defined(__SYMBIAN32__) +typedef unsigned short stbi__uint16; +typedef signed short stbi__int16; +typedef unsigned int stbi__uint32; +typedef signed int stbi__int32; +#else +#include +typedef uint16_t stbi__uint16; +typedef int16_t stbi__int16; +typedef uint32_t stbi__uint32; +typedef int32_t stbi__int32; +#endif + +// should produce compiler error if size is wrong +typedef unsigned char validate_uint32[sizeof(stbi__uint32)==4 ? 1 : -1]; + +#ifdef _MSC_VER +#define STBI_NOTUSED(v) (void)(v) +#else +#define STBI_NOTUSED(v) (void)sizeof(v) +#endif + +#ifdef _MSC_VER +#define STBI_HAS_LROTL +#endif + +#ifdef STBI_HAS_LROTL + #define stbi_lrot(x,y) _lrotl(x,y) +#else + #define stbi_lrot(x,y) (((x) << (y)) | ((x) >> (-(y) & 31))) +#endif + +#if defined(STBI_MALLOC) && defined(STBI_FREE) && (defined(STBI_REALLOC) || defined(STBI_REALLOC_SIZED)) +// ok +#elif !defined(STBI_MALLOC) && !defined(STBI_FREE) && !defined(STBI_REALLOC) && !defined(STBI_REALLOC_SIZED) +// ok +#else +#error "Must define all or none of STBI_MALLOC, STBI_FREE, and STBI_REALLOC (or STBI_REALLOC_SIZED)." +#endif + +#ifndef STBI_MALLOC +#define STBI_MALLOC(sz) malloc(sz) +#define STBI_REALLOC(p,newsz) realloc(p,newsz) +#define STBI_FREE(p) free(p) +#endif + +#ifndef STBI_REALLOC_SIZED +#define STBI_REALLOC_SIZED(p,oldsz,newsz) STBI_REALLOC(p,newsz) +#endif + +// x86/x64 detection +#if defined(__x86_64__) || defined(_M_X64) +#define STBI__X64_TARGET +#elif defined(__i386) || defined(_M_IX86) +#define STBI__X86_TARGET +#endif + +#if defined(__GNUC__) && defined(STBI__X86_TARGET) && !defined(__SSE2__) && !defined(STBI_NO_SIMD) +// gcc doesn't support sse2 intrinsics unless you compile with -msse2, +// which in turn means it gets to use SSE2 everywhere. This is unfortunate, +// but previous attempts to provide the SSE2 functions with runtime +// detection caused numerous issues. The way architecture extensions are +// exposed in GCC/Clang is, sadly, not really suited for one-file libs. +// New behavior: if compiled with -msse2, we use SSE2 without any +// detection; if not, we don't use it at all. +#define STBI_NO_SIMD +#endif + +#if defined(__MINGW32__) && defined(STBI__X86_TARGET) && !defined(STBI_MINGW_ENABLE_SSE2) && !defined(STBI_NO_SIMD) +// Note that __MINGW32__ doesn't actually mean 32-bit, so we have to avoid STBI__X64_TARGET +// +// 32-bit MinGW wants ESP to be 16-byte aligned, but this is not in the +// Windows ABI and VC++ as well as Windows DLLs don't maintain that invariant. +// As a result, enabling SSE2 on 32-bit MinGW is dangerous when not +// simultaneously enabling "-mstackrealign". +// +// See https://github.com/nothings/stb/issues/81 for more information. +// +// So default to no SSE2 on 32-bit MinGW. If you've read this far and added +// -mstackrealign to your build settings, feel free to #define STBI_MINGW_ENABLE_SSE2. +#define STBI_NO_SIMD +#endif + +#if !defined(STBI_NO_SIMD) && (defined(STBI__X86_TARGET) || defined(STBI__X64_TARGET)) +#define STBI_SSE2 +#include + +#ifdef _MSC_VER + +#if _MSC_VER >= 1400 // not VC6 +#include // __cpuid +static int stbi__cpuid3(void) +{ + int info[4]; + __cpuid(info,1); + return info[3]; +} +#else +static int stbi__cpuid3(void) +{ + int res; + __asm { + mov eax,1 + cpuid + mov res,edx + } + return res; +} +#endif + +#define STBI_SIMD_ALIGN(type, name) __declspec(align(16)) type name + +#if !defined(STBI_NO_JPEG) && defined(STBI_SSE2) +static int stbi__sse2_available(void) +{ + int info3 = stbi__cpuid3(); + return ((info3 >> 26) & 1) != 0; +} +#endif + +#else // assume GCC-style if not VC++ +#define STBI_SIMD_ALIGN(type, name) type name __attribute__((aligned(16))) + +#if !defined(STBI_NO_JPEG) && defined(STBI_SSE2) +static int stbi__sse2_available(void) +{ + // If we're even attempting to compile this on GCC/Clang, that means + // -msse2 is on, which means the compiler is allowed to use SSE2 + // instructions at will, and so are we. + return 1; +} +#endif + +#endif +#endif + +// ARM NEON +#if defined(STBI_NO_SIMD) && defined(STBI_NEON) +#undef STBI_NEON +#endif + +#ifdef STBI_NEON +#include +#ifdef _MSC_VER +#define STBI_SIMD_ALIGN(type, name) __declspec(align(16)) type name +#else +#define STBI_SIMD_ALIGN(type, name) type name __attribute__((aligned(16))) +#endif +#endif + +#ifndef STBI_SIMD_ALIGN +#define STBI_SIMD_ALIGN(type, name) type name +#endif + +#ifndef STBI_MAX_DIMENSIONS +#define STBI_MAX_DIMENSIONS (1 << 24) +#endif + +/////////////////////////////////////////////// +// +// stbi__context struct and start_xxx functions + +// stbi__context structure is our basic context used by all images, so it +// contains all the IO context, plus some basic image information +typedef struct +{ + stbi__uint32 img_x, img_y; + int img_n, img_out_n; + + stbi_io_callbacks io; + void *io_user_data; + + int read_from_callbacks; + int buflen; + stbi_uc buffer_start[128]; + int callback_already_read; + + stbi_uc *img_buffer, *img_buffer_end; + stbi_uc *img_buffer_original, *img_buffer_original_end; +} stbi__context; + + +static void stbi__refill_buffer(stbi__context *s); + +// initialize a memory-decode context +static void stbi__start_mem(stbi__context *s, stbi_uc const *buffer, int len) +{ + s->io.read = NULL; + s->read_from_callbacks = 0; + s->callback_already_read = 0; + s->img_buffer = s->img_buffer_original = (stbi_uc *) buffer; + s->img_buffer_end = s->img_buffer_original_end = (stbi_uc *) buffer+len; +} + +// initialize a callback-based context +static void stbi__start_callbacks(stbi__context *s, stbi_io_callbacks *c, void *user) +{ + s->io = *c; + s->io_user_data = user; + s->buflen = sizeof(s->buffer_start); + s->read_from_callbacks = 1; + s->callback_already_read = 0; + s->img_buffer = s->img_buffer_original = s->buffer_start; + stbi__refill_buffer(s); + s->img_buffer_original_end = s->img_buffer_end; +} + +#ifndef STBI_NO_STDIO + +static int stbi__stdio_read(void *user, char *data, int size) +{ + return (int) fread(data,1,size,(FILE*) user); +} + +static void stbi__stdio_skip(void *user, int n) +{ + int ch; + fseek((FILE*) user, n, SEEK_CUR); + ch = fgetc((FILE*) user); /* have to read a byte to reset feof()'s flag */ + if (ch != EOF) { + ungetc(ch, (FILE *) user); /* push byte back onto stream if valid. */ + } +} + +static int stbi__stdio_eof(void *user) +{ + return feof((FILE*) user) || ferror((FILE *) user); +} + +static stbi_io_callbacks stbi__stdio_callbacks = +{ + stbi__stdio_read, + stbi__stdio_skip, + stbi__stdio_eof, +}; + +static void stbi__start_file(stbi__context *s, FILE *f) +{ + stbi__start_callbacks(s, &stbi__stdio_callbacks, (void *) f); +} + +//static void stop_file(stbi__context *s) { } + +#endif // !STBI_NO_STDIO + +static void stbi__rewind(stbi__context *s) +{ + // conceptually rewind SHOULD rewind to the beginning of the stream, + // but we just rewind to the beginning of the initial buffer, because + // we only use it after doing 'test', which only ever looks at at most 92 bytes + s->img_buffer = s->img_buffer_original; + s->img_buffer_end = s->img_buffer_original_end; +} + +enum +{ + STBI_ORDER_RGB, + STBI_ORDER_BGR +}; + +typedef struct +{ + int bits_per_channel; + int num_channels; + int channel_order; +} stbi__result_info; + +#ifndef STBI_NO_JPEG +static int stbi__jpeg_test(stbi__context *s); +static void *stbi__jpeg_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); +static int stbi__jpeg_info(stbi__context *s, int *x, int *y, int *comp); +#endif + +#ifndef STBI_NO_PNG +static int stbi__png_test(stbi__context *s); +static void *stbi__png_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); +static int stbi__png_info(stbi__context *s, int *x, int *y, int *comp); +static int stbi__png_is16(stbi__context *s); +#endif + +#ifndef STBI_NO_BMP +static int stbi__bmp_test(stbi__context *s); +static void *stbi__bmp_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); +static int stbi__bmp_info(stbi__context *s, int *x, int *y, int *comp); +#endif + +#ifndef STBI_NO_TGA +static int stbi__tga_test(stbi__context *s); +static void *stbi__tga_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); +static int stbi__tga_info(stbi__context *s, int *x, int *y, int *comp); +#endif + +#ifndef STBI_NO_PSD +static int stbi__psd_test(stbi__context *s); +static void *stbi__psd_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri, int bpc); +static int stbi__psd_info(stbi__context *s, int *x, int *y, int *comp); +static int stbi__psd_is16(stbi__context *s); +#endif + +#ifndef STBI_NO_HDR +static int stbi__hdr_test(stbi__context *s); +static float *stbi__hdr_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); +static int stbi__hdr_info(stbi__context *s, int *x, int *y, int *comp); +#endif + +#ifndef STBI_NO_PIC +static int stbi__pic_test(stbi__context *s); +static void *stbi__pic_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); +static int stbi__pic_info(stbi__context *s, int *x, int *y, int *comp); +#endif + +#ifndef STBI_NO_GIF +static int stbi__gif_test(stbi__context *s); +static void *stbi__gif_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); +static void *stbi__load_gif_main(stbi__context *s, int **delays, int *x, int *y, int *z, int *comp, int req_comp); +static int stbi__gif_info(stbi__context *s, int *x, int *y, int *comp); +#endif + +#ifndef STBI_NO_PNM +static int stbi__pnm_test(stbi__context *s); +static void *stbi__pnm_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri); +static int stbi__pnm_info(stbi__context *s, int *x, int *y, int *comp); +static int stbi__pnm_is16(stbi__context *s); +#endif + +static +#ifdef STBI_THREAD_LOCAL +STBI_THREAD_LOCAL +#endif +const char *stbi__g_failure_reason; + +STBIDEF const char *stbi_failure_reason(void) +{ + return stbi__g_failure_reason; +} + +#ifndef STBI_NO_FAILURE_STRINGS +static int stbi__err(const char *str) +{ + stbi__g_failure_reason = str; + return 0; +} +#endif + +static void *stbi__malloc(size_t size) +{ + return STBI_MALLOC(size); +} + +// stb_image uses ints pervasively, including for offset calculations. +// therefore the largest decoded image size we can support with the +// current code, even on 64-bit targets, is INT_MAX. this is not a +// significant limitation for the intended use case. +// +// we do, however, need to make sure our size calculations don't +// overflow. hence a few helper functions for size calculations that +// multiply integers together, making sure that they're non-negative +// and no overflow occurs. + +// return 1 if the sum is valid, 0 on overflow. +// negative terms are considered invalid. +static int stbi__addsizes_valid(int a, int b) +{ + if (b < 0) return 0; + // now 0 <= b <= INT_MAX, hence also + // 0 <= INT_MAX - b <= INTMAX. + // And "a + b <= INT_MAX" (which might overflow) is the + // same as a <= INT_MAX - b (no overflow) + return a <= INT_MAX - b; +} + +// returns 1 if the product is valid, 0 on overflow. +// negative factors are considered invalid. +static int stbi__mul2sizes_valid(int a, int b) +{ + if (a < 0 || b < 0) return 0; + if (b == 0) return 1; // mul-by-0 is always safe + // portable way to check for no overflows in a*b + return a <= INT_MAX/b; +} + +#if !defined(STBI_NO_JPEG) || !defined(STBI_NO_PNG) || !defined(STBI_NO_TGA) || !defined(STBI_NO_HDR) +// returns 1 if "a*b + add" has no negative terms/factors and doesn't overflow +static int stbi__mad2sizes_valid(int a, int b, int add) +{ + return stbi__mul2sizes_valid(a, b) && stbi__addsizes_valid(a*b, add); +} +#endif + +// returns 1 if "a*b*c + add" has no negative terms/factors and doesn't overflow +static int stbi__mad3sizes_valid(int a, int b, int c, int add) +{ + return stbi__mul2sizes_valid(a, b) && stbi__mul2sizes_valid(a*b, c) && + stbi__addsizes_valid(a*b*c, add); +} + +// returns 1 if "a*b*c*d + add" has no negative terms/factors and doesn't overflow +#if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) || !defined(STBI_NO_PNM) +static int stbi__mad4sizes_valid(int a, int b, int c, int d, int add) +{ + return stbi__mul2sizes_valid(a, b) && stbi__mul2sizes_valid(a*b, c) && + stbi__mul2sizes_valid(a*b*c, d) && stbi__addsizes_valid(a*b*c*d, add); +} +#endif + +#if !defined(STBI_NO_JPEG) || !defined(STBI_NO_PNG) || !defined(STBI_NO_TGA) || !defined(STBI_NO_HDR) +// mallocs with size overflow checking +static void *stbi__malloc_mad2(int a, int b, int add) +{ + if (!stbi__mad2sizes_valid(a, b, add)) return NULL; + return stbi__malloc(a*b + add); +} +#endif + +static void *stbi__malloc_mad3(int a, int b, int c, int add) +{ + if (!stbi__mad3sizes_valid(a, b, c, add)) return NULL; + return stbi__malloc(a*b*c + add); +} + +#if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) || !defined(STBI_NO_PNM) +static void *stbi__malloc_mad4(int a, int b, int c, int d, int add) +{ + if (!stbi__mad4sizes_valid(a, b, c, d, add)) return NULL; + return stbi__malloc(a*b*c*d + add); +} +#endif + +// returns 1 if the sum of two signed ints is valid (between -2^31 and 2^31-1 inclusive), 0 on overflow. +static int stbi__addints_valid(int a, int b) +{ + if ((a >= 0) != (b >= 0)) return 1; // a and b have different signs, so no overflow + if (a < 0 && b < 0) return a >= INT_MIN - b; // same as a + b >= INT_MIN; INT_MIN - b cannot overflow since b < 0. + return a <= INT_MAX - b; +} + +// returns 1 if the product of two ints fits in a signed short, 0 on overflow. +static int stbi__mul2shorts_valid(int a, int b) +{ + if (b == 0 || b == -1) return 1; // multiplication by 0 is always 0; check for -1 so SHRT_MIN/b doesn't overflow + if ((a >= 0) == (b >= 0)) return a <= SHRT_MAX/b; // product is positive, so similar to mul2sizes_valid + if (b < 0) return a <= SHRT_MIN / b; // same as a * b >= SHRT_MIN + return a >= SHRT_MIN / b; +} + +// stbi__err - error +// stbi__errpf - error returning pointer to float +// stbi__errpuc - error returning pointer to unsigned char + +#ifdef STBI_NO_FAILURE_STRINGS + #define stbi__err(x,y) 0 +#elif defined(STBI_FAILURE_USERMSG) + #define stbi__err(x,y) stbi__err(y) +#else + #define stbi__err(x,y) stbi__err(x) +#endif + +#define stbi__errpf(x,y) ((float *)(size_t) (stbi__err(x,y)?NULL:NULL)) +#define stbi__errpuc(x,y) ((unsigned char *)(size_t) (stbi__err(x,y)?NULL:NULL)) + +STBIDEF void stbi_image_free(void *retval_from_stbi_load) +{ + STBI_FREE(retval_from_stbi_load); +} + +#ifndef STBI_NO_LINEAR +static float *stbi__ldr_to_hdr(stbi_uc *data, int x, int y, int comp); +#endif + +#ifndef STBI_NO_HDR +static stbi_uc *stbi__hdr_to_ldr(float *data, int x, int y, int comp); +#endif + +static int stbi__vertically_flip_on_load_global = 0; + +STBIDEF void stbi_set_flip_vertically_on_load(int flag_true_if_should_flip) +{ + stbi__vertically_flip_on_load_global = flag_true_if_should_flip; +} + +#ifndef STBI_THREAD_LOCAL +#define stbi__vertically_flip_on_load stbi__vertically_flip_on_load_global +#else +static STBI_THREAD_LOCAL int stbi__vertically_flip_on_load_local, stbi__vertically_flip_on_load_set; + +STBIDEF void stbi_set_flip_vertically_on_load_thread(int flag_true_if_should_flip) +{ + stbi__vertically_flip_on_load_local = flag_true_if_should_flip; + stbi__vertically_flip_on_load_set = 1; +} + +#define stbi__vertically_flip_on_load (stbi__vertically_flip_on_load_set \ + ? stbi__vertically_flip_on_load_local \ + : stbi__vertically_flip_on_load_global) +#endif // STBI_THREAD_LOCAL + +static void *stbi__load_main(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri, int bpc) +{ + memset(ri, 0, sizeof(*ri)); // make sure it's initialized if we add new fields + ri->bits_per_channel = 8; // default is 8 so most paths don't have to be changed + ri->channel_order = STBI_ORDER_RGB; // all current input & output are this, but this is here so we can add BGR order + ri->num_channels = 0; + + // test the formats with a very explicit header first (at least a FOURCC + // or distinctive magic number first) + #ifndef STBI_NO_PNG + if (stbi__png_test(s)) return stbi__png_load(s,x,y,comp,req_comp, ri); + #endif + #ifndef STBI_NO_BMP + if (stbi__bmp_test(s)) return stbi__bmp_load(s,x,y,comp,req_comp, ri); + #endif + #ifndef STBI_NO_GIF + if (stbi__gif_test(s)) return stbi__gif_load(s,x,y,comp,req_comp, ri); + #endif + #ifndef STBI_NO_PSD + if (stbi__psd_test(s)) return stbi__psd_load(s,x,y,comp,req_comp, ri, bpc); + #else + STBI_NOTUSED(bpc); + #endif + #ifndef STBI_NO_PIC + if (stbi__pic_test(s)) return stbi__pic_load(s,x,y,comp,req_comp, ri); + #endif + + // then the formats that can end up attempting to load with just 1 or 2 + // bytes matching expectations; these are prone to false positives, so + // try them later + #ifndef STBI_NO_JPEG + if (stbi__jpeg_test(s)) return stbi__jpeg_load(s,x,y,comp,req_comp, ri); + #endif + #ifndef STBI_NO_PNM + if (stbi__pnm_test(s)) return stbi__pnm_load(s,x,y,comp,req_comp, ri); + #endif + + #ifndef STBI_NO_HDR + if (stbi__hdr_test(s)) { + float *hdr = stbi__hdr_load(s, x,y,comp,req_comp, ri); + return stbi__hdr_to_ldr(hdr, *x, *y, req_comp ? req_comp : *comp); + } + #endif + + #ifndef STBI_NO_TGA + // test tga last because it's a crappy test! + if (stbi__tga_test(s)) + return stbi__tga_load(s,x,y,comp,req_comp, ri); + #endif + + return stbi__errpuc("unknown image type", "Image not of any known type, or corrupt"); +} + +static stbi_uc *stbi__convert_16_to_8(stbi__uint16 *orig, int w, int h, int channels) +{ + int i; + int img_len = w * h * channels; + stbi_uc *reduced; + + reduced = (stbi_uc *) stbi__malloc(img_len); + if (reduced == NULL) return stbi__errpuc("outofmem", "Out of memory"); + + for (i = 0; i < img_len; ++i) + reduced[i] = (stbi_uc)((orig[i] >> 8) & 0xFF); // top half of each byte is sufficient approx of 16->8 bit scaling + + STBI_FREE(orig); + return reduced; +} + +static stbi__uint16 *stbi__convert_8_to_16(stbi_uc *orig, int w, int h, int channels) +{ + int i; + int img_len = w * h * channels; + stbi__uint16 *enlarged; + + enlarged = (stbi__uint16 *) stbi__malloc(img_len*2); + if (enlarged == NULL) return (stbi__uint16 *) stbi__errpuc("outofmem", "Out of memory"); + + for (i = 0; i < img_len; ++i) + enlarged[i] = (stbi__uint16)((orig[i] << 8) + orig[i]); // replicate to high and low byte, maps 0->0, 255->0xffff + + STBI_FREE(orig); + return enlarged; +} + +static void stbi__vertical_flip(void *image, int w, int h, int bytes_per_pixel) +{ + int row; + size_t bytes_per_row = (size_t)w * bytes_per_pixel; + stbi_uc temp[2048]; + stbi_uc *bytes = (stbi_uc *)image; + + for (row = 0; row < (h>>1); row++) { + stbi_uc *row0 = bytes + row*bytes_per_row; + stbi_uc *row1 = bytes + (h - row - 1)*bytes_per_row; + // swap row0 with row1 + size_t bytes_left = bytes_per_row; + while (bytes_left) { + size_t bytes_copy = (bytes_left < sizeof(temp)) ? bytes_left : sizeof(temp); + memcpy(temp, row0, bytes_copy); + memcpy(row0, row1, bytes_copy); + memcpy(row1, temp, bytes_copy); + row0 += bytes_copy; + row1 += bytes_copy; + bytes_left -= bytes_copy; + } + } +} + +#ifndef STBI_NO_GIF +static void stbi__vertical_flip_slices(void *image, int w, int h, int z, int bytes_per_pixel) +{ + int slice; + int slice_size = w * h * bytes_per_pixel; + + stbi_uc *bytes = (stbi_uc *)image; + for (slice = 0; slice < z; ++slice) { + stbi__vertical_flip(bytes, w, h, bytes_per_pixel); + bytes += slice_size; + } +} +#endif + +static unsigned char *stbi__load_and_postprocess_8bit(stbi__context *s, int *x, int *y, int *comp, int req_comp) +{ + stbi__result_info ri; + void *result = stbi__load_main(s, x, y, comp, req_comp, &ri, 8); + + if (result == NULL) + return NULL; + + // it is the responsibility of the loaders to make sure we get either 8 or 16 bit. + STBI_ASSERT(ri.bits_per_channel == 8 || ri.bits_per_channel == 16); + + if (ri.bits_per_channel != 8) { + result = stbi__convert_16_to_8((stbi__uint16 *) result, *x, *y, req_comp == 0 ? *comp : req_comp); + ri.bits_per_channel = 8; + } + + // @TODO: move stbi__convert_format to here + + if (stbi__vertically_flip_on_load) { + int channels = req_comp ? req_comp : *comp; + stbi__vertical_flip(result, *x, *y, channels * sizeof(stbi_uc)); + } + + return (unsigned char *) result; +} + +static stbi__uint16 *stbi__load_and_postprocess_16bit(stbi__context *s, int *x, int *y, int *comp, int req_comp) +{ + stbi__result_info ri; + void *result = stbi__load_main(s, x, y, comp, req_comp, &ri, 16); + + if (result == NULL) + return NULL; + + // it is the responsibility of the loaders to make sure we get either 8 or 16 bit. + STBI_ASSERT(ri.bits_per_channel == 8 || ri.bits_per_channel == 16); + + if (ri.bits_per_channel != 16) { + result = stbi__convert_8_to_16((stbi_uc *) result, *x, *y, req_comp == 0 ? *comp : req_comp); + ri.bits_per_channel = 16; + } + + // @TODO: move stbi__convert_format16 to here + // @TODO: special case RGB-to-Y (and RGBA-to-YA) for 8-bit-to-16-bit case to keep more precision + + if (stbi__vertically_flip_on_load) { + int channels = req_comp ? req_comp : *comp; + stbi__vertical_flip(result, *x, *y, channels * sizeof(stbi__uint16)); + } + + return (stbi__uint16 *) result; +} + +#if !defined(STBI_NO_HDR) && !defined(STBI_NO_LINEAR) +static void stbi__float_postprocess(float *result, int *x, int *y, int *comp, int req_comp) +{ + if (stbi__vertically_flip_on_load && result != NULL) { + int channels = req_comp ? req_comp : *comp; + stbi__vertical_flip(result, *x, *y, channels * sizeof(float)); + } +} +#endif + +#ifndef STBI_NO_STDIO + +#if defined(_WIN32) && defined(STBI_WINDOWS_UTF8) +STBI_EXTERN __declspec(dllimport) int __stdcall MultiByteToWideChar(unsigned int cp, unsigned long flags, const char *str, int cbmb, wchar_t *widestr, int cchwide); +STBI_EXTERN __declspec(dllimport) int __stdcall WideCharToMultiByte(unsigned int cp, unsigned long flags, const wchar_t *widestr, int cchwide, char *str, int cbmb, const char *defchar, int *used_default); +#endif + +#if defined(_WIN32) && defined(STBI_WINDOWS_UTF8) +STBIDEF int stbi_convert_wchar_to_utf8(char *buffer, size_t bufferlen, const wchar_t* input) +{ + return WideCharToMultiByte(65001 /* UTF8 */, 0, input, -1, buffer, (int) bufferlen, NULL, NULL); +} +#endif + +static FILE *stbi__fopen(char const *filename, char const *mode) +{ + FILE *f; +#if defined(_WIN32) && defined(STBI_WINDOWS_UTF8) + wchar_t wMode[64]; + wchar_t wFilename[1024]; + if (0 == MultiByteToWideChar(65001 /* UTF8 */, 0, filename, -1, wFilename, sizeof(wFilename)/sizeof(*wFilename))) + return 0; + + if (0 == MultiByteToWideChar(65001 /* UTF8 */, 0, mode, -1, wMode, sizeof(wMode)/sizeof(*wMode))) + return 0; + +#if defined(_MSC_VER) && _MSC_VER >= 1400 + if (0 != _wfopen_s(&f, wFilename, wMode)) + f = 0; +#else + f = _wfopen(wFilename, wMode); +#endif + +#elif defined(_MSC_VER) && _MSC_VER >= 1400 + if (0 != fopen_s(&f, filename, mode)) + f=0; +#else + f = fopen(filename, mode); +#endif + return f; +} + + +STBIDEF stbi_uc *stbi_load(char const *filename, int *x, int *y, int *comp, int req_comp) +{ + FILE *f = stbi__fopen(filename, "rb"); + unsigned char *result; + if (!f) return stbi__errpuc("can't fopen", "Unable to open file"); + result = stbi_load_from_file(f,x,y,comp,req_comp); + fclose(f); + return result; +} + +STBIDEF stbi_uc *stbi_load_from_file(FILE *f, int *x, int *y, int *comp, int req_comp) +{ + unsigned char *result; + stbi__context s; + stbi__start_file(&s,f); + result = stbi__load_and_postprocess_8bit(&s,x,y,comp,req_comp); + if (result) { + // need to 'unget' all the characters in the IO buffer + fseek(f, - (int) (s.img_buffer_end - s.img_buffer), SEEK_CUR); + } + return result; +} + +STBIDEF stbi__uint16 *stbi_load_from_file_16(FILE *f, int *x, int *y, int *comp, int req_comp) +{ + stbi__uint16 *result; + stbi__context s; + stbi__start_file(&s,f); + result = stbi__load_and_postprocess_16bit(&s,x,y,comp,req_comp); + if (result) { + // need to 'unget' all the characters in the IO buffer + fseek(f, - (int) (s.img_buffer_end - s.img_buffer), SEEK_CUR); + } + return result; +} + +STBIDEF stbi_us *stbi_load_16(char const *filename, int *x, int *y, int *comp, int req_comp) +{ + FILE *f = stbi__fopen(filename, "rb"); + stbi__uint16 *result; + if (!f) return (stbi_us *) stbi__errpuc("can't fopen", "Unable to open file"); + result = stbi_load_from_file_16(f,x,y,comp,req_comp); + fclose(f); + return result; +} + + +#endif //!STBI_NO_STDIO + +STBIDEF stbi_us *stbi_load_16_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *channels_in_file, int desired_channels) +{ + stbi__context s; + stbi__start_mem(&s,buffer,len); + return stbi__load_and_postprocess_16bit(&s,x,y,channels_in_file,desired_channels); +} + +STBIDEF stbi_us *stbi_load_16_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *channels_in_file, int desired_channels) +{ + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *)clbk, user); + return stbi__load_and_postprocess_16bit(&s,x,y,channels_in_file,desired_channels); +} + +STBIDEF stbi_uc *stbi_load_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp) +{ + stbi__context s; + stbi__start_mem(&s,buffer,len); + return stbi__load_and_postprocess_8bit(&s,x,y,comp,req_comp); +} + +STBIDEF stbi_uc *stbi_load_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp, int req_comp) +{ + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *) clbk, user); + return stbi__load_and_postprocess_8bit(&s,x,y,comp,req_comp); +} + +#ifndef STBI_NO_GIF +STBIDEF stbi_uc *stbi_load_gif_from_memory(stbi_uc const *buffer, int len, int **delays, int *x, int *y, int *z, int *comp, int req_comp) +{ + unsigned char *result; + stbi__context s; + stbi__start_mem(&s,buffer,len); + + result = (unsigned char*) stbi__load_gif_main(&s, delays, x, y, z, comp, req_comp); + if (stbi__vertically_flip_on_load) { + stbi__vertical_flip_slices( result, *x, *y, *z, *comp ); + } + + return result; +} +#endif + +#ifndef STBI_NO_LINEAR +static float *stbi__loadf_main(stbi__context *s, int *x, int *y, int *comp, int req_comp) +{ + unsigned char *data; + #ifndef STBI_NO_HDR + if (stbi__hdr_test(s)) { + stbi__result_info ri; + float *hdr_data = stbi__hdr_load(s,x,y,comp,req_comp, &ri); + if (hdr_data) + stbi__float_postprocess(hdr_data,x,y,comp,req_comp); + return hdr_data; + } + #endif + data = stbi__load_and_postprocess_8bit(s, x, y, comp, req_comp); + if (data) + return stbi__ldr_to_hdr(data, *x, *y, req_comp ? req_comp : *comp); + return stbi__errpf("unknown image type", "Image not of any known type, or corrupt"); +} + +STBIDEF float *stbi_loadf_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp) +{ + stbi__context s; + stbi__start_mem(&s,buffer,len); + return stbi__loadf_main(&s,x,y,comp,req_comp); +} + +STBIDEF float *stbi_loadf_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp, int req_comp) +{ + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *) clbk, user); + return stbi__loadf_main(&s,x,y,comp,req_comp); +} + +#ifndef STBI_NO_STDIO +STBIDEF float *stbi_loadf(char const *filename, int *x, int *y, int *comp, int req_comp) +{ + float *result; + FILE *f = stbi__fopen(filename, "rb"); + if (!f) return stbi__errpf("can't fopen", "Unable to open file"); + result = stbi_loadf_from_file(f,x,y,comp,req_comp); + fclose(f); + return result; +} + +STBIDEF float *stbi_loadf_from_file(FILE *f, int *x, int *y, int *comp, int req_comp) +{ + stbi__context s; + stbi__start_file(&s,f); + return stbi__loadf_main(&s,x,y,comp,req_comp); +} +#endif // !STBI_NO_STDIO + +#endif // !STBI_NO_LINEAR + +// these is-hdr-or-not is defined independent of whether STBI_NO_LINEAR is +// defined, for API simplicity; if STBI_NO_LINEAR is defined, it always +// reports false! + +STBIDEF int stbi_is_hdr_from_memory(stbi_uc const *buffer, int len) +{ + #ifndef STBI_NO_HDR + stbi__context s; + stbi__start_mem(&s,buffer,len); + return stbi__hdr_test(&s); + #else + STBI_NOTUSED(buffer); + STBI_NOTUSED(len); + return 0; + #endif +} + +#ifndef STBI_NO_STDIO +STBIDEF int stbi_is_hdr (char const *filename) +{ + FILE *f = stbi__fopen(filename, "rb"); + int result=0; + if (f) { + result = stbi_is_hdr_from_file(f); + fclose(f); + } + return result; +} + +STBIDEF int stbi_is_hdr_from_file(FILE *f) +{ + #ifndef STBI_NO_HDR + long pos = ftell(f); + int res; + stbi__context s; + stbi__start_file(&s,f); + res = stbi__hdr_test(&s); + fseek(f, pos, SEEK_SET); + return res; + #else + STBI_NOTUSED(f); + return 0; + #endif +} +#endif // !STBI_NO_STDIO + +STBIDEF int stbi_is_hdr_from_callbacks(stbi_io_callbacks const *clbk, void *user) +{ + #ifndef STBI_NO_HDR + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *) clbk, user); + return stbi__hdr_test(&s); + #else + STBI_NOTUSED(clbk); + STBI_NOTUSED(user); + return 0; + #endif +} + +#ifndef STBI_NO_LINEAR +static float stbi__l2h_gamma=2.2f, stbi__l2h_scale=1.0f; + +STBIDEF void stbi_ldr_to_hdr_gamma(float gamma) { stbi__l2h_gamma = gamma; } +STBIDEF void stbi_ldr_to_hdr_scale(float scale) { stbi__l2h_scale = scale; } +#endif + +static float stbi__h2l_gamma_i=1.0f/2.2f, stbi__h2l_scale_i=1.0f; + +STBIDEF void stbi_hdr_to_ldr_gamma(float gamma) { stbi__h2l_gamma_i = 1/gamma; } +STBIDEF void stbi_hdr_to_ldr_scale(float scale) { stbi__h2l_scale_i = 1/scale; } + + +////////////////////////////////////////////////////////////////////////////// +// +// Common code used by all image loaders +// + +enum +{ + STBI__SCAN_load=0, + STBI__SCAN_type, + STBI__SCAN_header +}; + +static void stbi__refill_buffer(stbi__context *s) +{ + int n = (s->io.read)(s->io_user_data,(char*)s->buffer_start,s->buflen); + s->callback_already_read += (int) (s->img_buffer - s->img_buffer_original); + if (n == 0) { + // at end of file, treat same as if from memory, but need to handle case + // where s->img_buffer isn't pointing to safe memory, e.g. 0-byte file + s->read_from_callbacks = 0; + s->img_buffer = s->buffer_start; + s->img_buffer_end = s->buffer_start+1; + *s->img_buffer = 0; + } else { + s->img_buffer = s->buffer_start; + s->img_buffer_end = s->buffer_start + n; + } +} + +stbi_inline static stbi_uc stbi__get8(stbi__context *s) +{ + if (s->img_buffer < s->img_buffer_end) + return *s->img_buffer++; + if (s->read_from_callbacks) { + stbi__refill_buffer(s); + return *s->img_buffer++; + } + return 0; +} + +#if defined(STBI_NO_JPEG) && defined(STBI_NO_HDR) && defined(STBI_NO_PIC) && defined(STBI_NO_PNM) +// nothing +#else +stbi_inline static int stbi__at_eof(stbi__context *s) +{ + if (s->io.read) { + if (!(s->io.eof)(s->io_user_data)) return 0; + // if feof() is true, check if buffer = end + // special case: we've only got the special 0 character at the end + if (s->read_from_callbacks == 0) return 1; + } + + return s->img_buffer >= s->img_buffer_end; +} +#endif + +#if defined(STBI_NO_JPEG) && defined(STBI_NO_PNG) && defined(STBI_NO_BMP) && defined(STBI_NO_PSD) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) && defined(STBI_NO_PIC) +// nothing +#else +static void stbi__skip(stbi__context *s, int n) +{ + if (n == 0) return; // already there! + if (n < 0) { + s->img_buffer = s->img_buffer_end; + return; + } + if (s->io.read) { + int blen = (int) (s->img_buffer_end - s->img_buffer); + if (blen < n) { + s->img_buffer = s->img_buffer_end; + (s->io.skip)(s->io_user_data, n - blen); + return; + } + } + s->img_buffer += n; +} +#endif + +#if defined(STBI_NO_PNG) && defined(STBI_NO_TGA) && defined(STBI_NO_HDR) && defined(STBI_NO_PNM) +// nothing +#else +static int stbi__getn(stbi__context *s, stbi_uc *buffer, int n) +{ + if (s->io.read) { + int blen = (int) (s->img_buffer_end - s->img_buffer); + if (blen < n) { + int res, count; + + memcpy(buffer, s->img_buffer, blen); + + count = (s->io.read)(s->io_user_data, (char*) buffer + blen, n - blen); + res = (count == (n-blen)); + s->img_buffer = s->img_buffer_end; + return res; + } + } + + if (s->img_buffer+n <= s->img_buffer_end) { + memcpy(buffer, s->img_buffer, n); + s->img_buffer += n; + return 1; + } else + return 0; +} +#endif + +#if defined(STBI_NO_JPEG) && defined(STBI_NO_PNG) && defined(STBI_NO_PSD) && defined(STBI_NO_PIC) +// nothing +#else +static int stbi__get16be(stbi__context *s) +{ + int z = stbi__get8(s); + return (z << 8) + stbi__get8(s); +} +#endif + +#if defined(STBI_NO_PNG) && defined(STBI_NO_PSD) && defined(STBI_NO_PIC) +// nothing +#else +static stbi__uint32 stbi__get32be(stbi__context *s) +{ + stbi__uint32 z = stbi__get16be(s); + return (z << 16) + stbi__get16be(s); +} +#endif + +#if defined(STBI_NO_BMP) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) +// nothing +#else +static int stbi__get16le(stbi__context *s) +{ + int z = stbi__get8(s); + return z + (stbi__get8(s) << 8); +} +#endif + +#ifndef STBI_NO_BMP +static stbi__uint32 stbi__get32le(stbi__context *s) +{ + stbi__uint32 z = stbi__get16le(s); + z += (stbi__uint32)stbi__get16le(s) << 16; + return z; +} +#endif + +#define STBI__BYTECAST(x) ((stbi_uc) ((x) & 255)) // truncate int to byte without warnings + +#if defined(STBI_NO_JPEG) && defined(STBI_NO_PNG) && defined(STBI_NO_BMP) && defined(STBI_NO_PSD) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) && defined(STBI_NO_PIC) && defined(STBI_NO_PNM) +// nothing +#else +////////////////////////////////////////////////////////////////////////////// +// +// generic converter from built-in img_n to req_comp +// individual types do this automatically as much as possible (e.g. jpeg +// does all cases internally since it needs to colorspace convert anyway, +// and it never has alpha, so very few cases ). png can automatically +// interleave an alpha=255 channel, but falls back to this for other cases +// +// assume data buffer is malloced, so malloc a new one and free that one +// only failure mode is malloc failing + +static stbi_uc stbi__compute_y(int r, int g, int b) +{ + return (stbi_uc) (((r*77) + (g*150) + (29*b)) >> 8); +} +#endif + +#if defined(STBI_NO_PNG) && defined(STBI_NO_BMP) && defined(STBI_NO_PSD) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) && defined(STBI_NO_PIC) && defined(STBI_NO_PNM) +// nothing +#else +static unsigned char *stbi__convert_format(unsigned char *data, int img_n, int req_comp, unsigned int x, unsigned int y) +{ + int i,j; + unsigned char *good; + + if (req_comp == img_n) return data; + STBI_ASSERT(req_comp >= 1 && req_comp <= 4); + + good = (unsigned char *) stbi__malloc_mad3(req_comp, x, y, 0); + if (good == NULL) { + STBI_FREE(data); + return stbi__errpuc("outofmem", "Out of memory"); + } + + for (j=0; j < (int) y; ++j) { + unsigned char *src = data + j * x * img_n ; + unsigned char *dest = good + j * x * req_comp; + + #define STBI__COMBO(a,b) ((a)*8+(b)) + #define STBI__CASE(a,b) case STBI__COMBO(a,b): for(i=x-1; i >= 0; --i, src += a, dest += b) + // convert source image with img_n components to one with req_comp components; + // avoid switch per pixel, so use switch per scanline and massive macros + switch (STBI__COMBO(img_n, req_comp)) { + STBI__CASE(1,2) { dest[0]=src[0]; dest[1]=255; } break; + STBI__CASE(1,3) { dest[0]=dest[1]=dest[2]=src[0]; } break; + STBI__CASE(1,4) { dest[0]=dest[1]=dest[2]=src[0]; dest[3]=255; } break; + STBI__CASE(2,1) { dest[0]=src[0]; } break; + STBI__CASE(2,3) { dest[0]=dest[1]=dest[2]=src[0]; } break; + STBI__CASE(2,4) { dest[0]=dest[1]=dest[2]=src[0]; dest[3]=src[1]; } break; + STBI__CASE(3,4) { dest[0]=src[0];dest[1]=src[1];dest[2]=src[2];dest[3]=255; } break; + STBI__CASE(3,1) { dest[0]=stbi__compute_y(src[0],src[1],src[2]); } break; + STBI__CASE(3,2) { dest[0]=stbi__compute_y(src[0],src[1],src[2]); dest[1] = 255; } break; + STBI__CASE(4,1) { dest[0]=stbi__compute_y(src[0],src[1],src[2]); } break; + STBI__CASE(4,2) { dest[0]=stbi__compute_y(src[0],src[1],src[2]); dest[1] = src[3]; } break; + STBI__CASE(4,3) { dest[0]=src[0];dest[1]=src[1];dest[2]=src[2]; } break; + default: STBI_ASSERT(0); STBI_FREE(data); STBI_FREE(good); return stbi__errpuc("unsupported", "Unsupported format conversion"); + } + #undef STBI__CASE + } + + STBI_FREE(data); + return good; +} +#endif + +#if defined(STBI_NO_PNG) && defined(STBI_NO_PSD) +// nothing +#else +static stbi__uint16 stbi__compute_y_16(int r, int g, int b) +{ + return (stbi__uint16) (((r*77) + (g*150) + (29*b)) >> 8); +} +#endif + +#if defined(STBI_NO_PNG) && defined(STBI_NO_PSD) +// nothing +#else +static stbi__uint16 *stbi__convert_format16(stbi__uint16 *data, int img_n, int req_comp, unsigned int x, unsigned int y) +{ + int i,j; + stbi__uint16 *good; + + if (req_comp == img_n) return data; + STBI_ASSERT(req_comp >= 1 && req_comp <= 4); + + good = (stbi__uint16 *) stbi__malloc(req_comp * x * y * 2); + if (good == NULL) { + STBI_FREE(data); + return (stbi__uint16 *) stbi__errpuc("outofmem", "Out of memory"); + } + + for (j=0; j < (int) y; ++j) { + stbi__uint16 *src = data + j * x * img_n ; + stbi__uint16 *dest = good + j * x * req_comp; + + #define STBI__COMBO(a,b) ((a)*8+(b)) + #define STBI__CASE(a,b) case STBI__COMBO(a,b): for(i=x-1; i >= 0; --i, src += a, dest += b) + // convert source image with img_n components to one with req_comp components; + // avoid switch per pixel, so use switch per scanline and massive macros + switch (STBI__COMBO(img_n, req_comp)) { + STBI__CASE(1,2) { dest[0]=src[0]; dest[1]=0xffff; } break; + STBI__CASE(1,3) { dest[0]=dest[1]=dest[2]=src[0]; } break; + STBI__CASE(1,4) { dest[0]=dest[1]=dest[2]=src[0]; dest[3]=0xffff; } break; + STBI__CASE(2,1) { dest[0]=src[0]; } break; + STBI__CASE(2,3) { dest[0]=dest[1]=dest[2]=src[0]; } break; + STBI__CASE(2,4) { dest[0]=dest[1]=dest[2]=src[0]; dest[3]=src[1]; } break; + STBI__CASE(3,4) { dest[0]=src[0];dest[1]=src[1];dest[2]=src[2];dest[3]=0xffff; } break; + STBI__CASE(3,1) { dest[0]=stbi__compute_y_16(src[0],src[1],src[2]); } break; + STBI__CASE(3,2) { dest[0]=stbi__compute_y_16(src[0],src[1],src[2]); dest[1] = 0xffff; } break; + STBI__CASE(4,1) { dest[0]=stbi__compute_y_16(src[0],src[1],src[2]); } break; + STBI__CASE(4,2) { dest[0]=stbi__compute_y_16(src[0],src[1],src[2]); dest[1] = src[3]; } break; + STBI__CASE(4,3) { dest[0]=src[0];dest[1]=src[1];dest[2]=src[2]; } break; + default: STBI_ASSERT(0); STBI_FREE(data); STBI_FREE(good); return (stbi__uint16*) stbi__errpuc("unsupported", "Unsupported format conversion"); + } + #undef STBI__CASE + } + + STBI_FREE(data); + return good; +} +#endif + +#ifndef STBI_NO_LINEAR +static float *stbi__ldr_to_hdr(stbi_uc *data, int x, int y, int comp) +{ + int i,k,n; + float *output; + if (!data) return NULL; + output = (float *) stbi__malloc_mad4(x, y, comp, sizeof(float), 0); + if (output == NULL) { STBI_FREE(data); return stbi__errpf("outofmem", "Out of memory"); } + // compute number of non-alpha components + if (comp & 1) n = comp; else n = comp-1; + for (i=0; i < x*y; ++i) { + for (k=0; k < n; ++k) { + output[i*comp + k] = (float) (pow(data[i*comp+k]/255.0f, stbi__l2h_gamma) * stbi__l2h_scale); + } + } + if (n < comp) { + for (i=0; i < x*y; ++i) { + output[i*comp + n] = data[i*comp + n]/255.0f; + } + } + STBI_FREE(data); + return output; +} +#endif + +#ifndef STBI_NO_HDR +#define stbi__float2int(x) ((int) (x)) +static stbi_uc *stbi__hdr_to_ldr(float *data, int x, int y, int comp) +{ + int i,k,n; + stbi_uc *output; + if (!data) return NULL; + output = (stbi_uc *) stbi__malloc_mad3(x, y, comp, 0); + if (output == NULL) { STBI_FREE(data); return stbi__errpuc("outofmem", "Out of memory"); } + // compute number of non-alpha components + if (comp & 1) n = comp; else n = comp-1; + for (i=0; i < x*y; ++i) { + for (k=0; k < n; ++k) { + float z = (float) pow(data[i*comp+k]*stbi__h2l_scale_i, stbi__h2l_gamma_i) * 255 + 0.5f; + if (z < 0) z = 0; + if (z > 255) z = 255; + output[i*comp + k] = (stbi_uc) stbi__float2int(z); + } + if (k < comp) { + float z = data[i*comp+k] * 255 + 0.5f; + if (z < 0) z = 0; + if (z > 255) z = 255; + output[i*comp + k] = (stbi_uc) stbi__float2int(z); + } + } + STBI_FREE(data); + return output; +} +#endif + +////////////////////////////////////////////////////////////////////////////// +// +// "baseline" JPEG/JFIF decoder +// +// simple implementation +// - doesn't support delayed output of y-dimension +// - simple interface (only one output format: 8-bit interleaved RGB) +// - doesn't try to recover corrupt jpegs +// - doesn't allow partial loading, loading multiple at once +// - still fast on x86 (copying globals into locals doesn't help x86) +// - allocates lots of intermediate memory (full size of all components) +// - non-interleaved case requires this anyway +// - allows good upsampling (see next) +// high-quality +// - upsampled channels are bilinearly interpolated, even across blocks +// - quality integer IDCT derived from IJG's 'slow' +// performance +// - fast huffman; reasonable integer IDCT +// - some SIMD kernels for common paths on targets with SSE2/NEON +// - uses a lot of intermediate memory, could cache poorly + +#ifndef STBI_NO_JPEG + +// huffman decoding acceleration +#define FAST_BITS 9 // larger handles more cases; smaller stomps less cache + +typedef struct +{ + stbi_uc fast[1 << FAST_BITS]; + // weirdly, repacking this into AoS is a 10% speed loss, instead of a win + stbi__uint16 code[256]; + stbi_uc values[256]; + stbi_uc size[257]; + unsigned int maxcode[18]; + int delta[17]; // old 'firstsymbol' - old 'firstcode' +} stbi__huffman; + +typedef struct +{ + stbi__context *s; + stbi__huffman huff_dc[4]; + stbi__huffman huff_ac[4]; + stbi__uint16 dequant[4][64]; + stbi__int16 fast_ac[4][1 << FAST_BITS]; + +// sizes for components, interleaved MCUs + int img_h_max, img_v_max; + int img_mcu_x, img_mcu_y; + int img_mcu_w, img_mcu_h; + +// definition of jpeg image component + struct + { + int id; + int h,v; + int tq; + int hd,ha; + int dc_pred; + + int x,y,w2,h2; + stbi_uc *data; + void *raw_data, *raw_coeff; + stbi_uc *linebuf; + short *coeff; // progressive only + int coeff_w, coeff_h; // number of 8x8 coefficient blocks + } img_comp[4]; + + stbi__uint32 code_buffer; // jpeg entropy-coded buffer + int code_bits; // number of valid bits + unsigned char marker; // marker seen while filling entropy buffer + int nomore; // flag if we saw a marker so must stop + + int progressive; + int spec_start; + int spec_end; + int succ_high; + int succ_low; + int eob_run; + int jfif; + int app14_color_transform; // Adobe APP14 tag + int rgb; + + int scan_n, order[4]; + int restart_interval, todo; + +// kernels + void (*idct_block_kernel)(stbi_uc *out, int out_stride, short data[64]); + void (*YCbCr_to_RGB_kernel)(stbi_uc *out, const stbi_uc *y, const stbi_uc *pcb, const stbi_uc *pcr, int count, int step); + stbi_uc *(*resample_row_hv_2_kernel)(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs); +} stbi__jpeg; + +static int stbi__build_huffman(stbi__huffman *h, int *count) +{ + int i,j,k=0; + unsigned int code; + // build size list for each symbol (from JPEG spec) + for (i=0; i < 16; ++i) { + for (j=0; j < count[i]; ++j) { + h->size[k++] = (stbi_uc) (i+1); + if(k >= 257) return stbi__err("bad size list","Corrupt JPEG"); + } + } + h->size[k] = 0; + + // compute actual symbols (from jpeg spec) + code = 0; + k = 0; + for(j=1; j <= 16; ++j) { + // compute delta to add to code to compute symbol id + h->delta[j] = k - code; + if (h->size[k] == j) { + while (h->size[k] == j) + h->code[k++] = (stbi__uint16) (code++); + if (code-1 >= (1u << j)) return stbi__err("bad code lengths","Corrupt JPEG"); + } + // compute largest code + 1 for this size, preshifted as needed later + h->maxcode[j] = code << (16-j); + code <<= 1; + } + h->maxcode[j] = 0xffffffff; + + // build non-spec acceleration table; 255 is flag for not-accelerated + memset(h->fast, 255, 1 << FAST_BITS); + for (i=0; i < k; ++i) { + int s = h->size[i]; + if (s <= FAST_BITS) { + int c = h->code[i] << (FAST_BITS-s); + int m = 1 << (FAST_BITS-s); + for (j=0; j < m; ++j) { + h->fast[c+j] = (stbi_uc) i; + } + } + } + return 1; +} + +// build a table that decodes both magnitude and value of small ACs in +// one go. +static void stbi__build_fast_ac(stbi__int16 *fast_ac, stbi__huffman *h) +{ + int i; + for (i=0; i < (1 << FAST_BITS); ++i) { + stbi_uc fast = h->fast[i]; + fast_ac[i] = 0; + if (fast < 255) { + int rs = h->values[fast]; + int run = (rs >> 4) & 15; + int magbits = rs & 15; + int len = h->size[fast]; + + if (magbits && len + magbits <= FAST_BITS) { + // magnitude code followed by receive_extend code + int k = ((i << len) & ((1 << FAST_BITS) - 1)) >> (FAST_BITS - magbits); + int m = 1 << (magbits - 1); + if (k < m) k += (~0U << magbits) + 1; + // if the result is small enough, we can fit it in fast_ac table + if (k >= -128 && k <= 127) + fast_ac[i] = (stbi__int16) ((k * 256) + (run * 16) + (len + magbits)); + } + } + } +} + +static void stbi__grow_buffer_unsafe(stbi__jpeg *j) +{ + do { + unsigned int b = j->nomore ? 0 : stbi__get8(j->s); + if (b == 0xff) { + int c = stbi__get8(j->s); + while (c == 0xff) c = stbi__get8(j->s); // consume fill bytes + if (c != 0) { + j->marker = (unsigned char) c; + j->nomore = 1; + return; + } + } + j->code_buffer |= b << (24 - j->code_bits); + j->code_bits += 8; + } while (j->code_bits <= 24); +} + +// (1 << n) - 1 +static const stbi__uint32 stbi__bmask[17]={0,1,3,7,15,31,63,127,255,511,1023,2047,4095,8191,16383,32767,65535}; + +// decode a jpeg huffman value from the bitstream +stbi_inline static int stbi__jpeg_huff_decode(stbi__jpeg *j, stbi__huffman *h) +{ + unsigned int temp; + int c,k; + + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + + // look at the top FAST_BITS and determine what symbol ID it is, + // if the code is <= FAST_BITS + c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS)-1); + k = h->fast[c]; + if (k < 255) { + int s = h->size[k]; + if (s > j->code_bits) + return -1; + j->code_buffer <<= s; + j->code_bits -= s; + return h->values[k]; + } + + // naive test is to shift the code_buffer down so k bits are + // valid, then test against maxcode. To speed this up, we've + // preshifted maxcode left so that it has (16-k) 0s at the + // end; in other words, regardless of the number of bits, it + // wants to be compared against something shifted to have 16; + // that way we don't need to shift inside the loop. + temp = j->code_buffer >> 16; + for (k=FAST_BITS+1 ; ; ++k) + if (temp < h->maxcode[k]) + break; + if (k == 17) { + // error! code not found + j->code_bits -= 16; + return -1; + } + + if (k > j->code_bits) + return -1; + + // convert the huffman code to the symbol id + c = ((j->code_buffer >> (32 - k)) & stbi__bmask[k]) + h->delta[k]; + if(c < 0 || c >= 256) // symbol id out of bounds! + return -1; + STBI_ASSERT((((j->code_buffer) >> (32 - h->size[c])) & stbi__bmask[h->size[c]]) == h->code[c]); + + // convert the id to a symbol + j->code_bits -= k; + j->code_buffer <<= k; + return h->values[c]; +} + +// bias[n] = (-1<code_bits < n) stbi__grow_buffer_unsafe(j); + if (j->code_bits < n) return 0; // ran out of bits from stream, return 0s intead of continuing + + sgn = j->code_buffer >> 31; // sign bit always in MSB; 0 if MSB clear (positive), 1 if MSB set (negative) + k = stbi_lrot(j->code_buffer, n); + j->code_buffer = k & ~stbi__bmask[n]; + k &= stbi__bmask[n]; + j->code_bits -= n; + return k + (stbi__jbias[n] & (sgn - 1)); +} + +// get some unsigned bits +stbi_inline static int stbi__jpeg_get_bits(stbi__jpeg *j, int n) +{ + unsigned int k; + if (j->code_bits < n) stbi__grow_buffer_unsafe(j); + if (j->code_bits < n) return 0; // ran out of bits from stream, return 0s intead of continuing + k = stbi_lrot(j->code_buffer, n); + j->code_buffer = k & ~stbi__bmask[n]; + k &= stbi__bmask[n]; + j->code_bits -= n; + return k; +} + +stbi_inline static int stbi__jpeg_get_bit(stbi__jpeg *j) +{ + unsigned int k; + if (j->code_bits < 1) stbi__grow_buffer_unsafe(j); + if (j->code_bits < 1) return 0; // ran out of bits from stream, return 0s intead of continuing + k = j->code_buffer; + j->code_buffer <<= 1; + --j->code_bits; + return k & 0x80000000; +} + +// given a value that's at position X in the zigzag stream, +// where does it appear in the 8x8 matrix coded as row-major? +static const stbi_uc stbi__jpeg_dezigzag[64+15] = +{ + 0, 1, 8, 16, 9, 2, 3, 10, + 17, 24, 32, 25, 18, 11, 4, 5, + 12, 19, 26, 33, 40, 48, 41, 34, + 27, 20, 13, 6, 7, 14, 21, 28, + 35, 42, 49, 56, 57, 50, 43, 36, + 29, 22, 15, 23, 30, 37, 44, 51, + 58, 59, 52, 45, 38, 31, 39, 46, + 53, 60, 61, 54, 47, 55, 62, 63, + // let corrupt input sample past end + 63, 63, 63, 63, 63, 63, 63, 63, + 63, 63, 63, 63, 63, 63, 63 +}; + +// decode one 64-entry block-- +static int stbi__jpeg_decode_block(stbi__jpeg *j, short data[64], stbi__huffman *hdc, stbi__huffman *hac, stbi__int16 *fac, int b, stbi__uint16 *dequant) +{ + int diff,dc,k; + int t; + + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + t = stbi__jpeg_huff_decode(j, hdc); + if (t < 0 || t > 15) return stbi__err("bad huffman code","Corrupt JPEG"); + + // 0 all the ac values now so we can do it 32-bits at a time + memset(data,0,64*sizeof(data[0])); + + diff = t ? stbi__extend_receive(j, t) : 0; + if (!stbi__addints_valid(j->img_comp[b].dc_pred, diff)) return stbi__err("bad delta","Corrupt JPEG"); + dc = j->img_comp[b].dc_pred + diff; + j->img_comp[b].dc_pred = dc; + if (!stbi__mul2shorts_valid(dc, dequant[0])) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); + data[0] = (short) (dc * dequant[0]); + + // decode AC components, see JPEG spec + k = 1; + do { + unsigned int zig; + int c,r,s; + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS)-1); + r = fac[c]; + if (r) { // fast-AC path + k += (r >> 4) & 15; // run + s = r & 15; // combined length + if (s > j->code_bits) return stbi__err("bad huffman code", "Combined length longer than code bits available"); + j->code_buffer <<= s; + j->code_bits -= s; + // decode into unzigzag'd location + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short) ((r >> 8) * dequant[zig]); + } else { + int rs = stbi__jpeg_huff_decode(j, hac); + if (rs < 0) return stbi__err("bad huffman code","Corrupt JPEG"); + s = rs & 15; + r = rs >> 4; + if (s == 0) { + if (rs != 0xf0) break; // end block + k += 16; + } else { + k += r; + // decode into unzigzag'd location + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short) (stbi__extend_receive(j,s) * dequant[zig]); + } + } + } while (k < 64); + return 1; +} + +static int stbi__jpeg_decode_block_prog_dc(stbi__jpeg *j, short data[64], stbi__huffman *hdc, int b) +{ + int diff,dc; + int t; + if (j->spec_end != 0) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); + + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + + if (j->succ_high == 0) { + // first scan for DC coefficient, must be first + memset(data,0,64*sizeof(data[0])); // 0 all the ac values now + t = stbi__jpeg_huff_decode(j, hdc); + if (t < 0 || t > 15) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); + diff = t ? stbi__extend_receive(j, t) : 0; + + if (!stbi__addints_valid(j->img_comp[b].dc_pred, diff)) return stbi__err("bad delta", "Corrupt JPEG"); + dc = j->img_comp[b].dc_pred + diff; + j->img_comp[b].dc_pred = dc; + if (!stbi__mul2shorts_valid(dc, 1 << j->succ_low)) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); + data[0] = (short) (dc * (1 << j->succ_low)); + } else { + // refinement scan for DC coefficient + if (stbi__jpeg_get_bit(j)) + data[0] += (short) (1 << j->succ_low); + } + return 1; +} + +// @OPTIMIZE: store non-zigzagged during the decode passes, +// and only de-zigzag when dequantizing +static int stbi__jpeg_decode_block_prog_ac(stbi__jpeg *j, short data[64], stbi__huffman *hac, stbi__int16 *fac) +{ + int k; + if (j->spec_start == 0) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); + + if (j->succ_high == 0) { + int shift = j->succ_low; + + if (j->eob_run) { + --j->eob_run; + return 1; + } + + k = j->spec_start; + do { + unsigned int zig; + int c,r,s; + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS)-1); + r = fac[c]; + if (r) { // fast-AC path + k += (r >> 4) & 15; // run + s = r & 15; // combined length + if (s > j->code_bits) return stbi__err("bad huffman code", "Combined length longer than code bits available"); + j->code_buffer <<= s; + j->code_bits -= s; + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short) ((r >> 8) * (1 << shift)); + } else { + int rs = stbi__jpeg_huff_decode(j, hac); + if (rs < 0) return stbi__err("bad huffman code","Corrupt JPEG"); + s = rs & 15; + r = rs >> 4; + if (s == 0) { + if (r < 15) { + j->eob_run = (1 << r); + if (r) + j->eob_run += stbi__jpeg_get_bits(j, r); + --j->eob_run; + break; + } + k += 16; + } else { + k += r; + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short) (stbi__extend_receive(j,s) * (1 << shift)); + } + } + } while (k <= j->spec_end); + } else { + // refinement scan for these AC coefficients + + short bit = (short) (1 << j->succ_low); + + if (j->eob_run) { + --j->eob_run; + for (k = j->spec_start; k <= j->spec_end; ++k) { + short *p = &data[stbi__jpeg_dezigzag[k]]; + if (*p != 0) + if (stbi__jpeg_get_bit(j)) + if ((*p & bit)==0) { + if (*p > 0) + *p += bit; + else + *p -= bit; + } + } + } else { + k = j->spec_start; + do { + int r,s; + int rs = stbi__jpeg_huff_decode(j, hac); // @OPTIMIZE see if we can use the fast path here, advance-by-r is so slow, eh + if (rs < 0) return stbi__err("bad huffman code","Corrupt JPEG"); + s = rs & 15; + r = rs >> 4; + if (s == 0) { + if (r < 15) { + j->eob_run = (1 << r) - 1; + if (r) + j->eob_run += stbi__jpeg_get_bits(j, r); + r = 64; // force end of block + } else { + // r=15 s=0 should write 16 0s, so we just do + // a run of 15 0s and then write s (which is 0), + // so we don't have to do anything special here + } + } else { + if (s != 1) return stbi__err("bad huffman code", "Corrupt JPEG"); + // sign bit + if (stbi__jpeg_get_bit(j)) + s = bit; + else + s = -bit; + } + + // advance by r + while (k <= j->spec_end) { + short *p = &data[stbi__jpeg_dezigzag[k++]]; + if (*p != 0) { + if (stbi__jpeg_get_bit(j)) + if ((*p & bit)==0) { + if (*p > 0) + *p += bit; + else + *p -= bit; + } + } else { + if (r == 0) { + *p = (short) s; + break; + } + --r; + } + } + } while (k <= j->spec_end); + } + } + return 1; +} + +// take a -128..127 value and stbi__clamp it and convert to 0..255 +stbi_inline static stbi_uc stbi__clamp(int x) +{ + // trick to use a single test to catch both cases + if ((unsigned int) x > 255) { + if (x < 0) return 0; + if (x > 255) return 255; + } + return (stbi_uc) x; +} + +#define stbi__f2f(x) ((int) (((x) * 4096 + 0.5))) +#define stbi__fsh(x) ((x) * 4096) + +// derived from jidctint -- DCT_ISLOW +#define STBI__IDCT_1D(s0,s1,s2,s3,s4,s5,s6,s7) \ + int t0,t1,t2,t3,p1,p2,p3,p4,p5,x0,x1,x2,x3; \ + p2 = s2; \ + p3 = s6; \ + p1 = (p2+p3) * stbi__f2f(0.5411961f); \ + t2 = p1 + p3*stbi__f2f(-1.847759065f); \ + t3 = p1 + p2*stbi__f2f( 0.765366865f); \ + p2 = s0; \ + p3 = s4; \ + t0 = stbi__fsh(p2+p3); \ + t1 = stbi__fsh(p2-p3); \ + x0 = t0+t3; \ + x3 = t0-t3; \ + x1 = t1+t2; \ + x2 = t1-t2; \ + t0 = s7; \ + t1 = s5; \ + t2 = s3; \ + t3 = s1; \ + p3 = t0+t2; \ + p4 = t1+t3; \ + p1 = t0+t3; \ + p2 = t1+t2; \ + p5 = (p3+p4)*stbi__f2f( 1.175875602f); \ + t0 = t0*stbi__f2f( 0.298631336f); \ + t1 = t1*stbi__f2f( 2.053119869f); \ + t2 = t2*stbi__f2f( 3.072711026f); \ + t3 = t3*stbi__f2f( 1.501321110f); \ + p1 = p5 + p1*stbi__f2f(-0.899976223f); \ + p2 = p5 + p2*stbi__f2f(-2.562915447f); \ + p3 = p3*stbi__f2f(-1.961570560f); \ + p4 = p4*stbi__f2f(-0.390180644f); \ + t3 += p1+p4; \ + t2 += p2+p3; \ + t1 += p2+p4; \ + t0 += p1+p3; + +static void stbi__idct_block(stbi_uc *out, int out_stride, short data[64]) +{ + int i,val[64],*v=val; + stbi_uc *o; + short *d = data; + + // columns + for (i=0; i < 8; ++i,++d, ++v) { + // if all zeroes, shortcut -- this avoids dequantizing 0s and IDCTing + if (d[ 8]==0 && d[16]==0 && d[24]==0 && d[32]==0 + && d[40]==0 && d[48]==0 && d[56]==0) { + // no shortcut 0 seconds + // (1|2|3|4|5|6|7)==0 0 seconds + // all separate -0.047 seconds + // 1 && 2|3 && 4|5 && 6|7: -0.047 seconds + int dcterm = d[0]*4; + v[0] = v[8] = v[16] = v[24] = v[32] = v[40] = v[48] = v[56] = dcterm; + } else { + STBI__IDCT_1D(d[ 0],d[ 8],d[16],d[24],d[32],d[40],d[48],d[56]) + // constants scaled things up by 1<<12; let's bring them back + // down, but keep 2 extra bits of precision + x0 += 512; x1 += 512; x2 += 512; x3 += 512; + v[ 0] = (x0+t3) >> 10; + v[56] = (x0-t3) >> 10; + v[ 8] = (x1+t2) >> 10; + v[48] = (x1-t2) >> 10; + v[16] = (x2+t1) >> 10; + v[40] = (x2-t1) >> 10; + v[24] = (x3+t0) >> 10; + v[32] = (x3-t0) >> 10; + } + } + + for (i=0, v=val, o=out; i < 8; ++i,v+=8,o+=out_stride) { + // no fast case since the first 1D IDCT spread components out + STBI__IDCT_1D(v[0],v[1],v[2],v[3],v[4],v[5],v[6],v[7]) + // constants scaled things up by 1<<12, plus we had 1<<2 from first + // loop, plus horizontal and vertical each scale by sqrt(8) so together + // we've got an extra 1<<3, so 1<<17 total we need to remove. + // so we want to round that, which means adding 0.5 * 1<<17, + // aka 65536. Also, we'll end up with -128 to 127 that we want + // to encode as 0..255 by adding 128, so we'll add that before the shift + x0 += 65536 + (128<<17); + x1 += 65536 + (128<<17); + x2 += 65536 + (128<<17); + x3 += 65536 + (128<<17); + // tried computing the shifts into temps, or'ing the temps to see + // if any were out of range, but that was slower + o[0] = stbi__clamp((x0+t3) >> 17); + o[7] = stbi__clamp((x0-t3) >> 17); + o[1] = stbi__clamp((x1+t2) >> 17); + o[6] = stbi__clamp((x1-t2) >> 17); + o[2] = stbi__clamp((x2+t1) >> 17); + o[5] = stbi__clamp((x2-t1) >> 17); + o[3] = stbi__clamp((x3+t0) >> 17); + o[4] = stbi__clamp((x3-t0) >> 17); + } +} + +#ifdef STBI_SSE2 +// sse2 integer IDCT. not the fastest possible implementation but it +// produces bit-identical results to the generic C version so it's +// fully "transparent". +static void stbi__idct_simd(stbi_uc *out, int out_stride, short data[64]) +{ + // This is constructed to match our regular (generic) integer IDCT exactly. + __m128i row0, row1, row2, row3, row4, row5, row6, row7; + __m128i tmp; + + // dot product constant: even elems=x, odd elems=y + #define dct_const(x,y) _mm_setr_epi16((x),(y),(x),(y),(x),(y),(x),(y)) + + // out(0) = c0[even]*x + c0[odd]*y (c0, x, y 16-bit, out 32-bit) + // out(1) = c1[even]*x + c1[odd]*y + #define dct_rot(out0,out1, x,y,c0,c1) \ + __m128i c0##lo = _mm_unpacklo_epi16((x),(y)); \ + __m128i c0##hi = _mm_unpackhi_epi16((x),(y)); \ + __m128i out0##_l = _mm_madd_epi16(c0##lo, c0); \ + __m128i out0##_h = _mm_madd_epi16(c0##hi, c0); \ + __m128i out1##_l = _mm_madd_epi16(c0##lo, c1); \ + __m128i out1##_h = _mm_madd_epi16(c0##hi, c1) + + // out = in << 12 (in 16-bit, out 32-bit) + #define dct_widen(out, in) \ + __m128i out##_l = _mm_srai_epi32(_mm_unpacklo_epi16(_mm_setzero_si128(), (in)), 4); \ + __m128i out##_h = _mm_srai_epi32(_mm_unpackhi_epi16(_mm_setzero_si128(), (in)), 4) + + // wide add + #define dct_wadd(out, a, b) \ + __m128i out##_l = _mm_add_epi32(a##_l, b##_l); \ + __m128i out##_h = _mm_add_epi32(a##_h, b##_h) + + // wide sub + #define dct_wsub(out, a, b) \ + __m128i out##_l = _mm_sub_epi32(a##_l, b##_l); \ + __m128i out##_h = _mm_sub_epi32(a##_h, b##_h) + + // butterfly a/b, add bias, then shift by "s" and pack + #define dct_bfly32o(out0, out1, a,b,bias,s) \ + { \ + __m128i abiased_l = _mm_add_epi32(a##_l, bias); \ + __m128i abiased_h = _mm_add_epi32(a##_h, bias); \ + dct_wadd(sum, abiased, b); \ + dct_wsub(dif, abiased, b); \ + out0 = _mm_packs_epi32(_mm_srai_epi32(sum_l, s), _mm_srai_epi32(sum_h, s)); \ + out1 = _mm_packs_epi32(_mm_srai_epi32(dif_l, s), _mm_srai_epi32(dif_h, s)); \ + } + + // 8-bit interleave step (for transposes) + #define dct_interleave8(a, b) \ + tmp = a; \ + a = _mm_unpacklo_epi8(a, b); \ + b = _mm_unpackhi_epi8(tmp, b) + + // 16-bit interleave step (for transposes) + #define dct_interleave16(a, b) \ + tmp = a; \ + a = _mm_unpacklo_epi16(a, b); \ + b = _mm_unpackhi_epi16(tmp, b) + + #define dct_pass(bias,shift) \ + { \ + /* even part */ \ + dct_rot(t2e,t3e, row2,row6, rot0_0,rot0_1); \ + __m128i sum04 = _mm_add_epi16(row0, row4); \ + __m128i dif04 = _mm_sub_epi16(row0, row4); \ + dct_widen(t0e, sum04); \ + dct_widen(t1e, dif04); \ + dct_wadd(x0, t0e, t3e); \ + dct_wsub(x3, t0e, t3e); \ + dct_wadd(x1, t1e, t2e); \ + dct_wsub(x2, t1e, t2e); \ + /* odd part */ \ + dct_rot(y0o,y2o, row7,row3, rot2_0,rot2_1); \ + dct_rot(y1o,y3o, row5,row1, rot3_0,rot3_1); \ + __m128i sum17 = _mm_add_epi16(row1, row7); \ + __m128i sum35 = _mm_add_epi16(row3, row5); \ + dct_rot(y4o,y5o, sum17,sum35, rot1_0,rot1_1); \ + dct_wadd(x4, y0o, y4o); \ + dct_wadd(x5, y1o, y5o); \ + dct_wadd(x6, y2o, y5o); \ + dct_wadd(x7, y3o, y4o); \ + dct_bfly32o(row0,row7, x0,x7,bias,shift); \ + dct_bfly32o(row1,row6, x1,x6,bias,shift); \ + dct_bfly32o(row2,row5, x2,x5,bias,shift); \ + dct_bfly32o(row3,row4, x3,x4,bias,shift); \ + } + + __m128i rot0_0 = dct_const(stbi__f2f(0.5411961f), stbi__f2f(0.5411961f) + stbi__f2f(-1.847759065f)); + __m128i rot0_1 = dct_const(stbi__f2f(0.5411961f) + stbi__f2f( 0.765366865f), stbi__f2f(0.5411961f)); + __m128i rot1_0 = dct_const(stbi__f2f(1.175875602f) + stbi__f2f(-0.899976223f), stbi__f2f(1.175875602f)); + __m128i rot1_1 = dct_const(stbi__f2f(1.175875602f), stbi__f2f(1.175875602f) + stbi__f2f(-2.562915447f)); + __m128i rot2_0 = dct_const(stbi__f2f(-1.961570560f) + stbi__f2f( 0.298631336f), stbi__f2f(-1.961570560f)); + __m128i rot2_1 = dct_const(stbi__f2f(-1.961570560f), stbi__f2f(-1.961570560f) + stbi__f2f( 3.072711026f)); + __m128i rot3_0 = dct_const(stbi__f2f(-0.390180644f) + stbi__f2f( 2.053119869f), stbi__f2f(-0.390180644f)); + __m128i rot3_1 = dct_const(stbi__f2f(-0.390180644f), stbi__f2f(-0.390180644f) + stbi__f2f( 1.501321110f)); + + // rounding biases in column/row passes, see stbi__idct_block for explanation. + __m128i bias_0 = _mm_set1_epi32(512); + __m128i bias_1 = _mm_set1_epi32(65536 + (128<<17)); + + // load + row0 = _mm_load_si128((const __m128i *) (data + 0*8)); + row1 = _mm_load_si128((const __m128i *) (data + 1*8)); + row2 = _mm_load_si128((const __m128i *) (data + 2*8)); + row3 = _mm_load_si128((const __m128i *) (data + 3*8)); + row4 = _mm_load_si128((const __m128i *) (data + 4*8)); + row5 = _mm_load_si128((const __m128i *) (data + 5*8)); + row6 = _mm_load_si128((const __m128i *) (data + 6*8)); + row7 = _mm_load_si128((const __m128i *) (data + 7*8)); + + // column pass + dct_pass(bias_0, 10); + + { + // 16bit 8x8 transpose pass 1 + dct_interleave16(row0, row4); + dct_interleave16(row1, row5); + dct_interleave16(row2, row6); + dct_interleave16(row3, row7); + + // transpose pass 2 + dct_interleave16(row0, row2); + dct_interleave16(row1, row3); + dct_interleave16(row4, row6); + dct_interleave16(row5, row7); + + // transpose pass 3 + dct_interleave16(row0, row1); + dct_interleave16(row2, row3); + dct_interleave16(row4, row5); + dct_interleave16(row6, row7); + } + + // row pass + dct_pass(bias_1, 17); + + { + // pack + __m128i p0 = _mm_packus_epi16(row0, row1); // a0a1a2a3...a7b0b1b2b3...b7 + __m128i p1 = _mm_packus_epi16(row2, row3); + __m128i p2 = _mm_packus_epi16(row4, row5); + __m128i p3 = _mm_packus_epi16(row6, row7); + + // 8bit 8x8 transpose pass 1 + dct_interleave8(p0, p2); // a0e0a1e1... + dct_interleave8(p1, p3); // c0g0c1g1... + + // transpose pass 2 + dct_interleave8(p0, p1); // a0c0e0g0... + dct_interleave8(p2, p3); // b0d0f0h0... + + // transpose pass 3 + dct_interleave8(p0, p2); // a0b0c0d0... + dct_interleave8(p1, p3); // a4b4c4d4... + + // store + _mm_storel_epi64((__m128i *) out, p0); out += out_stride; + _mm_storel_epi64((__m128i *) out, _mm_shuffle_epi32(p0, 0x4e)); out += out_stride; + _mm_storel_epi64((__m128i *) out, p2); out += out_stride; + _mm_storel_epi64((__m128i *) out, _mm_shuffle_epi32(p2, 0x4e)); out += out_stride; + _mm_storel_epi64((__m128i *) out, p1); out += out_stride; + _mm_storel_epi64((__m128i *) out, _mm_shuffle_epi32(p1, 0x4e)); out += out_stride; + _mm_storel_epi64((__m128i *) out, p3); out += out_stride; + _mm_storel_epi64((__m128i *) out, _mm_shuffle_epi32(p3, 0x4e)); + } + +#undef dct_const +#undef dct_rot +#undef dct_widen +#undef dct_wadd +#undef dct_wsub +#undef dct_bfly32o +#undef dct_interleave8 +#undef dct_interleave16 +#undef dct_pass +} + +#endif // STBI_SSE2 + +#ifdef STBI_NEON + +// NEON integer IDCT. should produce bit-identical +// results to the generic C version. +static void stbi__idct_simd(stbi_uc *out, int out_stride, short data[64]) +{ + int16x8_t row0, row1, row2, row3, row4, row5, row6, row7; + + int16x4_t rot0_0 = vdup_n_s16(stbi__f2f(0.5411961f)); + int16x4_t rot0_1 = vdup_n_s16(stbi__f2f(-1.847759065f)); + int16x4_t rot0_2 = vdup_n_s16(stbi__f2f( 0.765366865f)); + int16x4_t rot1_0 = vdup_n_s16(stbi__f2f( 1.175875602f)); + int16x4_t rot1_1 = vdup_n_s16(stbi__f2f(-0.899976223f)); + int16x4_t rot1_2 = vdup_n_s16(stbi__f2f(-2.562915447f)); + int16x4_t rot2_0 = vdup_n_s16(stbi__f2f(-1.961570560f)); + int16x4_t rot2_1 = vdup_n_s16(stbi__f2f(-0.390180644f)); + int16x4_t rot3_0 = vdup_n_s16(stbi__f2f( 0.298631336f)); + int16x4_t rot3_1 = vdup_n_s16(stbi__f2f( 2.053119869f)); + int16x4_t rot3_2 = vdup_n_s16(stbi__f2f( 3.072711026f)); + int16x4_t rot3_3 = vdup_n_s16(stbi__f2f( 1.501321110f)); + +#define dct_long_mul(out, inq, coeff) \ + int32x4_t out##_l = vmull_s16(vget_low_s16(inq), coeff); \ + int32x4_t out##_h = vmull_s16(vget_high_s16(inq), coeff) + +#define dct_long_mac(out, acc, inq, coeff) \ + int32x4_t out##_l = vmlal_s16(acc##_l, vget_low_s16(inq), coeff); \ + int32x4_t out##_h = vmlal_s16(acc##_h, vget_high_s16(inq), coeff) + +#define dct_widen(out, inq) \ + int32x4_t out##_l = vshll_n_s16(vget_low_s16(inq), 12); \ + int32x4_t out##_h = vshll_n_s16(vget_high_s16(inq), 12) + +// wide add +#define dct_wadd(out, a, b) \ + int32x4_t out##_l = vaddq_s32(a##_l, b##_l); \ + int32x4_t out##_h = vaddq_s32(a##_h, b##_h) + +// wide sub +#define dct_wsub(out, a, b) \ + int32x4_t out##_l = vsubq_s32(a##_l, b##_l); \ + int32x4_t out##_h = vsubq_s32(a##_h, b##_h) + +// butterfly a/b, then shift using "shiftop" by "s" and pack +#define dct_bfly32o(out0,out1, a,b,shiftop,s) \ + { \ + dct_wadd(sum, a, b); \ + dct_wsub(dif, a, b); \ + out0 = vcombine_s16(shiftop(sum_l, s), shiftop(sum_h, s)); \ + out1 = vcombine_s16(shiftop(dif_l, s), shiftop(dif_h, s)); \ + } + +#define dct_pass(shiftop, shift) \ + { \ + /* even part */ \ + int16x8_t sum26 = vaddq_s16(row2, row6); \ + dct_long_mul(p1e, sum26, rot0_0); \ + dct_long_mac(t2e, p1e, row6, rot0_1); \ + dct_long_mac(t3e, p1e, row2, rot0_2); \ + int16x8_t sum04 = vaddq_s16(row0, row4); \ + int16x8_t dif04 = vsubq_s16(row0, row4); \ + dct_widen(t0e, sum04); \ + dct_widen(t1e, dif04); \ + dct_wadd(x0, t0e, t3e); \ + dct_wsub(x3, t0e, t3e); \ + dct_wadd(x1, t1e, t2e); \ + dct_wsub(x2, t1e, t2e); \ + /* odd part */ \ + int16x8_t sum15 = vaddq_s16(row1, row5); \ + int16x8_t sum17 = vaddq_s16(row1, row7); \ + int16x8_t sum35 = vaddq_s16(row3, row5); \ + int16x8_t sum37 = vaddq_s16(row3, row7); \ + int16x8_t sumodd = vaddq_s16(sum17, sum35); \ + dct_long_mul(p5o, sumodd, rot1_0); \ + dct_long_mac(p1o, p5o, sum17, rot1_1); \ + dct_long_mac(p2o, p5o, sum35, rot1_2); \ + dct_long_mul(p3o, sum37, rot2_0); \ + dct_long_mul(p4o, sum15, rot2_1); \ + dct_wadd(sump13o, p1o, p3o); \ + dct_wadd(sump24o, p2o, p4o); \ + dct_wadd(sump23o, p2o, p3o); \ + dct_wadd(sump14o, p1o, p4o); \ + dct_long_mac(x4, sump13o, row7, rot3_0); \ + dct_long_mac(x5, sump24o, row5, rot3_1); \ + dct_long_mac(x6, sump23o, row3, rot3_2); \ + dct_long_mac(x7, sump14o, row1, rot3_3); \ + dct_bfly32o(row0,row7, x0,x7,shiftop,shift); \ + dct_bfly32o(row1,row6, x1,x6,shiftop,shift); \ + dct_bfly32o(row2,row5, x2,x5,shiftop,shift); \ + dct_bfly32o(row3,row4, x3,x4,shiftop,shift); \ + } + + // load + row0 = vld1q_s16(data + 0*8); + row1 = vld1q_s16(data + 1*8); + row2 = vld1q_s16(data + 2*8); + row3 = vld1q_s16(data + 3*8); + row4 = vld1q_s16(data + 4*8); + row5 = vld1q_s16(data + 5*8); + row6 = vld1q_s16(data + 6*8); + row7 = vld1q_s16(data + 7*8); + + // add DC bias + row0 = vaddq_s16(row0, vsetq_lane_s16(1024, vdupq_n_s16(0), 0)); + + // column pass + dct_pass(vrshrn_n_s32, 10); + + // 16bit 8x8 transpose + { +// these three map to a single VTRN.16, VTRN.32, and VSWP, respectively. +// whether compilers actually get this is another story, sadly. +#define dct_trn16(x, y) { int16x8x2_t t = vtrnq_s16(x, y); x = t.val[0]; y = t.val[1]; } +#define dct_trn32(x, y) { int32x4x2_t t = vtrnq_s32(vreinterpretq_s32_s16(x), vreinterpretq_s32_s16(y)); x = vreinterpretq_s16_s32(t.val[0]); y = vreinterpretq_s16_s32(t.val[1]); } +#define dct_trn64(x, y) { int16x8_t x0 = x; int16x8_t y0 = y; x = vcombine_s16(vget_low_s16(x0), vget_low_s16(y0)); y = vcombine_s16(vget_high_s16(x0), vget_high_s16(y0)); } + + // pass 1 + dct_trn16(row0, row1); // a0b0a2b2a4b4a6b6 + dct_trn16(row2, row3); + dct_trn16(row4, row5); + dct_trn16(row6, row7); + + // pass 2 + dct_trn32(row0, row2); // a0b0c0d0a4b4c4d4 + dct_trn32(row1, row3); + dct_trn32(row4, row6); + dct_trn32(row5, row7); + + // pass 3 + dct_trn64(row0, row4); // a0b0c0d0e0f0g0h0 + dct_trn64(row1, row5); + dct_trn64(row2, row6); + dct_trn64(row3, row7); + +#undef dct_trn16 +#undef dct_trn32 +#undef dct_trn64 + } + + // row pass + // vrshrn_n_s32 only supports shifts up to 16, we need + // 17. so do a non-rounding shift of 16 first then follow + // up with a rounding shift by 1. + dct_pass(vshrn_n_s32, 16); + + { + // pack and round + uint8x8_t p0 = vqrshrun_n_s16(row0, 1); + uint8x8_t p1 = vqrshrun_n_s16(row1, 1); + uint8x8_t p2 = vqrshrun_n_s16(row2, 1); + uint8x8_t p3 = vqrshrun_n_s16(row3, 1); + uint8x8_t p4 = vqrshrun_n_s16(row4, 1); + uint8x8_t p5 = vqrshrun_n_s16(row5, 1); + uint8x8_t p6 = vqrshrun_n_s16(row6, 1); + uint8x8_t p7 = vqrshrun_n_s16(row7, 1); + + // again, these can translate into one instruction, but often don't. +#define dct_trn8_8(x, y) { uint8x8x2_t t = vtrn_u8(x, y); x = t.val[0]; y = t.val[1]; } +#define dct_trn8_16(x, y) { uint16x4x2_t t = vtrn_u16(vreinterpret_u16_u8(x), vreinterpret_u16_u8(y)); x = vreinterpret_u8_u16(t.val[0]); y = vreinterpret_u8_u16(t.val[1]); } +#define dct_trn8_32(x, y) { uint32x2x2_t t = vtrn_u32(vreinterpret_u32_u8(x), vreinterpret_u32_u8(y)); x = vreinterpret_u8_u32(t.val[0]); y = vreinterpret_u8_u32(t.val[1]); } + + // sadly can't use interleaved stores here since we only write + // 8 bytes to each scan line! + + // 8x8 8-bit transpose pass 1 + dct_trn8_8(p0, p1); + dct_trn8_8(p2, p3); + dct_trn8_8(p4, p5); + dct_trn8_8(p6, p7); + + // pass 2 + dct_trn8_16(p0, p2); + dct_trn8_16(p1, p3); + dct_trn8_16(p4, p6); + dct_trn8_16(p5, p7); + + // pass 3 + dct_trn8_32(p0, p4); + dct_trn8_32(p1, p5); + dct_trn8_32(p2, p6); + dct_trn8_32(p3, p7); + + // store + vst1_u8(out, p0); out += out_stride; + vst1_u8(out, p1); out += out_stride; + vst1_u8(out, p2); out += out_stride; + vst1_u8(out, p3); out += out_stride; + vst1_u8(out, p4); out += out_stride; + vst1_u8(out, p5); out += out_stride; + vst1_u8(out, p6); out += out_stride; + vst1_u8(out, p7); + +#undef dct_trn8_8 +#undef dct_trn8_16 +#undef dct_trn8_32 + } + +#undef dct_long_mul +#undef dct_long_mac +#undef dct_widen +#undef dct_wadd +#undef dct_wsub +#undef dct_bfly32o +#undef dct_pass +} + +#endif // STBI_NEON + +#define STBI__MARKER_none 0xff +// if there's a pending marker from the entropy stream, return that +// otherwise, fetch from the stream and get a marker. if there's no +// marker, return 0xff, which is never a valid marker value +static stbi_uc stbi__get_marker(stbi__jpeg *j) +{ + stbi_uc x; + if (j->marker != STBI__MARKER_none) { x = j->marker; j->marker = STBI__MARKER_none; return x; } + x = stbi__get8(j->s); + if (x != 0xff) return STBI__MARKER_none; + while (x == 0xff) + x = stbi__get8(j->s); // consume repeated 0xff fill bytes + return x; +} + +// in each scan, we'll have scan_n components, and the order +// of the components is specified by order[] +#define STBI__RESTART(x) ((x) >= 0xd0 && (x) <= 0xd7) + +// after a restart interval, stbi__jpeg_reset the entropy decoder and +// the dc prediction +static void stbi__jpeg_reset(stbi__jpeg *j) +{ + j->code_bits = 0; + j->code_buffer = 0; + j->nomore = 0; + j->img_comp[0].dc_pred = j->img_comp[1].dc_pred = j->img_comp[2].dc_pred = j->img_comp[3].dc_pred = 0; + j->marker = STBI__MARKER_none; + j->todo = j->restart_interval ? j->restart_interval : 0x7fffffff; + j->eob_run = 0; + // no more than 1<<31 MCUs if no restart_interal? that's plenty safe, + // since we don't even allow 1<<30 pixels +} + +static int stbi__parse_entropy_coded_data(stbi__jpeg *z) +{ + stbi__jpeg_reset(z); + if (!z->progressive) { + if (z->scan_n == 1) { + int i,j; + STBI_SIMD_ALIGN(short, data[64]); + int n = z->order[0]; + // non-interleaved data, we just need to process one block at a time, + // in trivial scanline order + // number of blocks to do just depends on how many actual "pixels" this + // component has, independent of interleaved MCU blocking and such + int w = (z->img_comp[n].x+7) >> 3; + int h = (z->img_comp[n].y+7) >> 3; + for (j=0; j < h; ++j) { + for (i=0; i < w; ++i) { + int ha = z->img_comp[n].ha; + if (!stbi__jpeg_decode_block(z, data, z->huff_dc+z->img_comp[n].hd, z->huff_ac+ha, z->fast_ac[ha], n, z->dequant[z->img_comp[n].tq])) return 0; + z->idct_block_kernel(z->img_comp[n].data+z->img_comp[n].w2*j*8+i*8, z->img_comp[n].w2, data); + // every data block is an MCU, so countdown the restart interval + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + // if it's NOT a restart, then just bail, so we get corrupt data + // rather than no data + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } else { // interleaved + int i,j,k,x,y; + STBI_SIMD_ALIGN(short, data[64]); + for (j=0; j < z->img_mcu_y; ++j) { + for (i=0; i < z->img_mcu_x; ++i) { + // scan an interleaved mcu... process scan_n components in order + for (k=0; k < z->scan_n; ++k) { + int n = z->order[k]; + // scan out an mcu's worth of this component; that's just determined + // by the basic H and V specified for the component + for (y=0; y < z->img_comp[n].v; ++y) { + for (x=0; x < z->img_comp[n].h; ++x) { + int x2 = (i*z->img_comp[n].h + x)*8; + int y2 = (j*z->img_comp[n].v + y)*8; + int ha = z->img_comp[n].ha; + if (!stbi__jpeg_decode_block(z, data, z->huff_dc+z->img_comp[n].hd, z->huff_ac+ha, z->fast_ac[ha], n, z->dequant[z->img_comp[n].tq])) return 0; + z->idct_block_kernel(z->img_comp[n].data+z->img_comp[n].w2*y2+x2, z->img_comp[n].w2, data); + } + } + } + // after all interleaved components, that's an interleaved MCU, + // so now count down the restart interval + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } + } else { + if (z->scan_n == 1) { + int i,j; + int n = z->order[0]; + // non-interleaved data, we just need to process one block at a time, + // in trivial scanline order + // number of blocks to do just depends on how many actual "pixels" this + // component has, independent of interleaved MCU blocking and such + int w = (z->img_comp[n].x+7) >> 3; + int h = (z->img_comp[n].y+7) >> 3; + for (j=0; j < h; ++j) { + for (i=0; i < w; ++i) { + short *data = z->img_comp[n].coeff + 64 * (i + j * z->img_comp[n].coeff_w); + if (z->spec_start == 0) { + if (!stbi__jpeg_decode_block_prog_dc(z, data, &z->huff_dc[z->img_comp[n].hd], n)) + return 0; + } else { + int ha = z->img_comp[n].ha; + if (!stbi__jpeg_decode_block_prog_ac(z, data, &z->huff_ac[ha], z->fast_ac[ha])) + return 0; + } + // every data block is an MCU, so countdown the restart interval + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } else { // interleaved + int i,j,k,x,y; + for (j=0; j < z->img_mcu_y; ++j) { + for (i=0; i < z->img_mcu_x; ++i) { + // scan an interleaved mcu... process scan_n components in order + for (k=0; k < z->scan_n; ++k) { + int n = z->order[k]; + // scan out an mcu's worth of this component; that's just determined + // by the basic H and V specified for the component + for (y=0; y < z->img_comp[n].v; ++y) { + for (x=0; x < z->img_comp[n].h; ++x) { + int x2 = (i*z->img_comp[n].h + x); + int y2 = (j*z->img_comp[n].v + y); + short *data = z->img_comp[n].coeff + 64 * (x2 + y2 * z->img_comp[n].coeff_w); + if (!stbi__jpeg_decode_block_prog_dc(z, data, &z->huff_dc[z->img_comp[n].hd], n)) + return 0; + } + } + } + // after all interleaved components, that's an interleaved MCU, + // so now count down the restart interval + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } + } +} + +static void stbi__jpeg_dequantize(short *data, stbi__uint16 *dequant) +{ + int i; + for (i=0; i < 64; ++i) + data[i] *= dequant[i]; +} + +static void stbi__jpeg_finish(stbi__jpeg *z) +{ + if (z->progressive) { + // dequantize and idct the data + int i,j,n; + for (n=0; n < z->s->img_n; ++n) { + int w = (z->img_comp[n].x+7) >> 3; + int h = (z->img_comp[n].y+7) >> 3; + for (j=0; j < h; ++j) { + for (i=0; i < w; ++i) { + short *data = z->img_comp[n].coeff + 64 * (i + j * z->img_comp[n].coeff_w); + stbi__jpeg_dequantize(data, z->dequant[z->img_comp[n].tq]); + z->idct_block_kernel(z->img_comp[n].data+z->img_comp[n].w2*j*8+i*8, z->img_comp[n].w2, data); + } + } + } + } +} + +static int stbi__process_marker(stbi__jpeg *z, int m) +{ + int L; + switch (m) { + case STBI__MARKER_none: // no marker found + return stbi__err("expected marker","Corrupt JPEG"); + + case 0xDD: // DRI - specify restart interval + if (stbi__get16be(z->s) != 4) return stbi__err("bad DRI len","Corrupt JPEG"); + z->restart_interval = stbi__get16be(z->s); + return 1; + + case 0xDB: // DQT - define quantization table + L = stbi__get16be(z->s)-2; + while (L > 0) { + int q = stbi__get8(z->s); + int p = q >> 4, sixteen = (p != 0); + int t = q & 15,i; + if (p != 0 && p != 1) return stbi__err("bad DQT type","Corrupt JPEG"); + if (t > 3) return stbi__err("bad DQT table","Corrupt JPEG"); + + for (i=0; i < 64; ++i) + z->dequant[t][stbi__jpeg_dezigzag[i]] = (stbi__uint16)(sixteen ? stbi__get16be(z->s) : stbi__get8(z->s)); + L -= (sixteen ? 129 : 65); + } + return L==0; + + case 0xC4: // DHT - define huffman table + L = stbi__get16be(z->s)-2; + while (L > 0) { + stbi_uc *v; + int sizes[16],i,n=0; + int q = stbi__get8(z->s); + int tc = q >> 4; + int th = q & 15; + if (tc > 1 || th > 3) return stbi__err("bad DHT header","Corrupt JPEG"); + for (i=0; i < 16; ++i) { + sizes[i] = stbi__get8(z->s); + n += sizes[i]; + } + if(n > 256) return stbi__err("bad DHT header","Corrupt JPEG"); // Loop over i < n would write past end of values! + L -= 17; + if (tc == 0) { + if (!stbi__build_huffman(z->huff_dc+th, sizes)) return 0; + v = z->huff_dc[th].values; + } else { + if (!stbi__build_huffman(z->huff_ac+th, sizes)) return 0; + v = z->huff_ac[th].values; + } + for (i=0; i < n; ++i) + v[i] = stbi__get8(z->s); + if (tc != 0) + stbi__build_fast_ac(z->fast_ac[th], z->huff_ac + th); + L -= n; + } + return L==0; + } + + // check for comment block or APP blocks + if ((m >= 0xE0 && m <= 0xEF) || m == 0xFE) { + L = stbi__get16be(z->s); + if (L < 2) { + if (m == 0xFE) + return stbi__err("bad COM len","Corrupt JPEG"); + else + return stbi__err("bad APP len","Corrupt JPEG"); + } + L -= 2; + + if (m == 0xE0 && L >= 5) { // JFIF APP0 segment + static const unsigned char tag[5] = {'J','F','I','F','\0'}; + int ok = 1; + int i; + for (i=0; i < 5; ++i) + if (stbi__get8(z->s) != tag[i]) + ok = 0; + L -= 5; + if (ok) + z->jfif = 1; + } else if (m == 0xEE && L >= 12) { // Adobe APP14 segment + static const unsigned char tag[6] = {'A','d','o','b','e','\0'}; + int ok = 1; + int i; + for (i=0; i < 6; ++i) + if (stbi__get8(z->s) != tag[i]) + ok = 0; + L -= 6; + if (ok) { + stbi__get8(z->s); // version + stbi__get16be(z->s); // flags0 + stbi__get16be(z->s); // flags1 + z->app14_color_transform = stbi__get8(z->s); // color transform + L -= 6; + } + } + + stbi__skip(z->s, L); + return 1; + } + + return stbi__err("unknown marker","Corrupt JPEG"); +} + +// after we see SOS +static int stbi__process_scan_header(stbi__jpeg *z) +{ + int i; + int Ls = stbi__get16be(z->s); + z->scan_n = stbi__get8(z->s); + if (z->scan_n < 1 || z->scan_n > 4 || z->scan_n > (int) z->s->img_n) return stbi__err("bad SOS component count","Corrupt JPEG"); + if (Ls != 6+2*z->scan_n) return stbi__err("bad SOS len","Corrupt JPEG"); + for (i=0; i < z->scan_n; ++i) { + int id = stbi__get8(z->s), which; + int q = stbi__get8(z->s); + for (which = 0; which < z->s->img_n; ++which) + if (z->img_comp[which].id == id) + break; + if (which == z->s->img_n) return 0; // no match + z->img_comp[which].hd = q >> 4; if (z->img_comp[which].hd > 3) return stbi__err("bad DC huff","Corrupt JPEG"); + z->img_comp[which].ha = q & 15; if (z->img_comp[which].ha > 3) return stbi__err("bad AC huff","Corrupt JPEG"); + z->order[i] = which; + } + + { + int aa; + z->spec_start = stbi__get8(z->s); + z->spec_end = stbi__get8(z->s); // should be 63, but might be 0 + aa = stbi__get8(z->s); + z->succ_high = (aa >> 4); + z->succ_low = (aa & 15); + if (z->progressive) { + if (z->spec_start > 63 || z->spec_end > 63 || z->spec_start > z->spec_end || z->succ_high > 13 || z->succ_low > 13) + return stbi__err("bad SOS", "Corrupt JPEG"); + } else { + if (z->spec_start != 0) return stbi__err("bad SOS","Corrupt JPEG"); + if (z->succ_high != 0 || z->succ_low != 0) return stbi__err("bad SOS","Corrupt JPEG"); + z->spec_end = 63; + } + } + + return 1; +} + +static int stbi__free_jpeg_components(stbi__jpeg *z, int ncomp, int why) +{ + int i; + for (i=0; i < ncomp; ++i) { + if (z->img_comp[i].raw_data) { + STBI_FREE(z->img_comp[i].raw_data); + z->img_comp[i].raw_data = NULL; + z->img_comp[i].data = NULL; + } + if (z->img_comp[i].raw_coeff) { + STBI_FREE(z->img_comp[i].raw_coeff); + z->img_comp[i].raw_coeff = 0; + z->img_comp[i].coeff = 0; + } + if (z->img_comp[i].linebuf) { + STBI_FREE(z->img_comp[i].linebuf); + z->img_comp[i].linebuf = NULL; + } + } + return why; +} + +static int stbi__process_frame_header(stbi__jpeg *z, int scan) +{ + stbi__context *s = z->s; + int Lf,p,i,q, h_max=1,v_max=1,c; + Lf = stbi__get16be(s); if (Lf < 11) return stbi__err("bad SOF len","Corrupt JPEG"); // JPEG + p = stbi__get8(s); if (p != 8) return stbi__err("only 8-bit","JPEG format not supported: 8-bit only"); // JPEG baseline + s->img_y = stbi__get16be(s); if (s->img_y == 0) return stbi__err("no header height", "JPEG format not supported: delayed height"); // Legal, but we don't handle it--but neither does IJG + s->img_x = stbi__get16be(s); if (s->img_x == 0) return stbi__err("0 width","Corrupt JPEG"); // JPEG requires + if (s->img_y > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); + if (s->img_x > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); + c = stbi__get8(s); + if (c != 3 && c != 1 && c != 4) return stbi__err("bad component count","Corrupt JPEG"); + s->img_n = c; + for (i=0; i < c; ++i) { + z->img_comp[i].data = NULL; + z->img_comp[i].linebuf = NULL; + } + + if (Lf != 8+3*s->img_n) return stbi__err("bad SOF len","Corrupt JPEG"); + + z->rgb = 0; + for (i=0; i < s->img_n; ++i) { + static const unsigned char rgb[3] = { 'R', 'G', 'B' }; + z->img_comp[i].id = stbi__get8(s); + if (s->img_n == 3 && z->img_comp[i].id == rgb[i]) + ++z->rgb; + q = stbi__get8(s); + z->img_comp[i].h = (q >> 4); if (!z->img_comp[i].h || z->img_comp[i].h > 4) return stbi__err("bad H","Corrupt JPEG"); + z->img_comp[i].v = q & 15; if (!z->img_comp[i].v || z->img_comp[i].v > 4) return stbi__err("bad V","Corrupt JPEG"); + z->img_comp[i].tq = stbi__get8(s); if (z->img_comp[i].tq > 3) return stbi__err("bad TQ","Corrupt JPEG"); + } + + if (scan != STBI__SCAN_load) return 1; + + if (!stbi__mad3sizes_valid(s->img_x, s->img_y, s->img_n, 0)) return stbi__err("too large", "Image too large to decode"); + + for (i=0; i < s->img_n; ++i) { + if (z->img_comp[i].h > h_max) h_max = z->img_comp[i].h; + if (z->img_comp[i].v > v_max) v_max = z->img_comp[i].v; + } + + // check that plane subsampling factors are integer ratios; our resamplers can't deal with fractional ratios + // and I've never seen a non-corrupted JPEG file actually use them + for (i=0; i < s->img_n; ++i) { + if (h_max % z->img_comp[i].h != 0) return stbi__err("bad H","Corrupt JPEG"); + if (v_max % z->img_comp[i].v != 0) return stbi__err("bad V","Corrupt JPEG"); + } + + // compute interleaved mcu info + z->img_h_max = h_max; + z->img_v_max = v_max; + z->img_mcu_w = h_max * 8; + z->img_mcu_h = v_max * 8; + // these sizes can't be more than 17 bits + z->img_mcu_x = (s->img_x + z->img_mcu_w-1) / z->img_mcu_w; + z->img_mcu_y = (s->img_y + z->img_mcu_h-1) / z->img_mcu_h; + + for (i=0; i < s->img_n; ++i) { + // number of effective pixels (e.g. for non-interleaved MCU) + z->img_comp[i].x = (s->img_x * z->img_comp[i].h + h_max-1) / h_max; + z->img_comp[i].y = (s->img_y * z->img_comp[i].v + v_max-1) / v_max; + // to simplify generation, we'll allocate enough memory to decode + // the bogus oversized data from using interleaved MCUs and their + // big blocks (e.g. a 16x16 iMCU on an image of width 33); we won't + // discard the extra data until colorspace conversion + // + // img_mcu_x, img_mcu_y: <=17 bits; comp[i].h and .v are <=4 (checked earlier) + // so these muls can't overflow with 32-bit ints (which we require) + z->img_comp[i].w2 = z->img_mcu_x * z->img_comp[i].h * 8; + z->img_comp[i].h2 = z->img_mcu_y * z->img_comp[i].v * 8; + z->img_comp[i].coeff = 0; + z->img_comp[i].raw_coeff = 0; + z->img_comp[i].linebuf = NULL; + z->img_comp[i].raw_data = stbi__malloc_mad2(z->img_comp[i].w2, z->img_comp[i].h2, 15); + if (z->img_comp[i].raw_data == NULL) + return stbi__free_jpeg_components(z, i+1, stbi__err("outofmem", "Out of memory")); + // align blocks for idct using mmx/sse + z->img_comp[i].data = (stbi_uc*) (((size_t) z->img_comp[i].raw_data + 15) & ~15); + if (z->progressive) { + // w2, h2 are multiples of 8 (see above) + z->img_comp[i].coeff_w = z->img_comp[i].w2 / 8; + z->img_comp[i].coeff_h = z->img_comp[i].h2 / 8; + z->img_comp[i].raw_coeff = stbi__malloc_mad3(z->img_comp[i].w2, z->img_comp[i].h2, sizeof(short), 15); + if (z->img_comp[i].raw_coeff == NULL) + return stbi__free_jpeg_components(z, i+1, stbi__err("outofmem", "Out of memory")); + z->img_comp[i].coeff = (short*) (((size_t) z->img_comp[i].raw_coeff + 15) & ~15); + } + } + + return 1; +} + +// use comparisons since in some cases we handle more than one case (e.g. SOF) +#define stbi__DNL(x) ((x) == 0xdc) +#define stbi__SOI(x) ((x) == 0xd8) +#define stbi__EOI(x) ((x) == 0xd9) +#define stbi__SOF(x) ((x) == 0xc0 || (x) == 0xc1 || (x) == 0xc2) +#define stbi__SOS(x) ((x) == 0xda) + +#define stbi__SOF_progressive(x) ((x) == 0xc2) + +static int stbi__decode_jpeg_header(stbi__jpeg *z, int scan) +{ + int m; + z->jfif = 0; + z->app14_color_transform = -1; // valid values are 0,1,2 + z->marker = STBI__MARKER_none; // initialize cached marker to empty + m = stbi__get_marker(z); + if (!stbi__SOI(m)) return stbi__err("no SOI","Corrupt JPEG"); + if (scan == STBI__SCAN_type) return 1; + m = stbi__get_marker(z); + while (!stbi__SOF(m)) { + if (!stbi__process_marker(z,m)) return 0; + m = stbi__get_marker(z); + while (m == STBI__MARKER_none) { + // some files have extra padding after their blocks, so ok, we'll scan + if (stbi__at_eof(z->s)) return stbi__err("no SOF", "Corrupt JPEG"); + m = stbi__get_marker(z); + } + } + z->progressive = stbi__SOF_progressive(m); + if (!stbi__process_frame_header(z, scan)) return 0; + return 1; +} + +static stbi_uc stbi__skip_jpeg_junk_at_end(stbi__jpeg *j) +{ + // some JPEGs have junk at end, skip over it but if we find what looks + // like a valid marker, resume there + while (!stbi__at_eof(j->s)) { + stbi_uc x = stbi__get8(j->s); + while (x == 0xff) { // might be a marker + if (stbi__at_eof(j->s)) return STBI__MARKER_none; + x = stbi__get8(j->s); + if (x != 0x00 && x != 0xff) { + // not a stuffed zero or lead-in to another marker, looks + // like an actual marker, return it + return x; + } + // stuffed zero has x=0 now which ends the loop, meaning we go + // back to regular scan loop. + // repeated 0xff keeps trying to read the next byte of the marker. + } + } + return STBI__MARKER_none; +} + +// decode image to YCbCr format +static int stbi__decode_jpeg_image(stbi__jpeg *j) +{ + int m; + for (m = 0; m < 4; m++) { + j->img_comp[m].raw_data = NULL; + j->img_comp[m].raw_coeff = NULL; + } + j->restart_interval = 0; + if (!stbi__decode_jpeg_header(j, STBI__SCAN_load)) return 0; + m = stbi__get_marker(j); + while (!stbi__EOI(m)) { + if (stbi__SOS(m)) { + if (!stbi__process_scan_header(j)) return 0; + if (!stbi__parse_entropy_coded_data(j)) return 0; + if (j->marker == STBI__MARKER_none ) { + j->marker = stbi__skip_jpeg_junk_at_end(j); + // if we reach eof without hitting a marker, stbi__get_marker() below will fail and we'll eventually return 0 + } + m = stbi__get_marker(j); + if (STBI__RESTART(m)) + m = stbi__get_marker(j); + } else if (stbi__DNL(m)) { + int Ld = stbi__get16be(j->s); + stbi__uint32 NL = stbi__get16be(j->s); + if (Ld != 4) return stbi__err("bad DNL len", "Corrupt JPEG"); + if (NL != j->s->img_y) return stbi__err("bad DNL height", "Corrupt JPEG"); + m = stbi__get_marker(j); + } else { + if (!stbi__process_marker(j, m)) return 1; + m = stbi__get_marker(j); + } + } + if (j->progressive) + stbi__jpeg_finish(j); + return 1; +} + +// static jfif-centered resampling (across block boundaries) + +typedef stbi_uc *(*resample_row_func)(stbi_uc *out, stbi_uc *in0, stbi_uc *in1, + int w, int hs); + +#define stbi__div4(x) ((stbi_uc) ((x) >> 2)) + +static stbi_uc *resample_row_1(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) +{ + STBI_NOTUSED(out); + STBI_NOTUSED(in_far); + STBI_NOTUSED(w); + STBI_NOTUSED(hs); + return in_near; +} + +static stbi_uc* stbi__resample_row_v_2(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) +{ + // need to generate two samples vertically for every one in input + int i; + STBI_NOTUSED(hs); + for (i=0; i < w; ++i) + out[i] = stbi__div4(3*in_near[i] + in_far[i] + 2); + return out; +} + +static stbi_uc* stbi__resample_row_h_2(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) +{ + // need to generate two samples horizontally for every one in input + int i; + stbi_uc *input = in_near; + + if (w == 1) { + // if only one sample, can't do any interpolation + out[0] = out[1] = input[0]; + return out; + } + + out[0] = input[0]; + out[1] = stbi__div4(input[0]*3 + input[1] + 2); + for (i=1; i < w-1; ++i) { + int n = 3*input[i]+2; + out[i*2+0] = stbi__div4(n+input[i-1]); + out[i*2+1] = stbi__div4(n+input[i+1]); + } + out[i*2+0] = stbi__div4(input[w-2]*3 + input[w-1] + 2); + out[i*2+1] = input[w-1]; + + STBI_NOTUSED(in_far); + STBI_NOTUSED(hs); + + return out; +} + +#define stbi__div16(x) ((stbi_uc) ((x) >> 4)) + +static stbi_uc *stbi__resample_row_hv_2(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) +{ + // need to generate 2x2 samples for every one in input + int i,t0,t1; + if (w == 1) { + out[0] = out[1] = stbi__div4(3*in_near[0] + in_far[0] + 2); + return out; + } + + t1 = 3*in_near[0] + in_far[0]; + out[0] = stbi__div4(t1+2); + for (i=1; i < w; ++i) { + t0 = t1; + t1 = 3*in_near[i]+in_far[i]; + out[i*2-1] = stbi__div16(3*t0 + t1 + 8); + out[i*2 ] = stbi__div16(3*t1 + t0 + 8); + } + out[w*2-1] = stbi__div4(t1+2); + + STBI_NOTUSED(hs); + + return out; +} + +#if defined(STBI_SSE2) || defined(STBI_NEON) +static stbi_uc *stbi__resample_row_hv_2_simd(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) +{ + // need to generate 2x2 samples for every one in input + int i=0,t0,t1; + + if (w == 1) { + out[0] = out[1] = stbi__div4(3*in_near[0] + in_far[0] + 2); + return out; + } + + t1 = 3*in_near[0] + in_far[0]; + // process groups of 8 pixels for as long as we can. + // note we can't handle the last pixel in a row in this loop + // because we need to handle the filter boundary conditions. + for (; i < ((w-1) & ~7); i += 8) { +#if defined(STBI_SSE2) + // load and perform the vertical filtering pass + // this uses 3*x + y = 4*x + (y - x) + __m128i zero = _mm_setzero_si128(); + __m128i farb = _mm_loadl_epi64((__m128i *) (in_far + i)); + __m128i nearb = _mm_loadl_epi64((__m128i *) (in_near + i)); + __m128i farw = _mm_unpacklo_epi8(farb, zero); + __m128i nearw = _mm_unpacklo_epi8(nearb, zero); + __m128i diff = _mm_sub_epi16(farw, nearw); + __m128i nears = _mm_slli_epi16(nearw, 2); + __m128i curr = _mm_add_epi16(nears, diff); // current row + + // horizontal filter works the same based on shifted vers of current + // row. "prev" is current row shifted right by 1 pixel; we need to + // insert the previous pixel value (from t1). + // "next" is current row shifted left by 1 pixel, with first pixel + // of next block of 8 pixels added in. + __m128i prv0 = _mm_slli_si128(curr, 2); + __m128i nxt0 = _mm_srli_si128(curr, 2); + __m128i prev = _mm_insert_epi16(prv0, t1, 0); + __m128i next = _mm_insert_epi16(nxt0, 3*in_near[i+8] + in_far[i+8], 7); + + // horizontal filter, polyphase implementation since it's convenient: + // even pixels = 3*cur + prev = cur*4 + (prev - cur) + // odd pixels = 3*cur + next = cur*4 + (next - cur) + // note the shared term. + __m128i bias = _mm_set1_epi16(8); + __m128i curs = _mm_slli_epi16(curr, 2); + __m128i prvd = _mm_sub_epi16(prev, curr); + __m128i nxtd = _mm_sub_epi16(next, curr); + __m128i curb = _mm_add_epi16(curs, bias); + __m128i even = _mm_add_epi16(prvd, curb); + __m128i odd = _mm_add_epi16(nxtd, curb); + + // interleave even and odd pixels, then undo scaling. + __m128i int0 = _mm_unpacklo_epi16(even, odd); + __m128i int1 = _mm_unpackhi_epi16(even, odd); + __m128i de0 = _mm_srli_epi16(int0, 4); + __m128i de1 = _mm_srli_epi16(int1, 4); + + // pack and write output + __m128i outv = _mm_packus_epi16(de0, de1); + _mm_storeu_si128((__m128i *) (out + i*2), outv); +#elif defined(STBI_NEON) + // load and perform the vertical filtering pass + // this uses 3*x + y = 4*x + (y - x) + uint8x8_t farb = vld1_u8(in_far + i); + uint8x8_t nearb = vld1_u8(in_near + i); + int16x8_t diff = vreinterpretq_s16_u16(vsubl_u8(farb, nearb)); + int16x8_t nears = vreinterpretq_s16_u16(vshll_n_u8(nearb, 2)); + int16x8_t curr = vaddq_s16(nears, diff); // current row + + // horizontal filter works the same based on shifted vers of current + // row. "prev" is current row shifted right by 1 pixel; we need to + // insert the previous pixel value (from t1). + // "next" is current row shifted left by 1 pixel, with first pixel + // of next block of 8 pixels added in. + int16x8_t prv0 = vextq_s16(curr, curr, 7); + int16x8_t nxt0 = vextq_s16(curr, curr, 1); + int16x8_t prev = vsetq_lane_s16(t1, prv0, 0); + int16x8_t next = vsetq_lane_s16(3*in_near[i+8] + in_far[i+8], nxt0, 7); + + // horizontal filter, polyphase implementation since it's convenient: + // even pixels = 3*cur + prev = cur*4 + (prev - cur) + // odd pixels = 3*cur + next = cur*4 + (next - cur) + // note the shared term. + int16x8_t curs = vshlq_n_s16(curr, 2); + int16x8_t prvd = vsubq_s16(prev, curr); + int16x8_t nxtd = vsubq_s16(next, curr); + int16x8_t even = vaddq_s16(curs, prvd); + int16x8_t odd = vaddq_s16(curs, nxtd); + + // undo scaling and round, then store with even/odd phases interleaved + uint8x8x2_t o; + o.val[0] = vqrshrun_n_s16(even, 4); + o.val[1] = vqrshrun_n_s16(odd, 4); + vst2_u8(out + i*2, o); +#endif + + // "previous" value for next iter + t1 = 3*in_near[i+7] + in_far[i+7]; + } + + t0 = t1; + t1 = 3*in_near[i] + in_far[i]; + out[i*2] = stbi__div16(3*t1 + t0 + 8); + + for (++i; i < w; ++i) { + t0 = t1; + t1 = 3*in_near[i]+in_far[i]; + out[i*2-1] = stbi__div16(3*t0 + t1 + 8); + out[i*2 ] = stbi__div16(3*t1 + t0 + 8); + } + out[w*2-1] = stbi__div4(t1+2); + + STBI_NOTUSED(hs); + + return out; +} +#endif + +static stbi_uc *stbi__resample_row_generic(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) +{ + // resample with nearest-neighbor + int i,j; + STBI_NOTUSED(in_far); + for (i=0; i < w; ++i) + for (j=0; j < hs; ++j) + out[i*hs+j] = in_near[i]; + return out; +} + +// this is a reduced-precision calculation of YCbCr-to-RGB introduced +// to make sure the code produces the same results in both SIMD and scalar +#define stbi__float2fixed(x) (((int) ((x) * 4096.0f + 0.5f)) << 8) +static void stbi__YCbCr_to_RGB_row(stbi_uc *out, const stbi_uc *y, const stbi_uc *pcb, const stbi_uc *pcr, int count, int step) +{ + int i; + for (i=0; i < count; ++i) { + int y_fixed = (y[i] << 20) + (1<<19); // rounding + int r,g,b; + int cr = pcr[i] - 128; + int cb = pcb[i] - 128; + r = y_fixed + cr* stbi__float2fixed(1.40200f); + g = y_fixed + (cr*-stbi__float2fixed(0.71414f)) + ((cb*-stbi__float2fixed(0.34414f)) & 0xffff0000); + b = y_fixed + cb* stbi__float2fixed(1.77200f); + r >>= 20; + g >>= 20; + b >>= 20; + if ((unsigned) r > 255) { if (r < 0) r = 0; else r = 255; } + if ((unsigned) g > 255) { if (g < 0) g = 0; else g = 255; } + if ((unsigned) b > 255) { if (b < 0) b = 0; else b = 255; } + out[0] = (stbi_uc)r; + out[1] = (stbi_uc)g; + out[2] = (stbi_uc)b; + out[3] = 255; + out += step; + } +} + +#if defined(STBI_SSE2) || defined(STBI_NEON) +static void stbi__YCbCr_to_RGB_simd(stbi_uc *out, stbi_uc const *y, stbi_uc const *pcb, stbi_uc const *pcr, int count, int step) +{ + int i = 0; + +#ifdef STBI_SSE2 + // step == 3 is pretty ugly on the final interleave, and i'm not convinced + // it's useful in practice (you wouldn't use it for textures, for example). + // so just accelerate step == 4 case. + if (step == 4) { + // this is a fairly straightforward implementation and not super-optimized. + __m128i signflip = _mm_set1_epi8(-0x80); + __m128i cr_const0 = _mm_set1_epi16( (short) ( 1.40200f*4096.0f+0.5f)); + __m128i cr_const1 = _mm_set1_epi16( - (short) ( 0.71414f*4096.0f+0.5f)); + __m128i cb_const0 = _mm_set1_epi16( - (short) ( 0.34414f*4096.0f+0.5f)); + __m128i cb_const1 = _mm_set1_epi16( (short) ( 1.77200f*4096.0f+0.5f)); + __m128i y_bias = _mm_set1_epi8((char) (unsigned char) 128); + __m128i xw = _mm_set1_epi16(255); // alpha channel + + for (; i+7 < count; i += 8) { + // load + __m128i y_bytes = _mm_loadl_epi64((__m128i *) (y+i)); + __m128i cr_bytes = _mm_loadl_epi64((__m128i *) (pcr+i)); + __m128i cb_bytes = _mm_loadl_epi64((__m128i *) (pcb+i)); + __m128i cr_biased = _mm_xor_si128(cr_bytes, signflip); // -128 + __m128i cb_biased = _mm_xor_si128(cb_bytes, signflip); // -128 + + // unpack to short (and left-shift cr, cb by 8) + __m128i yw = _mm_unpacklo_epi8(y_bias, y_bytes); + __m128i crw = _mm_unpacklo_epi8(_mm_setzero_si128(), cr_biased); + __m128i cbw = _mm_unpacklo_epi8(_mm_setzero_si128(), cb_biased); + + // color transform + __m128i yws = _mm_srli_epi16(yw, 4); + __m128i cr0 = _mm_mulhi_epi16(cr_const0, crw); + __m128i cb0 = _mm_mulhi_epi16(cb_const0, cbw); + __m128i cb1 = _mm_mulhi_epi16(cbw, cb_const1); + __m128i cr1 = _mm_mulhi_epi16(crw, cr_const1); + __m128i rws = _mm_add_epi16(cr0, yws); + __m128i gwt = _mm_add_epi16(cb0, yws); + __m128i bws = _mm_add_epi16(yws, cb1); + __m128i gws = _mm_add_epi16(gwt, cr1); + + // descale + __m128i rw = _mm_srai_epi16(rws, 4); + __m128i bw = _mm_srai_epi16(bws, 4); + __m128i gw = _mm_srai_epi16(gws, 4); + + // back to byte, set up for transpose + __m128i brb = _mm_packus_epi16(rw, bw); + __m128i gxb = _mm_packus_epi16(gw, xw); + + // transpose to interleave channels + __m128i t0 = _mm_unpacklo_epi8(brb, gxb); + __m128i t1 = _mm_unpackhi_epi8(brb, gxb); + __m128i o0 = _mm_unpacklo_epi16(t0, t1); + __m128i o1 = _mm_unpackhi_epi16(t0, t1); + + // store + _mm_storeu_si128((__m128i *) (out + 0), o0); + _mm_storeu_si128((__m128i *) (out + 16), o1); + out += 32; + } + } +#endif + +#ifdef STBI_NEON + // in this version, step=3 support would be easy to add. but is there demand? + if (step == 4) { + // this is a fairly straightforward implementation and not super-optimized. + uint8x8_t signflip = vdup_n_u8(0x80); + int16x8_t cr_const0 = vdupq_n_s16( (short) ( 1.40200f*4096.0f+0.5f)); + int16x8_t cr_const1 = vdupq_n_s16( - (short) ( 0.71414f*4096.0f+0.5f)); + int16x8_t cb_const0 = vdupq_n_s16( - (short) ( 0.34414f*4096.0f+0.5f)); + int16x8_t cb_const1 = vdupq_n_s16( (short) ( 1.77200f*4096.0f+0.5f)); + + for (; i+7 < count; i += 8) { + // load + uint8x8_t y_bytes = vld1_u8(y + i); + uint8x8_t cr_bytes = vld1_u8(pcr + i); + uint8x8_t cb_bytes = vld1_u8(pcb + i); + int8x8_t cr_biased = vreinterpret_s8_u8(vsub_u8(cr_bytes, signflip)); + int8x8_t cb_biased = vreinterpret_s8_u8(vsub_u8(cb_bytes, signflip)); + + // expand to s16 + int16x8_t yws = vreinterpretq_s16_u16(vshll_n_u8(y_bytes, 4)); + int16x8_t crw = vshll_n_s8(cr_biased, 7); + int16x8_t cbw = vshll_n_s8(cb_biased, 7); + + // color transform + int16x8_t cr0 = vqdmulhq_s16(crw, cr_const0); + int16x8_t cb0 = vqdmulhq_s16(cbw, cb_const0); + int16x8_t cr1 = vqdmulhq_s16(crw, cr_const1); + int16x8_t cb1 = vqdmulhq_s16(cbw, cb_const1); + int16x8_t rws = vaddq_s16(yws, cr0); + int16x8_t gws = vaddq_s16(vaddq_s16(yws, cb0), cr1); + int16x8_t bws = vaddq_s16(yws, cb1); + + // undo scaling, round, convert to byte + uint8x8x4_t o; + o.val[0] = vqrshrun_n_s16(rws, 4); + o.val[1] = vqrshrun_n_s16(gws, 4); + o.val[2] = vqrshrun_n_s16(bws, 4); + o.val[3] = vdup_n_u8(255); + + // store, interleaving r/g/b/a + vst4_u8(out, o); + out += 8*4; + } + } +#endif + + for (; i < count; ++i) { + int y_fixed = (y[i] << 20) + (1<<19); // rounding + int r,g,b; + int cr = pcr[i] - 128; + int cb = pcb[i] - 128; + r = y_fixed + cr* stbi__float2fixed(1.40200f); + g = y_fixed + cr*-stbi__float2fixed(0.71414f) + ((cb*-stbi__float2fixed(0.34414f)) & 0xffff0000); + b = y_fixed + cb* stbi__float2fixed(1.77200f); + r >>= 20; + g >>= 20; + b >>= 20; + if ((unsigned) r > 255) { if (r < 0) r = 0; else r = 255; } + if ((unsigned) g > 255) { if (g < 0) g = 0; else g = 255; } + if ((unsigned) b > 255) { if (b < 0) b = 0; else b = 255; } + out[0] = (stbi_uc)r; + out[1] = (stbi_uc)g; + out[2] = (stbi_uc)b; + out[3] = 255; + out += step; + } +} +#endif + +// set up the kernels +static void stbi__setup_jpeg(stbi__jpeg *j) +{ + j->idct_block_kernel = stbi__idct_block; + j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_row; + j->resample_row_hv_2_kernel = stbi__resample_row_hv_2; + +#ifdef STBI_SSE2 + if (stbi__sse2_available()) { + j->idct_block_kernel = stbi__idct_simd; + j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_simd; + j->resample_row_hv_2_kernel = stbi__resample_row_hv_2_simd; + } +#endif + +#ifdef STBI_NEON + j->idct_block_kernel = stbi__idct_simd; + j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_simd; + j->resample_row_hv_2_kernel = stbi__resample_row_hv_2_simd; +#endif +} + +// clean up the temporary component buffers +static void stbi__cleanup_jpeg(stbi__jpeg *j) +{ + stbi__free_jpeg_components(j, j->s->img_n, 0); +} + +typedef struct +{ + resample_row_func resample; + stbi_uc *line0,*line1; + int hs,vs; // expansion factor in each axis + int w_lores; // horizontal pixels pre-expansion + int ystep; // how far through vertical expansion we are + int ypos; // which pre-expansion row we're on +} stbi__resample; + +// fast 0..255 * 0..255 => 0..255 rounded multiplication +static stbi_uc stbi__blinn_8x8(stbi_uc x, stbi_uc y) +{ + unsigned int t = x*y + 128; + return (stbi_uc) ((t + (t >>8)) >> 8); +} + +static stbi_uc *load_jpeg_image(stbi__jpeg *z, int *out_x, int *out_y, int *comp, int req_comp) +{ + int n, decode_n, is_rgb; + z->s->img_n = 0; // make stbi__cleanup_jpeg safe + + // validate req_comp + if (req_comp < 0 || req_comp > 4) return stbi__errpuc("bad req_comp", "Internal error"); + + // load a jpeg image from whichever source, but leave in YCbCr format + if (!stbi__decode_jpeg_image(z)) { stbi__cleanup_jpeg(z); return NULL; } + + // determine actual number of components to generate + n = req_comp ? req_comp : z->s->img_n >= 3 ? 3 : 1; + + is_rgb = z->s->img_n == 3 && (z->rgb == 3 || (z->app14_color_transform == 0 && !z->jfif)); + + if (z->s->img_n == 3 && n < 3 && !is_rgb) + decode_n = 1; + else + decode_n = z->s->img_n; + + // nothing to do if no components requested; check this now to avoid + // accessing uninitialized coutput[0] later + if (decode_n <= 0) { stbi__cleanup_jpeg(z); return NULL; } + + // resample and color-convert + { + int k; + unsigned int i,j; + stbi_uc *output; + stbi_uc *coutput[4] = { NULL, NULL, NULL, NULL }; + + stbi__resample res_comp[4]; + + for (k=0; k < decode_n; ++k) { + stbi__resample *r = &res_comp[k]; + + // allocate line buffer big enough for upsampling off the edges + // with upsample factor of 4 + z->img_comp[k].linebuf = (stbi_uc *) stbi__malloc(z->s->img_x + 3); + if (!z->img_comp[k].linebuf) { stbi__cleanup_jpeg(z); return stbi__errpuc("outofmem", "Out of memory"); } + + r->hs = z->img_h_max / z->img_comp[k].h; + r->vs = z->img_v_max / z->img_comp[k].v; + r->ystep = r->vs >> 1; + r->w_lores = (z->s->img_x + r->hs-1) / r->hs; + r->ypos = 0; + r->line0 = r->line1 = z->img_comp[k].data; + + if (r->hs == 1 && r->vs == 1) r->resample = resample_row_1; + else if (r->hs == 1 && r->vs == 2) r->resample = stbi__resample_row_v_2; + else if (r->hs == 2 && r->vs == 1) r->resample = stbi__resample_row_h_2; + else if (r->hs == 2 && r->vs == 2) r->resample = z->resample_row_hv_2_kernel; + else r->resample = stbi__resample_row_generic; + } + + // can't error after this so, this is safe + output = (stbi_uc *) stbi__malloc_mad3(n, z->s->img_x, z->s->img_y, 1); + if (!output) { stbi__cleanup_jpeg(z); return stbi__errpuc("outofmem", "Out of memory"); } + + // now go ahead and resample + for (j=0; j < z->s->img_y; ++j) { + stbi_uc *out = output + n * z->s->img_x * j; + for (k=0; k < decode_n; ++k) { + stbi__resample *r = &res_comp[k]; + int y_bot = r->ystep >= (r->vs >> 1); + coutput[k] = r->resample(z->img_comp[k].linebuf, + y_bot ? r->line1 : r->line0, + y_bot ? r->line0 : r->line1, + r->w_lores, r->hs); + if (++r->ystep >= r->vs) { + r->ystep = 0; + r->line0 = r->line1; + if (++r->ypos < z->img_comp[k].y) + r->line1 += z->img_comp[k].w2; + } + } + if (n >= 3) { + stbi_uc *y = coutput[0]; + if (z->s->img_n == 3) { + if (is_rgb) { + for (i=0; i < z->s->img_x; ++i) { + out[0] = y[i]; + out[1] = coutput[1][i]; + out[2] = coutput[2][i]; + out[3] = 255; + out += n; + } + } else { + z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, n); + } + } else if (z->s->img_n == 4) { + if (z->app14_color_transform == 0) { // CMYK + for (i=0; i < z->s->img_x; ++i) { + stbi_uc m = coutput[3][i]; + out[0] = stbi__blinn_8x8(coutput[0][i], m); + out[1] = stbi__blinn_8x8(coutput[1][i], m); + out[2] = stbi__blinn_8x8(coutput[2][i], m); + out[3] = 255; + out += n; + } + } else if (z->app14_color_transform == 2) { // YCCK + z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, n); + for (i=0; i < z->s->img_x; ++i) { + stbi_uc m = coutput[3][i]; + out[0] = stbi__blinn_8x8(255 - out[0], m); + out[1] = stbi__blinn_8x8(255 - out[1], m); + out[2] = stbi__blinn_8x8(255 - out[2], m); + out += n; + } + } else { // YCbCr + alpha? Ignore the fourth channel for now + z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, n); + } + } else + for (i=0; i < z->s->img_x; ++i) { + out[0] = out[1] = out[2] = y[i]; + out[3] = 255; // not used if n==3 + out += n; + } + } else { + if (is_rgb) { + if (n == 1) + for (i=0; i < z->s->img_x; ++i) + *out++ = stbi__compute_y(coutput[0][i], coutput[1][i], coutput[2][i]); + else { + for (i=0; i < z->s->img_x; ++i, out += 2) { + out[0] = stbi__compute_y(coutput[0][i], coutput[1][i], coutput[2][i]); + out[1] = 255; + } + } + } else if (z->s->img_n == 4 && z->app14_color_transform == 0) { + for (i=0; i < z->s->img_x; ++i) { + stbi_uc m = coutput[3][i]; + stbi_uc r = stbi__blinn_8x8(coutput[0][i], m); + stbi_uc g = stbi__blinn_8x8(coutput[1][i], m); + stbi_uc b = stbi__blinn_8x8(coutput[2][i], m); + out[0] = stbi__compute_y(r, g, b); + out[1] = 255; + out += n; + } + } else if (z->s->img_n == 4 && z->app14_color_transform == 2) { + for (i=0; i < z->s->img_x; ++i) { + out[0] = stbi__blinn_8x8(255 - coutput[0][i], coutput[3][i]); + out[1] = 255; + out += n; + } + } else { + stbi_uc *y = coutput[0]; + if (n == 1) + for (i=0; i < z->s->img_x; ++i) out[i] = y[i]; + else + for (i=0; i < z->s->img_x; ++i) { *out++ = y[i]; *out++ = 255; } + } + } + } + stbi__cleanup_jpeg(z); + *out_x = z->s->img_x; + *out_y = z->s->img_y; + if (comp) *comp = z->s->img_n >= 3 ? 3 : 1; // report original components, not output + return output; + } +} + +static void *stbi__jpeg_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) +{ + unsigned char* result; + stbi__jpeg* j = (stbi__jpeg*) stbi__malloc(sizeof(stbi__jpeg)); + if (!j) return stbi__errpuc("outofmem", "Out of memory"); + memset(j, 0, sizeof(stbi__jpeg)); + STBI_NOTUSED(ri); + j->s = s; + stbi__setup_jpeg(j); + result = load_jpeg_image(j, x,y,comp,req_comp); + STBI_FREE(j); + return result; +} + +static int stbi__jpeg_test(stbi__context *s) +{ + int r; + stbi__jpeg* j = (stbi__jpeg*)stbi__malloc(sizeof(stbi__jpeg)); + if (!j) return stbi__err("outofmem", "Out of memory"); + memset(j, 0, sizeof(stbi__jpeg)); + j->s = s; + stbi__setup_jpeg(j); + r = stbi__decode_jpeg_header(j, STBI__SCAN_type); + stbi__rewind(s); + STBI_FREE(j); + return r; +} + +static int stbi__jpeg_info_raw(stbi__jpeg *j, int *x, int *y, int *comp) +{ + if (!stbi__decode_jpeg_header(j, STBI__SCAN_header)) { + stbi__rewind( j->s ); + return 0; + } + if (x) *x = j->s->img_x; + if (y) *y = j->s->img_y; + if (comp) *comp = j->s->img_n >= 3 ? 3 : 1; + return 1; +} + +static int stbi__jpeg_info(stbi__context *s, int *x, int *y, int *comp) +{ + int result; + stbi__jpeg* j = (stbi__jpeg*) (stbi__malloc(sizeof(stbi__jpeg))); + if (!j) return stbi__err("outofmem", "Out of memory"); + memset(j, 0, sizeof(stbi__jpeg)); + j->s = s; + result = stbi__jpeg_info_raw(j, x, y, comp); + STBI_FREE(j); + return result; +} +#endif + +// public domain zlib decode v0.2 Sean Barrett 2006-11-18 +// simple implementation +// - all input must be provided in an upfront buffer +// - all output is written to a single output buffer (can malloc/realloc) +// performance +// - fast huffman + +#ifndef STBI_NO_ZLIB + +// fast-way is faster to check than jpeg huffman, but slow way is slower +#define STBI__ZFAST_BITS 9 // accelerate all cases in default tables +#define STBI__ZFAST_MASK ((1 << STBI__ZFAST_BITS) - 1) +#define STBI__ZNSYMS 288 // number of symbols in literal/length alphabet + +// zlib-style huffman encoding +// (jpegs packs from left, zlib from right, so can't share code) +typedef struct +{ + stbi__uint16 fast[1 << STBI__ZFAST_BITS]; + stbi__uint16 firstcode[16]; + int maxcode[17]; + stbi__uint16 firstsymbol[16]; + stbi_uc size[STBI__ZNSYMS]; + stbi__uint16 value[STBI__ZNSYMS]; +} stbi__zhuffman; + +stbi_inline static int stbi__bitreverse16(int n) +{ + n = ((n & 0xAAAA) >> 1) | ((n & 0x5555) << 1); + n = ((n & 0xCCCC) >> 2) | ((n & 0x3333) << 2); + n = ((n & 0xF0F0) >> 4) | ((n & 0x0F0F) << 4); + n = ((n & 0xFF00) >> 8) | ((n & 0x00FF) << 8); + return n; +} + +stbi_inline static int stbi__bit_reverse(int v, int bits) +{ + STBI_ASSERT(bits <= 16); + // to bit reverse n bits, reverse 16 and shift + // e.g. 11 bits, bit reverse and shift away 5 + return stbi__bitreverse16(v) >> (16-bits); +} + +static int stbi__zbuild_huffman(stbi__zhuffman *z, const stbi_uc *sizelist, int num) +{ + int i,k=0; + int code, next_code[16], sizes[17]; + + // DEFLATE spec for generating codes + memset(sizes, 0, sizeof(sizes)); + memset(z->fast, 0, sizeof(z->fast)); + for (i=0; i < num; ++i) + ++sizes[sizelist[i]]; + sizes[0] = 0; + for (i=1; i < 16; ++i) + if (sizes[i] > (1 << i)) + return stbi__err("bad sizes", "Corrupt PNG"); + code = 0; + for (i=1; i < 16; ++i) { + next_code[i] = code; + z->firstcode[i] = (stbi__uint16) code; + z->firstsymbol[i] = (stbi__uint16) k; + code = (code + sizes[i]); + if (sizes[i]) + if (code-1 >= (1 << i)) return stbi__err("bad codelengths","Corrupt PNG"); + z->maxcode[i] = code << (16-i); // preshift for inner loop + code <<= 1; + k += sizes[i]; + } + z->maxcode[16] = 0x10000; // sentinel + for (i=0; i < num; ++i) { + int s = sizelist[i]; + if (s) { + int c = next_code[s] - z->firstcode[s] + z->firstsymbol[s]; + stbi__uint16 fastv = (stbi__uint16) ((s << 9) | i); + z->size [c] = (stbi_uc ) s; + z->value[c] = (stbi__uint16) i; + if (s <= STBI__ZFAST_BITS) { + int j = stbi__bit_reverse(next_code[s],s); + while (j < (1 << STBI__ZFAST_BITS)) { + z->fast[j] = fastv; + j += (1 << s); + } + } + ++next_code[s]; + } + } + return 1; +} + +// zlib-from-memory implementation for PNG reading +// because PNG allows splitting the zlib stream arbitrarily, +// and it's annoying structurally to have PNG call ZLIB call PNG, +// we require PNG read all the IDATs and combine them into a single +// memory buffer + +typedef struct +{ + stbi_uc *zbuffer, *zbuffer_end; + int num_bits; + int hit_zeof_once; + stbi__uint32 code_buffer; + + char *zout; + char *zout_start; + char *zout_end; + int z_expandable; + + stbi__zhuffman z_length, z_distance; +} stbi__zbuf; + +stbi_inline static int stbi__zeof(stbi__zbuf *z) +{ + return (z->zbuffer >= z->zbuffer_end); +} + +stbi_inline static stbi_uc stbi__zget8(stbi__zbuf *z) +{ + return stbi__zeof(z) ? 0 : *z->zbuffer++; +} + +static void stbi__fill_bits(stbi__zbuf *z) +{ + do { + if (z->code_buffer >= (1U << z->num_bits)) { + z->zbuffer = z->zbuffer_end; /* treat this as EOF so we fail. */ + return; + } + z->code_buffer |= (unsigned int) stbi__zget8(z) << z->num_bits; + z->num_bits += 8; + } while (z->num_bits <= 24); +} + +stbi_inline static unsigned int stbi__zreceive(stbi__zbuf *z, int n) +{ + unsigned int k; + if (z->num_bits < n) stbi__fill_bits(z); + k = z->code_buffer & ((1 << n) - 1); + z->code_buffer >>= n; + z->num_bits -= n; + return k; +} + +static int stbi__zhuffman_decode_slowpath(stbi__zbuf *a, stbi__zhuffman *z) +{ + int b,s,k; + // not resolved by fast table, so compute it the slow way + // use jpeg approach, which requires MSbits at top + k = stbi__bit_reverse(a->code_buffer, 16); + for (s=STBI__ZFAST_BITS+1; ; ++s) + if (k < z->maxcode[s]) + break; + if (s >= 16) return -1; // invalid code! + // code size is s, so: + b = (k >> (16-s)) - z->firstcode[s] + z->firstsymbol[s]; + if (b >= STBI__ZNSYMS) return -1; // some data was corrupt somewhere! + if (z->size[b] != s) return -1; // was originally an assert, but report failure instead. + a->code_buffer >>= s; + a->num_bits -= s; + return z->value[b]; +} + +stbi_inline static int stbi__zhuffman_decode(stbi__zbuf *a, stbi__zhuffman *z) +{ + int b,s; + if (a->num_bits < 16) { + if (stbi__zeof(a)) { + if (!a->hit_zeof_once) { + // This is the first time we hit eof, insert 16 extra padding btis + // to allow us to keep going; if we actually consume any of them + // though, that is invalid data. This is caught later. + a->hit_zeof_once = 1; + a->num_bits += 16; // add 16 implicit zero bits + } else { + // We already inserted our extra 16 padding bits and are again + // out, this stream is actually prematurely terminated. + return -1; + } + } else { + stbi__fill_bits(a); + } + } + b = z->fast[a->code_buffer & STBI__ZFAST_MASK]; + if (b) { + s = b >> 9; + a->code_buffer >>= s; + a->num_bits -= s; + return b & 511; + } + return stbi__zhuffman_decode_slowpath(a, z); +} + +static int stbi__zexpand(stbi__zbuf *z, char *zout, int n) // need to make room for n bytes +{ + char *q; + unsigned int cur, limit, old_limit; + z->zout = zout; + if (!z->z_expandable) return stbi__err("output buffer limit","Corrupt PNG"); + cur = (unsigned int) (z->zout - z->zout_start); + limit = old_limit = (unsigned) (z->zout_end - z->zout_start); + if (UINT_MAX - cur < (unsigned) n) return stbi__err("outofmem", "Out of memory"); + while (cur + n > limit) { + if(limit > UINT_MAX / 2) return stbi__err("outofmem", "Out of memory"); + limit *= 2; + } + q = (char *) STBI_REALLOC_SIZED(z->zout_start, old_limit, limit); + STBI_NOTUSED(old_limit); + if (q == NULL) return stbi__err("outofmem", "Out of memory"); + z->zout_start = q; + z->zout = q + cur; + z->zout_end = q + limit; + return 1; +} + +static const int stbi__zlength_base[31] = { + 3,4,5,6,7,8,9,10,11,13, + 15,17,19,23,27,31,35,43,51,59, + 67,83,99,115,131,163,195,227,258,0,0 }; + +static const int stbi__zlength_extra[31]= +{ 0,0,0,0,0,0,0,0,1,1,1,1,2,2,2,2,3,3,3,3,4,4,4,4,5,5,5,5,0,0,0 }; + +static const int stbi__zdist_base[32] = { 1,2,3,4,5,7,9,13,17,25,33,49,65,97,129,193, +257,385,513,769,1025,1537,2049,3073,4097,6145,8193,12289,16385,24577,0,0}; + +static const int stbi__zdist_extra[32] = +{ 0,0,0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,12,12,13,13}; + +static int stbi__parse_huffman_block(stbi__zbuf *a) +{ + char *zout = a->zout; + for(;;) { + int z = stbi__zhuffman_decode(a, &a->z_length); + if (z < 256) { + if (z < 0) return stbi__err("bad huffman code","Corrupt PNG"); // error in huffman codes + if (zout >= a->zout_end) { + if (!stbi__zexpand(a, zout, 1)) return 0; + zout = a->zout; + } + *zout++ = (char) z; + } else { + stbi_uc *p; + int len,dist; + if (z == 256) { + a->zout = zout; + if (a->hit_zeof_once && a->num_bits < 16) { + // The first time we hit zeof, we inserted 16 extra zero bits into our bit + // buffer so the decoder can just do its speculative decoding. But if we + // actually consumed any of those bits (which is the case when num_bits < 16), + // the stream actually read past the end so it is malformed. + return stbi__err("unexpected end","Corrupt PNG"); + } + return 1; + } + if (z >= 286) return stbi__err("bad huffman code","Corrupt PNG"); // per DEFLATE, length codes 286 and 287 must not appear in compressed data + z -= 257; + len = stbi__zlength_base[z]; + if (stbi__zlength_extra[z]) len += stbi__zreceive(a, stbi__zlength_extra[z]); + z = stbi__zhuffman_decode(a, &a->z_distance); + if (z < 0 || z >= 30) return stbi__err("bad huffman code","Corrupt PNG"); // per DEFLATE, distance codes 30 and 31 must not appear in compressed data + dist = stbi__zdist_base[z]; + if (stbi__zdist_extra[z]) dist += stbi__zreceive(a, stbi__zdist_extra[z]); + if (zout - a->zout_start < dist) return stbi__err("bad dist","Corrupt PNG"); + if (len > a->zout_end - zout) { + if (!stbi__zexpand(a, zout, len)) return 0; + zout = a->zout; + } + p = (stbi_uc *) (zout - dist); + if (dist == 1) { // run of one byte; common in images. + stbi_uc v = *p; + if (len) { do *zout++ = v; while (--len); } + } else { + if (len) { do *zout++ = *p++; while (--len); } + } + } + } +} + +static int stbi__compute_huffman_codes(stbi__zbuf *a) +{ + static const stbi_uc length_dezigzag[19] = { 16,17,18,0,8,7,9,6,10,5,11,4,12,3,13,2,14,1,15 }; + stbi__zhuffman z_codelength; + stbi_uc lencodes[286+32+137];//padding for maximum single op + stbi_uc codelength_sizes[19]; + int i,n; + + int hlit = stbi__zreceive(a,5) + 257; + int hdist = stbi__zreceive(a,5) + 1; + int hclen = stbi__zreceive(a,4) + 4; + int ntot = hlit + hdist; + + memset(codelength_sizes, 0, sizeof(codelength_sizes)); + for (i=0; i < hclen; ++i) { + int s = stbi__zreceive(a,3); + codelength_sizes[length_dezigzag[i]] = (stbi_uc) s; + } + if (!stbi__zbuild_huffman(&z_codelength, codelength_sizes, 19)) return 0; + + n = 0; + while (n < ntot) { + int c = stbi__zhuffman_decode(a, &z_codelength); + if (c < 0 || c >= 19) return stbi__err("bad codelengths", "Corrupt PNG"); + if (c < 16) + lencodes[n++] = (stbi_uc) c; + else { + stbi_uc fill = 0; + if (c == 16) { + c = stbi__zreceive(a,2)+3; + if (n == 0) return stbi__err("bad codelengths", "Corrupt PNG"); + fill = lencodes[n-1]; + } else if (c == 17) { + c = stbi__zreceive(a,3)+3; + } else if (c == 18) { + c = stbi__zreceive(a,7)+11; + } else { + return stbi__err("bad codelengths", "Corrupt PNG"); + } + if (ntot - n < c) return stbi__err("bad codelengths", "Corrupt PNG"); + memset(lencodes+n, fill, c); + n += c; + } + } + if (n != ntot) return stbi__err("bad codelengths","Corrupt PNG"); + if (!stbi__zbuild_huffman(&a->z_length, lencodes, hlit)) return 0; + if (!stbi__zbuild_huffman(&a->z_distance, lencodes+hlit, hdist)) return 0; + return 1; +} + +static int stbi__parse_uncompressed_block(stbi__zbuf *a) +{ + stbi_uc header[4]; + int len,nlen,k; + if (a->num_bits & 7) + stbi__zreceive(a, a->num_bits & 7); // discard + // drain the bit-packed data into header + k = 0; + while (a->num_bits > 0) { + header[k++] = (stbi_uc) (a->code_buffer & 255); // suppress MSVC run-time check + a->code_buffer >>= 8; + a->num_bits -= 8; + } + if (a->num_bits < 0) return stbi__err("zlib corrupt","Corrupt PNG"); + // now fill header the normal way + while (k < 4) + header[k++] = stbi__zget8(a); + len = header[1] * 256 + header[0]; + nlen = header[3] * 256 + header[2]; + if (nlen != (len ^ 0xffff)) return stbi__err("zlib corrupt","Corrupt PNG"); + if (a->zbuffer + len > a->zbuffer_end) return stbi__err("read past buffer","Corrupt PNG"); + if (a->zout + len > a->zout_end) + if (!stbi__zexpand(a, a->zout, len)) return 0; + memcpy(a->zout, a->zbuffer, len); + a->zbuffer += len; + a->zout += len; + return 1; +} + +static int stbi__parse_zlib_header(stbi__zbuf *a) +{ + int cmf = stbi__zget8(a); + int cm = cmf & 15; + /* int cinfo = cmf >> 4; */ + int flg = stbi__zget8(a); + if (stbi__zeof(a)) return stbi__err("bad zlib header","Corrupt PNG"); // zlib spec + if ((cmf*256+flg) % 31 != 0) return stbi__err("bad zlib header","Corrupt PNG"); // zlib spec + if (flg & 32) return stbi__err("no preset dict","Corrupt PNG"); // preset dictionary not allowed in png + if (cm != 8) return stbi__err("bad compression","Corrupt PNG"); // DEFLATE required for png + // window = 1 << (8 + cinfo)... but who cares, we fully buffer output + return 1; +} + +static const stbi_uc stbi__zdefault_length[STBI__ZNSYMS] = +{ + 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, + 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, + 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, + 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, + 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, + 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, + 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, + 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, + 7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7, 7,7,7,7,7,7,7,7,8,8,8,8,8,8,8,8 +}; +static const stbi_uc stbi__zdefault_distance[32] = +{ + 5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5 +}; +/* +Init algorithm: +{ + int i; // use <= to match clearly with spec + for (i=0; i <= 143; ++i) stbi__zdefault_length[i] = 8; + for ( ; i <= 255; ++i) stbi__zdefault_length[i] = 9; + for ( ; i <= 279; ++i) stbi__zdefault_length[i] = 7; + for ( ; i <= 287; ++i) stbi__zdefault_length[i] = 8; + + for (i=0; i <= 31; ++i) stbi__zdefault_distance[i] = 5; +} +*/ + +static int stbi__parse_zlib(stbi__zbuf *a, int parse_header) +{ + int final, type; + if (parse_header) + if (!stbi__parse_zlib_header(a)) return 0; + a->num_bits = 0; + a->code_buffer = 0; + a->hit_zeof_once = 0; + do { + final = stbi__zreceive(a,1); + type = stbi__zreceive(a,2); + if (type == 0) { + if (!stbi__parse_uncompressed_block(a)) return 0; + } else if (type == 3) { + return 0; + } else { + if (type == 1) { + // use fixed code lengths + if (!stbi__zbuild_huffman(&a->z_length , stbi__zdefault_length , STBI__ZNSYMS)) return 0; + if (!stbi__zbuild_huffman(&a->z_distance, stbi__zdefault_distance, 32)) return 0; + } else { + if (!stbi__compute_huffman_codes(a)) return 0; + } + if (!stbi__parse_huffman_block(a)) return 0; + } + } while (!final); + return 1; +} + +static int stbi__do_zlib(stbi__zbuf *a, char *obuf, int olen, int exp, int parse_header) +{ + a->zout_start = obuf; + a->zout = obuf; + a->zout_end = obuf + olen; + a->z_expandable = exp; + + return stbi__parse_zlib(a, parse_header); +} + +STBIDEF char *stbi_zlib_decode_malloc_guesssize(const char *buffer, int len, int initial_size, int *outlen) +{ + stbi__zbuf a; + char *p = (char *) stbi__malloc(initial_size); + if (p == NULL) return NULL; + a.zbuffer = (stbi_uc *) buffer; + a.zbuffer_end = (stbi_uc *) buffer + len; + if (stbi__do_zlib(&a, p, initial_size, 1, 1)) { + if (outlen) *outlen = (int) (a.zout - a.zout_start); + return a.zout_start; + } else { + STBI_FREE(a.zout_start); + return NULL; + } +} + +STBIDEF char *stbi_zlib_decode_malloc(char const *buffer, int len, int *outlen) +{ + return stbi_zlib_decode_malloc_guesssize(buffer, len, 16384, outlen); +} + +STBIDEF char *stbi_zlib_decode_malloc_guesssize_headerflag(const char *buffer, int len, int initial_size, int *outlen, int parse_header) +{ + stbi__zbuf a; + char *p = (char *) stbi__malloc(initial_size); + if (p == NULL) return NULL; + a.zbuffer = (stbi_uc *) buffer; + a.zbuffer_end = (stbi_uc *) buffer + len; + if (stbi__do_zlib(&a, p, initial_size, 1, parse_header)) { + if (outlen) *outlen = (int) (a.zout - a.zout_start); + return a.zout_start; + } else { + STBI_FREE(a.zout_start); + return NULL; + } +} + +STBIDEF int stbi_zlib_decode_buffer(char *obuffer, int olen, char const *ibuffer, int ilen) +{ + stbi__zbuf a; + a.zbuffer = (stbi_uc *) ibuffer; + a.zbuffer_end = (stbi_uc *) ibuffer + ilen; + if (stbi__do_zlib(&a, obuffer, olen, 0, 1)) + return (int) (a.zout - a.zout_start); + else + return -1; +} + +STBIDEF char *stbi_zlib_decode_noheader_malloc(char const *buffer, int len, int *outlen) +{ + stbi__zbuf a; + char *p = (char *) stbi__malloc(16384); + if (p == NULL) return NULL; + a.zbuffer = (stbi_uc *) buffer; + a.zbuffer_end = (stbi_uc *) buffer+len; + if (stbi__do_zlib(&a, p, 16384, 1, 0)) { + if (outlen) *outlen = (int) (a.zout - a.zout_start); + return a.zout_start; + } else { + STBI_FREE(a.zout_start); + return NULL; + } +} + +STBIDEF int stbi_zlib_decode_noheader_buffer(char *obuffer, int olen, const char *ibuffer, int ilen) +{ + stbi__zbuf a; + a.zbuffer = (stbi_uc *) ibuffer; + a.zbuffer_end = (stbi_uc *) ibuffer + ilen; + if (stbi__do_zlib(&a, obuffer, olen, 0, 0)) + return (int) (a.zout - a.zout_start); + else + return -1; +} +#endif + +// public domain "baseline" PNG decoder v0.10 Sean Barrett 2006-11-18 +// simple implementation +// - only 8-bit samples +// - no CRC checking +// - allocates lots of intermediate memory +// - avoids problem of streaming data between subsystems +// - avoids explicit window management +// performance +// - uses stb_zlib, a PD zlib implementation with fast huffman decoding + +#ifndef STBI_NO_PNG +typedef struct +{ + stbi__uint32 length; + stbi__uint32 type; +} stbi__pngchunk; + +static stbi__pngchunk stbi__get_chunk_header(stbi__context *s) +{ + stbi__pngchunk c; + c.length = stbi__get32be(s); + c.type = stbi__get32be(s); + return c; +} + +static int stbi__check_png_header(stbi__context *s) +{ + static const stbi_uc png_sig[8] = { 137,80,78,71,13,10,26,10 }; + int i; + for (i=0; i < 8; ++i) + if (stbi__get8(s) != png_sig[i]) return stbi__err("bad png sig","Not a PNG"); + return 1; +} + +typedef struct +{ + stbi__context *s; + stbi_uc *idata, *expanded, *out; + int depth; +} stbi__png; + + +enum { + STBI__F_none=0, + STBI__F_sub=1, + STBI__F_up=2, + STBI__F_avg=3, + STBI__F_paeth=4, + // synthetic filter used for first scanline to avoid needing a dummy row of 0s + STBI__F_avg_first +}; + +static stbi_uc first_row_filter[5] = +{ + STBI__F_none, + STBI__F_sub, + STBI__F_none, + STBI__F_avg_first, + STBI__F_sub // Paeth with b=c=0 turns out to be equivalent to sub +}; + +static int stbi__paeth(int a, int b, int c) +{ + // This formulation looks very different from the reference in the PNG spec, but is + // actually equivalent and has favorable data dependencies and admits straightforward + // generation of branch-free code, which helps performance significantly. + int thresh = c*3 - (a + b); + int lo = a < b ? a : b; + int hi = a < b ? b : a; + int t0 = (hi <= thresh) ? lo : c; + int t1 = (thresh <= lo) ? hi : t0; + return t1; +} + +static const stbi_uc stbi__depth_scale_table[9] = { 0, 0xff, 0x55, 0, 0x11, 0,0,0, 0x01 }; + +// adds an extra all-255 alpha channel +// dest == src is legal +// img_n must be 1 or 3 +static void stbi__create_png_alpha_expand8(stbi_uc *dest, stbi_uc *src, stbi__uint32 x, int img_n) +{ + int i; + // must process data backwards since we allow dest==src + if (img_n == 1) { + for (i=x-1; i >= 0; --i) { + dest[i*2+1] = 255; + dest[i*2+0] = src[i]; + } + } else { + STBI_ASSERT(img_n == 3); + for (i=x-1; i >= 0; --i) { + dest[i*4+3] = 255; + dest[i*4+2] = src[i*3+2]; + dest[i*4+1] = src[i*3+1]; + dest[i*4+0] = src[i*3+0]; + } + } +} + +// create the png data from post-deflated data +static int stbi__create_png_image_raw(stbi__png *a, stbi_uc *raw, stbi__uint32 raw_len, int out_n, stbi__uint32 x, stbi__uint32 y, int depth, int color) +{ + int bytes = (depth == 16 ? 2 : 1); + stbi__context *s = a->s; + stbi__uint32 i,j,stride = x*out_n*bytes; + stbi__uint32 img_len, img_width_bytes; + stbi_uc *filter_buf; + int all_ok = 1; + int k; + int img_n = s->img_n; // copy it into a local for later + + int output_bytes = out_n*bytes; + int filter_bytes = img_n*bytes; + int width = x; + + STBI_ASSERT(out_n == s->img_n || out_n == s->img_n+1); + a->out = (stbi_uc *) stbi__malloc_mad3(x, y, output_bytes, 0); // extra bytes to write off the end into + if (!a->out) return stbi__err("outofmem", "Out of memory"); + + // note: error exits here don't need to clean up a->out individually, + // stbi__do_png always does on error. + if (!stbi__mad3sizes_valid(img_n, x, depth, 7)) return stbi__err("too large", "Corrupt PNG"); + img_width_bytes = (((img_n * x * depth) + 7) >> 3); + if (!stbi__mad2sizes_valid(img_width_bytes, y, img_width_bytes)) return stbi__err("too large", "Corrupt PNG"); + img_len = (img_width_bytes + 1) * y; + + // we used to check for exact match between raw_len and img_len on non-interlaced PNGs, + // but issue #276 reported a PNG in the wild that had extra data at the end (all zeros), + // so just check for raw_len < img_len always. + if (raw_len < img_len) return stbi__err("not enough pixels","Corrupt PNG"); + + // Allocate two scan lines worth of filter workspace buffer. + filter_buf = (stbi_uc *) stbi__malloc_mad2(img_width_bytes, 2, 0); + if (!filter_buf) return stbi__err("outofmem", "Out of memory"); + + // Filtering for low-bit-depth images + if (depth < 8) { + filter_bytes = 1; + width = img_width_bytes; + } + + for (j=0; j < y; ++j) { + // cur/prior filter buffers alternate + stbi_uc *cur = filter_buf + (j & 1)*img_width_bytes; + stbi_uc *prior = filter_buf + (~j & 1)*img_width_bytes; + stbi_uc *dest = a->out + stride*j; + int nk = width * filter_bytes; + int filter = *raw++; + + // check filter type + if (filter > 4) { + all_ok = stbi__err("invalid filter","Corrupt PNG"); + break; + } + + // if first row, use special filter that doesn't sample previous row + if (j == 0) filter = first_row_filter[filter]; + + // perform actual filtering + switch (filter) { + case STBI__F_none: + memcpy(cur, raw, nk); + break; + case STBI__F_sub: + memcpy(cur, raw, filter_bytes); + for (k = filter_bytes; k < nk; ++k) + cur[k] = STBI__BYTECAST(raw[k] + cur[k-filter_bytes]); + break; + case STBI__F_up: + for (k = 0; k < nk; ++k) + cur[k] = STBI__BYTECAST(raw[k] + prior[k]); + break; + case STBI__F_avg: + for (k = 0; k < filter_bytes; ++k) + cur[k] = STBI__BYTECAST(raw[k] + (prior[k]>>1)); + for (k = filter_bytes; k < nk; ++k) + cur[k] = STBI__BYTECAST(raw[k] + ((prior[k] + cur[k-filter_bytes])>>1)); + break; + case STBI__F_paeth: + for (k = 0; k < filter_bytes; ++k) + cur[k] = STBI__BYTECAST(raw[k] + prior[k]); // prior[k] == stbi__paeth(0,prior[k],0) + for (k = filter_bytes; k < nk; ++k) + cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k-filter_bytes], prior[k], prior[k-filter_bytes])); + break; + case STBI__F_avg_first: + memcpy(cur, raw, filter_bytes); + for (k = filter_bytes; k < nk; ++k) + cur[k] = STBI__BYTECAST(raw[k] + (cur[k-filter_bytes] >> 1)); + break; + } + + raw += nk; + + // expand decoded bits in cur to dest, also adding an extra alpha channel if desired + if (depth < 8) { + stbi_uc scale = (color == 0) ? stbi__depth_scale_table[depth] : 1; // scale grayscale values to 0..255 range + stbi_uc *in = cur; + stbi_uc *out = dest; + stbi_uc inb = 0; + stbi__uint32 nsmp = x*img_n; + + // expand bits to bytes first + if (depth == 4) { + for (i=0; i < nsmp; ++i) { + if ((i & 1) == 0) inb = *in++; + *out++ = scale * (inb >> 4); + inb <<= 4; + } + } else if (depth == 2) { + for (i=0; i < nsmp; ++i) { + if ((i & 3) == 0) inb = *in++; + *out++ = scale * (inb >> 6); + inb <<= 2; + } + } else { + STBI_ASSERT(depth == 1); + for (i=0; i < nsmp; ++i) { + if ((i & 7) == 0) inb = *in++; + *out++ = scale * (inb >> 7); + inb <<= 1; + } + } + + // insert alpha=255 values if desired + if (img_n != out_n) + stbi__create_png_alpha_expand8(dest, dest, x, img_n); + } else if (depth == 8) { + if (img_n == out_n) + memcpy(dest, cur, x*img_n); + else + stbi__create_png_alpha_expand8(dest, cur, x, img_n); + } else if (depth == 16) { + // convert the image data from big-endian to platform-native + stbi__uint16 *dest16 = (stbi__uint16*)dest; + stbi__uint32 nsmp = x*img_n; + + if (img_n == out_n) { + for (i = 0; i < nsmp; ++i, ++dest16, cur += 2) + *dest16 = (cur[0] << 8) | cur[1]; + } else { + STBI_ASSERT(img_n+1 == out_n); + if (img_n == 1) { + for (i = 0; i < x; ++i, dest16 += 2, cur += 2) { + dest16[0] = (cur[0] << 8) | cur[1]; + dest16[1] = 0xffff; + } + } else { + STBI_ASSERT(img_n == 3); + for (i = 0; i < x; ++i, dest16 += 4, cur += 6) { + dest16[0] = (cur[0] << 8) | cur[1]; + dest16[1] = (cur[2] << 8) | cur[3]; + dest16[2] = (cur[4] << 8) | cur[5]; + dest16[3] = 0xffff; + } + } + } + } + } + + STBI_FREE(filter_buf); + if (!all_ok) return 0; + + return 1; +} + +static int stbi__create_png_image(stbi__png *a, stbi_uc *image_data, stbi__uint32 image_data_len, int out_n, int depth, int color, int interlaced) +{ + int bytes = (depth == 16 ? 2 : 1); + int out_bytes = out_n * bytes; + stbi_uc *final; + int p; + if (!interlaced) + return stbi__create_png_image_raw(a, image_data, image_data_len, out_n, a->s->img_x, a->s->img_y, depth, color); + + // de-interlacing + final = (stbi_uc *) stbi__malloc_mad3(a->s->img_x, a->s->img_y, out_bytes, 0); + if (!final) return stbi__err("outofmem", "Out of memory"); + for (p=0; p < 7; ++p) { + int xorig[] = { 0,4,0,2,0,1,0 }; + int yorig[] = { 0,0,4,0,2,0,1 }; + int xspc[] = { 8,8,4,4,2,2,1 }; + int yspc[] = { 8,8,8,4,4,2,2 }; + int i,j,x,y; + // pass1_x[4] = 0, pass1_x[5] = 1, pass1_x[12] = 1 + x = (a->s->img_x - xorig[p] + xspc[p]-1) / xspc[p]; + y = (a->s->img_y - yorig[p] + yspc[p]-1) / yspc[p]; + if (x && y) { + stbi__uint32 img_len = ((((a->s->img_n * x * depth) + 7) >> 3) + 1) * y; + if (!stbi__create_png_image_raw(a, image_data, image_data_len, out_n, x, y, depth, color)) { + STBI_FREE(final); + return 0; + } + for (j=0; j < y; ++j) { + for (i=0; i < x; ++i) { + int out_y = j*yspc[p]+yorig[p]; + int out_x = i*xspc[p]+xorig[p]; + memcpy(final + out_y*a->s->img_x*out_bytes + out_x*out_bytes, + a->out + (j*x+i)*out_bytes, out_bytes); + } + } + STBI_FREE(a->out); + image_data += img_len; + image_data_len -= img_len; + } + } + a->out = final; + + return 1; +} + +static int stbi__compute_transparency(stbi__png *z, stbi_uc tc[3], int out_n) +{ + stbi__context *s = z->s; + stbi__uint32 i, pixel_count = s->img_x * s->img_y; + stbi_uc *p = z->out; + + // compute color-based transparency, assuming we've + // already got 255 as the alpha value in the output + STBI_ASSERT(out_n == 2 || out_n == 4); + + if (out_n == 2) { + for (i=0; i < pixel_count; ++i) { + p[1] = (p[0] == tc[0] ? 0 : 255); + p += 2; + } + } else { + for (i=0; i < pixel_count; ++i) { + if (p[0] == tc[0] && p[1] == tc[1] && p[2] == tc[2]) + p[3] = 0; + p += 4; + } + } + return 1; +} + +static int stbi__compute_transparency16(stbi__png *z, stbi__uint16 tc[3], int out_n) +{ + stbi__context *s = z->s; + stbi__uint32 i, pixel_count = s->img_x * s->img_y; + stbi__uint16 *p = (stbi__uint16*) z->out; + + // compute color-based transparency, assuming we've + // already got 65535 as the alpha value in the output + STBI_ASSERT(out_n == 2 || out_n == 4); + + if (out_n == 2) { + for (i = 0; i < pixel_count; ++i) { + p[1] = (p[0] == tc[0] ? 0 : 65535); + p += 2; + } + } else { + for (i = 0; i < pixel_count; ++i) { + if (p[0] == tc[0] && p[1] == tc[1] && p[2] == tc[2]) + p[3] = 0; + p += 4; + } + } + return 1; +} + +static int stbi__expand_png_palette(stbi__png *a, stbi_uc *palette, int len, int pal_img_n) +{ + stbi__uint32 i, pixel_count = a->s->img_x * a->s->img_y; + stbi_uc *p, *temp_out, *orig = a->out; + + p = (stbi_uc *) stbi__malloc_mad2(pixel_count, pal_img_n, 0); + if (p == NULL) return stbi__err("outofmem", "Out of memory"); + + // between here and free(out) below, exitting would leak + temp_out = p; + + if (pal_img_n == 3) { + for (i=0; i < pixel_count; ++i) { + int n = orig[i]*4; + p[0] = palette[n ]; + p[1] = palette[n+1]; + p[2] = palette[n+2]; + p += 3; + } + } else { + for (i=0; i < pixel_count; ++i) { + int n = orig[i]*4; + p[0] = palette[n ]; + p[1] = palette[n+1]; + p[2] = palette[n+2]; + p[3] = palette[n+3]; + p += 4; + } + } + STBI_FREE(a->out); + a->out = temp_out; + + STBI_NOTUSED(len); + + return 1; +} + +static int stbi__unpremultiply_on_load_global = 0; +static int stbi__de_iphone_flag_global = 0; + +STBIDEF void stbi_set_unpremultiply_on_load(int flag_true_if_should_unpremultiply) +{ + stbi__unpremultiply_on_load_global = flag_true_if_should_unpremultiply; +} + +STBIDEF void stbi_convert_iphone_png_to_rgb(int flag_true_if_should_convert) +{ + stbi__de_iphone_flag_global = flag_true_if_should_convert; +} + +#ifndef STBI_THREAD_LOCAL +#define stbi__unpremultiply_on_load stbi__unpremultiply_on_load_global +#define stbi__de_iphone_flag stbi__de_iphone_flag_global +#else +static STBI_THREAD_LOCAL int stbi__unpremultiply_on_load_local, stbi__unpremultiply_on_load_set; +static STBI_THREAD_LOCAL int stbi__de_iphone_flag_local, stbi__de_iphone_flag_set; + +STBIDEF void stbi_set_unpremultiply_on_load_thread(int flag_true_if_should_unpremultiply) +{ + stbi__unpremultiply_on_load_local = flag_true_if_should_unpremultiply; + stbi__unpremultiply_on_load_set = 1; +} + +STBIDEF void stbi_convert_iphone_png_to_rgb_thread(int flag_true_if_should_convert) +{ + stbi__de_iphone_flag_local = flag_true_if_should_convert; + stbi__de_iphone_flag_set = 1; +} + +#define stbi__unpremultiply_on_load (stbi__unpremultiply_on_load_set \ + ? stbi__unpremultiply_on_load_local \ + : stbi__unpremultiply_on_load_global) +#define stbi__de_iphone_flag (stbi__de_iphone_flag_set \ + ? stbi__de_iphone_flag_local \ + : stbi__de_iphone_flag_global) +#endif // STBI_THREAD_LOCAL + +static void stbi__de_iphone(stbi__png *z) +{ + stbi__context *s = z->s; + stbi__uint32 i, pixel_count = s->img_x * s->img_y; + stbi_uc *p = z->out; + + if (s->img_out_n == 3) { // convert bgr to rgb + for (i=0; i < pixel_count; ++i) { + stbi_uc t = p[0]; + p[0] = p[2]; + p[2] = t; + p += 3; + } + } else { + STBI_ASSERT(s->img_out_n == 4); + if (stbi__unpremultiply_on_load) { + // convert bgr to rgb and unpremultiply + for (i=0; i < pixel_count; ++i) { + stbi_uc a = p[3]; + stbi_uc t = p[0]; + if (a) { + stbi_uc half = a / 2; + p[0] = (p[2] * 255 + half) / a; + p[1] = (p[1] * 255 + half) / a; + p[2] = ( t * 255 + half) / a; + } else { + p[0] = p[2]; + p[2] = t; + } + p += 4; + } + } else { + // convert bgr to rgb + for (i=0; i < pixel_count; ++i) { + stbi_uc t = p[0]; + p[0] = p[2]; + p[2] = t; + p += 4; + } + } + } +} + +#define STBI__PNG_TYPE(a,b,c,d) (((unsigned) (a) << 24) + ((unsigned) (b) << 16) + ((unsigned) (c) << 8) + (unsigned) (d)) + +static int stbi__parse_png_file(stbi__png *z, int scan, int req_comp) +{ + stbi_uc palette[1024], pal_img_n=0; + stbi_uc has_trans=0, tc[3]={0}; + stbi__uint16 tc16[3]; + stbi__uint32 ioff=0, idata_limit=0, i, pal_len=0; + int first=1,k,interlace=0, color=0, is_iphone=0; + stbi__context *s = z->s; + + z->expanded = NULL; + z->idata = NULL; + z->out = NULL; + + if (!stbi__check_png_header(s)) return 0; + + if (scan == STBI__SCAN_type) return 1; + + for (;;) { + stbi__pngchunk c = stbi__get_chunk_header(s); + switch (c.type) { + case STBI__PNG_TYPE('C','g','B','I'): + is_iphone = 1; + stbi__skip(s, c.length); + break; + case STBI__PNG_TYPE('I','H','D','R'): { + int comp,filter; + if (!first) return stbi__err("multiple IHDR","Corrupt PNG"); + first = 0; + if (c.length != 13) return stbi__err("bad IHDR len","Corrupt PNG"); + s->img_x = stbi__get32be(s); + s->img_y = stbi__get32be(s); + if (s->img_y > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); + if (s->img_x > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); + z->depth = stbi__get8(s); if (z->depth != 1 && z->depth != 2 && z->depth != 4 && z->depth != 8 && z->depth != 16) return stbi__err("1/2/4/8/16-bit only","PNG not supported: 1/2/4/8/16-bit only"); + color = stbi__get8(s); if (color > 6) return stbi__err("bad ctype","Corrupt PNG"); + if (color == 3 && z->depth == 16) return stbi__err("bad ctype","Corrupt PNG"); + if (color == 3) pal_img_n = 3; else if (color & 1) return stbi__err("bad ctype","Corrupt PNG"); + comp = stbi__get8(s); if (comp) return stbi__err("bad comp method","Corrupt PNG"); + filter= stbi__get8(s); if (filter) return stbi__err("bad filter method","Corrupt PNG"); + interlace = stbi__get8(s); if (interlace>1) return stbi__err("bad interlace method","Corrupt PNG"); + if (!s->img_x || !s->img_y) return stbi__err("0-pixel image","Corrupt PNG"); + if (!pal_img_n) { + s->img_n = (color & 2 ? 3 : 1) + (color & 4 ? 1 : 0); + if ((1 << 30) / s->img_x / s->img_n < s->img_y) return stbi__err("too large", "Image too large to decode"); + } else { + // if paletted, then pal_n is our final components, and + // img_n is # components to decompress/filter. + s->img_n = 1; + if ((1 << 30) / s->img_x / 4 < s->img_y) return stbi__err("too large","Corrupt PNG"); + } + // even with SCAN_header, have to scan to see if we have a tRNS + break; + } + + case STBI__PNG_TYPE('P','L','T','E'): { + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (c.length > 256*3) return stbi__err("invalid PLTE","Corrupt PNG"); + pal_len = c.length / 3; + if (pal_len * 3 != c.length) return stbi__err("invalid PLTE","Corrupt PNG"); + for (i=0; i < pal_len; ++i) { + palette[i*4+0] = stbi__get8(s); + palette[i*4+1] = stbi__get8(s); + palette[i*4+2] = stbi__get8(s); + palette[i*4+3] = 255; + } + break; + } + + case STBI__PNG_TYPE('t','R','N','S'): { + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (z->idata) return stbi__err("tRNS after IDAT","Corrupt PNG"); + if (pal_img_n) { + if (scan == STBI__SCAN_header) { s->img_n = 4; return 1; } + if (pal_len == 0) return stbi__err("tRNS before PLTE","Corrupt PNG"); + if (c.length > pal_len) return stbi__err("bad tRNS len","Corrupt PNG"); + pal_img_n = 4; + for (i=0; i < c.length; ++i) + palette[i*4+3] = stbi__get8(s); + } else { + if (!(s->img_n & 1)) return stbi__err("tRNS with alpha","Corrupt PNG"); + if (c.length != (stbi__uint32) s->img_n*2) return stbi__err("bad tRNS len","Corrupt PNG"); + has_trans = 1; + // non-paletted with tRNS = constant alpha. if header-scanning, we can stop now. + if (scan == STBI__SCAN_header) { ++s->img_n; return 1; } + if (z->depth == 16) { + for (k = 0; k < s->img_n && k < 3; ++k) // extra loop test to suppress false GCC warning + tc16[k] = (stbi__uint16)stbi__get16be(s); // copy the values as-is + } else { + for (k = 0; k < s->img_n && k < 3; ++k) + tc[k] = (stbi_uc)(stbi__get16be(s) & 255) * stbi__depth_scale_table[z->depth]; // non 8-bit images will be larger + } + } + break; + } + + case STBI__PNG_TYPE('I','D','A','T'): { + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (pal_img_n && !pal_len) return stbi__err("no PLTE","Corrupt PNG"); + if (scan == STBI__SCAN_header) { + // header scan definitely stops at first IDAT + if (pal_img_n) + s->img_n = pal_img_n; + return 1; + } + if (c.length > (1u << 30)) return stbi__err("IDAT size limit", "IDAT section larger than 2^30 bytes"); + if ((int)(ioff + c.length) < (int)ioff) return 0; + if (ioff + c.length > idata_limit) { + stbi__uint32 idata_limit_old = idata_limit; + stbi_uc *p; + if (idata_limit == 0) idata_limit = c.length > 4096 ? c.length : 4096; + while (ioff + c.length > idata_limit) + idata_limit *= 2; + STBI_NOTUSED(idata_limit_old); + p = (stbi_uc *) STBI_REALLOC_SIZED(z->idata, idata_limit_old, idata_limit); if (p == NULL) return stbi__err("outofmem", "Out of memory"); + z->idata = p; + } + if (!stbi__getn(s, z->idata+ioff,c.length)) return stbi__err("outofdata","Corrupt PNG"); + ioff += c.length; + break; + } + + case STBI__PNG_TYPE('I','E','N','D'): { + stbi__uint32 raw_len, bpl; + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (scan != STBI__SCAN_load) return 1; + if (z->idata == NULL) return stbi__err("no IDAT","Corrupt PNG"); + // initial guess for decoded data size to avoid unnecessary reallocs + bpl = (s->img_x * z->depth + 7) / 8; // bytes per line, per component + raw_len = bpl * s->img_y * s->img_n /* pixels */ + s->img_y /* filter mode per row */; + z->expanded = (stbi_uc *) stbi_zlib_decode_malloc_guesssize_headerflag((char *) z->idata, ioff, raw_len, (int *) &raw_len, !is_iphone); + if (z->expanded == NULL) return 0; // zlib should set error + STBI_FREE(z->idata); z->idata = NULL; + if ((req_comp == s->img_n+1 && req_comp != 3 && !pal_img_n) || has_trans) + s->img_out_n = s->img_n+1; + else + s->img_out_n = s->img_n; + if (!stbi__create_png_image(z, z->expanded, raw_len, s->img_out_n, z->depth, color, interlace)) return 0; + if (has_trans) { + if (z->depth == 16) { + if (!stbi__compute_transparency16(z, tc16, s->img_out_n)) return 0; + } else { + if (!stbi__compute_transparency(z, tc, s->img_out_n)) return 0; + } + } + if (is_iphone && stbi__de_iphone_flag && s->img_out_n > 2) + stbi__de_iphone(z); + if (pal_img_n) { + // pal_img_n == 3 or 4 + s->img_n = pal_img_n; // record the actual colors we had + s->img_out_n = pal_img_n; + if (req_comp >= 3) s->img_out_n = req_comp; + if (!stbi__expand_png_palette(z, palette, pal_len, s->img_out_n)) + return 0; + } else if (has_trans) { + // non-paletted image with tRNS -> source image has (constant) alpha + ++s->img_n; + } + STBI_FREE(z->expanded); z->expanded = NULL; + // end of PNG chunk, read and skip CRC + stbi__get32be(s); + return 1; + } + + default: + // if critical, fail + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if ((c.type & (1 << 29)) == 0) { + #ifndef STBI_NO_FAILURE_STRINGS + // not threadsafe + static char invalid_chunk[] = "XXXX PNG chunk not known"; + invalid_chunk[0] = STBI__BYTECAST(c.type >> 24); + invalid_chunk[1] = STBI__BYTECAST(c.type >> 16); + invalid_chunk[2] = STBI__BYTECAST(c.type >> 8); + invalid_chunk[3] = STBI__BYTECAST(c.type >> 0); + #endif + return stbi__err(invalid_chunk, "PNG not supported: unknown PNG chunk type"); + } + stbi__skip(s, c.length); + break; + } + // end of PNG chunk, read and skip CRC + stbi__get32be(s); + } +} + +static void *stbi__do_png(stbi__png *p, int *x, int *y, int *n, int req_comp, stbi__result_info *ri) +{ + void *result=NULL; + if (req_comp < 0 || req_comp > 4) return stbi__errpuc("bad req_comp", "Internal error"); + if (stbi__parse_png_file(p, STBI__SCAN_load, req_comp)) { + if (p->depth <= 8) + ri->bits_per_channel = 8; + else if (p->depth == 16) + ri->bits_per_channel = 16; + else + return stbi__errpuc("bad bits_per_channel", "PNG not supported: unsupported color depth"); + result = p->out; + p->out = NULL; + if (req_comp && req_comp != p->s->img_out_n) { + if (ri->bits_per_channel == 8) + result = stbi__convert_format((unsigned char *) result, p->s->img_out_n, req_comp, p->s->img_x, p->s->img_y); + else + result = stbi__convert_format16((stbi__uint16 *) result, p->s->img_out_n, req_comp, p->s->img_x, p->s->img_y); + p->s->img_out_n = req_comp; + if (result == NULL) return result; + } + *x = p->s->img_x; + *y = p->s->img_y; + if (n) *n = p->s->img_n; + } + STBI_FREE(p->out); p->out = NULL; + STBI_FREE(p->expanded); p->expanded = NULL; + STBI_FREE(p->idata); p->idata = NULL; + + return result; +} + +static void *stbi__png_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) +{ + stbi__png p; + p.s = s; + return stbi__do_png(&p, x,y,comp,req_comp, ri); +} + +static int stbi__png_test(stbi__context *s) +{ + int r; + r = stbi__check_png_header(s); + stbi__rewind(s); + return r; +} + +static int stbi__png_info_raw(stbi__png *p, int *x, int *y, int *comp) +{ + if (!stbi__parse_png_file(p, STBI__SCAN_header, 0)) { + stbi__rewind( p->s ); + return 0; + } + if (x) *x = p->s->img_x; + if (y) *y = p->s->img_y; + if (comp) *comp = p->s->img_n; + return 1; +} + +static int stbi__png_info(stbi__context *s, int *x, int *y, int *comp) +{ + stbi__png p; + p.s = s; + return stbi__png_info_raw(&p, x, y, comp); +} + +static int stbi__png_is16(stbi__context *s) +{ + stbi__png p; + p.s = s; + if (!stbi__png_info_raw(&p, NULL, NULL, NULL)) + return 0; + if (p.depth != 16) { + stbi__rewind(p.s); + return 0; + } + return 1; +} +#endif + +// Microsoft/Windows BMP image + +#ifndef STBI_NO_BMP +static int stbi__bmp_test_raw(stbi__context *s) +{ + int r; + int sz; + if (stbi__get8(s) != 'B') return 0; + if (stbi__get8(s) != 'M') return 0; + stbi__get32le(s); // discard filesize + stbi__get16le(s); // discard reserved + stbi__get16le(s); // discard reserved + stbi__get32le(s); // discard data offset + sz = stbi__get32le(s); + r = (sz == 12 || sz == 40 || sz == 56 || sz == 108 || sz == 124); + return r; +} + +static int stbi__bmp_test(stbi__context *s) +{ + int r = stbi__bmp_test_raw(s); + stbi__rewind(s); + return r; +} + + +// returns 0..31 for the highest set bit +static int stbi__high_bit(unsigned int z) +{ + int n=0; + if (z == 0) return -1; + if (z >= 0x10000) { n += 16; z >>= 16; } + if (z >= 0x00100) { n += 8; z >>= 8; } + if (z >= 0x00010) { n += 4; z >>= 4; } + if (z >= 0x00004) { n += 2; z >>= 2; } + if (z >= 0x00002) { n += 1;/* >>= 1;*/ } + return n; +} + +static int stbi__bitcount(unsigned int a) +{ + a = (a & 0x55555555) + ((a >> 1) & 0x55555555); // max 2 + a = (a & 0x33333333) + ((a >> 2) & 0x33333333); // max 4 + a = (a + (a >> 4)) & 0x0f0f0f0f; // max 8 per 4, now 8 bits + a = (a + (a >> 8)); // max 16 per 8 bits + a = (a + (a >> 16)); // max 32 per 8 bits + return a & 0xff; +} + +// extract an arbitrarily-aligned N-bit value (N=bits) +// from v, and then make it 8-bits long and fractionally +// extend it to full full range. +static int stbi__shiftsigned(unsigned int v, int shift, int bits) +{ + static unsigned int mul_table[9] = { + 0, + 0xff/*0b11111111*/, 0x55/*0b01010101*/, 0x49/*0b01001001*/, 0x11/*0b00010001*/, + 0x21/*0b00100001*/, 0x41/*0b01000001*/, 0x81/*0b10000001*/, 0x01/*0b00000001*/, + }; + static unsigned int shift_table[9] = { + 0, 0,0,1,0,2,4,6,0, + }; + if (shift < 0) + v <<= -shift; + else + v >>= shift; + STBI_ASSERT(v < 256); + v >>= (8-bits); + STBI_ASSERT(bits >= 0 && bits <= 8); + return (int) ((unsigned) v * mul_table[bits]) >> shift_table[bits]; +} + +typedef struct +{ + int bpp, offset, hsz; + unsigned int mr,mg,mb,ma, all_a; + int extra_read; +} stbi__bmp_data; + +static int stbi__bmp_set_mask_defaults(stbi__bmp_data *info, int compress) +{ + // BI_BITFIELDS specifies masks explicitly, don't override + if (compress == 3) + return 1; + + if (compress == 0) { + if (info->bpp == 16) { + info->mr = 31u << 10; + info->mg = 31u << 5; + info->mb = 31u << 0; + } else if (info->bpp == 32) { + info->mr = 0xffu << 16; + info->mg = 0xffu << 8; + info->mb = 0xffu << 0; + info->ma = 0xffu << 24; + info->all_a = 0; // if all_a is 0 at end, then we loaded alpha channel but it was all 0 + } else { + // otherwise, use defaults, which is all-0 + info->mr = info->mg = info->mb = info->ma = 0; + } + return 1; + } + return 0; // error +} + +static void *stbi__bmp_parse_header(stbi__context *s, stbi__bmp_data *info) +{ + int hsz; + if (stbi__get8(s) != 'B' || stbi__get8(s) != 'M') return stbi__errpuc("not BMP", "Corrupt BMP"); + stbi__get32le(s); // discard filesize + stbi__get16le(s); // discard reserved + stbi__get16le(s); // discard reserved + info->offset = stbi__get32le(s); + info->hsz = hsz = stbi__get32le(s); + info->mr = info->mg = info->mb = info->ma = 0; + info->extra_read = 14; + + if (info->offset < 0) return stbi__errpuc("bad BMP", "bad BMP"); + + if (hsz != 12 && hsz != 40 && hsz != 56 && hsz != 108 && hsz != 124) return stbi__errpuc("unknown BMP", "BMP type not supported: unknown"); + if (hsz == 12) { + s->img_x = stbi__get16le(s); + s->img_y = stbi__get16le(s); + } else { + s->img_x = stbi__get32le(s); + s->img_y = stbi__get32le(s); + } + if (stbi__get16le(s) != 1) return stbi__errpuc("bad BMP", "bad BMP"); + info->bpp = stbi__get16le(s); + if (hsz != 12) { + int compress = stbi__get32le(s); + if (compress == 1 || compress == 2) return stbi__errpuc("BMP RLE", "BMP type not supported: RLE"); + if (compress >= 4) return stbi__errpuc("BMP JPEG/PNG", "BMP type not supported: unsupported compression"); // this includes PNG/JPEG modes + if (compress == 3 && info->bpp != 16 && info->bpp != 32) return stbi__errpuc("bad BMP", "bad BMP"); // bitfields requires 16 or 32 bits/pixel + stbi__get32le(s); // discard sizeof + stbi__get32le(s); // discard hres + stbi__get32le(s); // discard vres + stbi__get32le(s); // discard colorsused + stbi__get32le(s); // discard max important + if (hsz == 40 || hsz == 56) { + if (hsz == 56) { + stbi__get32le(s); + stbi__get32le(s); + stbi__get32le(s); + stbi__get32le(s); + } + if (info->bpp == 16 || info->bpp == 32) { + if (compress == 0) { + stbi__bmp_set_mask_defaults(info, compress); + } else if (compress == 3) { + info->mr = stbi__get32le(s); + info->mg = stbi__get32le(s); + info->mb = stbi__get32le(s); + info->extra_read += 12; + // not documented, but generated by photoshop and handled by mspaint + if (info->mr == info->mg && info->mg == info->mb) { + // ?!?!? + return stbi__errpuc("bad BMP", "bad BMP"); + } + } else + return stbi__errpuc("bad BMP", "bad BMP"); + } + } else { + // V4/V5 header + int i; + if (hsz != 108 && hsz != 124) + return stbi__errpuc("bad BMP", "bad BMP"); + info->mr = stbi__get32le(s); + info->mg = stbi__get32le(s); + info->mb = stbi__get32le(s); + info->ma = stbi__get32le(s); + if (compress != 3) // override mr/mg/mb unless in BI_BITFIELDS mode, as per docs + stbi__bmp_set_mask_defaults(info, compress); + stbi__get32le(s); // discard color space + for (i=0; i < 12; ++i) + stbi__get32le(s); // discard color space parameters + if (hsz == 124) { + stbi__get32le(s); // discard rendering intent + stbi__get32le(s); // discard offset of profile data + stbi__get32le(s); // discard size of profile data + stbi__get32le(s); // discard reserved + } + } + } + return (void *) 1; +} + + +static void *stbi__bmp_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) +{ + stbi_uc *out; + unsigned int mr=0,mg=0,mb=0,ma=0, all_a; + stbi_uc pal[256][4]; + int psize=0,i,j,width; + int flip_vertically, pad, target; + stbi__bmp_data info; + STBI_NOTUSED(ri); + + info.all_a = 255; + if (stbi__bmp_parse_header(s, &info) == NULL) + return NULL; // error code already set + + flip_vertically = ((int) s->img_y) > 0; + s->img_y = abs((int) s->img_y); + + if (s->img_y > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + if (s->img_x > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + + mr = info.mr; + mg = info.mg; + mb = info.mb; + ma = info.ma; + all_a = info.all_a; + + if (info.hsz == 12) { + if (info.bpp < 24) + psize = (info.offset - info.extra_read - 24) / 3; + } else { + if (info.bpp < 16) + psize = (info.offset - info.extra_read - info.hsz) >> 2; + } + if (psize == 0) { + // accept some number of extra bytes after the header, but if the offset points either to before + // the header ends or implies a large amount of extra data, reject the file as malformed + int bytes_read_so_far = s->callback_already_read + (int)(s->img_buffer - s->img_buffer_original); + int header_limit = 1024; // max we actually read is below 256 bytes currently. + int extra_data_limit = 256*4; // what ordinarily goes here is a palette; 256 entries*4 bytes is its max size. + if (bytes_read_so_far <= 0 || bytes_read_so_far > header_limit) { + return stbi__errpuc("bad header", "Corrupt BMP"); + } + // we established that bytes_read_so_far is positive and sensible. + // the first half of this test rejects offsets that are either too small positives, or + // negative, and guarantees that info.offset >= bytes_read_so_far > 0. this in turn + // ensures the number computed in the second half of the test can't overflow. + if (info.offset < bytes_read_so_far || info.offset - bytes_read_so_far > extra_data_limit) { + return stbi__errpuc("bad offset", "Corrupt BMP"); + } else { + stbi__skip(s, info.offset - bytes_read_so_far); + } + } + + if (info.bpp == 24 && ma == 0xff000000) + s->img_n = 3; + else + s->img_n = ma ? 4 : 3; + if (req_comp && req_comp >= 3) // we can directly decode 3 or 4 + target = req_comp; + else + target = s->img_n; // if they want monochrome, we'll post-convert + + // sanity-check size + if (!stbi__mad3sizes_valid(target, s->img_x, s->img_y, 0)) + return stbi__errpuc("too large", "Corrupt BMP"); + + out = (stbi_uc *) stbi__malloc_mad3(target, s->img_x, s->img_y, 0); + if (!out) return stbi__errpuc("outofmem", "Out of memory"); + if (info.bpp < 16) { + int z=0; + if (psize == 0 || psize > 256) { STBI_FREE(out); return stbi__errpuc("invalid", "Corrupt BMP"); } + for (i=0; i < psize; ++i) { + pal[i][2] = stbi__get8(s); + pal[i][1] = stbi__get8(s); + pal[i][0] = stbi__get8(s); + if (info.hsz != 12) stbi__get8(s); + pal[i][3] = 255; + } + stbi__skip(s, info.offset - info.extra_read - info.hsz - psize * (info.hsz == 12 ? 3 : 4)); + if (info.bpp == 1) width = (s->img_x + 7) >> 3; + else if (info.bpp == 4) width = (s->img_x + 1) >> 1; + else if (info.bpp == 8) width = s->img_x; + else { STBI_FREE(out); return stbi__errpuc("bad bpp", "Corrupt BMP"); } + pad = (-width)&3; + if (info.bpp == 1) { + for (j=0; j < (int) s->img_y; ++j) { + int bit_offset = 7, v = stbi__get8(s); + for (i=0; i < (int) s->img_x; ++i) { + int color = (v>>bit_offset)&0x1; + out[z++] = pal[color][0]; + out[z++] = pal[color][1]; + out[z++] = pal[color][2]; + if (target == 4) out[z++] = 255; + if (i+1 == (int) s->img_x) break; + if((--bit_offset) < 0) { + bit_offset = 7; + v = stbi__get8(s); + } + } + stbi__skip(s, pad); + } + } else { + for (j=0; j < (int) s->img_y; ++j) { + for (i=0; i < (int) s->img_x; i += 2) { + int v=stbi__get8(s),v2=0; + if (info.bpp == 4) { + v2 = v & 15; + v >>= 4; + } + out[z++] = pal[v][0]; + out[z++] = pal[v][1]; + out[z++] = pal[v][2]; + if (target == 4) out[z++] = 255; + if (i+1 == (int) s->img_x) break; + v = (info.bpp == 8) ? stbi__get8(s) : v2; + out[z++] = pal[v][0]; + out[z++] = pal[v][1]; + out[z++] = pal[v][2]; + if (target == 4) out[z++] = 255; + } + stbi__skip(s, pad); + } + } + } else { + int rshift=0,gshift=0,bshift=0,ashift=0,rcount=0,gcount=0,bcount=0,acount=0; + int z = 0; + int easy=0; + stbi__skip(s, info.offset - info.extra_read - info.hsz); + if (info.bpp == 24) width = 3 * s->img_x; + else if (info.bpp == 16) width = 2*s->img_x; + else /* bpp = 32 and pad = 0 */ width=0; + pad = (-width) & 3; + if (info.bpp == 24) { + easy = 1; + } else if (info.bpp == 32) { + if (mb == 0xff && mg == 0xff00 && mr == 0x00ff0000 && ma == 0xff000000) + easy = 2; + } + if (!easy) { + if (!mr || !mg || !mb) { STBI_FREE(out); return stbi__errpuc("bad masks", "Corrupt BMP"); } + // right shift amt to put high bit in position #7 + rshift = stbi__high_bit(mr)-7; rcount = stbi__bitcount(mr); + gshift = stbi__high_bit(mg)-7; gcount = stbi__bitcount(mg); + bshift = stbi__high_bit(mb)-7; bcount = stbi__bitcount(mb); + ashift = stbi__high_bit(ma)-7; acount = stbi__bitcount(ma); + if (rcount > 8 || gcount > 8 || bcount > 8 || acount > 8) { STBI_FREE(out); return stbi__errpuc("bad masks", "Corrupt BMP"); } + } + for (j=0; j < (int) s->img_y; ++j) { + if (easy) { + for (i=0; i < (int) s->img_x; ++i) { + unsigned char a; + out[z+2] = stbi__get8(s); + out[z+1] = stbi__get8(s); + out[z+0] = stbi__get8(s); + z += 3; + a = (easy == 2 ? stbi__get8(s) : 255); + all_a |= a; + if (target == 4) out[z++] = a; + } + } else { + int bpp = info.bpp; + for (i=0; i < (int) s->img_x; ++i) { + stbi__uint32 v = (bpp == 16 ? (stbi__uint32) stbi__get16le(s) : stbi__get32le(s)); + unsigned int a; + out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mr, rshift, rcount)); + out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mg, gshift, gcount)); + out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mb, bshift, bcount)); + a = (ma ? stbi__shiftsigned(v & ma, ashift, acount) : 255); + all_a |= a; + if (target == 4) out[z++] = STBI__BYTECAST(a); + } + } + stbi__skip(s, pad); + } + } + + // if alpha channel is all 0s, replace with all 255s + if (target == 4 && all_a == 0) + for (i=4*s->img_x*s->img_y-1; i >= 0; i -= 4) + out[i] = 255; + + if (flip_vertically) { + stbi_uc t; + for (j=0; j < (int) s->img_y>>1; ++j) { + stbi_uc *p1 = out + j *s->img_x*target; + stbi_uc *p2 = out + (s->img_y-1-j)*s->img_x*target; + for (i=0; i < (int) s->img_x*target; ++i) { + t = p1[i]; p1[i] = p2[i]; p2[i] = t; + } + } + } + + if (req_comp && req_comp != target) { + out = stbi__convert_format(out, target, req_comp, s->img_x, s->img_y); + if (out == NULL) return out; // stbi__convert_format frees input on failure + } + + *x = s->img_x; + *y = s->img_y; + if (comp) *comp = s->img_n; + return out; +} +#endif + +// Targa Truevision - TGA +// by Jonathan Dummer +#ifndef STBI_NO_TGA +// returns STBI_rgb or whatever, 0 on error +static int stbi__tga_get_comp(int bits_per_pixel, int is_grey, int* is_rgb16) +{ + // only RGB or RGBA (incl. 16bit) or grey allowed + if (is_rgb16) *is_rgb16 = 0; + switch(bits_per_pixel) { + case 8: return STBI_grey; + case 16: if(is_grey) return STBI_grey_alpha; + // fallthrough + case 15: if(is_rgb16) *is_rgb16 = 1; + return STBI_rgb; + case 24: // fallthrough + case 32: return bits_per_pixel/8; + default: return 0; + } +} + +static int stbi__tga_info(stbi__context *s, int *x, int *y, int *comp) +{ + int tga_w, tga_h, tga_comp, tga_image_type, tga_bits_per_pixel, tga_colormap_bpp; + int sz, tga_colormap_type; + stbi__get8(s); // discard Offset + tga_colormap_type = stbi__get8(s); // colormap type + if( tga_colormap_type > 1 ) { + stbi__rewind(s); + return 0; // only RGB or indexed allowed + } + tga_image_type = stbi__get8(s); // image type + if ( tga_colormap_type == 1 ) { // colormapped (paletted) image + if (tga_image_type != 1 && tga_image_type != 9) { + stbi__rewind(s); + return 0; + } + stbi__skip(s,4); // skip index of first colormap entry and number of entries + sz = stbi__get8(s); // check bits per palette color entry + if ( (sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32) ) { + stbi__rewind(s); + return 0; + } + stbi__skip(s,4); // skip image x and y origin + tga_colormap_bpp = sz; + } else { // "normal" image w/o colormap - only RGB or grey allowed, +/- RLE + if ( (tga_image_type != 2) && (tga_image_type != 3) && (tga_image_type != 10) && (tga_image_type != 11) ) { + stbi__rewind(s); + return 0; // only RGB or grey allowed, +/- RLE + } + stbi__skip(s,9); // skip colormap specification and image x/y origin + tga_colormap_bpp = 0; + } + tga_w = stbi__get16le(s); + if( tga_w < 1 ) { + stbi__rewind(s); + return 0; // test width + } + tga_h = stbi__get16le(s); + if( tga_h < 1 ) { + stbi__rewind(s); + return 0; // test height + } + tga_bits_per_pixel = stbi__get8(s); // bits per pixel + stbi__get8(s); // ignore alpha bits + if (tga_colormap_bpp != 0) { + if((tga_bits_per_pixel != 8) && (tga_bits_per_pixel != 16)) { + // when using a colormap, tga_bits_per_pixel is the size of the indexes + // I don't think anything but 8 or 16bit indexes makes sense + stbi__rewind(s); + return 0; + } + tga_comp = stbi__tga_get_comp(tga_colormap_bpp, 0, NULL); + } else { + tga_comp = stbi__tga_get_comp(tga_bits_per_pixel, (tga_image_type == 3) || (tga_image_type == 11), NULL); + } + if(!tga_comp) { + stbi__rewind(s); + return 0; + } + if (x) *x = tga_w; + if (y) *y = tga_h; + if (comp) *comp = tga_comp; + return 1; // seems to have passed everything +} + +static int stbi__tga_test(stbi__context *s) +{ + int res = 0; + int sz, tga_color_type; + stbi__get8(s); // discard Offset + tga_color_type = stbi__get8(s); // color type + if ( tga_color_type > 1 ) goto errorEnd; // only RGB or indexed allowed + sz = stbi__get8(s); // image type + if ( tga_color_type == 1 ) { // colormapped (paletted) image + if (sz != 1 && sz != 9) goto errorEnd; // colortype 1 demands image type 1 or 9 + stbi__skip(s,4); // skip index of first colormap entry and number of entries + sz = stbi__get8(s); // check bits per palette color entry + if ( (sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32) ) goto errorEnd; + stbi__skip(s,4); // skip image x and y origin + } else { // "normal" image w/o colormap + if ( (sz != 2) && (sz != 3) && (sz != 10) && (sz != 11) ) goto errorEnd; // only RGB or grey allowed, +/- RLE + stbi__skip(s,9); // skip colormap specification and image x/y origin + } + if ( stbi__get16le(s) < 1 ) goto errorEnd; // test width + if ( stbi__get16le(s) < 1 ) goto errorEnd; // test height + sz = stbi__get8(s); // bits per pixel + if ( (tga_color_type == 1) && (sz != 8) && (sz != 16) ) goto errorEnd; // for colormapped images, bpp is size of an index + if ( (sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32) ) goto errorEnd; + + res = 1; // if we got this far, everything's good and we can return 1 instead of 0 + +errorEnd: + stbi__rewind(s); + return res; +} + +// read 16bit value and convert to 24bit RGB +static void stbi__tga_read_rgb16(stbi__context *s, stbi_uc* out) +{ + stbi__uint16 px = (stbi__uint16)stbi__get16le(s); + stbi__uint16 fiveBitMask = 31; + // we have 3 channels with 5bits each + int r = (px >> 10) & fiveBitMask; + int g = (px >> 5) & fiveBitMask; + int b = px & fiveBitMask; + // Note that this saves the data in RGB(A) order, so it doesn't need to be swapped later + out[0] = (stbi_uc)((r * 255)/31); + out[1] = (stbi_uc)((g * 255)/31); + out[2] = (stbi_uc)((b * 255)/31); + + // some people claim that the most significant bit might be used for alpha + // (possibly if an alpha-bit is set in the "image descriptor byte") + // but that only made 16bit test images completely translucent.. + // so let's treat all 15 and 16bit TGAs as RGB with no alpha. +} + +static void *stbi__tga_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) +{ + // read in the TGA header stuff + int tga_offset = stbi__get8(s); + int tga_indexed = stbi__get8(s); + int tga_image_type = stbi__get8(s); + int tga_is_RLE = 0; + int tga_palette_start = stbi__get16le(s); + int tga_palette_len = stbi__get16le(s); + int tga_palette_bits = stbi__get8(s); + int tga_x_origin = stbi__get16le(s); + int tga_y_origin = stbi__get16le(s); + int tga_width = stbi__get16le(s); + int tga_height = stbi__get16le(s); + int tga_bits_per_pixel = stbi__get8(s); + int tga_comp, tga_rgb16=0; + int tga_inverted = stbi__get8(s); + // int tga_alpha_bits = tga_inverted & 15; // the 4 lowest bits - unused (useless?) + // image data + unsigned char *tga_data; + unsigned char *tga_palette = NULL; + int i, j; + unsigned char raw_data[4] = {0}; + int RLE_count = 0; + int RLE_repeating = 0; + int read_next_pixel = 1; + STBI_NOTUSED(ri); + STBI_NOTUSED(tga_x_origin); // @TODO + STBI_NOTUSED(tga_y_origin); // @TODO + + if (tga_height > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + if (tga_width > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + + // do a tiny bit of precessing + if ( tga_image_type >= 8 ) + { + tga_image_type -= 8; + tga_is_RLE = 1; + } + tga_inverted = 1 - ((tga_inverted >> 5) & 1); + + // If I'm paletted, then I'll use the number of bits from the palette + if ( tga_indexed ) tga_comp = stbi__tga_get_comp(tga_palette_bits, 0, &tga_rgb16); + else tga_comp = stbi__tga_get_comp(tga_bits_per_pixel, (tga_image_type == 3), &tga_rgb16); + + if(!tga_comp) // shouldn't really happen, stbi__tga_test() should have ensured basic consistency + return stbi__errpuc("bad format", "Can't find out TGA pixelformat"); + + // tga info + *x = tga_width; + *y = tga_height; + if (comp) *comp = tga_comp; + + if (!stbi__mad3sizes_valid(tga_width, tga_height, tga_comp, 0)) + return stbi__errpuc("too large", "Corrupt TGA"); + + tga_data = (unsigned char*)stbi__malloc_mad3(tga_width, tga_height, tga_comp, 0); + if (!tga_data) return stbi__errpuc("outofmem", "Out of memory"); + + // skip to the data's starting position (offset usually = 0) + stbi__skip(s, tga_offset ); + + if ( !tga_indexed && !tga_is_RLE && !tga_rgb16 ) { + for (i=0; i < tga_height; ++i) { + int row = tga_inverted ? tga_height -i - 1 : i; + stbi_uc *tga_row = tga_data + row*tga_width*tga_comp; + stbi__getn(s, tga_row, tga_width * tga_comp); + } + } else { + // do I need to load a palette? + if ( tga_indexed) + { + if (tga_palette_len == 0) { /* you have to have at least one entry! */ + STBI_FREE(tga_data); + return stbi__errpuc("bad palette", "Corrupt TGA"); + } + + // any data to skip? (offset usually = 0) + stbi__skip(s, tga_palette_start ); + // load the palette + tga_palette = (unsigned char*)stbi__malloc_mad2(tga_palette_len, tga_comp, 0); + if (!tga_palette) { + STBI_FREE(tga_data); + return stbi__errpuc("outofmem", "Out of memory"); + } + if (tga_rgb16) { + stbi_uc *pal_entry = tga_palette; + STBI_ASSERT(tga_comp == STBI_rgb); + for (i=0; i < tga_palette_len; ++i) { + stbi__tga_read_rgb16(s, pal_entry); + pal_entry += tga_comp; + } + } else if (!stbi__getn(s, tga_palette, tga_palette_len * tga_comp)) { + STBI_FREE(tga_data); + STBI_FREE(tga_palette); + return stbi__errpuc("bad palette", "Corrupt TGA"); + } + } + // load the data + for (i=0; i < tga_width * tga_height; ++i) + { + // if I'm in RLE mode, do I need to get a RLE stbi__pngchunk? + if ( tga_is_RLE ) + { + if ( RLE_count == 0 ) + { + // yep, get the next byte as a RLE command + int RLE_cmd = stbi__get8(s); + RLE_count = 1 + (RLE_cmd & 127); + RLE_repeating = RLE_cmd >> 7; + read_next_pixel = 1; + } else if ( !RLE_repeating ) + { + read_next_pixel = 1; + } + } else + { + read_next_pixel = 1; + } + // OK, if I need to read a pixel, do it now + if ( read_next_pixel ) + { + // load however much data we did have + if ( tga_indexed ) + { + // read in index, then perform the lookup + int pal_idx = (tga_bits_per_pixel == 8) ? stbi__get8(s) : stbi__get16le(s); + if ( pal_idx >= tga_palette_len ) { + // invalid index + pal_idx = 0; + } + pal_idx *= tga_comp; + for (j = 0; j < tga_comp; ++j) { + raw_data[j] = tga_palette[pal_idx+j]; + } + } else if(tga_rgb16) { + STBI_ASSERT(tga_comp == STBI_rgb); + stbi__tga_read_rgb16(s, raw_data); + } else { + // read in the data raw + for (j = 0; j < tga_comp; ++j) { + raw_data[j] = stbi__get8(s); + } + } + // clear the reading flag for the next pixel + read_next_pixel = 0; + } // end of reading a pixel + + // copy data + for (j = 0; j < tga_comp; ++j) + tga_data[i*tga_comp+j] = raw_data[j]; + + // in case we're in RLE mode, keep counting down + --RLE_count; + } + // do I need to invert the image? + if ( tga_inverted ) + { + for (j = 0; j*2 < tga_height; ++j) + { + int index1 = j * tga_width * tga_comp; + int index2 = (tga_height - 1 - j) * tga_width * tga_comp; + for (i = tga_width * tga_comp; i > 0; --i) + { + unsigned char temp = tga_data[index1]; + tga_data[index1] = tga_data[index2]; + tga_data[index2] = temp; + ++index1; + ++index2; + } + } + } + // clear my palette, if I had one + if ( tga_palette != NULL ) + { + STBI_FREE( tga_palette ); + } + } + + // swap RGB - if the source data was RGB16, it already is in the right order + if (tga_comp >= 3 && !tga_rgb16) + { + unsigned char* tga_pixel = tga_data; + for (i=0; i < tga_width * tga_height; ++i) + { + unsigned char temp = tga_pixel[0]; + tga_pixel[0] = tga_pixel[2]; + tga_pixel[2] = temp; + tga_pixel += tga_comp; + } + } + + // convert to target component count + if (req_comp && req_comp != tga_comp) + tga_data = stbi__convert_format(tga_data, tga_comp, req_comp, tga_width, tga_height); + + // the things I do to get rid of an error message, and yet keep + // Microsoft's C compilers happy... [8^( + tga_palette_start = tga_palette_len = tga_palette_bits = + tga_x_origin = tga_y_origin = 0; + STBI_NOTUSED(tga_palette_start); + // OK, done + return tga_data; +} +#endif + +// ************************************************************************************************* +// Photoshop PSD loader -- PD by Thatcher Ulrich, integration by Nicolas Schulz, tweaked by STB + +#ifndef STBI_NO_PSD +static int stbi__psd_test(stbi__context *s) +{ + int r = (stbi__get32be(s) == 0x38425053); + stbi__rewind(s); + return r; +} + +static int stbi__psd_decode_rle(stbi__context *s, stbi_uc *p, int pixelCount) +{ + int count, nleft, len; + + count = 0; + while ((nleft = pixelCount - count) > 0) { + len = stbi__get8(s); + if (len == 128) { + // No-op. + } else if (len < 128) { + // Copy next len+1 bytes literally. + len++; + if (len > nleft) return 0; // corrupt data + count += len; + while (len) { + *p = stbi__get8(s); + p += 4; + len--; + } + } else if (len > 128) { + stbi_uc val; + // Next -len+1 bytes in the dest are replicated from next source byte. + // (Interpret len as a negative 8-bit int.) + len = 257 - len; + if (len > nleft) return 0; // corrupt data + val = stbi__get8(s); + count += len; + while (len) { + *p = val; + p += 4; + len--; + } + } + } + + return 1; +} + +static void *stbi__psd_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri, int bpc) +{ + int pixelCount; + int channelCount, compression; + int channel, i; + int bitdepth; + int w,h; + stbi_uc *out; + STBI_NOTUSED(ri); + + // Check identifier + if (stbi__get32be(s) != 0x38425053) // "8BPS" + return stbi__errpuc("not PSD", "Corrupt PSD image"); + + // Check file type version. + if (stbi__get16be(s) != 1) + return stbi__errpuc("wrong version", "Unsupported version of PSD image"); + + // Skip 6 reserved bytes. + stbi__skip(s, 6 ); + + // Read the number of channels (R, G, B, A, etc). + channelCount = stbi__get16be(s); + if (channelCount < 0 || channelCount > 16) + return stbi__errpuc("wrong channel count", "Unsupported number of channels in PSD image"); + + // Read the rows and columns of the image. + h = stbi__get32be(s); + w = stbi__get32be(s); + + if (h > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + if (w > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + + // Make sure the depth is 8 bits. + bitdepth = stbi__get16be(s); + if (bitdepth != 8 && bitdepth != 16) + return stbi__errpuc("unsupported bit depth", "PSD bit depth is not 8 or 16 bit"); + + // Make sure the color mode is RGB. + // Valid options are: + // 0: Bitmap + // 1: Grayscale + // 2: Indexed color + // 3: RGB color + // 4: CMYK color + // 7: Multichannel + // 8: Duotone + // 9: Lab color + if (stbi__get16be(s) != 3) + return stbi__errpuc("wrong color format", "PSD is not in RGB color format"); + + // Skip the Mode Data. (It's the palette for indexed color; other info for other modes.) + stbi__skip(s,stbi__get32be(s) ); + + // Skip the image resources. (resolution, pen tool paths, etc) + stbi__skip(s, stbi__get32be(s) ); + + // Skip the reserved data. + stbi__skip(s, stbi__get32be(s) ); + + // Find out if the data is compressed. + // Known values: + // 0: no compression + // 1: RLE compressed + compression = stbi__get16be(s); + if (compression > 1) + return stbi__errpuc("bad compression", "PSD has an unknown compression format"); + + // Check size + if (!stbi__mad3sizes_valid(4, w, h, 0)) + return stbi__errpuc("too large", "Corrupt PSD"); + + // Create the destination image. + + if (!compression && bitdepth == 16 && bpc == 16) { + out = (stbi_uc *) stbi__malloc_mad3(8, w, h, 0); + ri->bits_per_channel = 16; + } else + out = (stbi_uc *) stbi__malloc(4 * w*h); + + if (!out) return stbi__errpuc("outofmem", "Out of memory"); + pixelCount = w*h; + + // Initialize the data to zero. + //memset( out, 0, pixelCount * 4 ); + + // Finally, the image data. + if (compression) { + // RLE as used by .PSD and .TIFF + // Loop until you get the number of unpacked bytes you are expecting: + // Read the next source byte into n. + // If n is between 0 and 127 inclusive, copy the next n+1 bytes literally. + // Else if n is between -127 and -1 inclusive, copy the next byte -n+1 times. + // Else if n is 128, noop. + // Endloop + + // The RLE-compressed data is preceded by a 2-byte data count for each row in the data, + // which we're going to just skip. + stbi__skip(s, h * channelCount * 2 ); + + // Read the RLE data by channel. + for (channel = 0; channel < 4; channel++) { + stbi_uc *p; + + p = out+channel; + if (channel >= channelCount) { + // Fill this channel with default data. + for (i = 0; i < pixelCount; i++, p += 4) + *p = (channel == 3 ? 255 : 0); + } else { + // Read the RLE data. + if (!stbi__psd_decode_rle(s, p, pixelCount)) { + STBI_FREE(out); + return stbi__errpuc("corrupt", "bad RLE data"); + } + } + } + + } else { + // We're at the raw image data. It's each channel in order (Red, Green, Blue, Alpha, ...) + // where each channel consists of an 8-bit (or 16-bit) value for each pixel in the image. + + // Read the data by channel. + for (channel = 0; channel < 4; channel++) { + if (channel >= channelCount) { + // Fill this channel with default data. + if (bitdepth == 16 && bpc == 16) { + stbi__uint16 *q = ((stbi__uint16 *) out) + channel; + stbi__uint16 val = channel == 3 ? 65535 : 0; + for (i = 0; i < pixelCount; i++, q += 4) + *q = val; + } else { + stbi_uc *p = out+channel; + stbi_uc val = channel == 3 ? 255 : 0; + for (i = 0; i < pixelCount; i++, p += 4) + *p = val; + } + } else { + if (ri->bits_per_channel == 16) { // output bpc + stbi__uint16 *q = ((stbi__uint16 *) out) + channel; + for (i = 0; i < pixelCount; i++, q += 4) + *q = (stbi__uint16) stbi__get16be(s); + } else { + stbi_uc *p = out+channel; + if (bitdepth == 16) { // input bpc + for (i = 0; i < pixelCount; i++, p += 4) + *p = (stbi_uc) (stbi__get16be(s) >> 8); + } else { + for (i = 0; i < pixelCount; i++, p += 4) + *p = stbi__get8(s); + } + } + } + } + } + + // remove weird white matte from PSD + if (channelCount >= 4) { + if (ri->bits_per_channel == 16) { + for (i=0; i < w*h; ++i) { + stbi__uint16 *pixel = (stbi__uint16 *) out + 4*i; + if (pixel[3] != 0 && pixel[3] != 65535) { + float a = pixel[3] / 65535.0f; + float ra = 1.0f / a; + float inv_a = 65535.0f * (1 - ra); + pixel[0] = (stbi__uint16) (pixel[0]*ra + inv_a); + pixel[1] = (stbi__uint16) (pixel[1]*ra + inv_a); + pixel[2] = (stbi__uint16) (pixel[2]*ra + inv_a); + } + } + } else { + for (i=0; i < w*h; ++i) { + unsigned char *pixel = out + 4*i; + if (pixel[3] != 0 && pixel[3] != 255) { + float a = pixel[3] / 255.0f; + float ra = 1.0f / a; + float inv_a = 255.0f * (1 - ra); + pixel[0] = (unsigned char) (pixel[0]*ra + inv_a); + pixel[1] = (unsigned char) (pixel[1]*ra + inv_a); + pixel[2] = (unsigned char) (pixel[2]*ra + inv_a); + } + } + } + } + + // convert to desired output format + if (req_comp && req_comp != 4) { + if (ri->bits_per_channel == 16) + out = (stbi_uc *) stbi__convert_format16((stbi__uint16 *) out, 4, req_comp, w, h); + else + out = stbi__convert_format(out, 4, req_comp, w, h); + if (out == NULL) return out; // stbi__convert_format frees input on failure + } + + if (comp) *comp = 4; + *y = h; + *x = w; + + return out; +} +#endif + +// ************************************************************************************************* +// Softimage PIC loader +// by Tom Seddon +// +// See http://softimage.wiki.softimage.com/index.php/INFO:_PIC_file_format +// See http://ozviz.wasp.uwa.edu.au/~pbourke/dataformats/softimagepic/ + +#ifndef STBI_NO_PIC +static int stbi__pic_is4(stbi__context *s,const char *str) +{ + int i; + for (i=0; i<4; ++i) + if (stbi__get8(s) != (stbi_uc)str[i]) + return 0; + + return 1; +} + +static int stbi__pic_test_core(stbi__context *s) +{ + int i; + + if (!stbi__pic_is4(s,"\x53\x80\xF6\x34")) + return 0; + + for(i=0;i<84;++i) + stbi__get8(s); + + if (!stbi__pic_is4(s,"PICT")) + return 0; + + return 1; +} + +typedef struct +{ + stbi_uc size,type,channel; +} stbi__pic_packet; + +static stbi_uc *stbi__readval(stbi__context *s, int channel, stbi_uc *dest) +{ + int mask=0x80, i; + + for (i=0; i<4; ++i, mask>>=1) { + if (channel & mask) { + if (stbi__at_eof(s)) return stbi__errpuc("bad file","PIC file too short"); + dest[i]=stbi__get8(s); + } + } + + return dest; +} + +static void stbi__copyval(int channel,stbi_uc *dest,const stbi_uc *src) +{ + int mask=0x80,i; + + for (i=0;i<4; ++i, mask>>=1) + if (channel&mask) + dest[i]=src[i]; +} + +static stbi_uc *stbi__pic_load_core(stbi__context *s,int width,int height,int *comp, stbi_uc *result) +{ + int act_comp=0,num_packets=0,y,chained; + stbi__pic_packet packets[10]; + + // this will (should...) cater for even some bizarre stuff like having data + // for the same channel in multiple packets. + do { + stbi__pic_packet *packet; + + if (num_packets==sizeof(packets)/sizeof(packets[0])) + return stbi__errpuc("bad format","too many packets"); + + packet = &packets[num_packets++]; + + chained = stbi__get8(s); + packet->size = stbi__get8(s); + packet->type = stbi__get8(s); + packet->channel = stbi__get8(s); + + act_comp |= packet->channel; + + if (stbi__at_eof(s)) return stbi__errpuc("bad file","file too short (reading packets)"); + if (packet->size != 8) return stbi__errpuc("bad format","packet isn't 8bpp"); + } while (chained); + + *comp = (act_comp & 0x10 ? 4 : 3); // has alpha channel? + + for(y=0; ytype) { + default: + return stbi__errpuc("bad format","packet has bad compression type"); + + case 0: {//uncompressed + int x; + + for(x=0;xchannel,dest)) + return 0; + break; + } + + case 1://Pure RLE + { + int left=width, i; + + while (left>0) { + stbi_uc count,value[4]; + + count=stbi__get8(s); + if (stbi__at_eof(s)) return stbi__errpuc("bad file","file too short (pure read count)"); + + if (count > left) + count = (stbi_uc) left; + + if (!stbi__readval(s,packet->channel,value)) return 0; + + for(i=0; ichannel,dest,value); + left -= count; + } + } + break; + + case 2: {//Mixed RLE + int left=width; + while (left>0) { + int count = stbi__get8(s), i; + if (stbi__at_eof(s)) return stbi__errpuc("bad file","file too short (mixed read count)"); + + if (count >= 128) { // Repeated + stbi_uc value[4]; + + if (count==128) + count = stbi__get16be(s); + else + count -= 127; + if (count > left) + return stbi__errpuc("bad file","scanline overrun"); + + if (!stbi__readval(s,packet->channel,value)) + return 0; + + for(i=0;ichannel,dest,value); + } else { // Raw + ++count; + if (count>left) return stbi__errpuc("bad file","scanline overrun"); + + for(i=0;ichannel,dest)) + return 0; + } + left-=count; + } + break; + } + } + } + } + + return result; +} + +static void *stbi__pic_load(stbi__context *s,int *px,int *py,int *comp,int req_comp, stbi__result_info *ri) +{ + stbi_uc *result; + int i, x,y, internal_comp; + STBI_NOTUSED(ri); + + if (!comp) comp = &internal_comp; + + for (i=0; i<92; ++i) + stbi__get8(s); + + x = stbi__get16be(s); + y = stbi__get16be(s); + + if (y > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + if (x > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + + if (stbi__at_eof(s)) return stbi__errpuc("bad file","file too short (pic header)"); + if (!stbi__mad3sizes_valid(x, y, 4, 0)) return stbi__errpuc("too large", "PIC image too large to decode"); + + stbi__get32be(s); //skip `ratio' + stbi__get16be(s); //skip `fields' + stbi__get16be(s); //skip `pad' + + // intermediate buffer is RGBA + result = (stbi_uc *) stbi__malloc_mad3(x, y, 4, 0); + if (!result) return stbi__errpuc("outofmem", "Out of memory"); + memset(result, 0xff, x*y*4); + + if (!stbi__pic_load_core(s,x,y,comp, result)) { + STBI_FREE(result); + result=0; + } + *px = x; + *py = y; + if (req_comp == 0) req_comp = *comp; + result=stbi__convert_format(result,4,req_comp,x,y); + + return result; +} + +static int stbi__pic_test(stbi__context *s) +{ + int r = stbi__pic_test_core(s); + stbi__rewind(s); + return r; +} +#endif + +// ************************************************************************************************* +// GIF loader -- public domain by Jean-Marc Lienher -- simplified/shrunk by stb + +#ifndef STBI_NO_GIF +typedef struct +{ + stbi__int16 prefix; + stbi_uc first; + stbi_uc suffix; +} stbi__gif_lzw; + +typedef struct +{ + int w,h; + stbi_uc *out; // output buffer (always 4 components) + stbi_uc *background; // The current "background" as far as a gif is concerned + stbi_uc *history; + int flags, bgindex, ratio, transparent, eflags; + stbi_uc pal[256][4]; + stbi_uc lpal[256][4]; + stbi__gif_lzw codes[8192]; + stbi_uc *color_table; + int parse, step; + int lflags; + int start_x, start_y; + int max_x, max_y; + int cur_x, cur_y; + int line_size; + int delay; +} stbi__gif; + +static int stbi__gif_test_raw(stbi__context *s) +{ + int sz; + if (stbi__get8(s) != 'G' || stbi__get8(s) != 'I' || stbi__get8(s) != 'F' || stbi__get8(s) != '8') return 0; + sz = stbi__get8(s); + if (sz != '9' && sz != '7') return 0; + if (stbi__get8(s) != 'a') return 0; + return 1; +} + +static int stbi__gif_test(stbi__context *s) +{ + int r = stbi__gif_test_raw(s); + stbi__rewind(s); + return r; +} + +static void stbi__gif_parse_colortable(stbi__context *s, stbi_uc pal[256][4], int num_entries, int transp) +{ + int i; + for (i=0; i < num_entries; ++i) { + pal[i][2] = stbi__get8(s); + pal[i][1] = stbi__get8(s); + pal[i][0] = stbi__get8(s); + pal[i][3] = transp == i ? 0 : 255; + } +} + +static int stbi__gif_header(stbi__context *s, stbi__gif *g, int *comp, int is_info) +{ + stbi_uc version; + if (stbi__get8(s) != 'G' || stbi__get8(s) != 'I' || stbi__get8(s) != 'F' || stbi__get8(s) != '8') + return stbi__err("not GIF", "Corrupt GIF"); + + version = stbi__get8(s); + if (version != '7' && version != '9') return stbi__err("not GIF", "Corrupt GIF"); + if (stbi__get8(s) != 'a') return stbi__err("not GIF", "Corrupt GIF"); + + stbi__g_failure_reason = ""; + g->w = stbi__get16le(s); + g->h = stbi__get16le(s); + g->flags = stbi__get8(s); + g->bgindex = stbi__get8(s); + g->ratio = stbi__get8(s); + g->transparent = -1; + + if (g->w > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); + if (g->h > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)"); + + if (comp != 0) *comp = 4; // can't actually tell whether it's 3 or 4 until we parse the comments + + if (is_info) return 1; + + if (g->flags & 0x80) + stbi__gif_parse_colortable(s,g->pal, 2 << (g->flags & 7), -1); + + return 1; +} + +static int stbi__gif_info_raw(stbi__context *s, int *x, int *y, int *comp) +{ + stbi__gif* g = (stbi__gif*) stbi__malloc(sizeof(stbi__gif)); + if (!g) return stbi__err("outofmem", "Out of memory"); + if (!stbi__gif_header(s, g, comp, 1)) { + STBI_FREE(g); + stbi__rewind( s ); + return 0; + } + if (x) *x = g->w; + if (y) *y = g->h; + STBI_FREE(g); + return 1; +} + +static void stbi__out_gif_code(stbi__gif *g, stbi__uint16 code) +{ + stbi_uc *p, *c; + int idx; + + // recurse to decode the prefixes, since the linked-list is backwards, + // and working backwards through an interleaved image would be nasty + if (g->codes[code].prefix >= 0) + stbi__out_gif_code(g, g->codes[code].prefix); + + if (g->cur_y >= g->max_y) return; + + idx = g->cur_x + g->cur_y; + p = &g->out[idx]; + g->history[idx / 4] = 1; + + c = &g->color_table[g->codes[code].suffix * 4]; + if (c[3] > 128) { // don't render transparent pixels; + p[0] = c[2]; + p[1] = c[1]; + p[2] = c[0]; + p[3] = c[3]; + } + g->cur_x += 4; + + if (g->cur_x >= g->max_x) { + g->cur_x = g->start_x; + g->cur_y += g->step; + + while (g->cur_y >= g->max_y && g->parse > 0) { + g->step = (1 << g->parse) * g->line_size; + g->cur_y = g->start_y + (g->step >> 1); + --g->parse; + } + } +} + +static stbi_uc *stbi__process_gif_raster(stbi__context *s, stbi__gif *g) +{ + stbi_uc lzw_cs; + stbi__int32 len, init_code; + stbi__uint32 first; + stbi__int32 codesize, codemask, avail, oldcode, bits, valid_bits, clear; + stbi__gif_lzw *p; + + lzw_cs = stbi__get8(s); + if (lzw_cs > 12) return NULL; + clear = 1 << lzw_cs; + first = 1; + codesize = lzw_cs + 1; + codemask = (1 << codesize) - 1; + bits = 0; + valid_bits = 0; + for (init_code = 0; init_code < clear; init_code++) { + g->codes[init_code].prefix = -1; + g->codes[init_code].first = (stbi_uc) init_code; + g->codes[init_code].suffix = (stbi_uc) init_code; + } + + // support no starting clear code + avail = clear+2; + oldcode = -1; + + len = 0; + for(;;) { + if (valid_bits < codesize) { + if (len == 0) { + len = stbi__get8(s); // start new block + if (len == 0) + return g->out; + } + --len; + bits |= (stbi__int32) stbi__get8(s) << valid_bits; + valid_bits += 8; + } else { + stbi__int32 code = bits & codemask; + bits >>= codesize; + valid_bits -= codesize; + // @OPTIMIZE: is there some way we can accelerate the non-clear path? + if (code == clear) { // clear code + codesize = lzw_cs + 1; + codemask = (1 << codesize) - 1; + avail = clear + 2; + oldcode = -1; + first = 0; + } else if (code == clear + 1) { // end of stream code + stbi__skip(s, len); + while ((len = stbi__get8(s)) > 0) + stbi__skip(s,len); + return g->out; + } else if (code <= avail) { + if (first) { + return stbi__errpuc("no clear code", "Corrupt GIF"); + } + + if (oldcode >= 0) { + p = &g->codes[avail++]; + if (avail > 8192) { + return stbi__errpuc("too many codes", "Corrupt GIF"); + } + + p->prefix = (stbi__int16) oldcode; + p->first = g->codes[oldcode].first; + p->suffix = (code == avail) ? p->first : g->codes[code].first; + } else if (code == avail) + return stbi__errpuc("illegal code in raster", "Corrupt GIF"); + + stbi__out_gif_code(g, (stbi__uint16) code); + + if ((avail & codemask) == 0 && avail <= 0x0FFF) { + codesize++; + codemask = (1 << codesize) - 1; + } + + oldcode = code; + } else { + return stbi__errpuc("illegal code in raster", "Corrupt GIF"); + } + } + } +} + +// this function is designed to support animated gifs, although stb_image doesn't support it +// two back is the image from two frames ago, used for a very specific disposal format +static stbi_uc *stbi__gif_load_next(stbi__context *s, stbi__gif *g, int *comp, int req_comp, stbi_uc *two_back) +{ + int dispose; + int first_frame; + int pi; + int pcount; + STBI_NOTUSED(req_comp); + + // on first frame, any non-written pixels get the background colour (non-transparent) + first_frame = 0; + if (g->out == 0) { + if (!stbi__gif_header(s, g, comp,0)) return 0; // stbi__g_failure_reason set by stbi__gif_header + if (!stbi__mad3sizes_valid(4, g->w, g->h, 0)) + return stbi__errpuc("too large", "GIF image is too large"); + pcount = g->w * g->h; + g->out = (stbi_uc *) stbi__malloc(4 * pcount); + g->background = (stbi_uc *) stbi__malloc(4 * pcount); + g->history = (stbi_uc *) stbi__malloc(pcount); + if (!g->out || !g->background || !g->history) + return stbi__errpuc("outofmem", "Out of memory"); + + // image is treated as "transparent" at the start - ie, nothing overwrites the current background; + // background colour is only used for pixels that are not rendered first frame, after that "background" + // color refers to the color that was there the previous frame. + memset(g->out, 0x00, 4 * pcount); + memset(g->background, 0x00, 4 * pcount); // state of the background (starts transparent) + memset(g->history, 0x00, pcount); // pixels that were affected previous frame + first_frame = 1; + } else { + // second frame - how do we dispose of the previous one? + dispose = (g->eflags & 0x1C) >> 2; + pcount = g->w * g->h; + + if ((dispose == 3) && (two_back == 0)) { + dispose = 2; // if I don't have an image to revert back to, default to the old background + } + + if (dispose == 3) { // use previous graphic + for (pi = 0; pi < pcount; ++pi) { + if (g->history[pi]) { + memcpy( &g->out[pi * 4], &two_back[pi * 4], 4 ); + } + } + } else if (dispose == 2) { + // restore what was changed last frame to background before that frame; + for (pi = 0; pi < pcount; ++pi) { + if (g->history[pi]) { + memcpy( &g->out[pi * 4], &g->background[pi * 4], 4 ); + } + } + } else { + // This is a non-disposal case eithe way, so just + // leave the pixels as is, and they will become the new background + // 1: do not dispose + // 0: not specified. + } + + // background is what out is after the undoing of the previou frame; + memcpy( g->background, g->out, 4 * g->w * g->h ); + } + + // clear my history; + memset( g->history, 0x00, g->w * g->h ); // pixels that were affected previous frame + + for (;;) { + int tag = stbi__get8(s); + switch (tag) { + case 0x2C: /* Image Descriptor */ + { + stbi__int32 x, y, w, h; + stbi_uc *o; + + x = stbi__get16le(s); + y = stbi__get16le(s); + w = stbi__get16le(s); + h = stbi__get16le(s); + if (((x + w) > (g->w)) || ((y + h) > (g->h))) + return stbi__errpuc("bad Image Descriptor", "Corrupt GIF"); + + g->line_size = g->w * 4; + g->start_x = x * 4; + g->start_y = y * g->line_size; + g->max_x = g->start_x + w * 4; + g->max_y = g->start_y + h * g->line_size; + g->cur_x = g->start_x; + g->cur_y = g->start_y; + + // if the width of the specified rectangle is 0, that means + // we may not see *any* pixels or the image is malformed; + // to make sure this is caught, move the current y down to + // max_y (which is what out_gif_code checks). + if (w == 0) + g->cur_y = g->max_y; + + g->lflags = stbi__get8(s); + + if (g->lflags & 0x40) { + g->step = 8 * g->line_size; // first interlaced spacing + g->parse = 3; + } else { + g->step = g->line_size; + g->parse = 0; + } + + if (g->lflags & 0x80) { + stbi__gif_parse_colortable(s,g->lpal, 2 << (g->lflags & 7), g->eflags & 0x01 ? g->transparent : -1); + g->color_table = (stbi_uc *) g->lpal; + } else if (g->flags & 0x80) { + g->color_table = (stbi_uc *) g->pal; + } else + return stbi__errpuc("missing color table", "Corrupt GIF"); + + o = stbi__process_gif_raster(s, g); + if (!o) return NULL; + + // if this was the first frame, + pcount = g->w * g->h; + if (first_frame && (g->bgindex > 0)) { + // if first frame, any pixel not drawn to gets the background color + for (pi = 0; pi < pcount; ++pi) { + if (g->history[pi] == 0) { + g->pal[g->bgindex][3] = 255; // just in case it was made transparent, undo that; It will be reset next frame if need be; + memcpy( &g->out[pi * 4], &g->pal[g->bgindex], 4 ); + } + } + } + + return o; + } + + case 0x21: // Comment Extension. + { + int len; + int ext = stbi__get8(s); + if (ext == 0xF9) { // Graphic Control Extension. + len = stbi__get8(s); + if (len == 4) { + g->eflags = stbi__get8(s); + g->delay = 10 * stbi__get16le(s); // delay - 1/100th of a second, saving as 1/1000ths. + + // unset old transparent + if (g->transparent >= 0) { + g->pal[g->transparent][3] = 255; + } + if (g->eflags & 0x01) { + g->transparent = stbi__get8(s); + if (g->transparent >= 0) { + g->pal[g->transparent][3] = 0; + } + } else { + // don't need transparent + stbi__skip(s, 1); + g->transparent = -1; + } + } else { + stbi__skip(s, len); + break; + } + } + while ((len = stbi__get8(s)) != 0) { + stbi__skip(s, len); + } + break; + } + + case 0x3B: // gif stream termination code + return (stbi_uc *) s; // using '1' causes warning on some compilers + + default: + return stbi__errpuc("unknown code", "Corrupt GIF"); + } + } +} + +static void *stbi__load_gif_main_outofmem(stbi__gif *g, stbi_uc *out, int **delays) +{ + STBI_FREE(g->out); + STBI_FREE(g->history); + STBI_FREE(g->background); + + if (out) STBI_FREE(out); + if (delays && *delays) STBI_FREE(*delays); + return stbi__errpuc("outofmem", "Out of memory"); +} + +static void *stbi__load_gif_main(stbi__context *s, int **delays, int *x, int *y, int *z, int *comp, int req_comp) +{ + if (stbi__gif_test(s)) { + int layers = 0; + stbi_uc *u = 0; + stbi_uc *out = 0; + stbi_uc *two_back = 0; + stbi__gif g; + int stride; + int out_size = 0; + int delays_size = 0; + + STBI_NOTUSED(out_size); + STBI_NOTUSED(delays_size); + + memset(&g, 0, sizeof(g)); + if (delays) { + *delays = 0; + } + + do { + u = stbi__gif_load_next(s, &g, comp, req_comp, two_back); + if (u == (stbi_uc *) s) u = 0; // end of animated gif marker + + if (u) { + *x = g.w; + *y = g.h; + ++layers; + stride = g.w * g.h * 4; + + if (out) { + void *tmp = (stbi_uc*) STBI_REALLOC_SIZED( out, out_size, layers * stride ); + if (!tmp) + return stbi__load_gif_main_outofmem(&g, out, delays); + else { + out = (stbi_uc*) tmp; + out_size = layers * stride; + } + + if (delays) { + int *new_delays = (int*) STBI_REALLOC_SIZED( *delays, delays_size, sizeof(int) * layers ); + if (!new_delays) + return stbi__load_gif_main_outofmem(&g, out, delays); + *delays = new_delays; + delays_size = layers * sizeof(int); + } + } else { + out = (stbi_uc*)stbi__malloc( layers * stride ); + if (!out) + return stbi__load_gif_main_outofmem(&g, out, delays); + out_size = layers * stride; + if (delays) { + *delays = (int*) stbi__malloc( layers * sizeof(int) ); + if (!*delays) + return stbi__load_gif_main_outofmem(&g, out, delays); + delays_size = layers * sizeof(int); + } + } + memcpy( out + ((layers - 1) * stride), u, stride ); + if (layers >= 2) { + two_back = out - 2 * stride; + } + + if (delays) { + (*delays)[layers - 1U] = g.delay; + } + } + } while (u != 0); + + // free temp buffer; + STBI_FREE(g.out); + STBI_FREE(g.history); + STBI_FREE(g.background); + + // do the final conversion after loading everything; + if (req_comp && req_comp != 4) + out = stbi__convert_format(out, 4, req_comp, layers * g.w, g.h); + + *z = layers; + return out; + } else { + return stbi__errpuc("not GIF", "Image was not as a gif type."); + } +} + +static void *stbi__gif_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) +{ + stbi_uc *u = 0; + stbi__gif g; + memset(&g, 0, sizeof(g)); + STBI_NOTUSED(ri); + + u = stbi__gif_load_next(s, &g, comp, req_comp, 0); + if (u == (stbi_uc *) s) u = 0; // end of animated gif marker + if (u) { + *x = g.w; + *y = g.h; + + // moved conversion to after successful load so that the same + // can be done for multiple frames. + if (req_comp && req_comp != 4) + u = stbi__convert_format(u, 4, req_comp, g.w, g.h); + } else if (g.out) { + // if there was an error and we allocated an image buffer, free it! + STBI_FREE(g.out); + } + + // free buffers needed for multiple frame loading; + STBI_FREE(g.history); + STBI_FREE(g.background); + + return u; +} + +static int stbi__gif_info(stbi__context *s, int *x, int *y, int *comp) +{ + return stbi__gif_info_raw(s,x,y,comp); +} +#endif + +// ************************************************************************************************* +// Radiance RGBE HDR loader +// originally by Nicolas Schulz +#ifndef STBI_NO_HDR +static int stbi__hdr_test_core(stbi__context *s, const char *signature) +{ + int i; + for (i=0; signature[i]; ++i) + if (stbi__get8(s) != signature[i]) + return 0; + stbi__rewind(s); + return 1; +} + +static int stbi__hdr_test(stbi__context* s) +{ + int r = stbi__hdr_test_core(s, "#?RADIANCE\n"); + stbi__rewind(s); + if(!r) { + r = stbi__hdr_test_core(s, "#?RGBE\n"); + stbi__rewind(s); + } + return r; +} + +#define STBI__HDR_BUFLEN 1024 +static char *stbi__hdr_gettoken(stbi__context *z, char *buffer) +{ + int len=0; + char c = '\0'; + + c = (char) stbi__get8(z); + + while (!stbi__at_eof(z) && c != '\n') { + buffer[len++] = c; + if (len == STBI__HDR_BUFLEN-1) { + // flush to end of line + while (!stbi__at_eof(z) && stbi__get8(z) != '\n') + ; + break; + } + c = (char) stbi__get8(z); + } + + buffer[len] = 0; + return buffer; +} + +static void stbi__hdr_convert(float *output, stbi_uc *input, int req_comp) +{ + if ( input[3] != 0 ) { + float f1; + // Exponent + f1 = (float) ldexp(1.0f, input[3] - (int)(128 + 8)); + if (req_comp <= 2) + output[0] = (input[0] + input[1] + input[2]) * f1 / 3; + else { + output[0] = input[0] * f1; + output[1] = input[1] * f1; + output[2] = input[2] * f1; + } + if (req_comp == 2) output[1] = 1; + if (req_comp == 4) output[3] = 1; + } else { + switch (req_comp) { + case 4: output[3] = 1; /* fallthrough */ + case 3: output[0] = output[1] = output[2] = 0; + break; + case 2: output[1] = 1; /* fallthrough */ + case 1: output[0] = 0; + break; + } + } +} + +static float *stbi__hdr_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) +{ + char buffer[STBI__HDR_BUFLEN]; + char *token; + int valid = 0; + int width, height; + stbi_uc *scanline; + float *hdr_data; + int len; + unsigned char count, value; + int i, j, k, c1,c2, z; + const char *headerToken; + STBI_NOTUSED(ri); + + // Check identifier + headerToken = stbi__hdr_gettoken(s,buffer); + if (strcmp(headerToken, "#?RADIANCE") != 0 && strcmp(headerToken, "#?RGBE") != 0) + return stbi__errpf("not HDR", "Corrupt HDR image"); + + // Parse header + for(;;) { + token = stbi__hdr_gettoken(s,buffer); + if (token[0] == 0) break; + if (strcmp(token, "FORMAT=32-bit_rle_rgbe") == 0) valid = 1; + } + + if (!valid) return stbi__errpf("unsupported format", "Unsupported HDR format"); + + // Parse width and height + // can't use sscanf() if we're not using stdio! + token = stbi__hdr_gettoken(s,buffer); + if (strncmp(token, "-Y ", 3)) return stbi__errpf("unsupported data layout", "Unsupported HDR format"); + token += 3; + height = (int) strtol(token, &token, 10); + while (*token == ' ') ++token; + if (strncmp(token, "+X ", 3)) return stbi__errpf("unsupported data layout", "Unsupported HDR format"); + token += 3; + width = (int) strtol(token, NULL, 10); + + if (height > STBI_MAX_DIMENSIONS) return stbi__errpf("too large","Very large image (corrupt?)"); + if (width > STBI_MAX_DIMENSIONS) return stbi__errpf("too large","Very large image (corrupt?)"); + + *x = width; + *y = height; + + if (comp) *comp = 3; + if (req_comp == 0) req_comp = 3; + + if (!stbi__mad4sizes_valid(width, height, req_comp, sizeof(float), 0)) + return stbi__errpf("too large", "HDR image is too large"); + + // Read data + hdr_data = (float *) stbi__malloc_mad4(width, height, req_comp, sizeof(float), 0); + if (!hdr_data) + return stbi__errpf("outofmem", "Out of memory"); + + // Load image data + // image data is stored as some number of sca + if ( width < 8 || width >= 32768) { + // Read flat data + for (j=0; j < height; ++j) { + for (i=0; i < width; ++i) { + stbi_uc rgbe[4]; + main_decode_loop: + stbi__getn(s, rgbe, 4); + stbi__hdr_convert(hdr_data + j * width * req_comp + i * req_comp, rgbe, req_comp); + } + } + } else { + // Read RLE-encoded data + scanline = NULL; + + for (j = 0; j < height; ++j) { + c1 = stbi__get8(s); + c2 = stbi__get8(s); + len = stbi__get8(s); + if (c1 != 2 || c2 != 2 || (len & 0x80)) { + // not run-length encoded, so we have to actually use THIS data as a decoded + // pixel (note this can't be a valid pixel--one of RGB must be >= 128) + stbi_uc rgbe[4]; + rgbe[0] = (stbi_uc) c1; + rgbe[1] = (stbi_uc) c2; + rgbe[2] = (stbi_uc) len; + rgbe[3] = (stbi_uc) stbi__get8(s); + stbi__hdr_convert(hdr_data, rgbe, req_comp); + i = 1; + j = 0; + STBI_FREE(scanline); + goto main_decode_loop; // yes, this makes no sense + } + len <<= 8; + len |= stbi__get8(s); + if (len != width) { STBI_FREE(hdr_data); STBI_FREE(scanline); return stbi__errpf("invalid decoded scanline length", "corrupt HDR"); } + if (scanline == NULL) { + scanline = (stbi_uc *) stbi__malloc_mad2(width, 4, 0); + if (!scanline) { + STBI_FREE(hdr_data); + return stbi__errpf("outofmem", "Out of memory"); + } + } + + for (k = 0; k < 4; ++k) { + int nleft; + i = 0; + while ((nleft = width - i) > 0) { + count = stbi__get8(s); + if (count > 128) { + // Run + value = stbi__get8(s); + count -= 128; + if ((count == 0) || (count > nleft)) { STBI_FREE(hdr_data); STBI_FREE(scanline); return stbi__errpf("corrupt", "bad RLE data in HDR"); } + for (z = 0; z < count; ++z) + scanline[i++ * 4 + k] = value; + } else { + // Dump + if ((count == 0) || (count > nleft)) { STBI_FREE(hdr_data); STBI_FREE(scanline); return stbi__errpf("corrupt", "bad RLE data in HDR"); } + for (z = 0; z < count; ++z) + scanline[i++ * 4 + k] = stbi__get8(s); + } + } + } + for (i=0; i < width; ++i) + stbi__hdr_convert(hdr_data+(j*width + i)*req_comp, scanline + i*4, req_comp); + } + if (scanline) + STBI_FREE(scanline); + } + + return hdr_data; +} + +static int stbi__hdr_info(stbi__context *s, int *x, int *y, int *comp) +{ + char buffer[STBI__HDR_BUFLEN]; + char *token; + int valid = 0; + int dummy; + + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; + + if (stbi__hdr_test(s) == 0) { + stbi__rewind( s ); + return 0; + } + + for(;;) { + token = stbi__hdr_gettoken(s,buffer); + if (token[0] == 0) break; + if (strcmp(token, "FORMAT=32-bit_rle_rgbe") == 0) valid = 1; + } + + if (!valid) { + stbi__rewind( s ); + return 0; + } + token = stbi__hdr_gettoken(s,buffer); + if (strncmp(token, "-Y ", 3)) { + stbi__rewind( s ); + return 0; + } + token += 3; + *y = (int) strtol(token, &token, 10); + while (*token == ' ') ++token; + if (strncmp(token, "+X ", 3)) { + stbi__rewind( s ); + return 0; + } + token += 3; + *x = (int) strtol(token, NULL, 10); + *comp = 3; + return 1; +} +#endif // STBI_NO_HDR + +#ifndef STBI_NO_BMP +static int stbi__bmp_info(stbi__context *s, int *x, int *y, int *comp) +{ + void *p; + stbi__bmp_data info; + + info.all_a = 255; + p = stbi__bmp_parse_header(s, &info); + if (p == NULL) { + stbi__rewind( s ); + return 0; + } + if (x) *x = s->img_x; + if (y) *y = s->img_y; + if (comp) { + if (info.bpp == 24 && info.ma == 0xff000000) + *comp = 3; + else + *comp = info.ma ? 4 : 3; + } + return 1; +} +#endif + +#ifndef STBI_NO_PSD +static int stbi__psd_info(stbi__context *s, int *x, int *y, int *comp) +{ + int channelCount, dummy, depth; + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; + if (stbi__get32be(s) != 0x38425053) { + stbi__rewind( s ); + return 0; + } + if (stbi__get16be(s) != 1) { + stbi__rewind( s ); + return 0; + } + stbi__skip(s, 6); + channelCount = stbi__get16be(s); + if (channelCount < 0 || channelCount > 16) { + stbi__rewind( s ); + return 0; + } + *y = stbi__get32be(s); + *x = stbi__get32be(s); + depth = stbi__get16be(s); + if (depth != 8 && depth != 16) { + stbi__rewind( s ); + return 0; + } + if (stbi__get16be(s) != 3) { + stbi__rewind( s ); + return 0; + } + *comp = 4; + return 1; +} + +static int stbi__psd_is16(stbi__context *s) +{ + int channelCount, depth; + if (stbi__get32be(s) != 0x38425053) { + stbi__rewind( s ); + return 0; + } + if (stbi__get16be(s) != 1) { + stbi__rewind( s ); + return 0; + } + stbi__skip(s, 6); + channelCount = stbi__get16be(s); + if (channelCount < 0 || channelCount > 16) { + stbi__rewind( s ); + return 0; + } + STBI_NOTUSED(stbi__get32be(s)); + STBI_NOTUSED(stbi__get32be(s)); + depth = stbi__get16be(s); + if (depth != 16) { + stbi__rewind( s ); + return 0; + } + return 1; +} +#endif + +#ifndef STBI_NO_PIC +static int stbi__pic_info(stbi__context *s, int *x, int *y, int *comp) +{ + int act_comp=0,num_packets=0,chained,dummy; + stbi__pic_packet packets[10]; + + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; + + if (!stbi__pic_is4(s,"\x53\x80\xF6\x34")) { + stbi__rewind(s); + return 0; + } + + stbi__skip(s, 88); + + *x = stbi__get16be(s); + *y = stbi__get16be(s); + if (stbi__at_eof(s)) { + stbi__rewind( s); + return 0; + } + if ( (*x) != 0 && (1 << 28) / (*x) < (*y)) { + stbi__rewind( s ); + return 0; + } + + stbi__skip(s, 8); + + do { + stbi__pic_packet *packet; + + if (num_packets==sizeof(packets)/sizeof(packets[0])) + return 0; + + packet = &packets[num_packets++]; + chained = stbi__get8(s); + packet->size = stbi__get8(s); + packet->type = stbi__get8(s); + packet->channel = stbi__get8(s); + act_comp |= packet->channel; + + if (stbi__at_eof(s)) { + stbi__rewind( s ); + return 0; + } + if (packet->size != 8) { + stbi__rewind( s ); + return 0; + } + } while (chained); + + *comp = (act_comp & 0x10 ? 4 : 3); + + return 1; +} +#endif + +// ************************************************************************************************* +// Portable Gray Map and Portable Pixel Map loader +// by Ken Miller +// +// PGM: http://netpbm.sourceforge.net/doc/pgm.html +// PPM: http://netpbm.sourceforge.net/doc/ppm.html +// +// Known limitations: +// Does not support comments in the header section +// Does not support ASCII image data (formats P2 and P3) + +#ifndef STBI_NO_PNM + +static int stbi__pnm_test(stbi__context *s) +{ + char p, t; + p = (char) stbi__get8(s); + t = (char) stbi__get8(s); + if (p != 'P' || (t != '5' && t != '6')) { + stbi__rewind( s ); + return 0; + } + return 1; +} + +static void *stbi__pnm_load(stbi__context *s, int *x, int *y, int *comp, int req_comp, stbi__result_info *ri) +{ + stbi_uc *out; + STBI_NOTUSED(ri); + + ri->bits_per_channel = stbi__pnm_info(s, (int *)&s->img_x, (int *)&s->img_y, (int *)&s->img_n); + if (ri->bits_per_channel == 0) + return 0; + + if (s->img_y > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + if (s->img_x > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)"); + + *x = s->img_x; + *y = s->img_y; + if (comp) *comp = s->img_n; + + if (!stbi__mad4sizes_valid(s->img_n, s->img_x, s->img_y, ri->bits_per_channel / 8, 0)) + return stbi__errpuc("too large", "PNM too large"); + + out = (stbi_uc *) stbi__malloc_mad4(s->img_n, s->img_x, s->img_y, ri->bits_per_channel / 8, 0); + if (!out) return stbi__errpuc("outofmem", "Out of memory"); + if (!stbi__getn(s, out, s->img_n * s->img_x * s->img_y * (ri->bits_per_channel / 8))) { + STBI_FREE(out); + return stbi__errpuc("bad PNM", "PNM file truncated"); + } + + if (req_comp && req_comp != s->img_n) { + if (ri->bits_per_channel == 16) { + out = (stbi_uc *) stbi__convert_format16((stbi__uint16 *) out, s->img_n, req_comp, s->img_x, s->img_y); + } else { + out = stbi__convert_format(out, s->img_n, req_comp, s->img_x, s->img_y); + } + if (out == NULL) return out; // stbi__convert_format frees input on failure + } + return out; +} + +static int stbi__pnm_isspace(char c) +{ + return c == ' ' || c == '\t' || c == '\n' || c == '\v' || c == '\f' || c == '\r'; +} + +static void stbi__pnm_skip_whitespace(stbi__context *s, char *c) +{ + for (;;) { + while (!stbi__at_eof(s) && stbi__pnm_isspace(*c)) + *c = (char) stbi__get8(s); + + if (stbi__at_eof(s) || *c != '#') + break; + + while (!stbi__at_eof(s) && *c != '\n' && *c != '\r' ) + *c = (char) stbi__get8(s); + } +} + +static int stbi__pnm_isdigit(char c) +{ + return c >= '0' && c <= '9'; +} + +static int stbi__pnm_getinteger(stbi__context *s, char *c) +{ + int value = 0; + + while (!stbi__at_eof(s) && stbi__pnm_isdigit(*c)) { + value = value*10 + (*c - '0'); + *c = (char) stbi__get8(s); + if((value > 214748364) || (value == 214748364 && *c > '7')) + return stbi__err("integer parse overflow", "Parsing an integer in the PPM header overflowed a 32-bit int"); + } + + return value; +} + +static int stbi__pnm_info(stbi__context *s, int *x, int *y, int *comp) +{ + int maxv, dummy; + char c, p, t; + + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; + + stbi__rewind(s); + + // Get identifier + p = (char) stbi__get8(s); + t = (char) stbi__get8(s); + if (p != 'P' || (t != '5' && t != '6')) { + stbi__rewind(s); + return 0; + } + + *comp = (t == '6') ? 3 : 1; // '5' is 1-component .pgm; '6' is 3-component .ppm + + c = (char) stbi__get8(s); + stbi__pnm_skip_whitespace(s, &c); + + *x = stbi__pnm_getinteger(s, &c); // read width + if(*x == 0) + return stbi__err("invalid width", "PPM image header had zero or overflowing width"); + stbi__pnm_skip_whitespace(s, &c); + + *y = stbi__pnm_getinteger(s, &c); // read height + if (*y == 0) + return stbi__err("invalid width", "PPM image header had zero or overflowing width"); + stbi__pnm_skip_whitespace(s, &c); + + maxv = stbi__pnm_getinteger(s, &c); // read max value + if (maxv > 65535) + return stbi__err("max value > 65535", "PPM image supports only 8-bit and 16-bit images"); + else if (maxv > 255) + return 16; + else + return 8; +} + +static int stbi__pnm_is16(stbi__context *s) +{ + if (stbi__pnm_info(s, NULL, NULL, NULL) == 16) + return 1; + return 0; +} +#endif + +static int stbi__info_main(stbi__context *s, int *x, int *y, int *comp) +{ + #ifndef STBI_NO_JPEG + if (stbi__jpeg_info(s, x, y, comp)) return 1; + #endif + + #ifndef STBI_NO_PNG + if (stbi__png_info(s, x, y, comp)) return 1; + #endif + + #ifndef STBI_NO_GIF + if (stbi__gif_info(s, x, y, comp)) return 1; + #endif + + #ifndef STBI_NO_BMP + if (stbi__bmp_info(s, x, y, comp)) return 1; + #endif + + #ifndef STBI_NO_PSD + if (stbi__psd_info(s, x, y, comp)) return 1; + #endif + + #ifndef STBI_NO_PIC + if (stbi__pic_info(s, x, y, comp)) return 1; + #endif + + #ifndef STBI_NO_PNM + if (stbi__pnm_info(s, x, y, comp)) return 1; + #endif + + #ifndef STBI_NO_HDR + if (stbi__hdr_info(s, x, y, comp)) return 1; + #endif + + // test tga last because it's a crappy test! + #ifndef STBI_NO_TGA + if (stbi__tga_info(s, x, y, comp)) + return 1; + #endif + return stbi__err("unknown image type", "Image not of any known type, or corrupt"); +} + +static int stbi__is_16_main(stbi__context *s) +{ + #ifndef STBI_NO_PNG + if (stbi__png_is16(s)) return 1; + #endif + + #ifndef STBI_NO_PSD + if (stbi__psd_is16(s)) return 1; + #endif + + #ifndef STBI_NO_PNM + if (stbi__pnm_is16(s)) return 1; + #endif + return 0; +} + +#ifndef STBI_NO_STDIO +STBIDEF int stbi_info(char const *filename, int *x, int *y, int *comp) +{ + FILE *f = stbi__fopen(filename, "rb"); + int result; + if (!f) return stbi__err("can't fopen", "Unable to open file"); + result = stbi_info_from_file(f, x, y, comp); + fclose(f); + return result; +} + +STBIDEF int stbi_info_from_file(FILE *f, int *x, int *y, int *comp) +{ + int r; + stbi__context s; + long pos = ftell(f); + stbi__start_file(&s, f); + r = stbi__info_main(&s,x,y,comp); + fseek(f,pos,SEEK_SET); + return r; +} + +STBIDEF int stbi_is_16_bit(char const *filename) +{ + FILE *f = stbi__fopen(filename, "rb"); + int result; + if (!f) return stbi__err("can't fopen", "Unable to open file"); + result = stbi_is_16_bit_from_file(f); + fclose(f); + return result; +} + +STBIDEF int stbi_is_16_bit_from_file(FILE *f) +{ + int r; + stbi__context s; + long pos = ftell(f); + stbi__start_file(&s, f); + r = stbi__is_16_main(&s); + fseek(f,pos,SEEK_SET); + return r; +} +#endif // !STBI_NO_STDIO + +STBIDEF int stbi_info_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp) +{ + stbi__context s; + stbi__start_mem(&s,buffer,len); + return stbi__info_main(&s,x,y,comp); +} + +STBIDEF int stbi_info_from_callbacks(stbi_io_callbacks const *c, void *user, int *x, int *y, int *comp) +{ + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *) c, user); + return stbi__info_main(&s,x,y,comp); +} + +STBIDEF int stbi_is_16_bit_from_memory(stbi_uc const *buffer, int len) +{ + stbi__context s; + stbi__start_mem(&s,buffer,len); + return stbi__is_16_main(&s); +} + +STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const *c, void *user) +{ + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *) c, user); + return stbi__is_16_main(&s); +} + +#endif // STB_IMAGE_IMPLEMENTATION + +/* + revision history: + 2.20 (2019-02-07) support utf8 filenames in Windows; fix warnings and platform ifdefs + 2.19 (2018-02-11) fix warning + 2.18 (2018-01-30) fix warnings + 2.17 (2018-01-29) change sbti__shiftsigned to avoid clang -O2 bug + 1-bit BMP + *_is_16_bit api + avoid warnings + 2.16 (2017-07-23) all functions have 16-bit variants; + STBI_NO_STDIO works again; + compilation fixes; + fix rounding in unpremultiply; + optimize vertical flip; + disable raw_len validation; + documentation fixes + 2.15 (2017-03-18) fix png-1,2,4 bug; now all Imagenet JPGs decode; + warning fixes; disable run-time SSE detection on gcc; + uniform handling of optional "return" values; + thread-safe initialization of zlib tables + 2.14 (2017-03-03) remove deprecated STBI_JPEG_OLD; fixes for Imagenet JPGs + 2.13 (2016-11-29) add 16-bit API, only supported for PNG right now + 2.12 (2016-04-02) fix typo in 2.11 PSD fix that caused crashes + 2.11 (2016-04-02) allocate large structures on the stack + remove white matting for transparent PSD + fix reported channel count for PNG & BMP + re-enable SSE2 in non-gcc 64-bit + support RGB-formatted JPEG + read 16-bit PNGs (only as 8-bit) + 2.10 (2016-01-22) avoid warning introduced in 2.09 by STBI_REALLOC_SIZED + 2.09 (2016-01-16) allow comments in PNM files + 16-bit-per-pixel TGA (not bit-per-component) + info() for TGA could break due to .hdr handling + info() for BMP to shares code instead of sloppy parse + can use STBI_REALLOC_SIZED if allocator doesn't support realloc + code cleanup + 2.08 (2015-09-13) fix to 2.07 cleanup, reading RGB PSD as RGBA + 2.07 (2015-09-13) fix compiler warnings + partial animated GIF support + limited 16-bpc PSD support + #ifdef unused functions + bug with < 92 byte PIC,PNM,HDR,TGA + 2.06 (2015-04-19) fix bug where PSD returns wrong '*comp' value + 2.05 (2015-04-19) fix bug in progressive JPEG handling, fix warning + 2.04 (2015-04-15) try to re-enable SIMD on MinGW 64-bit + 2.03 (2015-04-12) extra corruption checking (mmozeiko) + stbi_set_flip_vertically_on_load (nguillemot) + fix NEON support; fix mingw support + 2.02 (2015-01-19) fix incorrect assert, fix warning + 2.01 (2015-01-17) fix various warnings; suppress SIMD on gcc 32-bit without -msse2 + 2.00b (2014-12-25) fix STBI_MALLOC in progressive JPEG + 2.00 (2014-12-25) optimize JPG, including x86 SSE2 & NEON SIMD (ryg) + progressive JPEG (stb) + PGM/PPM support (Ken Miller) + STBI_MALLOC,STBI_REALLOC,STBI_FREE + GIF bugfix -- seemingly never worked + STBI_NO_*, STBI_ONLY_* + 1.48 (2014-12-14) fix incorrectly-named assert() + 1.47 (2014-12-14) 1/2/4-bit PNG support, both direct and paletted (Omar Cornut & stb) + optimize PNG (ryg) + fix bug in interlaced PNG with user-specified channel count (stb) + 1.46 (2014-08-26) + fix broken tRNS chunk (colorkey-style transparency) in non-paletted PNG + 1.45 (2014-08-16) + fix MSVC-ARM internal compiler error by wrapping malloc + 1.44 (2014-08-07) + various warning fixes from Ronny Chevalier + 1.43 (2014-07-15) + fix MSVC-only compiler problem in code changed in 1.42 + 1.42 (2014-07-09) + don't define _CRT_SECURE_NO_WARNINGS (affects user code) + fixes to stbi__cleanup_jpeg path + added STBI_ASSERT to avoid requiring assert.h + 1.41 (2014-06-25) + fix search&replace from 1.36 that messed up comments/error messages + 1.40 (2014-06-22) + fix gcc struct-initialization warning + 1.39 (2014-06-15) + fix to TGA optimization when req_comp != number of components in TGA; + fix to GIF loading because BMP wasn't rewinding (whoops, no GIFs in my test suite) + add support for BMP version 5 (more ignored fields) + 1.38 (2014-06-06) + suppress MSVC warnings on integer casts truncating values + fix accidental rename of 'skip' field of I/O + 1.37 (2014-06-04) + remove duplicate typedef + 1.36 (2014-06-03) + convert to header file single-file library + if de-iphone isn't set, load iphone images color-swapped instead of returning NULL + 1.35 (2014-05-27) + various warnings + fix broken STBI_SIMD path + fix bug where stbi_load_from_file no longer left file pointer in correct place + fix broken non-easy path for 32-bit BMP (possibly never used) + TGA optimization by Arseny Kapoulkine + 1.34 (unknown) + use STBI_NOTUSED in stbi__resample_row_generic(), fix one more leak in tga failure case + 1.33 (2011-07-14) + make stbi_is_hdr work in STBI_NO_HDR (as specified), minor compiler-friendly improvements + 1.32 (2011-07-13) + support for "info" function for all supported filetypes (SpartanJ) + 1.31 (2011-06-20) + a few more leak fixes, bug in PNG handling (SpartanJ) + 1.30 (2011-06-11) + added ability to load files via callbacks to accomidate custom input streams (Ben Wenger) + removed deprecated format-specific test/load functions + removed support for installable file formats (stbi_loader) -- would have been broken for IO callbacks anyway + error cases in bmp and tga give messages and don't leak (Raymond Barbiero, grisha) + fix inefficiency in decoding 32-bit BMP (David Woo) + 1.29 (2010-08-16) + various warning fixes from Aurelien Pocheville + 1.28 (2010-08-01) + fix bug in GIF palette transparency (SpartanJ) + 1.27 (2010-08-01) + cast-to-stbi_uc to fix warnings + 1.26 (2010-07-24) + fix bug in file buffering for PNG reported by SpartanJ + 1.25 (2010-07-17) + refix trans_data warning (Won Chun) + 1.24 (2010-07-12) + perf improvements reading from files on platforms with lock-heavy fgetc() + minor perf improvements for jpeg + deprecated type-specific functions so we'll get feedback if they're needed + attempt to fix trans_data warning (Won Chun) + 1.23 fixed bug in iPhone support + 1.22 (2010-07-10) + removed image *writing* support + stbi_info support from Jetro Lauha + GIF support from Jean-Marc Lienher + iPhone PNG-extensions from James Brown + warning-fixes from Nicolas Schulz and Janez Zemva (i.stbi__err. Janez (U+017D)emva) + 1.21 fix use of 'stbi_uc' in header (reported by jon blow) + 1.20 added support for Softimage PIC, by Tom Seddon + 1.19 bug in interlaced PNG corruption check (found by ryg) + 1.18 (2008-08-02) + fix a threading bug (local mutable static) + 1.17 support interlaced PNG + 1.16 major bugfix - stbi__convert_format converted one too many pixels + 1.15 initialize some fields for thread safety + 1.14 fix threadsafe conversion bug + header-file-only version (#define STBI_HEADER_FILE_ONLY before including) + 1.13 threadsafe + 1.12 const qualifiers in the API + 1.11 Support installable IDCT, colorspace conversion routines + 1.10 Fixes for 64-bit (don't use "unsigned long") + optimized upsampling by Fabian "ryg" Giesen + 1.09 Fix format-conversion for PSD code (bad global variables!) + 1.08 Thatcher Ulrich's PSD code integrated by Nicolas Schulz + 1.07 attempt to fix C++ warning/errors again + 1.06 attempt to fix C++ warning/errors again + 1.05 fix TGA loading to return correct *comp and use good luminance calc + 1.04 default float alpha is 1, not 255; use 'void *' for stbi_image_free + 1.03 bugfixes to STBI_NO_STDIO, STBI_NO_HDR + 1.02 support for (subset of) HDR files, float interface for preferred access to them + 1.01 fix bug: possible bug in handling right-side up bmps... not sure + fix bug: the stbi__bmp_load() and stbi__tga_load() functions didn't work at all + 1.00 interface to zlib that skips zlib header + 0.99 correct handling of alpha in palette + 0.98 TGA loader by lonesock; dynamically add loaders (untested) + 0.97 jpeg errors on too large a file; also catch another malloc failure + 0.96 fix detection of invalid v value - particleman@mollyrocket forum + 0.95 during header scan, seek to markers in case of padding + 0.94 STBI_NO_STDIO to disable stdio usage; rename all #defines the same + 0.93 handle jpegtran output; verbose errors + 0.92 read 4,8,16,24,32-bit BMP files of several formats + 0.91 output 24-bit Windows 3.0 BMP files + 0.90 fix a few more warnings; bump version number to approach 1.0 + 0.61 bugfixes due to Marc LeBlanc, Christopher Lloyd + 0.60 fix compiling as c++ + 0.59 fix warnings: merge Dave Moore's -Wall fixes + 0.58 fix bug: zlib uncompressed mode len/nlen was wrong endian + 0.57 fix bug: jpg last huffman symbol before marker was >9 bits but less than 16 available + 0.56 fix bug: zlib uncompressed mode len vs. nlen + 0.55 fix bug: restart_interval not initialized to 0 + 0.54 allow NULL for 'int *comp' + 0.53 fix bug in png 3->4; speedup png decoding + 0.52 png handles req_comp=3,4 directly; minor cleanup; jpeg comments + 0.51 obey req_comp requests, 1-component jpegs return as 1-component, + on 'test' only check type, not whether we support this variant + 0.50 (2006-11-19) + first released version +*/ + + +/* +------------------------------------------------------------------------------ +This software is available under 2 licenses -- choose whichever you prefer. +------------------------------------------------------------------------------ +ALTERNATIVE A - MIT License +Copyright (c) 2017 Sean Barrett +Permission is hereby granted, free of charge, to any person obtaining a copy of +this software and associated documentation files (the "Software"), to deal in +the Software without restriction, including without limitation the rights to +use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies +of the Software, and to permit persons to whom the Software is furnished to do +so, subject to the following conditions: +The above copyright notice and this permission notice shall be included in all +copies or substantial portions of the Software. +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +SOFTWARE. +------------------------------------------------------------------------------ +ALTERNATIVE B - Public Domain (www.unlicense.org) +This is free and unencumbered software released into the public domain. +Anyone is free to copy, modify, publish, use, compile, sell, or distribute this +software, either in source code form or as a compiled binary, for any purpose, +commercial or non-commercial, and by any means. +In jurisdictions that recognize copyright laws, the author or authors of this +software dedicate any and all copyright interest in the software to the public +domain. We make this dedication for the benefit of the public at large and to +the detriment of our heirs and successors. We intend this dedication to be an +overt act of relinquishment in perpetuity of all present and future rights to +this software under copyright law. +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN +ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION +WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. +------------------------------------------------------------------------------ +*/ diff --git a/Spear/old/Spear/Assets/Image.hsc b/Spear/old/Spear/Assets/Image.hsc new file mode 100644 index 0000000..db90afe --- /dev/null +++ b/Spear/old/Spear/Assets/Image.hsc @@ -0,0 +1,126 @@ +{-# LANGUAGE CPP, ForeignFunctionInterface #-} + +module Spear.Assets.Image +( + -- * Data types + Image + -- * Loading and unloading +, loadImage + -- * Accessors +, width +, height +, bpp +, pixels +) +where + +import Spear.Game +import Foreign.Ptr +import Foreign.Storable +import Foreign.C.Types +import Foreign.C.String +import Foreign.Marshal.Utils as Foreign (with) +import Foreign.Marshal.Alloc (alloca) +import Data.List (splitAt, elemIndex) +import Data.Char (toLower) + +#include "Image.h" +#include "BMP/BMP_load.h" + +data ImageErrorCode + = ImageSuccess + | ImageReadError + | ImageMemoryAllocationError + | ImageFileNotFound + | ImageInvalidFormat + | ImageNoSuitableLoader + deriving (Eq, Enum, Show) + +data CImage = CImage + { cwidth :: CInt + , cheight :: CInt + , cbpp :: CInt + , cpixels :: Ptr CUChar + } + +instance Storable CImage where + sizeOf _ = #{size Image} + alignment _ = alignment (undefined :: CInt) + + peek ptr = do + width <- #{peek Image, width} ptr + height <- #{peek Image, height} ptr + bpp <- #{peek Image, bpp} ptr + pixels <- #{peek Image, pixels} ptr + return $ CImage width height bpp pixels + + poke ptr (CImage width height bpp pixels) = do + #{poke Image, width} ptr width + #{poke Image, height} ptr height + #{poke Image, bpp} ptr bpp + #{poke Image, pixels} ptr pixels + +-- | An image resource. +data Image = Image + { imageData :: CImage + , rkey :: ReleaseKey + } + +instance ResourceClass Image where + getResource = rkey + +foreign import ccall "Image.h image_free" + image_free :: Ptr CImage -> IO () + +foreign import ccall "BMP_load.h BMP_load" + bmp_load' :: Ptr CChar -> Ptr CImage -> IO Int + +bmp_load :: Ptr CChar -> Ptr CImage -> IO ImageErrorCode +bmp_load file image = bmp_load' file image >>= \code -> return . toEnum $ code + +-- | Load the image specified by the given file. +loadImage :: FilePath -> Game s Image +loadImage file = do + dotPos <- case elemIndex '.' file of + Nothing -> gameError $ "file name has no extension: " ++ file + Just p -> return p + + let ext = map toLower . tail . snd $ splitAt dotPos file + + result <- liftIO . alloca $ \ptr -> do + status <- withCString file $ \fileCstr -> do + case ext of + "bmp" -> bmp_load fileCstr ptr + _ -> return ImageNoSuitableLoader + + case status of + ImageSuccess -> peek ptr >>= return . Right + ImageReadError -> return . Left $ "read error" + ImageMemoryAllocationError -> return . Left $ "memory allocation error" + ImageFileNotFound -> return . Left $ "file not found" + ImageInvalidFormat -> return . Left $ "invalid format" + ImageNoSuitableLoader -> return . Left $ "no suitable loader for extension " ++ ext + + case result of + Right image -> register (freeImage image) >>= return . Image image + Left err -> gameError $ "loadImage: " ++ err + +-- | Free the given 'CImage'. +freeImage :: CImage -> IO () +freeImage image = Foreign.with image image_free + +-- | Return the given image's width. +width :: Image -> Int +width = fromIntegral . cwidth . imageData + +-- | Return the given image's height. +height :: Image -> Int +height = fromIntegral . cheight . imageData + +-- | Return the given image's bits per pixel. +bpp :: Image -> Int +bpp = fromIntegral . cbpp . imageData + +-- | Return the given image's pixels. +pixels :: Image -> Ptr CUChar +pixels = cpixels . imageData diff --git a/Spear/old/Spear/Assets/Image/BMP/BMP_load.c b/Spear/old/Spear/Assets/Image/BMP/BMP_load.c new file mode 100644 index 0000000..5c1b195 --- /dev/null +++ b/Spear/old/Spear/Assets/Image/BMP/BMP_load.c @@ -0,0 +1,257 @@ +#include "BMP_load.h" +#include +#include +#include +#include + +#define BITMAP_ID 0x4D42 + + +/// Bitmap file header structure. +typedef struct +{ + U16 type; // Specifies the image type; must be BM (0x4D42). + U32 size; // Specifies the size in bytes of the bitmap file. + U32 reserved; // Reserved; must be zero. + U32 offBits; // Specifies the offset, in bytes, from the BitmapFileHeader structure to the bitmap bits. +} +BitmapFileHeader; + + +/// Bitmap info header structure. +typedef struct +{ + U32 size; // Specifies the number of bytes required by the structure. + U32 width; // Specifies the width of the bitmap, in pixels. + U32 height; // Specifies the height of the bitmap, in pixels. + U16 planes; // Specifies the number of color planes; must be 1. + U16 bitCount; // Specifies the number of bits per pixel; must be 1, 4, 16, 24, or 32. + U32 compression; // Specifies the type of compression. + U32 imageSize; // Specifies the size of the image in bytes. + U32 xPelsPerMeter; // Specifies the number of pixels per meter on the x axis. + U32 yPelsPerMeter; // Specifies the number of pixels per meter on the y axis. + U32 clrUsed; // Specifies the number of colours used by the bitmap. + U32 clrImportant; // Specifies the number of colours that are important. +} +BitmapInfoHeader; + + +static void safe_free (void* ptr) +{ + if (ptr) free (ptr); +} + + +static Image_error_code read_raw_data( + FILE* filePtr, const BitmapFileHeader* bitmapFileHeader, + const BitmapInfoHeader* bitmapInfoHeader, U8** data) +{ + U8* bitmapImage; + U8* auxrow; + size_t row_size = bitmapInfoHeader->width * 3; + size_t numBytes = bitmapInfoHeader->height * row_size; + size_t bytes_read; + + // Allocate memory for the bitmap data and the auxiliary row. + bitmapImage = (U8*) malloc (numBytes); + auxrow = (U8*) malloc (row_size); + if (!bitmapImage || !auxrow) + { + safe_free (bitmapImage); + safe_free (auxrow); + return Image_Memory_Allocation_Error; + } + + // Move the file pointer to the beginning of bitmap data and read the data. + fseek(filePtr, bitmapFileHeader->offBits, SEEK_SET); + bytes_read = fread(bitmapImage, 1, numBytes, filePtr); + if (bytes_read != numBytes) + { + free(bitmapImage); + return Image_Read_Error; + } + + size_t i; + + // Reverse rows. + /*size_t h = bitmapInfoHeader->height / 2; + for (i = 0; i < h; ++i) + { + U8* row1 = bitmapImage + i * row_size; + U8* row2 = bitmapImage + (bitmapInfoHeader->height - i - 1) * row_size; + + memcpy (auxrow, row1, row_size); + memcpy (row1, row2, row_size); + memcpy (row2, auxrow, row_size); + }*/ + + // Swap B and R channels. BGR -> RGB. + for (i = 0; i < bytes_read; i+= 3) + { + U8 tmp = bitmapImage[i]; + bitmapImage[i] = bitmapImage[i+2]; + bitmapImage[i+2] = tmp; + } + + *data = bitmapImage; + + return Image_Success; +} + + +static Image_error_code read_paletised_data8 + (FILE* filePtr, const BitmapInfoHeader *bitmapInfoHeader, U8** data) +{ + U8* bitmapImage; + U8* palette; + U8* colourIndices; + size_t bytes_read; + + size_t paletteSize = pow(2, bitmapInfoHeader->bitCount) * 4; + size_t colourIndicesSize = bitmapInfoHeader->width * bitmapInfoHeader->height; + int bitmapImageSize = colourIndicesSize * 3; + + // Save memory for the palette, colour indices and bitmap image. + palette = (U8*) malloc (paletteSize); + colourIndices = (U8*) malloc(colourIndicesSize); + bitmapImage = (U8*) malloc(bitmapImageSize); + if (!palette | !colourIndices || !bitmapImage) + { + safe_free (palette); + safe_free (colourIndices); + safe_free (bitmapImage); + return Image_Memory_Allocation_Error; + } + + // Read the colour palette. + bytes_read = fread(palette, 1, paletteSize, filePtr); + if (bytes_read != paletteSize) return Image_Read_Error; + + // Read the colour indices. + bytes_read = fread(colourIndices, 1, colourIndicesSize, filePtr); + if (bytes_read != colourIndicesSize) return Image_Read_Error; + + // Decode the image data. + U8* imgptr = &bitmapImage[bitmapImageSize - (bitmapInfoHeader->width * 4)]; + + size_t i; + for (i = 0; i < colourIndicesSize; i++) + { + int index = colourIndices[i]; + + memcpy(imgptr, (const void*) &palette[index * 4], 3); + imgptr += 3; + + if (!((i+1) % bitmapInfoHeader->width)) + { + imgptr -= (bitmapInfoHeader->width * 4 * 2); + } + } + + free(palette); + free(colourIndices); + + *data = bitmapImage; + + return Image_Success; +} + + +Image_error_code BMP_load (const char* filename, Image* image) +{ + FILE* filePtr; + BitmapFileHeader bitmapFileHeader; + BitmapInfoHeader bitmapInfoHeader; + U8 buf[40]; + U8* data; + + // Open the file in binary read-only mode. + filePtr = fopen(filename, "rb"); + if (!filePtr) return Image_File_Not_Found; + + if ((fread(buf, 14, 1, filePtr)) != 1) + { + fclose(filePtr); + return Image_Read_Error; + } + + bitmapFileHeader.type = *((U16*)buf); + bitmapFileHeader.size = *((U32*)(buf+2)); + bitmapFileHeader.reserved = *((U32*)(buf+6)); + bitmapFileHeader.offBits = *((U32*)(buf+10)); + + // Check that this is in fact a BMP file. + if (bitmapFileHeader.type != BITMAP_ID) + { + fprintf(stderr, "Not a valid BMP file\n"); + fclose(filePtr); + return Image_Invalid_Format; + } + + if ((fread(&buf, 40, 1, filePtr)) != 1) + { + fclose(filePtr); + return Image_Read_Error; + } + + bitmapInfoHeader.size = *((U32*)(buf)); + bitmapInfoHeader.width = *((U32*)(buf+4)); + bitmapInfoHeader.height = *((U32*)(buf+8)); + bitmapInfoHeader.planes = *((U16*)(buf+12)); + bitmapInfoHeader.bitCount = *((U16*)(buf+14)); + bitmapInfoHeader.compression = *((U32*)(buf+16)); + bitmapInfoHeader.imageSize = *((U32*)(buf+20)); + bitmapInfoHeader.xPelsPerMeter = *((U32*)(buf+24)); + bitmapInfoHeader.yPelsPerMeter = *((U32*)(buf+28)); + bitmapInfoHeader.clrUsed = *((U32*)(buf+32)); + bitmapInfoHeader.clrImportant = *((U32*)(buf+36)); + + // Check that no compression is used. + // Compression is not supported at the moment. + if (bitmapInfoHeader.compression != 0) + { + fprintf(stderr, "Compression not supported\n"); + fclose(filePtr); + return Image_Invalid_Format; + } + + // Check that this is a Windows BMP file. + // Other formats are not supported. + if (bitmapInfoHeader.size != 40) + { + fprintf(stderr, "Only Windows BMP files supported\n"); + fclose(filePtr); + return Image_Invalid_Format; + } + + Image_error_code status; + + if (bitmapInfoHeader.bitCount == 8) + { + // The BMP file uses a colour palette. + // We are already positioned at the colour palette. + status = read_paletised_data8 (filePtr, &bitmapInfoHeader, &data); + } + else if (bitmapInfoHeader.bitCount >= 16) + { + // The BMP file uses no colour palette. + status = read_raw_data (filePtr, &bitmapFileHeader, &bitmapInfoHeader, &data); + } + else + { + fprintf(stderr, "Only 24-bit and 16-bit palette images supported\n"); + fclose(filePtr); + return Image_Invalid_Format; + } + + fclose(filePtr); + + if (data == 0) return status; + + image->width = bitmapInfoHeader.width; + image->height = bitmapInfoHeader.height; + image->bpp = 3; + image->pixels = data; + + return Image_Success; +} diff --git a/Spear/old/Spear/Assets/Image/BMP/BMP_load.h b/Spear/old/Spear/Assets/Image/BMP/BMP_load.h new file mode 100644 index 0000000..f4ad32f --- /dev/null +++ b/Spear/old/Spear/Assets/Image/BMP/BMP_load.h @@ -0,0 +1,23 @@ +#ifndef _BMP_LOAD_H +#define _BMP_LOAD_H + + +#include "../Image.h" +#include "../Image_error_code.h" + + +#ifdef __cplusplus +extern "C" { +#endif + + +/// Loads the BMP file specified by the given string. +/// (0,0) corresponds to the top left corner of the image. +Image_error_code BMP_load (const char* filename, Image* image); + + +#ifdef __cplusplus +} +#endif + +#endif // _BMP_LOAD_H diff --git a/Spear/old/Spear/Assets/Image/Image.c b/Spear/old/Spear/Assets/Image/Image.c new file mode 100644 index 0000000..f4150e1 --- /dev/null +++ b/Spear/old/Spear/Assets/Image/Image.c @@ -0,0 +1,8 @@ +#include "Image.h" +#include + + +void image_free (Image* image) +{ + free (image->pixels); +} diff --git a/Spear/old/Spear/Assets/Image/Image.h b/Spear/old/Spear/Assets/Image/Image.h new file mode 100644 index 0000000..aaca5e9 --- /dev/null +++ b/Spear/old/Spear/Assets/Image/Image.h @@ -0,0 +1,32 @@ +#ifndef _SPEAR_IMAGE_H +#define _SPEAR_IMAGE_H + +#include "sys_types.h" + + +typedef struct +{ + int width; + int height; + int bpp; // Bits per pixel. + // If bpp = 3 then format = RGB. + // If bpp = 4 then format = RGBA. + U8* pixels; +} +Image; + + +#ifdef __cplusplus +extern "C" { +#endif + +/// Frees the given Image from memory. +/// The 'image' pointer itself is not freed. +void image_free (Image* image); + +#ifdef __cplusplus +} +#endif + + +#endif // _SPEAR_IMAGE_H diff --git a/Spear/old/Spear/Assets/Image/Image_error_code.h b/Spear/old/Spear/Assets/Image/Image_error_code.h new file mode 100644 index 0000000..dc54fc2 --- /dev/null +++ b/Spear/old/Spear/Assets/Image/Image_error_code.h @@ -0,0 +1,15 @@ +#ifndef _SPEAR_IMAGE_ERROR_CODE_H +#define _SPEAR_IMAGE_ERROR_CODE_H + +typedef enum +{ + Image_Success, + Image_Read_Error, + Image_Memory_Allocation_Error, + Image_File_Not_Found, + Image_Invalid_Format, + Image_No_Suitable_Loader, +} +Image_error_code; + +#endif // _SPEAR_IMAGE_ERROR_CODE_H diff --git a/Spear/old/Spear/Assets/Image/sys_types.h b/Spear/old/Spear/Assets/Image/sys_types.h new file mode 100644 index 0000000..6aca9e9 --- /dev/null +++ b/Spear/old/Spear/Assets/Image/sys_types.h @@ -0,0 +1,16 @@ +#ifndef _SPEAR_SYS_TYPES_H +#define _SPEAR_SYS_TYPES_H + +#include + +typedef int8_t I8; +typedef int16_t I16; +typedef int32_t I32; +typedef int64_t I64; +typedef uint8_t U8; +typedef uint16_t U16; +typedef uint32_t U32; +typedef uint64_t U64; + +#endif // _SPEAR_SYS_TYPES_H + diff --git a/Spear/old/Spear/Assets/Model.hsc b/Spear/old/Spear/Assets/Model.hsc new file mode 100644 index 0000000..02e1edf --- /dev/null +++ b/Spear/old/Spear/Assets/Model.hsc @@ -0,0 +1,440 @@ +{-# LANGUAGE CPP, ForeignFunctionInterface #-} + +module Spear.Assets.Model +( + -- * Data types + Vec2(..) +, Vec3(..) +, TexCoord(..) +, CTriangle(..) +, Box(..) +, Skin(..) +, Animation(..) +, Triangle(..) +, Model(..) + -- * Loading +, loadModel + -- * Accessors +, animated +, animation +, animationByName +, triangles' + -- * Manipulation +, transformVerts +, transformNormals +, toGround +, modelBoxes +) +where + +import Spear.Game + +import qualified Data.ByteString.Char8 as B +import Data.Char (toLower) +import Data.List (splitAt, elemIndex) +import qualified Data.Vector as V +import qualified Data.Vector.Storable as S +import Foreign.Ptr +import Foreign.Storable +import Foreign.C.Types +import Foreign.C.String +import Foreign.Marshal.Utils as Foreign (with) +import Foreign.Marshal.Alloc (alloca, allocaBytes) +import Foreign.Marshal.Array (allocaArray, copyArray, peekArray) +import Unsafe.Coerce (unsafeCoerce) + +#include "Model.h" +#include "MD2/MD2_load.h" +#include "OBJ/OBJ_load.h" + +data ModelErrorCode + = ModelSuccess + | ModelReadError + | ModelMemoryAllocationError + | ModelFileNotFound + | ModelFileMismatch + | ModelNoSuitableLoader + deriving (Eq, Enum, Show) + +sizeFloat = #{size float} +sizePtr = #{size int*} + +-- | A 2D vector. +data Vec2 = Vec2 {-# UNPACK #-} !Float {-# UNPACK #-} !Float + +instance Storable Vec2 where + sizeOf _ = 2*sizeFloat + alignment _ = alignment (undefined :: CFloat) + + peek ptr = do + f0 <- peekByteOff ptr 0 + f1 <- peekByteOff ptr sizeFloat + return $ Vec2 f0 f1 + + poke ptr (Vec2 f0 f1) = do + pokeByteOff ptr 0 f0 + pokeByteOff ptr sizeFloat f1 + +-- | A 3D vector. +data Vec3 = Vec3 {-# UNPACK #-} !Float {-# UNPACK #-} !Float {-# UNPACK #-} !Float + +instance Storable Vec3 where + sizeOf _ = 3*sizeFloat + alignment _ = alignment (undefined :: CFloat) + + peek ptr = do + f0 <- peekByteOff ptr 0 + f1 <- peekByteOff ptr sizeFloat + f2 <- peekByteOff ptr (2*sizeFloat) + return $ Vec3 f0 f1 f2 + + poke ptr (Vec3 f0 f1 f2) = do + pokeByteOff ptr 0 f0 + pokeByteOff ptr sizeFloat f1 + pokeByteOff ptr (2*sizeFloat) f2 + +-- | A 2D texture coordinate. +data TexCoord = TexCoord {-# UNPACK #-} !Float {-# UNPACK #-} !Float + +instance Storable TexCoord where + sizeOf _ = 2*sizeFloat + alignment _ = alignment (undefined :: CFloat) + + peek ptr = do + f0 <- peekByteOff ptr 0 + f1 <- peekByteOff ptr sizeFloat + return $ TexCoord f0 f1 + + poke ptr (TexCoord f0 f1) = do + pokeByteOff ptr 0 f0 + pokeByteOff ptr sizeFloat f1 + +-- | A raw triangle holding vertex/normal and texture indices. +data CTriangle = CTriangle + { vertexIndex0 :: {-# UNPACK #-} !CUShort + , vertexIndex1 :: {-# UNPACK #-} !CUShort + , vertexIndex2 :: {-# UNPACK #-} !CUShort + , textureIndex1 :: {-# UNPACK #-} !CUShort + , textureIndex2 :: {-# UNPACK #-} !CUShort + , textureIndex3 :: {-# UNPACK #-} !CUShort + } + +instance Storable CTriangle where + sizeOf _ = #{size triangle} + alignment _ = alignment (undefined :: CUShort) + + peek ptr = do + v0 <- #{peek triangle, vertexIndices[0]} ptr + v1 <- #{peek triangle, vertexIndices[1]} ptr + v2 <- #{peek triangle, vertexIndices[2]} ptr + + t0 <- #{peek triangle, textureIndices[0]} ptr + t1 <- #{peek triangle, textureIndices[1]} ptr + t2 <- #{peek triangle, textureIndices[2]} ptr + + return $ CTriangle v0 v1 v2 t0 t1 t2 + + poke ptr (CTriangle v0 v1 v2 t0 t1 t2) = do + #{poke triangle, vertexIndices[0]} ptr v0 + #{poke triangle, vertexIndices[1]} ptr v1 + #{poke triangle, vertexIndices[2]} ptr v2 + + #{poke triangle, textureIndices[0]} ptr t0 + #{poke triangle, textureIndices[1]} ptr t1 + #{poke triangle, textureIndices[2]} ptr t2 + +-- | A 3D axis-aligned bounding box. +data Box = Box {-# UNPACK #-} !Vec3 {-# UNPACK #-} !Vec3 + +instance Storable Box where + sizeOf _ = 6 * sizeFloat + alignment _ = alignment (undefined :: CFloat) + + peek ptr = do + xmin <- peekByteOff ptr 0 + ymin <- peekByteOff ptr sizeFloat + zmin <- peekByteOff ptr $ 2*sizeFloat + xmax <- peekByteOff ptr $ 3*sizeFloat + ymax <- peekByteOff ptr $ 4*sizeFloat + zmax <- peekByteOff ptr $ 5*sizeFloat + return $ Box (Vec3 xmin ymin zmin) (Vec3 xmax ymax zmax) + + poke ptr (Box (Vec3 xmin ymin zmin) (Vec3 xmax ymax zmax)) = do + pokeByteOff ptr 0 xmin + pokeByteOff ptr sizeFloat ymin + pokeByteOff ptr (2*sizeFloat) zmin + pokeByteOff ptr (3*sizeFloat) xmax + pokeByteOff ptr (4*sizeFloat) ymax + pokeByteOff ptr (5*sizeFloat) zmax + +-- | A model skin. +newtype Skin = Skin { skinName :: B.ByteString } + +instance Storable Skin where + sizeOf (Skin s) = 64 + alignment _ = 1 + + peek ptr = do + s <- B.packCString $ unsafeCoerce ptr + return $ Skin s + + poke ptr (Skin s) = do + B.useAsCStringLen s $ \(sptr, len) -> copyArray (unsafeCoerce ptr) sptr len + +-- | A model animation. +-- +-- See also: 'animation', 'animationByName', 'numAnimations'. +data Animation = Animation + { name :: B.ByteString + , start :: Int + , end :: Int + } + +instance Storable Animation where + sizeOf _ = #{size animation} + alignment _ = alignment (undefined :: CUInt) + + peek ptr = do + name <- B.packCString (unsafeCoerce ptr) + start <- #{peek animation, start} ptr + end <- #{peek animation, end} ptr + return $ Animation name start end + + poke ptr (Animation name start end) = do + B.useAsCStringLen name $ \(sptr, len) -> copyArray (unsafeCoerce ptr) sptr len + #{poke animation, start} ptr start + #{poke animation, end} ptr end + +-- | A 3D model. +data Model = Model + { vertices :: S.Vector Vec3 -- ^ Array of 'numFrames' * 'numVerts' vertices. + , normals :: S.Vector Vec3 -- ^ Array of 'numFrames' * 'numVerts' normals. + , texCoords :: S.Vector TexCoord -- ^ Array of 'numTexCoords' texture coordinates. + , triangles :: S.Vector CTriangle -- ^ Array of 'numTriangles' triangles. + , skins :: S.Vector Skin -- ^ Array of 'numSkins' skins. + , animations :: S.Vector Animation -- ^ Array of 'numAnimations' animations. + , numFrames :: Int -- ^ Number of frames. + , numVerts :: Int -- ^ Number of vertices (and normals) per frame. + , numTriangles :: Int -- ^ Number of triangles in one frame. + , numTexCoords :: Int -- ^ Number of texture coordinates in one frame. + , numSkins :: Int -- ^ Number of skins. + , numAnimations :: Int -- ^ Number of animations. + } + +instance Storable Model where + sizeOf _ = #{size Model} + alignment _ = alignment (undefined :: CUInt) + + peek ptr = do + numFrames <- #{peek Model, numFrames} ptr + numVertices <- #{peek Model, numVertices} ptr + numTriangles <- #{peek Model, numTriangles} ptr + numTexCoords <- #{peek Model, numTexCoords} ptr + numSkins <- #{peek Model, numSkins} ptr + numAnimations <- #{peek Model, numAnimations} ptr + pVerts <- peek (unsafeCoerce ptr) + pNormals <- peekByteOff ptr sizePtr + pTexCoords <- peekByteOff ptr (2*sizePtr) + pTriangles <- peekByteOff ptr (3*sizePtr) + pSkins <- peekByteOff ptr (4*sizePtr) + pAnimations <- peekByteOff ptr (5*sizePtr) + vertices <- fmap S.fromList $ peekArray (numVertices*numFrames) pVerts + normals <- fmap S.fromList $ peekArray (numVertices*numFrames) pNormals + texCoords <- fmap S.fromList $ peekArray numTexCoords pTexCoords + triangles <- fmap S.fromList $ peekArray numTriangles pTriangles + skins <- fmap S.fromList $ peekArray numSkins pSkins + animations <- fmap S.fromList $ peekArray numAnimations pAnimations + return $ + Model vertices normals texCoords triangles skins animations + numFrames numVertices numTriangles numTexCoords numSkins numAnimations + + poke ptr + (Model verts normals texCoords tris skins animations + numFrames numVerts numTris numTex numSkins numAnimations) = + S.unsafeWith verts $ \pVerts -> + S.unsafeWith normals $ \pNormals -> + S.unsafeWith texCoords $ \pTexCoords -> + S.unsafeWith tris $ \pTris -> + S.unsafeWith skins $ \pSkins -> + S.unsafeWith animations $ \pAnimations -> do + #{poke Model, vertices} ptr pVerts + #{poke Model, normals} ptr pNormals + #{poke Model, texCoords} ptr pTexCoords + #{poke Model, triangles} ptr pTris + #{poke Model, skins} ptr pSkins + #{poke Model, animations} ptr pAnimations + #{poke Model, numFrames} ptr numFrames + #{poke Model, numVertices} ptr numVerts + #{poke Model, numTriangles} ptr numTris + #{poke Model, numTexCoords} ptr numTex + #{poke Model, numSkins} ptr numSkins + #{poke Model, numAnimations} ptr numAnimations + +-- | A model triangle. +-- +-- See also: 'triangles''. +data Triangle = Triangle + { v0 :: Vec3 + , v1 :: Vec3 + , v2 :: Vec3 + , n0 :: Vec3 + , n1 :: Vec3 + , n2 :: Vec3 + , t0 :: TexCoord + , t1 :: TexCoord + , t2 :: TexCoord + } + +instance Storable Triangle where + sizeOf _ = #{size model_triangle} + alignment _ = alignment (undefined :: Float) + + peek ptr = do + v0 <- #{peek model_triangle, v0} ptr + v1 <- #{peek model_triangle, v1} ptr + v2 <- #{peek model_triangle, v2} ptr + n0 <- #{peek model_triangle, n0} ptr + n1 <- #{peek model_triangle, n1} ptr + n2 <- #{peek model_triangle, n2} ptr + t0 <- #{peek model_triangle, t0} ptr + t1 <- #{peek model_triangle, t1} ptr + t2 <- #{peek model_triangle, t2} ptr + return $ Triangle v0 v1 v2 n0 n1 n2 t0 t1 t2 + + poke ptr (Triangle v0 v1 v2 n0 n1 n2 t0 t1 t2) = do + #{poke model_triangle, v0} ptr v0 + #{poke model_triangle, v1} ptr v1 + #{poke model_triangle, v2} ptr v2 + #{poke model_triangle, n0} ptr n0 + #{poke model_triangle, n1} ptr n1 + #{poke model_triangle, n2} ptr n2 + #{poke model_triangle, t0} ptr t0 + #{poke model_triangle, t1} ptr t1 + #{poke model_triangle, t2} ptr t2 + +foreign import ccall "Model.h model_free" + model_free :: Ptr Model -> IO () + +foreign import ccall "MD2_load.h MD2_load" + md2_load' :: Ptr CChar -> CChar -> CChar -> Ptr Model -> IO Int + +foreign import ccall "OBJ_load.h OBJ_load" + obj_load' :: Ptr CChar -> CChar -> CChar -> Ptr Model -> IO Int + +md2_load :: Ptr CChar -> CChar -> CChar -> Ptr Model -> IO ModelErrorCode +md2_load file clockwise leftHanded model = + md2_load' file clockwise leftHanded model >>= \code -> return . toEnum $ code + +obj_load :: Ptr CChar -> CChar -> CChar -> Ptr Model -> IO ModelErrorCode +obj_load file clockwise leftHanded model = + obj_load' file clockwise leftHanded model >>= \code -> return . toEnum $ code + +-- | Load the model specified by the given file. +loadModel :: FilePath -> Game s Model +loadModel file = do + dotPos <- case elemIndex '.' file of + Nothing -> gameError $ "file name has no extension: " ++ file + Just p -> return p + + let ext = map toLower . tail . snd $ splitAt dotPos file + + result <- liftIO . alloca $ \ptr -> do + status <- withCString file $ \fileCstr -> do + case ext of + "md2" -> md2_load fileCstr 0 0 ptr + "obj" -> obj_load fileCstr 0 0 ptr + _ -> return ModelNoSuitableLoader + + case status of + ModelSuccess -> do + model <- peek ptr + model_free ptr + return . Right $ model + ModelReadError -> return . Left $ "read error" + ModelMemoryAllocationError -> return . Left $ "memory allocation error" + ModelFileNotFound -> return . Left $ "file not found" + ModelFileMismatch -> return . Left $ "file mismatch" + ModelNoSuitableLoader -> return . Left $ "no suitable loader for extension " ++ ext + + case result of + Right model -> return model + Left err -> gameError $ "loadModel: " ++ err + +-- | Return 'True' if the model is animated, 'False' otherwise. +animated :: Model -> Bool +animated = (>1) . numFrames + +-- | Return the model's ith animation. +animation :: Model -> Int -> Animation +animation model i = animations model S.! i + +-- | Return the animation specified by the given string. +animationByName :: Model -> String -> Maybe Animation +animationByName model anim = + let anim' = B.pack anim in S.find ((==) anim' . name) $ animations model + +-- | Return a copy of the model's triangles. +triangles' :: Model -> IO [Triangle] +triangles' model = + let n = numVerts model * numFrames model + in with model $ \modelPtr -> + allocaArray n $ \arrayPtr -> do + model_copy_triangles modelPtr arrayPtr + tris <- peekArray n arrayPtr + return tris + +foreign import ccall "Model.h model_copy_triangles" + model_copy_triangles :: Ptr Model -> Ptr Triangle -> IO () + +-- | Transform the model's vertices. +transformVerts :: Model -> (Vec3 -> Vec3) -> Model +transformVerts model f = model { vertices = vertices' } + where + n = numVerts model * numFrames model + vertices' = S.generate n f' + f' i = f $ vertices model S.! i + +-- | Transform the model's normals. +transformNormals :: Model -> (Vec3 -> Vec3) -> Model +transformNormals model f = model { normals = normals' } + where + n = numVerts model * numFrames model + normals' = S.generate n f' + f' i = f $ normals model S.! i + +-- | Translate the model such that its lowest point has y = 0. +toGround :: Model -> IO Model +toGround model = + let model' = model { vertices = S.generate n $ \i -> vertices model S.! i } + n = numVerts model * numFrames model + in + with model' model_to_ground >> return model' + +foreign import ccall "Model.h model_to_ground" + model_to_ground :: Ptr Model -> IO () + +-- | Get the model's 3D bounding boxes. +modelBoxes :: Model -> IO (V.Vector Box) +modelBoxes model = + with model $ \modelPtr -> + allocaArray (numVerts model * numFrames model * 6) $ \pointsPtr -> do + model_compute_boxes modelPtr pointsPtr + let n = numFrames model + getBoxes = peekBoxes pointsPtr n 0 0 $ return [] + peekBoxes ptr n cur off l + | cur == n = l + | otherwise = do + xmin <- peekByteOff ptr off + ymin <- peekByteOff ptr $ off + sizeFloat + zmin <- peekByteOff ptr $ off + 2*sizeFloat + xmax <- peekByteOff ptr $ off + 3*sizeFloat + ymax <- peekByteOff ptr $ off + 4*sizeFloat + zmax <- peekByteOff ptr $ off + 5*sizeFloat + let pmin = Vec3 xmin ymin zmin + pmax = Vec3 xmax ymax zmax + box = Box pmin pmax + peekBoxes ptr n (cur+1) (off + 6*sizeFloat) $ fmap (box:) l + fmap (V.fromList . reverse) getBoxes + +foreign import ccall "Model.h model_compute_boxes" + model_compute_boxes :: Ptr Model -> Ptr Vec2 -> IO () diff --git a/Spear/old/Spear/Assets/Model/MD2/MD2_load.c b/Spear/old/Spear/Assets/Model/MD2/MD2_load.c new file mode 100644 index 0000000..92b1ac2 --- /dev/null +++ b/Spear/old/Spear/Assets/Model/MD2/MD2_load.c @@ -0,0 +1,480 @@ +#include "MD2_load.h" +#include +#include +#include // malloc +#include // sqrt + +//! The MD2 magic number used to identify MD2 files. +#define MD2_ID 0x32504449 + +//! Limit values for the MD2 file format. +#define MD2_MAX_TRIANGLES 4096 +#define MD2_MAX_VERTICES 2048 +#define MD2_MAX_TEXCOORDS 2048 +#define MD2_MAX_FRAMES 512 +#define MD2_MAX_SKINS 32 + + +/// MD2 file header. +typedef struct +{ + I32 magic; /// The magic number "IDP2"; 844121161 in decimal; 0x32504449 + I32 version; /// Version number, always 8. + I32 skinWidth; /// Width of the skin(s) in pixels. + I32 skinHeight; /// Height of the skin(s) in pixels. + I32 frameSize; /// Size of a single frame in bytes. + I32 numSkins; /// Number of skins. + I32 numVertices; /// Number of vertices in a single frame. + I32 numTexCoords; /// Number of texture coordinates. + I32 numTriangles; /// Number of triangles. + I32 numGlCommands; /// Number of dwords in the Gl command list. + I32 numFrames; /// Number of frames. + I32 offsetSkins; /// Offset from the start of the file to the array of skins. + I32 offsetTexCoords; /// Offset from the start of the file to the array of texture coordinates. + I32 offsetTriangles; /// Offset from the start of the file to the array of triangles. + I32 offsetFrames; /// Offset from the start of the file to the array of frames. + I32 offsetGlCommands; /// Offset from the start of the file to the array of Gl commands. + I32 offsetEnd; /// Offset from the start of the file to the end of the file (the file size). +} +md2Header_t; + + +/// Represents a texture coordinate index. +typedef struct +{ + I16 s; + I16 t; +} +texCoord_t; + + +/// Represents a frame point. +typedef struct +{ + U8 x, y, z; + U8 lightNormalIndex; +} +vertex_t; + + +/// Represents a single frame. +typedef struct +{ + vec3 scale; + vec3 translate; + I8 name[16]; + vertex_t vertices[1]; +} +frame_t; + + +static void normalise (vec3* v) +{ + float x = v->x; + float y = v->y; + float z = v->z; + float mag = sqrt (x*x + y*y + z*z); + mag = mag == 0 ? 1 : mag; + v->x = x / mag; + v->y = y / mag; + v->z = z / mag; +} + + +static void cross (const vec3* a, const vec3* b, vec3* c) +{ + c->x = a->y * b->z - a->z * b->y; + c->y = a->z * b->x - a->x * b->z; + c->z = a->x * b->y - a->y * b->x; +} + + +static void vec3_sub (const vec3* a, const vec3* b, vec3* out) +{ + out->x = a->x - b->x; + out->y = a->y - b->y; + out->z = a->z - b->z; +} + + +static void normal (char clockwise, const vec3* p1, const vec3* p2, const vec3* p3, vec3* n) +{ + vec3 v1, v2; + if (clockwise) + { + vec3_sub (p3, p2, &v1); + vec3_sub (p1, p2, &v2); + } + else + { + vec3_sub (p1, p2, &v1); + vec3_sub (p3, p2, &v2); + } + cross (&v1, &v2, n); + normalise (n); +} + + +typedef struct +{ + vec3* normals; + vec3* base; + unsigned int N; +} +normal_map; + + +static void normal_map_initialise (normal_map* m, unsigned int N) +{ + m->N = N; +} + + +static void normal_map_clear (normal_map* m, vec3* normals, vec3* base) +{ + memset (normals, 0, m->N * sizeof(vec3)); + m->normals = normals; + m->base = base; +} + + +static void normal_map_insert (normal_map* m, vec3* vec, vec3 normal) +{ + unsigned int i = vec - m->base; + vec3* n = m->normals + i; + n->x += normal.x; + n->y += normal.y; + n->z += normal.z; +} + + +static void compute_normals (normal_map* m, char left_handed) +{ + vec3* n = m->normals; + unsigned int i; + for (i = 0; i < m->N; ++i) + { + if (!left_handed) + { + n->x = -n->x; + n->y = -n->y; + n->z = -n->z; + } + normalise (n); + n++; + } +} + + +static void safe_free (void* ptr) +{ + if (ptr) free (ptr); +} + + +static char frame_equal (const char* name1, const char* name2) +{ + char equal = 1; + int i; + + if (((name1 == 0) && (name2 != 0)) || ((name1 != 0) && (name2 == 0))) + { + return 0; + } + + for (i = 0; i < 16; ++i) + { + char c1 = *name1; + char c2 = *name2; + if ((c1 >= '0' && c1 <= '9') || (c2 >= '0' && c2 <= '9')) break; + if (c1 != c2) + { + equal = 0; + break; + } + if (c1 == '_' || c2 == '_') break; + name1++; + name2++; + } + return equal; +} + + +static void animation_remove_numbers (char* name) +{ + int i; + for (i = 0; i < 16; ++i) + { + char c = *name; + if (c == 0) break; + if (c >= '0' && c <= '9') *name = 0; + name++; + } +} + + +Model_error_code MD2_load (const char* filename, char clockwise, char left_handed, Model* model) +{ + FILE* filePtr; + vec3* vertices; + vec3* normals; + texCoord* texCoords; + triangle* triangles; + skin* skins; + animation* animations; + int i; + + // Open the file for reading. + filePtr = fopen(filename, "rb"); + if (!filePtr) return Model_File_Not_Found; + + // Make sure it is an MD2 file. + int magic; + if ((fread(&magic, 4, 1, filePtr)) != 1) + { + fclose(filePtr); + return Model_Read_Error; + } + + if (magic != MD2_ID) return Model_File_Mismatch; + + // Find out the file size. + long int fileSize; + fseek(filePtr, 0, SEEK_END); + fileSize = ftell(filePtr); + fseek(filePtr, 0, SEEK_SET); + + // Allocate a chunk of data to store the file in. + char *buffer = (char*) malloc(fileSize); + if (!buffer) + { + fclose(filePtr); + return Model_Memory_Allocation_Error; + } + + // Read the entire file into memory. + if ((fread(buffer, 1, fileSize, filePtr)) != (unsigned int)fileSize) + { + fclose(filePtr); + free(buffer); + return Model_Read_Error; + } + + // File stream is no longer needed. + fclose(filePtr); + + // Set a pointer to the header for parsing. + md2Header_t* header = (md2Header_t*) buffer; + + // Compute the number of animations. + unsigned numAnimations = 1; + int currentFrame; + const char* name = 0; + for (currentFrame = 0; currentFrame < header->numFrames; ++currentFrame) + { + frame_t* frame = (frame_t*) &buffer[header->offsetFrames + currentFrame * header->frameSize]; + if (name == 0) + { + name = frame->name; + } + else if (!frame_equal(name, frame->name)) + { + numAnimations++; + name = frame->name; + } + } + + // Allocate memory for arrays. + vertices = (vec3*) malloc(sizeof(vec3) * header->numVertices * header->numFrames); + normals = (vec3*) malloc(sizeof(vec3) * header->numVertices * header->numFrames); + texCoords = (texCoord*) malloc(sizeof(texCoord) * header->numTexCoords); + triangles = (triangle*) malloc(sizeof(triangle) * header->numTriangles); + skins = (skin*) malloc(sizeof(skin) * header->numSkins); + animations = (animation*) malloc (numAnimations * sizeof(animation)); + + if (!vertices || !normals || !texCoords || !triangles || !skins || !animations) + { + safe_free (animations); + safe_free (skins); + safe_free (triangles); + safe_free (texCoords); + safe_free (normals); + safe_free (vertices); + free (buffer); + return Model_Memory_Allocation_Error; + } + + // Load the model's vertices. + // Loop through each frame, grab the vertices that make it up, transform them back + // to their real coordinates and store them in the model's vertex array. + for (currentFrame = 0; currentFrame < header->numFrames; ++currentFrame) + { + // Set a frame pointer to the current frame. + frame_t* frame = (frame_t*) &buffer[header->offsetFrames + currentFrame * header->frameSize]; + + // Set a vertex pointer to the model's vertex array, at the appropiate position. + vec3* vert = &vertices[currentFrame * header->numVertices]; + + // Now parse those vertices and transform them back. + int currentVertex; + for (currentVertex = 0; currentVertex != header->numVertices; ++currentVertex) + { + vert[currentVertex].x = frame->vertices[currentVertex].x * frame->scale.x + frame->translate.x; + vert[currentVertex].y = frame->vertices[currentVertex].y * frame->scale.y + frame->translate.y; + vert[currentVertex].z = frame->vertices[currentVertex].z * frame->scale.z + frame->translate.z; + } + } + + // Load the model's triangles. + + // Set a pointer to the triangles array in the buffer. + triangle* t = (triangle*) &buffer[header->offsetTriangles]; + + if (clockwise) + { + for (i = 0; i < header->numTriangles; ++i) + { + triangles[i].vertexIndices[0] = t[i].vertexIndices[0]; + triangles[i].vertexIndices[1] = t[i].vertexIndices[1]; + triangles[i].vertexIndices[2] = t[i].vertexIndices[2]; + + triangles[i].textureIndices[0] = t[i].textureIndices[0]; + triangles[i].textureIndices[1] = t[i].textureIndices[1]; + triangles[i].textureIndices[2] = t[i].textureIndices[2]; + } + } + else + { + for (i = 0; i < header->numTriangles; ++i) + { + triangles[i].vertexIndices[0] = t[i].vertexIndices[0]; + triangles[i].vertexIndices[1] = t[i].vertexIndices[2]; + triangles[i].vertexIndices[2] = t[i].vertexIndices[1]; + + triangles[i].textureIndices[0] = t[i].textureIndices[0]; + triangles[i].textureIndices[1] = t[i].textureIndices[2]; + triangles[i].textureIndices[2] = t[i].textureIndices[1]; + } + } + + // Load the texture coordinates. + float sw = (float) header->skinWidth; + float sh = (float) header->skinHeight; + texCoord_t* texc = (texCoord_t*) &buffer[header->offsetTexCoords]; + for (i = 0; i < header->numTexCoords; ++i) + { + texCoords[i].s = (float)texc->s / sw; + texCoords[i].t = 1.0f - (float)texc->t / sh; + texc++; + } + + // Iterate over every frame and compute normals for every triangle. + vec3 n; + + normal_map map; + normal_map_initialise (&map, header->numVertices); + + for (currentFrame = 0; currentFrame < header->numFrames; ++currentFrame) + { + // Set a pointer to the triangle array. + triangle* t = triangles; + + // Set a pointer to the vertex array at the appropiate position. + vec3* vertex_array = vertices + header->numVertices * currentFrame; + + // Set a pointer to the normals array at the appropiate position. + vec3* normals_ptr = normals + header->numVertices * currentFrame; + + normal_map_clear (&map, normals_ptr, vertex_array); + + for (i = 0; i < header->numTriangles; ++i) + { + // Compute face normal. + vec3* v0 = &vertex_array[t->vertexIndices[0]]; + vec3* v1 = &vertex_array[t->vertexIndices[1]]; + vec3* v2 = &vertex_array[t->vertexIndices[2]]; + normal (clockwise, v0, v1, v2, &n); + + // Add face normal to each of the face's vertices. + normal_map_insert (&map, v0, n); + normal_map_insert (&map, v1, n); + normal_map_insert (&map, v2, n); + + t++; + } + + compute_normals (&map, left_handed); + } + + // Load the model's skins. + const skin* s = (const skin*) &buffer[header->offsetSkins]; + for (i = 0; i < header->numSkins; ++i) + { + memcpy (skins[i].name, s->name, 64); + s++; + } + + // Load the model's animations. + unsigned start = 0; + name = 0; + animation* currentAnimation = animations; + for (currentFrame = 0; currentFrame < header->numFrames; ++currentFrame) + { + frame_t* frame = (frame_t*) &buffer[header->offsetFrames + currentFrame * header->frameSize]; + if (name == 0) + { + name = frame->name; + } + else if (!frame_equal(name, frame->name)) + { + memcpy (currentAnimation->name, name, 16); + animation_remove_numbers (currentAnimation->name); + currentAnimation->start = start; + currentAnimation->end = currentFrame-1; + if (currentAnimation != animations) + { + animation* prev = currentAnimation; + prev--; + prev->end = start-1; + } + name = frame->name; + currentAnimation++; + start = currentFrame; + } + } + currentAnimation = animations + numAnimations - 1; + memcpy (currentAnimation->name, name, 16); + animation_remove_numbers (currentAnimation->name); + currentAnimation->start = start; + currentAnimation->end = header->numFrames-1; + + /*printf ("finished loading model %s\n", filename); + printf ("numAnimations: %u\n", numAnimations); + printf ("animations: %p\n", animations); + + currentAnimation = animations; + for (i = 0; i < numAnimations; ++i) + { + printf ("Animation %d, name: %s, start: %d, end %d\n", + i, currentAnimation->name, currentAnimation->start, currentAnimation->end); + currentAnimation++; + }*/ + + model->vertices = vertices; + model->normals = normals; + model->texCoords = texCoords; + model->triangles = triangles; + model->skins = skins; + model->animations = animations; + + model->numFrames = header->numFrames; + model->numVertices = header->numVertices; + model->numTriangles = header->numTriangles; + model->numTexCoords = header->numTexCoords; + model->numSkins = header->numSkins; + model->numAnimations = numAnimations; + + free(buffer); + + return Model_Success; +} diff --git a/Spear/old/Spear/Assets/Model/MD2/MD2_load.h b/Spear/old/Spear/Assets/Model/MD2/MD2_load.h new file mode 100644 index 0000000..689f9d1 --- /dev/null +++ b/Spear/old/Spear/Assets/Model/MD2/MD2_load.h @@ -0,0 +1,22 @@ +#ifndef _MD2_LOAD_H +#define _MD2_LOAD_H + +#include "../Model.h" +#include "../Model_error_code.h" + + +#ifdef __cplusplus +extern "C" { +#endif + +/// Loads the MD2 file specified by the given string. +/// 'clockwise' should be 1 if you plan to render the model in a clockwise environment, 0 otherwise. +/// 'smooth_normals' should be 1 if you want the loader to compute smooth normals, 0 otherwise. +Model_error_code MD2_load (const char* filename, char clockwise, char left_handed, Model* model); + +#ifdef __cplusplus +} +#endif + + +#endif // _MD2_LOAD_H diff --git a/Spear/old/Spear/Assets/Model/Model.c b/Spear/old/Spear/Assets/Model/Model.c new file mode 100644 index 0000000..fd588ec --- /dev/null +++ b/Spear/old/Spear/Assets/Model/Model.c @@ -0,0 +1,112 @@ +#include "Model.h" +#include // free +#include + + +#define TO_RAD M_PI / 180.0 + + +static void safe_free (void* ptr) +{ + if (ptr) + { + free (ptr); + ptr = 0; + } +} + + +void model_free (Model* model) +{ + safe_free (model->vertices); + safe_free (model->normals); + safe_free (model->texCoords); + safe_free (model->triangles); + safe_free (model->skins); + safe_free (model->animations); +} + + +void model_to_ground (Model* model) +{ + unsigned i, f; + vec3* v = model->vertices; + + // Compute the minimum y coordinate for each frame and translate + // the model appropriately. + for (f = 0; f < model->numFrames; ++f) + { + vec3* w = v; + float y = v->y; + + for (i = 0; i < model->numVertices; ++i, ++v) + { + y = fmin (y, v->y); + } + + v = w; + for (i = 0; i < model->numVertices; ++i, ++v) + { + v->y -= y; + } + } +} + + +void model_copy_triangles (Model* model, unsigned frame, model_triangle* tris) +{ + int i; + int j = model->numVertices; + + vec3* v = model->vertices + j * frame; + vec3* n = model->normals + j * frame; + texCoord* t = model->texCoords; + triangle* tri = model->triangles; + + + for (i = 0; i < j; ++i, ++tri, ++tris) + { + tris->v0 = v[tri->vertexIndices[0]]; + tris->v1 = v[tri->vertexIndices[1]]; + tris->v2 = v[tri->vertexIndices[2]]; + + tris->n0 = n[tri->vertexIndices[0]]; + tris->n1 = n[tri->vertexIndices[1]]; + tris->n2 = n[tri->vertexIndices[2]]; + + tris->t0 = t[tri->textureIndices[0]]; + tris->t1 = t[tri->textureIndices[1]]; + tris->t2 = t[tri->textureIndices[2]]; + } +} + + +void model_compute_boxes (Model* model, float* points) +{ + vec3* v = model->vertices; + + unsigned f; + for (f = 0; f < model->numFrames; ++f) + { + float xmin = v->x; + float xmax = v->x; + float ymin = v->y; + float ymax = v->y; + float zmin = v->z; + float zmax = v->z; + + unsigned i; + for (i = 0; i < model->numVertices; ++i, ++v) + { + xmin = fmin (xmin, v->x); + ymin = fmin (ymin, v->y); + zmin = fmin (zmin, v->z); + xmax = fmax (xmax, v->x); + ymax = fmax (ymax, v->y); + zmax = fmax (zmax, v->z); + } + + *points++ = xmin; *points++ = ymin; *points++ = zmin; + *points++ = xmax; *points++ = ymax; *points++ = zmax; + } +} diff --git a/Spear/old/Spear/Assets/Model/Model.h b/Spear/old/Spear/Assets/Model/Model.h new file mode 100644 index 0000000..0f2aece --- /dev/null +++ b/Spear/old/Spear/Assets/Model/Model.h @@ -0,0 +1,100 @@ +#ifndef _SPEAR_MODEL_H +#define _SPEAR_MODEL_H + +#include "sys_types.h" + + +typedef struct +{ + char name[64]; +} +skin; + + +typedef struct +{ + float x, y, z; +} +vec3; + + +typedef struct +{ + float s, t; +} +texCoord; + + +typedef struct +{ + U16 vertexIndices[3]; + U16 textureIndices[3]; +} +triangle; + + +typedef struct +{ + char name[16]; + unsigned int start; + unsigned int end; +} +animation; + + +typedef struct +{ + vec3* vertices; // One array per frame. + vec3* normals; // One array per frame. One normal per vertex per frame. + texCoord* texCoords; // One array for all frames. + triangle* triangles; // One array for all frames. + skin* skins; // Holds the model's texture files. + animation* animations; // Holds the model's animations. + + unsigned int numFrames; + unsigned int numVertices; // Number of vertices per frame. + unsigned int numTriangles; // Number of triangles in one frame. + unsigned int numTexCoords; // Number of texture coordinates in one frame. + unsigned int numSkins; + unsigned int numAnimations; +} +Model; + + +typedef struct +{ + vec3 v0; + vec3 v1; + vec3 v2; + vec3 n0; + vec3 n1; + vec3 n2; + texCoord t0; + texCoord t1; + texCoord t2; +} +model_triangle; + + +#ifdef __cplusplus +extern "C" { +#endif + +/// Frees the given Model from memory. +/// The 'model' pointer itself is not freed. +void model_free (Model* model); + +/// Translate the Model such that its lowest point has y = 0. +void model_to_ground (Model* model); + +/// Copy the triangles of the given frame from the Model into the given array. +void model_copy_triangles (Model* model, unsigned frame, model_triangle* tris); + +/// Compute the model's 2d AABBs. +void model_compute_boxes (Model* model, float* points); + +#ifdef __cplusplus +} +#endif + +#endif // _SPEAR_MODEL_H diff --git a/Spear/old/Spear/Assets/Model/Model_error_code.h b/Spear/old/Spear/Assets/Model/Model_error_code.h new file mode 100644 index 0000000..d306052 --- /dev/null +++ b/Spear/old/Spear/Assets/Model/Model_error_code.h @@ -0,0 +1,16 @@ +#ifndef _SPEAR_MODEL_ERROR_CODE_H +#define _SPEAR_MODEL_ERROR_CODE_H + +typedef enum +{ + Model_Success, + Model_Read_Error, + Model_Memory_Allocation_Error, + Model_File_Not_Found, + Model_File_Mismatch, + Model_No_Suitable_Loader, +} +Model_error_code; + +#endif // _SPEAR_MODEL_ERROR_CODE_H + diff --git a/Spear/old/Spear/Assets/Model/OBJ/Makefile b/Spear/old/Spear/Assets/Model/OBJ/Makefile new file mode 100644 index 0000000..9630c9d --- /dev/null +++ b/Spear/old/Spear/Assets/Model/OBJ/Makefile @@ -0,0 +1,15 @@ +test: ../Model.o OBJ_load.o cvector.o test.o + $(CC) Model.o OBJ_load.o cvector.o test.o -o $@ -lm + +vector: cvector.o vector-test.o + $(CC) cvector.o vector-test.o -o vector + + +%.o: %.c %.h + $(CC) -g -c $< + + +clean: + @rm -f test vector + @rm -f *.o + diff --git a/Spear/old/Spear/Assets/Model/OBJ/OBJ_load.c b/Spear/old/Spear/Assets/Model/OBJ/OBJ_load.c new file mode 100644 index 0000000..cdd39c9 --- /dev/null +++ b/Spear/old/Spear/Assets/Model/OBJ/OBJ_load.c @@ -0,0 +1,274 @@ +#include "OBJ_load.h" +#include "cvector.h" +#include +#include // free +#include // memcpy +#include // sqrt + + +char lastError [128]; + + +static void safe_free (void* ptr) +{ + if (ptr) + { + free (ptr); + ptr = 0; + } +} + + +static void cross (vec3 a, vec3 b, vec3* c) +{ + c->x = a.y * b.z - a.z * b.y; + c->y = a.z * b.x - a.x * b.z; + c->z = a.x * b.y - a.y * b.x; +} + + +static void vec3_sub (vec3 a, vec3 b, vec3* out) +{ + out->x = a.x - b.x; + out->y = a.y - b.y; + out->z = a.z - b.z; +} + + +static void compute_normal (char clockwise, vec3 p1, vec3 p2, vec3 p3, vec3* n) +{ + vec3 v1, v2; + if (!clockwise) + { + vec3_sub (p3, p2, &v1); + vec3_sub (p1, p2, &v2); + } + else + { + vec3_sub (p1, p2, &v1); + vec3_sub (p3, p2, &v2); + } + cross (v1, v2, n); +} + + +static void normalise (vec3* v) +{ + float x = v->x; + float y = v->y; + float z = v->z; + float mag = sqrt (x*x + y*y + z*z); + mag = mag == 0.0f ? 1.0f : mag; + v->x /= mag; + v->y /= mag; + v->z /= mag; +} + + +static void vec3_add (vec3 a, vec3* b) +{ + b->x += a.x; + b->y += a.y; + b->z += a.z; +} + + +static void read_vertex (FILE* file, vec3* vert) +{ + fscanf (file, "%f %f", &vert->x, &vert->y); + if (fscanf(file, "%f", &vert->z) == 0) vert->z = 0.0f; +} + + +static void read_normal (FILE* file, vec3* normal) +{ + fscanf (file, "%f %f %f", &normal->x, &normal->y, &normal->z); +} + + +static void read_tex_coord (FILE* file, texCoord* texc) +{ + fscanf (file, "%f %f", &texc->s, &texc->t); +} + + +static void read_face (FILE* file, + char clockwise, + vector* vertices, + vector* normals, + vector* triangles) +{ + vector idxs; + vector texCoords; + + vector_new (&idxs, sizeof(int), 4); + vector_new (&texCoords, sizeof(int), 4); + + unsigned int index; + unsigned int normal; + unsigned int texc; + + fscanf (file, "f"); + + while (!feof(file) && fscanf(file, "%d", &index) > 0) + { + vector_append (&idxs, &index); + + if (fgetc (file) == '/') + { + fscanf (file, "%d", &texc); + vector_append (&texCoords, &texc); + } + else fseek (file, -1, SEEK_CUR); + + if (fgetc (file) == '/') + { + fscanf (file, "%d", &normal); + } + else fseek (file, -1, SEEK_CUR); + } + + // Triangulate the face and add its triangles to the triangles vector. + triangle tri; + tri.vertexIndices[0] = *((int*) vector_ith (&idxs, 0)) - 1; + tri.textureIndices[0] = *((int*) vector_ith (&texCoords, 0)) - 1; + + int i; + for (i = 1; i < vector_size(&idxs)-1; i++) + { + tri.vertexIndices[1] = *((int*) vector_ith (&idxs, i)) - 1; + tri.textureIndices[1] = *((int*) vector_ith (&texCoords, i)) - 1; + tri.vertexIndices[2] = *((int*) vector_ith (&idxs, i+1)) - 1; + tri.textureIndices[2] = *((int*) vector_ith (&texCoords, i+1)) - 1; + vector_append (triangles, &tri); + } + + // Compute face normal and add contribution to each of the face's vertices. + unsigned int i0 = tri.vertexIndices[0]; + unsigned int i1 = tri.vertexIndices[1]; + unsigned int i2 = tri.vertexIndices[2]; + + vec3 n; + vec3 v0 = *((vec3*) vector_ith (vertices, i0)); + vec3 v1 = *((vec3*) vector_ith (vertices, i1)); + vec3 v2 = *((vec3*) vector_ith (vertices, i2)); + compute_normal (clockwise, v0, v1, v2, &n); + + for (i = 0; i < vector_size (&idxs); i++) + { + int j = *((int*) vector_ith (&idxs, i)) - 1; + vec3* normal = (vec3*) vector_ith (normals, j); + vec3_add (n, normal); + } + + vector_free (&idxs); + vector_free (&texCoords); +} + + +Model_error_code OBJ_load (const char* filename, char clockwise, char left_handed, Model* model) +{ + vec3* norms = 0; + vec3* verts = 0; + texCoord* texcs = 0; + triangle* tris = 0; + + FILE* file = fopen (filename, "r"); + if (file == NULL) return Model_File_Not_Found; + + vec3 vert; + vec3 normal; + texCoord texc; + + vector vertices; + vector normals; + vector texCoords; + vector triangles; + + int result = vector_new (&vertices, sizeof(vec3), 0) + | vector_new (&normals, sizeof(vec3), 0) + | vector_new (&texCoords, sizeof(texCoord), 0) + | vector_new (&triangles, sizeof(triangle), 0); + + if (result != 0) + { + safe_free (vertices.data); + safe_free (normals.data); + safe_free (texCoords.data); + safe_free (triangles.data); + return Model_Memory_Allocation_Error; + } + + while (!feof(file)) + { + switch (fgetc(file)) + { + case 'v': + switch (fgetc(file)) + { + case 't': + read_tex_coord (file, &texc); + vector_append (&texCoords, &texc); + break; + + case 'n': + read_normal (file, &normal); + vector_append (&normals, &normal); + break; + + default: + read_vertex (file, &vert); + vector_append (&vertices, &vert); + break; + } + break; + + case 'f': + // Initialise the normals vector if it is empty. + if (vector_size(&normals) == 0) + { + vec3 zero; + zero.x = 0.0f; zero.y = 0.0f; zero.z = 0.0f; + vector_new (&normals, sizeof(vec3), vector_size(&vertices)); + vector_initialise (&normals, &zero); + } + read_face (file, clockwise, &vertices, &normals, &triangles); + break; + + case '#': + { + int x = 17; + while (x != '\n' && x != EOF) x = fgetc(file); + break; + } + + default: break; + } + } + + fclose (file); + + unsigned numVertices = vector_size (&vertices); + + // Normalise normals. + unsigned i; + for (i = 0; i < numVertices; ++i) + { + normalise (vector_ith (&normals, i)); + } + + model->vertices = (vec3*) vertices.data; + model->normals = (vec3*) normals.data; + model->texCoords = (texCoord*) texCoords.data; + model->triangles = (triangle*) triangles.data; + model->skins = 0; + model->animations = 0; + model->numFrames = 1; + model->numVertices = numVertices; + model->numTriangles = vector_size (&triangles); + model->numTexCoords = vector_size (&texCoords); + model->numSkins = 0; + model->numAnimations = 0; + + return Model_Success; +} diff --git a/Spear/old/Spear/Assets/Model/OBJ/OBJ_load.h b/Spear/old/Spear/Assets/Model/OBJ/OBJ_load.h new file mode 100644 index 0000000..485d3cc --- /dev/null +++ b/Spear/old/Spear/Assets/Model/OBJ/OBJ_load.h @@ -0,0 +1,25 @@ +#ifndef _OBJ_LOAD_H +#define _OBJ_LOAD_H + +#include "../Model.h" +#include "../Model_error_code.h" + + +#ifdef __cplusplus +extern "C" { +#endif + +/// Loads the OBJ file specified by the given string. +/// 'clockwise' should be 1 if you plan to render the model in a clockwise environment, 0 otherwise. +/// 'smooth_normals' should be 1 if you want the loader to compute smooth normals, 0 otherwise. +Model_error_code OBJ_load (const char* filename, char clockwise, char left_handed, Model* model); + +/// Gets the last error generated by the OBJ loader. +char* get_last_error (); + +#ifdef __cplusplus +} +#endif + + +#endif // _OBJ_LOAD_H diff --git a/Spear/old/Spear/Assets/Model/OBJ/cvector.c b/Spear/old/Spear/Assets/Model/OBJ/cvector.c new file mode 100644 index 0000000..9213d8d --- /dev/null +++ b/Spear/old/Spear/Assets/Model/OBJ/cvector.c @@ -0,0 +1,90 @@ +#include "cvector.h" +#include // malloc, realloc, free +#include // memcpy + + +int max (int a, int b) +{ + if (a > b) return a; + return b; +} + + +int vector_new (vector* v, int elem_size, int num_elems) +{ + int n = num_elems * elem_size; + + char* data = 0; + if (num_elems > 0) + { + data = (char*) malloc (n); + if (data == NULL) return 1; + } + + v->data = data; + v->next = data; + v->chunk_size = n; + v->elem_size = elem_size; + + return 0; +} + + +void vector_free (vector* v) +{ + if (v->data != 0) free (v->data); +} + + +void vector_initialise (vector* v, void* value) +{ + char* ptr = v->data; + int esize = v->elem_size; + int n = vector_size (v); + + int i; + for (i = 0; i < n; ++i) + { + memcpy (ptr, value, esize); + ptr += esize; + } +} + + +int vector_append (vector* v, void* elem) +{ + // Realloc a bigger chunk when the vector runs out of space. + if (v->next == v->data + v->chunk_size) + { + int old_chunk_size = v->chunk_size; + int n = max (v->elem_size, 2 * old_chunk_size); + + char* data = (char*) realloc (v->data, n); + if (data == NULL) return 1; + + v->data = data; + v->next = data + old_chunk_size; + v->chunk_size = n; + } + + memcpy ((void*)v->next, elem, v->elem_size); + v->next += v->elem_size; +} + + +void* vector_ith (vector* v, int i) +{ + return (void*) (v->data + i*v->elem_size); +} + + +int vector_size (vector* v) +{ + return (v->next - v->data) / v->elem_size; +} + + +int vector_capacity (vector* v) +{ + return v->chunk_size / v->elem_size; +} diff --git a/Spear/old/Spear/Assets/Model/OBJ/cvector.h b/Spear/old/Spear/Assets/Model/OBJ/cvector.h new file mode 100644 index 0000000..2c269e4 --- /dev/null +++ b/Spear/old/Spear/Assets/Model/OBJ/cvector.h @@ -0,0 +1,36 @@ +#ifndef _C_SPEAR_VECTOR_H +#define _C_SPEAR_VECTOR_H + +typedef struct +{ + char* data; + char* next; + int chunk_size; + int elem_size; +} +vector; + +/// Construct a new vector. +/// Returns non-zero on error. +int vector_new (vector* v, int elem_size, int num_elems); + +/// Free the vector. +void vector_free (vector* v); + +/// Initialise every position to the given value. +void vector_initialise (vector* v, void* value); + +/// Append an element. +/// Returns non-zero on error. +int vector_append (vector* v, void* elem); + +/// Access the ith element. +void* vector_ith (vector* v, int i); + +/// Return the number of elements in the vector. +int vector_size (vector* v); + +/// Return the vector's capacity. +int vector_capacity (vector* v); + +#endif // _C_SPEAR_VECTOR_H diff --git a/Spear/old/Spear/Assets/Model/sys_types.h b/Spear/old/Spear/Assets/Model/sys_types.h new file mode 100644 index 0000000..6aca9e9 --- /dev/null +++ b/Spear/old/Spear/Assets/Model/sys_types.h @@ -0,0 +1,16 @@ +#ifndef _SPEAR_SYS_TYPES_H +#define _SPEAR_SYS_TYPES_H + +#include + +typedef int8_t I8; +typedef int16_t I16; +typedef int32_t I32; +typedef int64_t I64; +typedef uint8_t U8; +typedef uint16_t U16; +typedef uint32_t U32; +typedef uint64_t U64; + +#endif // _SPEAR_SYS_TYPES_H + diff --git a/Spear/old/Spear/GL.hs b/Spear/old/Spear/GL.hs new file mode 100644 index 0000000..3c1734b --- /dev/null +++ b/Spear/old/Spear/GL.hs @@ -0,0 +1,800 @@ +{-# LANGUAGE FlexibleInstances #-} + +module Spear.GL + ( -- * Programs + GLSLProgram, + newProgram, + linkProgram, + useProgram, + unuseProgram, + withGLSLProgram, + + -- ** Locations + attribLocation, + fragLocation, + uniformLocation, + + -- ** Uniforms + Uniform (..), + + -- * Shaders + GLSLShader, + ShaderType (..), + attachShader, + detachShader, + loadShader, + newShader, + + -- ** Source loading + loadSource, + shaderSource, + readSource, + compile, + + -- * Helper functions + ($=), + Data.StateVar.get, + + -- * VAOs + VAO(..), + newVAO, + bindVAO, + unbindVAO, + enableVAOAttrib, + attribVAOPointer, + + -- ** Rendering + drawArrays, + drawElements, + + -- * Buffers + GLBuffer(..), + TargetBuffer (..), + BufferUsage (..), + newBuffer, + bindBuffer, + unbindBuffer, + BufferData (..), + bufferData', + withGLBuffer, + + -- * Textures + Texture, + SettableStateVar, + ($), + + -- ** Creation and destruction + newTexture, + loadTextureImage, + + -- ** Manipulation + bindTexture, + unbindTexture, + loadTextureData, + texParami, + texParamf, + activeTexture, + + -- * Error Handling + getGLError, + printGLError, + assertGL, + + -- * OpenGL + module Graphics.GL.Core46, + Ptr, + nullPtr, + ) +where + +import Spear.Assets.Image +import Spear.Game +import Spear.Math.Algebra +import Spear.Math.Matrix3 (Matrix3) +import Spear.Math.Matrix4 (Matrix4) +import Spear.Math.Vector + +import Control.Monad +import Control.Monad.Trans.Class +import Control.Monad.Trans.State as State +import qualified Data.ByteString.Char8 as B +import Data.StateVar +import Data.Word +import Foreign.C.String +import Foreign.C.Types +import Foreign.Marshal.Alloc (alloca) +import Foreign.Marshal.Array (withArray) +import Foreign.Marshal.Utils as Foreign (with) +import Foreign.Ptr +import Foreign.Storable +import Foreign.Storable (peek) +import Graphics.GL.Core46 +import Prelude hiding ((*)) +import System.Directory (doesFileExist, getCurrentDirectory, + setCurrentDirectory) +import System.IO (hPutStrLn, stderr) +import Unsafe.Coerce + +-- +-- MANAGEMENT +-- + +-- | A GLSL shader handle. +data GLSLShader = GLSLShader + { getShader :: GLuint, + getShaderKey :: ReleaseKey + } + +instance ResourceClass GLSLShader where + getResource = getShaderKey + +-- | A GLSL program handle. +data GLSLProgram = GLSLProgram + { getProgram :: GLuint, + getProgramKey :: ReleaseKey + } + +instance ResourceClass GLSLProgram where + getResource = getProgramKey + +-- | Supported shader types. +data ShaderType = VertexShader | FragmentShader | GeometryShader deriving (Eq, Show) + +toGLShader :: ShaderType -> GLenum +toGLShader VertexShader = GL_VERTEX_SHADER +toGLShader FragmentShader = GL_FRAGMENT_SHADER +toGLShader GeometryShader = GL_GEOMETRY_SHADER + +-- | Apply the given function to the program's id. +withGLSLProgram :: GLSLProgram -> (GLuint -> a) -> a +withGLSLProgram prog f = f $ getProgram prog + +-- | Get the location of the given uniform variable within the given program. +uniformLocation :: GLSLProgram -> String -> GettableStateVar GLint +uniformLocation prog var = makeGettableStateVar $ + withCString var $ \str -> glGetUniformLocation (getProgram prog) (unsafeCoerce str) + +-- | Get or set the location of the given variable to a fragment shader colour number. +fragLocation :: GLSLProgram -> String -> StateVar GLint +fragLocation prog var = makeStateVar get set + where + get = withCString var $ \str -> glGetFragDataLocation (getProgram prog) (unsafeCoerce str) + set idx = withCString var $ \str -> + glBindFragDataLocation (getProgram prog) (unsafeCoerce idx) (unsafeCoerce str) + +-- | Get or set the location of the given attribute within the given program. +attribLocation :: GLSLProgram -> String -> StateVar GLint +attribLocation prog var = makeStateVar get set + where + get = withCString var $ \str -> glGetAttribLocation (getProgram prog) (unsafeCoerce str) + set idx = withCString var $ \str -> + glBindAttribLocation (getProgram prog) (unsafeCoerce idx) (unsafeCoerce str) + +-- | Create a new program. +newProgram :: [GLSLShader] -> Game s GLSLProgram +newProgram shaders = do + h <- liftIO glCreateProgram + when (h == 0) $ gameError "glCreateProgram failed" + rkey <- register $ deleteProgram h + let program = GLSLProgram h rkey + mapM_ (liftIO . attachShader program) shaders + linkProgram program + return program + +-- Delete the program. +deleteProgram :: GLuint -> IO () +--deleteProgram = glDeleteProgram +deleteProgram prog = do + putStrLn $ "Deleting shader program " ++ show prog + glDeleteProgram prog + +-- | Link the program. +linkProgram :: GLSLProgram -> Game s () +linkProgram prog = do + let h = getProgram prog + err <- liftIO $ do + glLinkProgram h + alloca $ \statptr -> do + glGetProgramiv h GL_LINK_STATUS statptr + status <- peek statptr + case status of + 0 -> getStatus glGetProgramiv glGetProgramInfoLog h + _ -> return "" + + case length err of + 0 -> return () + _ -> gameError err + +-- | Use the program. +useProgram :: GLSLProgram -> IO () +useProgram prog = glUseProgram $ getProgram prog + +-- | Deactivate the active program. +unuseProgram :: IO () +unuseProgram = glUseProgram 0 + +-- | Attach the given shader to the given program. +attachShader :: GLSLProgram -> GLSLShader -> IO () +attachShader prog shader = glAttachShader (getProgram prog) (getShader shader) + +-- | Detach the given GLSL from the given program. +detachShader :: GLSLProgram -> GLSLShader -> IO () +detachShader prog shader = glDetachShader (getProgram prog) (getShader shader) + +-- | Load a shader from the file specified by the given string. +-- +-- This function creates a new shader. To load source code into an existing shader, +-- see 'loadSource', 'shaderSource' and 'readSource'. +loadShader :: ShaderType -> FilePath -> Game s GLSLShader +loadShader shaderType file = do + shader <- newShader shaderType + loadSource file shader + compile file shader + return shader + +-- | Create a new shader. +newShader :: ShaderType -> Game s GLSLShader +newShader shaderType = do + h <- liftIO $ glCreateShader (toGLShader shaderType) + case h of + 0 -> gameError "glCreateShader failed" + _ -> do + rkey <- register $ deleteShader h + return $ GLSLShader h rkey + +-- | Free the shader. +deleteShader :: GLuint -> IO () +--deleteShader = glDeleteShader +deleteShader shader = do + putStrLn $ "Deleting shader " ++ show shader + glDeleteShader shader + +-- | Load a shader source from the file specified by the given string +-- into the shader. +loadSource :: FilePath -> GLSLShader -> Game s () +loadSource file h = do + exists <- liftIO $ doesFileExist file + case exists of + False -> gameError "the specified shader file does not exist" + True -> liftIO $ do + code <- readSource file + withCString code $ shaderSource h + +-- | Load the given shader source into the shader. +shaderSource :: GLSLShader -> CString -> IO () +shaderSource shader str = + let ptr = unsafeCoerce str + in withArray [ptr] $ flip (glShaderSource (getShader shader) 1) nullPtr + +-- | Compile the shader. +compile :: FilePath -> GLSLShader -> Game s () +compile file shader = do + let h = getShader shader + + -- Compile + liftIO $ glCompileShader h + + -- Verify status + err <- liftIO $ + alloca $ \statusPtr -> do + glGetShaderiv h GL_COMPILE_STATUS statusPtr + result <- peek statusPtr + case result of + 0 -> getStatus glGetShaderiv glGetShaderInfoLog h + _ -> return "" + + case length err of + 0 -> return () + _ -> gameError $ "Unable to compile shader " ++ file ++ ":\n" ++ err + +type StatusCall = GLuint -> GLenum -> Ptr GLint -> IO () + +type LogCall = GLuint -> GLsizei -> Ptr GLsizei -> Ptr GLchar -> IO () + +getStatus :: StatusCall -> LogCall -> GLuint -> IO String +getStatus getStatus getLog h = do + alloca $ \lenPtr -> do + getStatus h GL_INFO_LOG_LENGTH lenPtr + len <- peek lenPtr + case len of + 0 -> return "" + _ -> withCString (replicate (unsafeCoerce len) '\0') $ getErrorString getLog h (unsafeCoerce len) + +getErrorString :: LogCall -> GLuint -> GLsizei -> CString -> IO String +getErrorString getLog h len str = do + let ptr = unsafeCoerce str + getLog h len nullPtr ptr + peekCString str + +-- | Load the shader source specified by the given file. +-- +-- This function implements an #include mechanism, so the given file can +-- refer to other files. +readSource :: FilePath -> IO String +readSource = fmap B.unpack . readSource' + +readSource' :: FilePath -> IO B.ByteString +readSource' file = do + let includeB = B.pack "#include" + newLineB = B.pack "\n" + isInclude = ((==) includeB) . B.take 8 + clean = B.dropWhile (\c -> c == ' ') + cleanInclude = B.filter (\c -> c /= '"') . B.dropWhile (\c -> c /= ' ') + toLines = B.splitWith (\c -> c == '\n' || c == '\r') + addNewLine s = if (B.last s /= '\n') then B.append s newLineB else s + parse = + fmap B.concat . (fmap . fmap $ flip B.append newLineB) . sequence + . fmap (processLine . clean) + . toLines + processLine l = + if isInclude l + then readSource' $ B.unpack . clean . cleanInclude $ l + else return l + + contents <- B.readFile file + + dir <- getCurrentDirectory + let dir' = dir ++ "/" ++ (reverse . dropWhile (\c -> c /= '/') . reverse) file + + setCurrentDirectory dir' + code <- parse contents + setCurrentDirectory dir + + return code + +class Uniform a where + -- | Load a list of uniform values. + uniform :: GLint -> a -> IO () + +instance Uniform Int where uniform loc a = glUniform1i loc (fromIntegral a) + +instance Uniform Float where uniform loc a = glUniform1f loc a + +instance Uniform CFloat where uniform loc a = glUniform1f loc (unsafeCoerce a) + +instance Uniform (Int, Int) where + uniform loc (x, y) = glUniform2i loc (fromIntegral x) (fromIntegral y) + +instance Uniform (Float, Float) where + uniform loc (x, y) = glUniform2f loc x y + +instance Uniform (Int, Int, Int) where + uniform loc (x, y, z) = glUniform3i loc (fromIntegral x) (fromIntegral y) (fromIntegral z) + +instance Uniform (Float, Float, Float) where + uniform loc (x, y, z) = glUniform3f loc x y z + +instance Uniform (Int, Int, Int, Int) where + uniform loc (x, y, z, w) = + glUniform4i + loc + (fromIntegral x) + (fromIntegral y) + (fromIntegral z) + (fromIntegral w) + +instance Uniform (Float, Float, Float, Float) where + uniform loc (x, y, z, w) = glUniform4f loc x y z w + +instance Uniform Vector2 where + uniform loc v = glUniform2f loc x' y' + where + x' = unsafeCoerce $ x v + y' = unsafeCoerce $ y v + +instance Uniform Vector3 where + uniform loc v = glUniform3f loc x' y' z' + where + x' = unsafeCoerce $ x v + y' = unsafeCoerce $ y v + z' = unsafeCoerce $ z v + +instance Uniform Vector4 where + uniform loc v = glUniform4f loc x' y' z' w' + where + x' = unsafeCoerce $ x v + y' = unsafeCoerce $ y v + z' = unsafeCoerce $ z v + w' = unsafeCoerce $ w v + +instance Uniform Matrix3 where + uniform loc mat = + with mat $ \ptrMat -> + glUniformMatrix3fv loc 1 (toEnum 0) (unsafeCoerce ptrMat) + +instance Uniform Matrix4 where + uniform loc mat = + with mat $ \ptrMat -> + glUniformMatrix4fv loc 1 (toEnum 0) (unsafeCoerce ptrMat) + +instance Uniform [Float] where + uniform loc vals = withArray (map unsafeCoerce vals) $ \ptr -> + case length vals of + 1 -> glUniform1fv loc 1 ptr + 2 -> glUniform2fv loc 1 ptr + 3 -> glUniform3fv loc 1 ptr + 4 -> glUniform4fv loc 1 ptr + +instance Uniform [CFloat] where + uniform loc vals = withArray vals $ \ptr -> + case length vals of + 1 -> glUniform1fv loc 1 $ castPtr ptr + 2 -> glUniform2fv loc 1 $ castPtr ptr + 3 -> glUniform3fv loc 1 $ castPtr ptr + 4 -> glUniform4fv loc 1 $ castPtr ptr + +instance Uniform [Int] where + uniform loc vals = withArray (map fromIntegral vals) $ \ptr -> + case length vals of + 1 -> glUniform1iv loc 1 ptr + 2 -> glUniform2iv loc 1 ptr + 3 -> glUniform3iv loc 1 ptr + 4 -> glUniform4iv loc 1 ptr + +-- +-- VAOs +-- + +-- | A vertex array object. +data VAO = VAO + { getVAO :: GLuint, + vaoKey :: ReleaseKey + } + +instance ResourceClass VAO where + getResource = vaoKey + +instance Eq VAO where + vao1 == vao2 = getVAO vao1 == getVAO vao2 + +instance Ord VAO where + vao1 < vao2 = getVAO vao1 < getVAO vao2 + vao1 <= vao2 = getVAO vao1 <= getVAO vao2 + +-- | Create a new vao. +newVAO :: Game s VAO +newVAO = do + h <- liftIO . alloca $ \ptr -> do + glGenVertexArrays 1 ptr + peek ptr + + rkey <- register $ deleteVAO h + return $ VAO h rkey + +-- | Delete the vao. +deleteVAO :: GLuint -> IO () +deleteVAO vao = Foreign.with vao $ glDeleteVertexArrays 1 + +-- | Bind the vao. +bindVAO :: VAO -> IO () +bindVAO = glBindVertexArray . getVAO + +-- | Unbind the bound vao. +unbindVAO :: IO () +unbindVAO = glBindVertexArray 0 + +-- | Enable the given vertex attribute of the bound vao. +-- +-- See also 'bindVAO'. +enableVAOAttrib :: + -- | Attribute index. + GLuint -> + IO () +enableVAOAttrib = glEnableVertexAttribArray + +-- | Bind the bound buffer to the given point. +attribVAOPointer :: + -- | The index of the generic vertex attribute to be modified. + GLuint -> + -- | The number of components per generic vertex attribute. Must be 1,2,3,4. + GLint -> + -- | The data type of each component in the array. + GLenum -> + -- | Whether fixed-point data values should be normalized. + Bool -> + -- | Stride. Byte offset between consecutive generic vertex attributes. + GLsizei -> + -- | Offset to the first component in the array. + Int -> + IO () +attribVAOPointer idx ncomp dattype normalise stride off = + glVertexAttribPointer idx ncomp dattype normalise' stride (unsafeCoerce off) + where + normalise' = if normalise then 1 else 0 + +-- | Draw the bound vao. +drawArrays :: + -- | The kind of primitives to render. + GLenum -> + -- | Starting index in the enabled arrays. + Int -> + -- | The number of indices to be rendered. + Int -> + IO () +drawArrays mode first count = glDrawArrays mode (unsafeCoerce first) (unsafeCoerce count) + +-- | Draw the bound vao, indexed mode. +drawElements :: + -- | The kind of primitives to render. + GLenum -> + -- | The number of elements to be rendered. + Int -> + -- | The type of the index values. Must be one of gl_UNSIGNED_BYTE, gl_UNSIGNED_SHORT, or gl_UNSIGNED_INT. + GLenum -> + -- | Pointer to the location where indices are stored, or offset into the index array when there is a bound ElementArrayBuffer. + Ptr a -> + IO () +drawElements mode count t idxs = glDrawElements mode (unsafeCoerce count) t idxs + +-- +-- BUFFER +-- + +-- | An OpenGL buffer. +data GLBuffer = GLBuffer + { getBuffer :: GLuint, + bufferKey :: ReleaseKey + } + +instance ResourceClass GLBuffer where + getResource = bufferKey + +-- | The type of target buffer. +data TargetBuffer + = ArrayBuffer + | ElementArrayBuffer + | PixelPackBuffer + | PixelUnpackBuffer + deriving (Eq, Show) + +fromTarget :: TargetBuffer -> GLenum +fromTarget ArrayBuffer = GL_ARRAY_BUFFER +fromTarget ElementArrayBuffer = GL_ELEMENT_ARRAY_BUFFER +fromTarget PixelPackBuffer = GL_PIXEL_PACK_BUFFER +fromTarget PixelUnpackBuffer = GL_PIXEL_UNPACK_BUFFER + +-- | A buffer usage. +data BufferUsage + = StreamDraw + | StreamRead + | StreamCopy + | StaticDraw + | StaticRead + | StaticCopy + | DynamicDraw + | DynamicRead + | DynamicCopy + deriving (Eq, Show) + +fromUsage :: BufferUsage -> GLenum +fromUsage StreamDraw = GL_STREAM_DRAW +fromUsage StreamRead = GL_STREAM_READ +fromUsage StreamCopy = GL_STREAM_COPY +fromUsage StaticDraw = GL_STATIC_DRAW +fromUsage StaticRead = GL_STATIC_READ +fromUsage StaticCopy = GL_STATIC_COPY +fromUsage DynamicDraw = GL_DYNAMIC_DRAW +fromUsage DynamicRead = GL_DYNAMIC_READ +fromUsage DynamicCopy = GL_DYNAMIC_COPY + +-- | Create a new buffer. +newBuffer :: Game s GLBuffer +newBuffer = do + h <- liftIO . alloca $ \ptr -> do + glGenBuffers 1 ptr + peek ptr + + rkey <- register $ deleteBuffer h + return $ GLBuffer h rkey + +-- | Delete the buffer. +deleteBuffer :: GLuint -> IO () +deleteBuffer buf = Foreign.with buf $ glDeleteBuffers 1 + +-- | Bind the buffer. +bindBuffer :: TargetBuffer -> GLBuffer -> IO () +bindBuffer target buf = glBindBuffer (fromTarget target) $ getBuffer buf + +-- | Unbind the bound buffer. +unbindBuffer :: TargetBuffer -> IO () +unbindBuffer target = glBindBuffer (fromTarget target) 0 + +class Storable a => BufferData a where + -- | Set the buffer's data. + bufferData :: TargetBuffer -> [a] -> BufferUsage -> IO () + bufferData tgt vals usage = + let n = sizeOf (head vals) * length vals + in withArray vals $ \ptr -> bufferData' tgt n ptr usage + +instance BufferData Word8 + +instance BufferData Word16 + +instance BufferData Word32 + +instance BufferData CChar + +instance BufferData CInt + +instance BufferData CFloat + +instance BufferData CDouble + +instance BufferData Int + +instance BufferData Float + +instance BufferData Double + +{-bufferData :: Storable a + => TargetBuffer + -> Int -- ^ The size in bytes of an element in the data list. + -> [a] -- ^ The data list. + -> BufferUsage + -> IO () +bufferData target n bufData usage = withArray bufData $ + \ptr -> bufferData target (n * length bufData) ptr usage-} + +-- | Set the buffer's data. +bufferData' :: + TargetBuffer -> + -- | Buffer size in bytes. + Int -> + Ptr a -> + BufferUsage -> + IO () +bufferData' target n bufData usage = glBufferData (fromTarget target) (unsafeCoerce n) bufData (fromUsage usage) + +-- | Apply the given function the buffer's id. +withGLBuffer :: GLBuffer -> (GLuint -> a) -> a +withGLBuffer buf f = f $ getBuffer buf + +-- +-- TEXTURE +-- + +-- | Represents a texture resource. +data Texture = Texture + { getTex :: GLuint, + texKey :: ReleaseKey + } + +instance Eq Texture where + t1 == t2 = getTex t1 == getTex t2 + +instance Ord Texture where + t1 < t2 = getTex t1 < getTex t2 + t1 <= t2 = getTex t1 <= getTex t2 + +instance ResourceClass Texture where + getResource = texKey + +-- | Create a new texture. +newTexture :: Game s Texture +newTexture = do + tex <- liftIO . alloca $ \ptr -> do + glGenTextures 1 ptr + peek ptr + + rkey <- register $ deleteTexture tex + return $ Texture tex rkey + +-- | Delete the texture. +deleteTexture :: GLuint -> IO () +--deleteTexture tex = with tex $ glDeleteTextures 1 +deleteTexture tex = do + putStrLn $ "Releasing texture " ++ show tex + with tex $ glDeleteTextures 1 + +-- | Load the 'Texture' specified by the given file. +loadTextureImage :: + FilePath -> + -- | Texture's min filter. + GLenum -> + -- | Texture's mag filter. + GLenum -> + Game s Texture +loadTextureImage file minFilter magFilter = do + image <- loadImage file + tex <- newTexture + liftIO $ do + let w = width image + h = height image + pix = pixels image + rgb = fromIntegral . fromEnum $ GL_RGB + + bindTexture tex + loadTextureData GL_TEXTURE_2D 0 rgb w h 0 GL_RGB GL_UNSIGNED_BYTE pix + texParami GL_TEXTURE_2D GL_TEXTURE_MIN_FILTER $= minFilter + texParami GL_TEXTURE_2D GL_TEXTURE_MAG_FILTER $= magFilter + + return tex + +-- | Bind the texture. +bindTexture :: Texture -> IO () +bindTexture = glBindTexture GL_TEXTURE_2D . getTex + +-- | Unbind the bound texture. +unbindTexture :: IO () +unbindTexture = glBindTexture GL_TEXTURE_2D 0 + +-- | Load data onto the bound texture. +-- +-- See also 'bindTexture'. +loadTextureData :: + GLenum -> + -- | Target + Int -> + -- | Level + Int -> + -- | Internal format + Int -> + -- | Width + Int -> + -- | Height + Int -> + -- | Border + GLenum -> + -- | Texture type + GLenum -> + -- | Texture data + Ptr a -> + IO () +loadTextureData target level internalFormat width height border format texType texData = do + glTexImage2D + target + (fromIntegral level) + (fromIntegral internalFormat) + (fromIntegral width) + (fromIntegral height) + (fromIntegral border) + (fromIntegral format) + texType + texData + +-- | Set the bound texture's parameter to the given value. +texParami :: GLenum -> GLenum -> SettableStateVar GLenum +texParami target param = makeSettableStateVar $ \val -> glTexParameteri target param $ fromIntegral . fromEnum $ val + +-- | Set the bound texture's parameter to the given value. +texParamf :: GLenum -> GLenum -> SettableStateVar Float +texParamf target param = makeSettableStateVar $ \val -> glTexParameterf target param (unsafeCoerce val) + +-- | Set the active texture unit. +activeTexture :: SettableStateVar GLenum +activeTexture = makeSettableStateVar glActiveTexture + +-- +-- ERROR +-- + +-- | Get the last OpenGL error. +getGLError :: IO (Maybe String) +getGLError = fmap translate glGetError + where + translate err + | err == GL_NO_ERROR = Nothing + | err == GL_INVALID_ENUM = Just "Invalid enum" + | err == GL_INVALID_VALUE = Just "Invalid value" + | err == GL_INVALID_OPERATION = Just "Invalid operation" + | err == GL_OUT_OF_MEMORY = Just "Out of memory" + | otherwise = Just "Unknown error" + +-- | Print the last OpenGL error. +printGLError :: IO () +printGLError = + getGLError >>= \err -> case err of + Nothing -> return () + Just str -> hPutStrLn stderr str + +-- | Run the given setup action and check for OpenGL errors. +-- +-- If an OpenGL error is produced, an exception is thrown containing +-- the given string appended to the string describing the error. +assertGL :: Game s a -> String -> Game s a +assertGL action err = do + result <- action + status <- liftIO getGLError + case status of + Just str -> gameError $ "OpenGL error raised: " ++ err ++ "; " ++ str + Nothing -> return result diff --git a/Spear/old/Spear/Math/Collision.hs b/Spear/old/Spear/Math/Collision.hs new file mode 100644 index 0000000..4412b10 --- /dev/null +++ b/Spear/old/Spear/Math/Collision.hs @@ -0,0 +1,239 @@ +{-# LANGUAGE NoImplicitPrelude #-} + +module Spear.Math.Collision +( + CollisionType(..) + -- * 2D Collision +, Collisionable2(..) +, Collisioner2(..) + -- ** Construction +, aabb2Collisioner +, circleCollisioner +, mkCols + -- ** Collision test +, collide + -- ** Manipulation +, move + -- ** Helpers +, buildAABB2 +, aabb2FromCircle +, circleFromAABB2 + -- * 3D Collision +, Collisionable3(..) + -- ** Helpers +, aabb3FromSphere +) +where + +import Spear.Assets.Model +import Spear.Math.AABB +import Spear.Math.Algebra +import Spear.Math.Circle +import qualified Spear.Math.Matrix4 as M4 +import Spear.Math.Plane +import Spear.Math.Sphere +import Spear.Math.Vector +import Spear.Prelude + +import Data.List (foldl') + +data CollisionType = NoCollision | Collision | FullyContains | FullyContainedBy + deriving (Eq, Show) + +-- 2D collision + +class Collisionable2 a where + -- | Collide the object with an AABB. + collideAABB2 :: AABB2 -> a -> CollisionType + + -- | Collide the object with a circle. + collideCircle :: Circle -> a -> CollisionType + +instance Collisionable2 AABB2 where + collideAABB2 box1@(AABB2 min1 max1) box2@(AABB2 min2 max2) + | (x max1) < (x min2) = NoCollision + | (x min1) > (x max2) = NoCollision + | (y max1) < (y min2) = NoCollision + | (y min1) > (y max2) = NoCollision + | box1 `aabb2pt` min2 && box1 `aabb2pt` max2 = FullyContains + | box2 `aabb2pt` min1 && box2 `aabb2pt` max1 = FullyContainedBy + | otherwise = Collision + + collideCircle circle@(Circle c r) aabb@(AABB2 min max) + | test == FullyContains || test == FullyContainedBy = test + | normSq (c - boxC) > (l + r)^2 = NoCollision + | otherwise = Collision + where + test = collideAABB2 aabb $ aabb2FromCircle circle + boxC = min + (max-min) / (2::Float) + l = norm $ min + (vec2 (x boxC) (y min)) - min + +instance Collisionable2 Circle where + collideAABB2 box circle = case collideCircle circle box of + FullyContains -> FullyContainedBy + FullyContainedBy -> FullyContains + x -> x + + collideCircle s1@(Circle c1 r1) s2@(Circle c2 r2) + | distance_centers <= sub_radii = if (r1 > r2) then FullyContains else FullyContainedBy + | distance_centers <= sum_radii = Collision + | otherwise = NoCollision + where + distance_centers = normSq $ c1 - c2 + sum_radii = (r1 + r2)^2 + sub_radii = (r1 - r2)^2 + +instance Collisionable2 Collisioner2 where + collideAABB2 box (AABB2Col self) = collideAABB2 box self + collideAABB2 box (CircleCol self) = collideAABB2 box self + + collideCircle circle (AABB2Col self) = collideCircle circle self + collideCircle circle (CircleCol self) = collideCircle circle self + + +aabbPoints :: AABB2 -> [Vector2] +aabbPoints (AABB2 min max) = [p1,p2,p3,p4,p5,p6,p7,p8] + where + p1 = vec2 (x min) (y min) + p2 = vec2 (x min) (y min) + p3 = vec2 (x min) (y max) + p4 = vec2 (x min) (y max) + p5 = vec2 (x max) (y min) + p6 = vec2 (x max) (y min) + p7 = vec2 (x max) (y max) + p8 = vec2 (x max) (y max) + + +-- | A collisioner component. +data Collisioner2 + -- | An axis-aligned bounding box. + = AABB2Col {-# UNPACK #-} !AABB2 + -- | A bounding circle. + | CircleCol {-# UNPACK #-} !Circle + + +-- | Create a collisioner from the specified box. +aabb2Collisioner :: AABB2 -> Collisioner2 +aabb2Collisioner = AABB2Col + +-- | Create a collisioner from the specified circle. +circleCollisioner :: Circle -> Collisioner2 +circleCollisioner = CircleCol + +-- | Compute AABB collisioners in view space from the given AABB. +mkCols :: M4.Matrix4 -- ^ Modelview matrix + -> Box + -> [Collisioner2] +mkCols modelview (Box (Vec3 xmin ymin zmin) (Vec3 xmax ymax zmax)) = + let + toVec2 v = vec2 (x v) (y v) + p1 = toVec2 $ modelview `M4.mulp` vec3 xmin ymin zmax + p2 = toVec2 $ modelview `M4.mulp` vec3 xmax ymin zmin + p3 = toVec2 $ modelview `M4.mulp` vec3 xmax ymax zmin + col1 = AABB2Col $ AABB2 p1 p2 + col2 = AABB2Col $ AABB2 p1 p3 + in + [col1, col2] + +-- | Create the minimal AABB fully containing the specified collisioners. +buildAABB2 :: [Collisioner2] -> AABB2 +buildAABB2 cols = aabb2 $ generatePoints cols + +-- | Create the minimal box fully containing the specified circle. +aabb2FromCircle :: Circle -> AABB2 +aabb2FromCircle (Circle c r) = AABB2 bot top + where + bot = c - vec2 r r + top = c + vec2 r r + +-- | Create the minimal circle fully containing the specified box. +circleFromAABB2 :: AABB2 -> Circle +circleFromAABB2 (AABB2 min max) = Circle c r + where + c = (0.5::Float) * (min + max) + r = norm . (*(0.5::Float)) $ max - min + +generatePoints :: [Collisioner2] -> [Vector2] +generatePoints = foldl' generate [] + where + generate acc (AABB2Col (AABB2 pmin pmax)) = p1:p2:p3:p4:p5:p6:p7:p8:acc + where + p1 = vec2 (x pmin) (y pmin) + p2 = vec2 (x pmin) (y pmin) + p3 = vec2 (x pmin) (y pmax) + p4 = vec2 (x pmin) (y pmax) + p5 = vec2 (x pmax) (y pmin) + p6 = vec2 (x pmax) (y pmin) + p7 = vec2 (x pmax) (y pmax) + p8 = vec2 (x pmax) (y pmax) + + generate acc (CircleCol (Circle c r)) = p1:p2:p3:p4:acc + where + p1 = c + unitx2 * vec2 r r + p2 = c - unitx2 * vec2 r r + p3 = c + unity2 * vec2 r r + p4 = c - unity2 * vec2 r r + +-- | Collide the given collisioners. +collide :: Collisioner2 -> Collisioner2 -> CollisionType +collide (AABB2Col box1) (AABB2Col box2) = collideAABB2 box1 box2 +collide (AABB2Col box) (CircleCol circle) = collideAABB2 box circle +collide (CircleCol s1) (CircleCol s2) = collideCircle s1 s2 +collide (CircleCol circle) (AABB2Col box) = collideCircle circle box + +-- | Move the collisioner. +move :: Vector2 -> Collisioner2 -> Collisioner2 +move v (AABB2Col (AABB2 min max)) = AABB2Col (AABB2 (min+v) (max+v)) +move v (CircleCol (Circle c r)) = CircleCol (Circle (c+v) r) + + +-- | 3D collision +class Collisionable3 a where + -- | Collide the object with an AABB. + collideAABB3 :: AABB3 -> a -> CollisionType + + -- | Collide the object with a sphere. + collideSphere :: Sphere -> a -> CollisionType + +instance Collisionable3 AABB3 where + collideAABB3 box1@(AABB3 min1 max1) box2@(AABB3 min2 max2) + | x max1 < x min2 = NoCollision + | x min1 > x max2 = NoCollision + | y max1 < y min2 = NoCollision + | y min1 > y max2 = NoCollision + | box1 `aabb3pt` min2 && box1 `aabb3pt` max2 = FullyContains + | box2 `aabb3pt` min1 && box2 `aabb3pt` max1 = FullyContainedBy + | otherwise = Collision + + collideSphere sphere@(Sphere c r) aabb@(AABB3 min max) + | test == FullyContains || test == FullyContainedBy = test + | normSq (c - boxC) > (l + r)^2 = NoCollision + | otherwise = Collision + where + test = collideAABB3 aabb $ aabb3FromSphere sphere + boxC = min + v + l = norm v + v = (max-min) / (2::Float) + +instance Collisionable3 Sphere where + collideAABB3 box sphere = case collideSphere sphere box of + FullyContains -> FullyContainedBy + FullyContainedBy -> FullyContains + x -> x + + collideSphere s1@(Sphere c1 r1) s2@(Sphere c2 r2) + | distance_centers <= sub_radii = + if r1 > r2 then FullyContains else FullyContainedBy + | distance_centers <= sum_radii = Collision + | otherwise = NoCollision + where + distance_centers = normSq $ c1 - c2 + sum_radii = (r1 + r2)^2 + sub_radii = (r1 - r2)^2 + +-- | Create the minimal box fully containing the specified sphere. +aabb3FromSphere :: Sphere -> AABB3 +aabb3FromSphere (Sphere c r) = AABB3 bot top + where + bot = c - vec3 r r r + top = c + vec3 r r r diff --git a/Spear/old/Spear/Math/Octree.hs b/Spear/old/Spear/Math/Octree.hs new file mode 100644 index 0000000..6c22468 --- /dev/null +++ b/Spear/old/Spear/Math/Octree.hs @@ -0,0 +1,228 @@ +module Spear.Math.Octree +( + Octree +, makeOctree +, clone +, Spear.Math.Octree.insert +, Spear.Math.Octree.map +, gmap +) +where + +import Spear.Math.AABB +import Spear.Math.Collision +import Spear.Math.Vector + +import Control.Applicative ((<*>)) +import Data.List +import Data.Functor +import Data.Monoid +import qualified Data.Foldable as F + +-- | An octree. +data Octree e + = Octree + { root :: !AABB2 + , ents :: ![e] + , c1 :: !(Octree e) + , c2 :: !(Octree e) + , c3 :: !(Octree e) + , c4 :: !(Octree e) + , c5 :: !(Octree e) + , c6 :: !(Octree e) + , c7 :: !(Octree e) + , c8 :: !(Octree e) + } + | + Leaf + { root :: !AABB2 + , ents :: ![e] + } + +-- | Construct an octree using the specified AABB as the root and having the specified depth. +makeOctree :: Int -> AABB2 -> Octree e +makeOctree d root@(AABB2 min max) + | d == 0 = Leaf root [] + | otherwise = Octree root [] c1 c2 c3 c4 c5 c6 c7 c8 + where + boxes = subdivide root + c1 = makeOctree (d-1) $ boxes !! 0 + c2 = makeOctree (d-1) $ boxes !! 1 + c3 = makeOctree (d-1) $ boxes !! 2 + c4 = makeOctree (d-1) $ boxes !! 3 + c5 = makeOctree (d-1) $ boxes !! 4 + c6 = makeOctree (d-1) $ boxes !! 5 + c7 = makeOctree (d-1) $ boxes !! 6 + c8 = makeOctree (d-1) $ boxes !! 7 + +subdivide :: AABB2 -> [AABB2] +subdivide (AABB2 min max) = [a1, a2, a3, a4, a5, a6, a7, a8] + where + v = (max-min) / 2 + c = vec2 (x min + x v) (y min + y v) + a1 = AABB2 min c + a2 = AABB2 ( vec2 (x min) (y min)) ( vec2 (x c) (y c) ) + a3 = AABB2 ( vec2 (x min) (y c) ) ( vec2 (x c) (y max)) + a4 = AABB2 ( vec2 (x min) (y c) ) ( vec2 (x c) (y max)) + a5 = AABB2 ( vec2 (x c) (y min)) ( vec2 (x max) (y c) ) + a6 = AABB2 ( vec2 (x c) (y min)) ( vec2 (x max) (y c) ) + a7 = AABB2 ( vec2 (x c) (y c) ) ( vec2 (x max) (y max)) + a8 = AABB2 c max + +-- | Clone the structure of the octree. The new octree has no entities. +clone :: Octree e -> Octree e +clone (Leaf root ents) = Leaf root [] +clone (Octree root ents c1 c2 c3 c4 c5 c6 c7 c8) = Octree root [] c1' c2' c3' c4' c5' c6' c7' c8' + where + c1' = clone c1 + c2' = clone c2 + c3' = clone c3 + c4' = clone c4 + c5' = clone c5 + c6' = clone c6 + c7' = clone c7 + c8' = clone c8 + +keep :: (e -> AABB2 -> CollisionType) -> AABB2 -> e -> Bool +keep testAABB2 aabb e = test == FullyContainedBy + where test = e `testAABB2` aabb + +-- | Insert a list of entities into the octree. +insert :: (e -> AABB2 -> CollisionType) -> Octree e -> [e] -> Octree e +insert testAABB2 octree es = octree' where (octree', _) = insert' testAABB2 es octree + +insert' :: (e -> AABB2 -> CollisionType) -> [e] -> Octree e -> (Octree e, [e]) + +insert' testAABB2 es (Leaf root ents) = (Leaf root ents', outliers) + where + ents' = ents ++ ents_kept + ents_kept = filter (keep testAABB2 root) es + outliers = filter (not . keep testAABB2 root) es + +insert' testAABB2 es (Octree root ents c1 c2 c3 c4 c5 c6 c7 c8) = + (Octree root ents' c1' c2' c3' c4' c5' c6' c7' c8', outliers) + where + ents' = ents ++ ents_kept + new_ents = es ++ ents1 ++ ents2 ++ ents3 ++ ents4 ++ ents5 ++ ents6 ++ ents7 ++ ents8 + ents_kept = filter (keep testAABB2 root) new_ents + outliers = filter (not . keep testAABB2 root) new_ents + (c1', ents1) = insert' testAABB2 es c1 + (c2', ents2) = insert' testAABB2 es c2 + (c3', ents3) = insert' testAABB2 es c3 + (c4', ents4) = insert' testAABB2 es c4 + (c5', ents5) = insert' testAABB2 es c5 + (c6', ents6) = insert' testAABB2 es c6 + (c7', ents7) = insert' testAABB2 es c7 + (c8', ents8) = insert' testAABB2 es c8 + +-- | Extract all entities from the octree. The resulting octree has no entities. +extract :: Octree e -> (Octree e, [e]) +extract (Leaf root ents) = (Leaf root [], ents) +extract (Octree root ents c1 c2 c3 c4 c5 c6 c7 c8) = (Octree root [] c1' c2' c3' c4' c5' c6' c7' c8', ents') + where + (c1', ents1) = extract c1 + (c2', ents2) = extract c2 + (c3', ents3) = extract c3 + (c4', ents4) = extract c4 + (c5', ents5) = extract c5 + (c6', ents6) = extract c6 + (c7', ents7) = extract c7 + (c8', ents8) = extract c8 + ents' = ents ++ ents1 ++ ents2 ++ ents3 ++ ents4 ++ ents5 ++ ents6 ++ ents7 ++ ents8 + +-- | Apply the given function to the entities in the octree. +-- +-- Entities that break out of their cell are reallocated appropriately. +map :: (e -> AABB2 -> CollisionType) -> (e -> e) -> Octree e -> Octree e +map testAABB2 f o = + let (o', outliers) = map' testAABB2 f o + in Spear.Math.Octree.insert testAABB2 o' outliers + +map' :: (e -> AABB2 -> CollisionType) -> (e -> e) -> Octree e -> (Octree e, [e]) + +map' testAABB2 f (Leaf root ents) = (Leaf root ents_kept, outliers) + where + ents' = fmap f ents + ents_kept = filter (keep testAABB2 root) ents' + outliers = filter (not . keep testAABB2 root) ents' + +map' testAABB2 f (Octree root ents c1 c2 c3 c4 c5 c6 c7 c8) = + (Octree root ents_kept c1' c2' c3' c4' c5' c6' c7' c8', outliers) + where + ents' = (fmap f ents) ++ out1 ++ out2 ++ out3 ++ out4 ++ out5 ++ out6 ++ out7 ++ out8 + ents_kept = filter (keep testAABB2 root) ents' + outliers = filter (not . keep testAABB2 root) ents' + (c1', out1) = map' testAABB2 f c1 + (c2', out2) = map' testAABB2 f c2 + (c3', out3) = map' testAABB2 f c3 + (c4', out4) = map' testAABB2 f c4 + (c5', out5) = map' testAABB2 f c5 + (c6', out6) = map' testAABB2 f c6 + (c7', out7) = map' testAABB2 f c7 + (c8', out8) = map' testAABB2 f c8 + + +-- | Apply a function to the entity groups in the octree. +-- +-- Entities that break out of their cell are reallocated appropriately. +gmap :: (e -> AABB2 -> CollisionType) -> (e -> e -> e) -> Octree e -> Octree e +gmap testAABB2 f o = + let (o', outliers) = gmap' testAABB2 f o + in Spear.Math.Octree.insert testAABB2 o' outliers + +gmap' :: (e -> AABB2 -> CollisionType) -> (e -> e -> e) -> Octree e -> (Octree e, [e]) + +gmap' testAABB2 f (Leaf root ents) = (Leaf root ents_kept, outliers) + where + ents' = f <$> ents <*> ents + ents_kept = filter (keep testAABB2 root) ents' + outliers = filter (not . keep testAABB2 root) ents' + +gmap' testAABB2 f (Octree root ents c1 c2 c3 c4 c5 c6 c7 c8) = + (Octree root ents_kept c1' c2' c3' c4' c5' c6' c7' c8', outliers) + where + ents'= (f <$> ents <*> ents) ++ out1 ++ out2 ++ out3 ++ out4 ++ out5 ++ out6 ++ out7 ++ out8 + ents_kept = filter (keep testAABB2 root) ents' + outliers = filter (not . keep testAABB2 root) ents' + (c1', out1) = gmap' testAABB2 f c1 + (c2', out2) = gmap' testAABB2 f c2 + (c3', out3) = gmap' testAABB2 f c3 + (c4', out4) = gmap' testAABB2 f c4 + (c5', out5) = gmap' testAABB2 f c5 + (c6', out6) = gmap' testAABB2 f c6 + (c7', out7) = gmap' testAABB2 f c7 + (c8', out8) = gmap' testAABB2 f c8 + +instance Functor Octree where + + fmap f (Leaf root ents) = Leaf root $ fmap f ents + + fmap f (Octree root ents c1 c2 c3 c4 c5 c6 c7 c8) = + Octree root (fmap f ents) c1' c2' c3' c4' c5' c6' c7' c8' + where + c1' = fmap f c1 + c2' = fmap f c2 + c3' = fmap f c3 + c4' = fmap f c4 + c5' = fmap f c5 + c6' = fmap f c6 + c7' = fmap f c7 + c8' = fmap f c8 + +instance F.Foldable Octree where + + foldMap f (Leaf root ents) = mconcat . fmap f $ ents + + foldMap f (Octree root ents c1 c2 c3 c4 c5 c6 c7 c8) = + mconcat (fmap f ents) `mappend` + c1' `mappend` c2' `mappend` c3' `mappend` c4' `mappend` + c5' `mappend` c6' `mappend` c7' `mappend` c8' + where + c1' = F.foldMap f c1 + c2' = F.foldMap f c2 + c3' = F.foldMap f c3 + c4' = F.foldMap f c4 + c5' = F.foldMap f c5 + c6' = F.foldMap f c6 + c7' = F.foldMap f c7 + c8' = F.foldMap f c8 diff --git a/Spear/old/Spear/Render/AnimatedModel.hs b/Spear/old/Spear/Render/AnimatedModel.hs new file mode 100644 index 0000000..e5b29ec --- /dev/null +++ b/Spear/old/Spear/Render/AnimatedModel.hs @@ -0,0 +1,262 @@ + +{-# LANGUAGE NoImplicitPrelude #-} + +module Spear.Render.AnimatedModel + ( -- * Data types + AnimatedModelResource, + AnimatedModelRenderer, + AnimationSpeed, + + -- * Construction and destruction + animatedModelResource, + animatedModelRenderer, + + -- * Accessors + animationSpeed, + box, + currentAnimation, + currentFrame, + frameProgress, + modelRes, + nextFrame, + + -- * Manipulation + update, + setAnimation, + setAnimationSpeed, + + -- * Rendering + bind, + render, + + -- * Collision + mkColsFromAnimated, + ) +where + +import Spear.Assets.Model +import Spear.Game +import Spear.GL +import Spear.Math.AABB +import Spear.Math.Algebra +import Spear.Math.Collision +import Spear.Math.Matrix4 (Matrix4) +import Spear.Math.Vector +import Spear.Prelude +import Spear.Render.Material +import Spear.Render.Model +import Spear.Render.Program + +import Control.Applicative ((<$>), (<*>)) +import qualified Data.Vector as V +import Foreign.C.Types +import Unsafe.Coerce (unsafeCoerce) + + +type AnimationSpeed = Float + +-- | An animated model resource. +-- +-- Contains model data necessary to render an animated model. +data AnimatedModelResource = AnimatedModelResource + { model :: Model, + vao :: VAO, + nFrames :: Int, + nVertices :: Int, + material :: Material, + texture :: Texture, + boxes :: V.Vector Box, + rkey :: ReleaseKey + } + +instance Eq AnimatedModelResource where + m1 == m2 = vao m1 == vao m2 + +instance Ord AnimatedModelResource where + m1 < m2 = vao m1 < vao m2 + m1 <= m2 = vao m1 <= vao m2 + +instance ResourceClass AnimatedModelResource where + getResource = rkey + +-- | An animated model renderer. +-- +-- Holds animation data necessary to render an animated model and a reference +-- to an 'AnimatedModelResource'. +-- +-- Model data is kept separate from animation data. This allows instances +-- of 'AnimatedModelRenderer' to share the underlying 'AnimatedModelResource', +-- minimising the amount of data in memory and allowing one to minimise OpenGL +-- state changes by sorting 'AnimatedModelRenderer's by their underlying +-- 'AnimatedModelResource' when rendering the scene. +data AnimatedModelRenderer = AnimatedModelRenderer + { modelResource :: AnimatedModelResource, + currentAnim :: Int, + frameStart :: Int, + frameEnd :: Int, + -- | Get the renderer's current frame. + currentFrame :: Int, + -- | Get the renderer's frame progress. + frameProgress :: Float, + -- | Get the renderer's animation speed. + animationSpeed :: Float + } + +instance Eq AnimatedModelRenderer where + m1 == m2 = modelResource m1 == modelResource m2 + +instance Ord AnimatedModelRenderer where + m1 < m2 = modelResource m1 < modelResource m2 + m1 <= m2 = modelResource m1 <= modelResource m2 + +-- | Create an model resource from the given model. +animatedModelResource :: + AnimatedProgramChannels -> + Material -> + Texture -> + Model -> + Game s AnimatedModelResource +animatedModelResource + (AnimatedProgramChannels vertChan1 vertChan2 normChan1 normChan2 texChan) + material + texture + model = do + RenderModel elements numFrames numVertices <- liftIO . renderModelFromModel $ model + elementBuf <- newBuffer + vao <- newVAO + boxes <- liftIO $ modelBoxes model + + liftIO $ do + let elemSize = 56::CUInt + elemSize' = fromIntegral elemSize + n = numVertices * numFrames + + bindVAO vao + + bindBuffer ArrayBuffer elementBuf + bufferData' ArrayBuffer (unsafeCoerce $ elemSize * n) elements StaticDraw + + attribVAOPointer vertChan1 3 GL_FLOAT False elemSize' 0 + attribVAOPointer vertChan2 3 GL_FLOAT False elemSize' 12 + attribVAOPointer normChan1 3 GL_FLOAT False elemSize' 24 + attribVAOPointer normChan2 3 GL_FLOAT False elemSize' 36 + attribVAOPointer texChan 2 GL_FLOAT False elemSize' 48 + + enableVAOAttrib vertChan1 + enableVAOAttrib vertChan2 + enableVAOAttrib normChan1 + enableVAOAttrib normChan2 + enableVAOAttrib texChan + + rkey <- register $ do + release $ vaoKey vao + release $ bufferKey elementBuf + + return $ + AnimatedModelResource + model + vao + (unsafeCoerce numFrames) + (unsafeCoerce numVertices) + material + texture + boxes + rkey + +-- | Create a renderer from the given model resource. +animatedModelRenderer :: AnimationSpeed -> AnimatedModelResource -> AnimatedModelRenderer +animatedModelRenderer animSpeed modelResource = + AnimatedModelRenderer modelResource 0 0 0 0 0 animSpeed + +-- | Update the renderer. +update :: Float -> AnimatedModelRenderer -> AnimatedModelRenderer +update dt (AnimatedModelRenderer model curAnim startFrame endFrame curFrame fp s) = + AnimatedModelRenderer model curAnim startFrame endFrame curFrame' fp' s + where + f = fp + dt * s + nextFrame = f >= 1.0 + fp' = if nextFrame then f - (1::Float) else f + curFrame' = + if nextFrame + then + let x = curFrame + (1::Int) + in if x > endFrame then startFrame else x + else curFrame + +-- | Get the model's ith bounding box. +box :: Int -> AnimatedModelResource -> Box +box i model = boxes model V.! i + +-- | Get the renderer's current animation. +currentAnimation :: Enum a => AnimatedModelRenderer -> a +currentAnimation = toEnum . currentAnim + +-- | Get the renderer's model resource. +modelRes :: AnimatedModelRenderer -> AnimatedModelResource +modelRes = modelResource + +-- | Get the renderer's next frame. +nextFrame :: AnimatedModelRenderer -> Int +nextFrame rend = + let curFrame = currentFrame rend + in if curFrame == frameEnd rend + then frameStart rend + else curFrame + (1::Int) + +-- | Set the active animation to the given one. +setAnimation :: Enum a => a -> AnimatedModelRenderer -> AnimatedModelRenderer +setAnimation anim modelRend = + let (Animation _ f1 f2) = animation (model . modelResource $ modelRend) anim' + anim' = fromEnum anim + in modelRend {currentAnim = anim', frameStart = f1, frameEnd = f2, currentFrame = f1} + +-- | Set the renderer's animation speed. +setAnimationSpeed :: AnimationSpeed -> AnimatedModelRenderer -> AnimatedModelRenderer +setAnimationSpeed s r = r {animationSpeed = s} + +-- | Bind the given renderer to prepare it for rendering. +bind :: AnimatedProgramUniforms -> AnimatedModelRenderer -> IO () +bind (AnimatedProgramUniforms kaLoc kdLoc ksLoc shiLoc texLoc _ _ _ _) modelRend = + let model' = modelResource modelRend + in do + bindVAO . vao $ model' + bindTexture $ texture model' + activeTexture $= GL_TEXTURE0 + glUniform1i texLoc 0 + +-- | Render the model described by the given renderer. +render :: AnimatedProgramUniforms -> AnimatedModelRenderer -> IO () +render uniforms (AnimatedModelRenderer model _ _ _ curFrame fp _) = + let n = nVertices model + (Material _ ka kd ks shi) = material model + in do + uniform (kaLoc uniforms) ka + uniform (kdLoc uniforms) kd + uniform (ksLoc uniforms) ks + glUniform1f (shiLoc uniforms) $ unsafeCoerce shi + glUniform1f (fpLoc uniforms) (unsafeCoerce fp) + drawArrays GL_TRIANGLES (n * curFrame) n + +-- | Compute AABB collisioners in view space from the given model. +mkColsFromAnimated :: + -- | Source frame + Int -> + -- | Dest frame + Int -> + -- | Frame progress + Float -> + -- | Modelview matrix + Matrix4 -> + AnimatedModelResource -> + [Collisioner2] +mkColsFromAnimated f1 f2 fp modelview modelRes = + let (Box (Vec3 xmin1 ymin1 zmin1) (Vec3 xmax1 ymax1 zmax1)) = box f1 modelRes + (Box (Vec3 xmin2 ymin2 zmin2) (Vec3 xmax2 ymax2 zmax2)) = box f2 modelRes + min1 = vec3 xmin1 ymin1 zmin1 + max1 = vec3 xmax1 ymax1 zmax1 + min2 = vec3 xmin2 ymin2 zmin2 + max2 = vec3 xmax2 ymax2 zmax2 + min = min1 + fp * (min2 - min1) + max = max1 + fp * (max2 - max1) + in mkCols modelview $ + Box (Vec3 (x min) (y min) (z min)) (Vec3 (x max) (y max) (z max)) diff --git a/Spear/old/Spear/Render/Box.hs b/Spear/old/Spear/Render/Box.hs new file mode 100644 index 0000000..305ef32 --- /dev/null +++ b/Spear/old/Spear/Render/Box.hs @@ -0,0 +1,193 @@ +module Spear.Render.Box +( + render +, renderOutwards +, renderInwards +, renderEdges +) +where + + +import Spear.Math.Vector3 +import Spear.Math.Matrix +import Graphics.Rendering.OpenGL.Raw +import Unsafe.Coerce +import Control.Monad.Instances + +type Center = Vector3 +type Colour = Vector4 +type Length = Float +type Normals = [Vector3] +type GenerateTexCoords = Bool + + +applyColour :: Colour -> IO () +--applyColour col = glColor4f (unsafeCoerce $ x col) (unsafeCoerce $ y col) (unsafeCoerce $ z col) (unsafeCoerce $ w col) +applyColour = do + ax <- unsafeCoerce . x + ay <- unsafeCoerce . y + az <- unsafeCoerce . z + aw <- unsafeCoerce . w + glColor4f ax ay az aw + + +applyNormal :: Vector3 -> IO () +--applyNormal v = glNormal3f (unsafeCoerce $ x v) (unsafeCoerce $ y v) (unsafeCoerce $ z v) +applyNormal = do + nx <- unsafeCoerce . x + ny <- unsafeCoerce . y + nz <- unsafeCoerce . z + glNormal3f nx ny nz + + +-- | Renders a box. +render :: Center -- ^ The box's center. + -> Length -- ^ The perpendicular distance from the box's center to any of its sides. + -> Colour -- ^ The box's colour. + -> Normals -- ^ The box's normals, of the form [front, back, right, left, top, bottom]. + -> IO () +render c l col normals = do + glPushMatrix + glTranslatef (unsafeCoerce $ x c) (unsafeCoerce $ y c) (unsafeCoerce $ z c) + applyColour col + + let d = unsafeCoerce l + glBegin gl_QUADS + + --Front + --glNormal3f 0 0 (-1) + applyNormal $ normals !! 0 + glVertex3f d (-d) (-d) + glVertex3f d d (-d) + glVertex3f (-d) d (-d) + glVertex3f (-d) (-d) (-d) + + --Back + --glNormal3f 0 0 1 + applyNormal $ normals !! 1 + glVertex3f (-d) (-d) d + glVertex3f (-d) d d + glVertex3f d d d + glVertex3f d (-d) d + + --Right + --glNormal3f 1 0 0 + applyNormal $ normals !! 2 + glVertex3f d (-d) (-d) + glVertex3f d (-d) d + glVertex3f d d d + glVertex3f d d (-d) + + --Left + --glNormal3f (-1) 0 0 + applyNormal $ normals !! 3 + glVertex3f (-d) (-d) (-d) + glVertex3f (-d) d (-d) + glVertex3f (-d) d d + glVertex3f (-d) (-d) d + + --Top + --glNormal3f 0 1 0 + applyNormal $ normals !! 4 + glVertex3f (-d) d (-d) + glVertex3f d d (-d) + glVertex3f d d d + glVertex3f (-d) d d + + --Bottom + --glNormal3f 0 (-1) 0 + applyNormal $ normals !! 5 + glVertex3f d (-d) d + glVertex3f d (-d) (-d) + glVertex3f (-d) (-d) (-d) + glVertex3f (-d) (-d) d + + glEnd + + glPopMatrix + + +normals = [vec3 0 0 (-1), vec3 0 0 1, vec3 1 0 0, vec3 (-1) 0 0, vec3 0 1 0, vec3 0 (-1) 0] + + +-- | Renders a box with normals facing outwards. +renderOutwards :: Center -- ^ The box's center. + -> Length -- ^ The perpendicular distance from the box's center to any of its sides. + -> Colour -- ^ The box's colour. + -> IO () +renderOutwards c l col = render c l col normals + + +-- | Renders a box with normals facing inwards. +renderInwards :: Center -- ^ The box's center. + -> Length -- ^ The perpendicular distance from the box's center to any of its sides. + -> Colour -- ^ The box's colour. + -> IO () +renderInwards c l col = do + glFrontFace gl_CW + render c l col $ Prelude.map neg normals + glFrontFace gl_CCW + + +renderEdges :: Center -- ^ The box's center. + -> Length -- ^ The perpendicular distance from the box's center to any of its sides. + -> Colour -- ^ The box's colour. + -> IO () +renderEdges c l col = do + glPushMatrix + glTranslatef (unsafeCoerce $ x c) (unsafeCoerce $ y c) (unsafeCoerce $ z c) + applyColour col + + let d = unsafeCoerce l + + --Front + glBegin gl_LINE_STRIP + glVertex3f d (-d) (-d) + glVertex3f d d (-d) + glVertex3f (-d) d (-d) + glVertex3f (-d) (-d) (-d) + glEnd + + --Back + glBegin gl_LINE_STRIP + glVertex3f (-d) (-d) d + glVertex3f (-d) d d + glVertex3f d d d + glVertex3f d (-d) d + glVertex3f (-d) (-d) d + glEnd + + --Right + glBegin gl_LINE_STRIP + glVertex3f d (-d) (-d) + glVertex3f d (-d) d + glVertex3f d d d + glVertex3f d d (-d) + glEnd + + --Left + glBegin gl_LINE_STRIP + glVertex3f (-d) (-d) (-d) + glVertex3f (-d) d (-d) + glVertex3f (-d) d d + glVertex3f (-d) (-d) d + glEnd + + --Top + glBegin gl_LINE_STRIP + glVertex3f (-d) d (-d) + glVertex3f d d (-d) + glVertex3f d d d + glVertex3f (-d) d d + glEnd + + --Bottom + glBegin gl_LINE_STRIP + glVertex3f d (-d) d + glVertex3f d (-d) (-d) + glVertex3f (-d) (-d) (-d) + glVertex3f (-d) (-d) d + glEnd + + glPopMatrix + \ No newline at end of file diff --git a/Spear/old/Spear/Render/Light.hs b/Spear/old/Spear/Render/Light.hs new file mode 100644 index 0000000..cc1de1f --- /dev/null +++ b/Spear/old/Spear/Render/Light.hs @@ -0,0 +1,25 @@ +module Spear.Render.Light +where + + +import Spear.Vector +import Graphics.Rendering.OpenGL.Raw + + +data LightData = LightData { + ambient :: Vector Float, + diffuse :: Vector Float, + spec :: Vector Float +} + + +data Light = + ParallelLight { + ld :: LightData + } +| PointLight { + ld :: LightData + } +| SpotLight { + ld :: LightData + } diff --git a/Spear/old/Spear/Render/Material.hs b/Spear/old/Spear/Render/Material.hs new file mode 100644 index 0000000..d9c60ea --- /dev/null +++ b/Spear/old/Spear/Render/Material.hs @@ -0,0 +1,16 @@ +module Spear.Render.Material +( Material(..) +) +where + + +import Spear.Math.Vector + + +data Material = Material + { ke :: Vector4 + , ka :: Vector4 + , kd :: Vector4 + , ks :: Vector4 + , shininess :: Float + } diff --git a/Spear/old/Spear/Render/Model.hsc b/Spear/old/Spear/Render/Model.hsc new file mode 100644 index 0000000..ba6bf39 --- /dev/null +++ b/Spear/old/Spear/Render/Model.hsc @@ -0,0 +1,54 @@ +{-# LANGUAGE CPP, ForeignFunctionInterface #-} + +module Spear.Render.Model +( + RenderModel(..) +, renderModelFromModel +) +where + +import qualified Spear.Assets.Model as Assets +import Spear.Game + +import Foreign.Ptr +import Foreign.C.Types +import Foreign.Marshal.Alloc +import Foreign.Marshal.Array +import Foreign.Marshal.Utils (with) +import Foreign.Storable + +#include "RenderModel.h" + +data Vec3 = Vec3 !CFloat !CFloat !CFloat + +data TexCoord = TexCoord !CFloat !CFloat + +data RenderModel = RenderModel + { elements :: Ptr CChar + , numFrames :: CUInt + , numVertices :: CUInt -- ^ Number of vertices per frame. + } + +instance Storable RenderModel where + sizeOf _ = #{size RenderModel} + alignment _ = alignment (undefined :: CUInt) + + peek ptr = do + elements <- #{peek RenderModel, elements} ptr + numFrames <- #{peek RenderModel, numFrames} ptr + numVertices <- #{peek RenderModel, numVertices} ptr + return $ RenderModel elements numFrames numVertices + + poke ptr (RenderModel elements numFrames numVertices) = do + #{poke RenderModel, elements} ptr elements + #{poke RenderModel, numFrames} ptr numFrames + #{poke RenderModel, numVertices} ptr numVertices + +foreign import ccall "RenderModel.h render_model_from_model_asset" + render_model_from_model_asset :: Ptr Assets.Model -> Ptr RenderModel -> IO Int + +-- | Convert the given 'Model' to a 'ModelData' instance. +renderModelFromModel :: Assets.Model -> IO RenderModel +renderModelFromModel m = with m $ \mPtr -> alloca $ \mdPtr -> do + render_model_from_model_asset mPtr mdPtr + peek mdPtr diff --git a/Spear/old/Spear/Render/Program.hs b/Spear/old/Spear/Render/Program.hs new file mode 100644 index 0000000..b5a8658 --- /dev/null +++ b/Spear/old/Spear/Render/Program.hs @@ -0,0 +1,102 @@ +module Spear.Render.Program +( + StaticProgram(..) +, AnimatedProgram(..) +, Program(..) +, ProgramUniforms(..) +, StaticProgramChannels(..) +, StaticProgramUniforms(..) +, AnimatedProgramChannels(..) +, AnimatedProgramUniforms(..) +) +where + +import Spear.GL + +data StaticProgram = StaticProgram + { staticProgram :: GLSLProgram + , staticProgramChannels :: StaticProgramChannels + , staticProgramUniforms :: StaticProgramUniforms + } + +data AnimatedProgram = AnimatedProgram + { animatedProgram :: GLSLProgram + , animatedProgramChannels :: AnimatedProgramChannels + , animatedProgramUniforms :: AnimatedProgramUniforms + } + +data StaticProgramChannels = StaticProgramChannels + { vertexChannel :: GLuint -- ^ Vertex channel. + , normalChannel :: GLuint -- ^ Normal channel. + , stexChannel :: GLuint -- ^ Texture channel. + } + +data AnimatedProgramChannels = AnimatedProgramChannels + { vertexChannel1 :: GLuint -- ^ Vertex channel 1. + , vertexChannel2 :: GLuint -- ^ Vertex channel 2. + , normalChannel1 :: GLuint -- ^ Normal channel 1. + , normalChannel2 :: GLuint -- ^ Normal channel 2. + , atexChannel :: GLuint -- ^ Texture channel. + } + +data StaticProgramUniforms = StaticProgramUniforms + { skaLoc :: GLint -- ^ Material ambient uniform location. + , skdLoc :: GLint -- ^ Material diffuse uniform location. + , sksLoc :: GLint -- ^ Material specular uniform location. + , sshiLoc :: GLint -- ^ Material shininess uniform location. + , stexLoc :: GLint -- ^ Texture sampler location. + , smodelviewLoc :: GLint -- ^ Modelview matrix location. + , snormalmatLoc :: GLint -- ^ Normal matrix location. + , sprojLoc :: GLint -- ^ Projection matrix location. + } + +data AnimatedProgramUniforms = AnimatedProgramUniforms + { akaLoc :: GLint -- ^ Material ambient uniform location. + , akdLoc :: GLint -- ^ Material diffuse uniform location. + , aksLoc :: GLint -- ^ Material specular uniform location. + , ashiLoc :: GLint -- ^ Material shininess uniform location. + , atexLoc :: GLint -- ^ Texture sampler location. + , fpLoc :: GLint -- ^ Frame progress uniform location. + , amodelviewLoc :: GLint -- ^ Modelview matrix location. + , anormalmatLoc :: GLint -- ^ Normal matrix location. + , aprojLoc :: GLint -- ^ Projection matrix location. + } + +class Program a where + program :: a -> GLSLProgram + +instance Program StaticProgram where + program = staticProgram + +instance Program AnimatedProgram where + program = animatedProgram + +class ProgramUniforms a where + kaLoc :: a -> GLint + kdLoc :: a -> GLint + ksLoc :: a -> GLint + shiLoc :: a -> GLint + texLoc :: a -> GLint + modelviewLoc :: a -> GLint + normalmatLoc :: a -> GLint + projLoc :: a -> GLint + +instance ProgramUniforms StaticProgramUniforms where + kaLoc = skaLoc + kdLoc = skdLoc + ksLoc = sksLoc + shiLoc = sshiLoc + texLoc = stexLoc + modelviewLoc = smodelviewLoc + normalmatLoc = snormalmatLoc + projLoc = sprojLoc + +instance ProgramUniforms AnimatedProgramUniforms where + kaLoc = akaLoc + kdLoc = akdLoc + ksLoc = aksLoc + shiLoc = ashiLoc + texLoc = atexLoc + modelviewLoc = amodelviewLoc + normalmatLoc = anormalmatLoc + projLoc = aprojLoc diff --git a/Spear/old/Spear/Render/RenderModel.c b/Spear/old/Spear/Render/RenderModel.c new file mode 100644 index 0000000..1543052 --- /dev/null +++ b/Spear/old/Spear/Render/RenderModel.c @@ -0,0 +1,232 @@ +#include "RenderModel.h" +#include // free +#include // memcpy +#include + + +static void safe_free (void* ptr) +{ + if (ptr) + { + free (ptr); + ptr = 0; + } +} + + +/// Populate elements of an animated model to be rendered from +/// start to end in a loop. +/*int populate_elements_animated (Model* model_asset, RenderModel* model) +{ + size_t nverts = model_asset->numVertices; + size_t ntriangles = model_asset->numTriangles; + size_t nframes = model_asset->numFrames; + size_t n = nframes * ntriangles * 3; + + model->elements = malloc (56 * n); + if (!model->elements) return -1; + + // Populate elements. + + size_t f, i; + + char* elem = (char*) model->elements; + vec3* v1 = model_asset->vertices; + vec3* v2 = v1 + nverts; + vec3* n1 = model_asset->normals; + vec3* n2 = n1 + nverts; + texCoord* tex = model_asset->texCoords; + + for (f = 0; f < nframes; ++f) + { + triangle* t = model_asset->triangles; + + for (i = 0; i < ntriangles; ++i) + { + *((vec3*) elem) = v1[t->vertexIndices[0]]; + *((vec3*) (elem + 12)) = v2[t->vertexIndices[0]]; + *((vec3*) (elem + 24)) = n1[t->vertexIndices[0]]; + *((vec3*) (elem + 36)) = n2[t->vertexIndices[0]]; + *((texCoord*) (elem + 48)) = tex[t->textureIndices[0]]; + elem += 56; + + *((vec3*) elem) = v1[t->vertexIndices[1]]; + *((vec3*) (elem + 12)) = v2[t->vertexIndices[1]]; + *((vec3*) (elem + 24)) = n1[t->vertexIndices[1]]; + *((vec3*) (elem + 36)) = n2[t->vertexIndices[1]]; + *((texCoord*) (elem + 48)) = tex[t->textureIndices[1]]; + elem += 56; + + *((vec3*) elem) = v1[t->vertexIndices[2]]; + *((vec3*) (elem + 12)) = v2[t->vertexIndices[2]]; + *((vec3*) (elem + 24)) = n1[t->vertexIndices[2]]; + *((vec3*) (elem + 36)) = n2[t->vertexIndices[2]]; + *((texCoord*) (elem + 48)) = tex[t->textureIndices[2]]; + elem += 56; + + t++; + } + + v1 += nverts; + v2 += nverts; + n1 += nverts; + n2 += nverts; + + if (f == nframes-2) + { + v2 = model_asset->vertices; + n2 = model_asset->normals; + } + } + + return 0; +}*/ + + +/// Populate elements of an animated model according to its frames +/// of animation. +int populate_elements_animated (Model* model_asset, RenderModel* model) +{ + size_t nverts = model_asset->numVertices; + size_t ntriangles = model_asset->numTriangles; + size_t nframes = model_asset->numFrames; + size_t n = nframes * ntriangles * 3; + + model->elements = malloc (56 * n); + if (!model->elements) return -1; + + // Populate elements. + + unsigned f, i, j, u; + + char* elem = (char*) model->elements; + animation* anim = model_asset->animations; + + for (i = 0; i < model_asset->numAnimations; ++i, anim++) + { + unsigned start = anim->start; + unsigned end = anim->end; + + char singleFrameAnim = start == end; + + vec3* v1 = model_asset->vertices + start*nverts; + vec3* v2 = singleFrameAnim ? v1 : v1 + nverts; + vec3* n1 = model_asset->normals + start*nverts; + vec3* n2 = singleFrameAnim ? n1 : n1 + nverts; + texCoord* tex = model_asset->texCoords; + + for (u = start; u <= end; ++u) + { + triangle* t = model_asset->triangles; + + for (j = 0; j < ntriangles; ++j, t++) + { + *((vec3*) elem) = v1[t->vertexIndices[0]]; + *((vec3*) (elem + 12)) = v2[t->vertexIndices[0]]; + *((vec3*) (elem + 24)) = n1[t->vertexIndices[0]]; + *((vec3*) (elem + 36)) = n2[t->vertexIndices[0]]; + *((texCoord*) (elem + 48)) = tex[t->textureIndices[0]]; + elem += 56; + + *((vec3*) elem) = v1[t->vertexIndices[1]]; + *((vec3*) (elem + 12)) = v2[t->vertexIndices[1]]; + *((vec3*) (elem + 24)) = n1[t->vertexIndices[1]]; + *((vec3*) (elem + 36)) = n2[t->vertexIndices[1]]; + *((texCoord*) (elem + 48)) = tex[t->textureIndices[1]]; + elem += 56; + + *((vec3*) elem) = v1[t->vertexIndices[2]]; + *((vec3*) (elem + 12)) = v2[t->vertexIndices[2]]; + *((vec3*) (elem + 24)) = n1[t->vertexIndices[2]]; + *((vec3*) (elem + 36)) = n2[t->vertexIndices[2]]; + *((texCoord*) (elem + 48)) = tex[t->textureIndices[2]]; + elem += 56; + } + + // Advance to the next frame of animation of the current + // animation. + v1 += nverts; + v2 += nverts; + n1 += nverts; + n2 += nverts; + + // Reset the secondary pointers to the beginning of the + // animation when we are about to reach the last frame. + if (u == end-1) + { + v2 = model_asset->vertices + start*nverts; + n2 = model_asset->normals + start*nverts; + } + } + } + + return 0; +} + + +int populate_elements_static (Model* model_asset, RenderModel* model) +{ + size_t nverts = model_asset->numVertices; + size_t ntriangles = model_asset->numTriangles; + size_t n = ntriangles * 3; + + model->elements = malloc (32 * n); + if (!model->elements) return -1; + + // Populate elements. + + size_t f, i; + + char* elem = (char*) model->elements; + vec3* vert = model_asset->vertices; + vec3* norm = model_asset->normals; + texCoord* tex = model_asset->texCoords; + + triangle* t = model_asset->triangles; + + for (i = 0; i < ntriangles; ++i) + { + *((vec3*) elem) = vert[t->vertexIndices[0]]; + *((vec3*) (elem + 12)) = norm[t->vertexIndices[0]]; + *((texCoord*) (elem + 24)) = tex[t->textureIndices[0]]; + elem += 32; + + *((vec3*) elem) = vert[t->vertexIndices[1]]; + *((vec3*) (elem + 12)) = norm[t->vertexIndices[1]]; + *((texCoord*) (elem + 24)) = tex[t->textureIndices[1]]; + elem += 32; + + *((vec3*) elem) = vert[t->vertexIndices[2]]; + *((vec3*) (elem + 12)) = norm[t->vertexIndices[2]]; + *((texCoord*) (elem + 24)) = tex[t->textureIndices[2]]; + elem += 32; + + t++; + } + + return 0; +} + + +int render_model_from_model_asset (Model* model_asset, RenderModel* model) +{ + U32 ntriangles = model_asset->numTriangles; + U32 nframes = model_asset->numFrames; + + int result; + if (nframes > 1) result = populate_elements_animated (model_asset, model); + else result = populate_elements_static (model_asset, model); + + if (result != 0) return result; + + model->numFrames = nframes; + model->numVertices = ntriangles * 3; // Number of vertices per frame. + + return 0; +} + + +void render_model_free (RenderModel* model) +{ + safe_free (model->elements); +} diff --git a/Spear/old/Spear/Render/RenderModel.h b/Spear/old/Spear/Render/RenderModel.h new file mode 100644 index 0000000..6a5fb5e --- /dev/null +++ b/Spear/old/Spear/Render/RenderModel.h @@ -0,0 +1,49 @@ +#ifndef _SPEAR_RENDER_MODEL_H +#define _SPEAR_RENDER_MODEL_H + +#include "Model.h" + + +/// Represents a renderable model. +/** + * If the model is animated: + * + * Buffer layout: + * vert1 vert2 norm1 norm2 texc + * + * element size = (3 + 3 + 3 + 3 + 2)*4 = 56 B + * buffer size = element size * num vertices = 56n + * + * If the model is static: + * + * Buffer layout: + * vert norm texc + * + * element size = (3 + 3 + 2)*4 = 32 B + * buffer size = element size * num vertices = 32n + * + **/ +typedef struct +{ + void* elements; + U32 numFrames; + U32 numVertices; // Number of vertices per frame. +} +RenderModel; + + +#ifdef __cplusplus +extern "C" { +#endif + +int render_model_from_model_asset (Model* model_asset, RenderModel* render_model); + +void render_model_free (RenderModel* model); + +#ifdef __cplusplus +} +#endif + + +#endif // _SPEAR_RENDER_MODEL_H + diff --git a/Spear/old/Spear/Render/Sphere.hs b/Spear/old/Spear/Render/Sphere.hs new file mode 100644 index 0000000..4e74375 --- /dev/null +++ b/Spear/old/Spear/Render/Sphere.hs @@ -0,0 +1,45 @@ +module Spear.Render.Sphere +( + render +) +where + + +import Spear.Math.Vector as Vector +import Spear.Math.Matrix +import Graphics.Rendering.OpenGL.Raw +import Graphics.Rendering.OpenGL.GL.Colors +import qualified Graphics.Rendering.OpenGL.GLU as GLU +import Unsafe.Coerce + + +type Center = Vector R +type Radius = R +type Colour = Vector R + + +applyColour :: Colour -> IO () +applyColour col = + if Vector.length col == 4 then + glColor4f (unsafeCoerce $ x col) (unsafeCoerce $ y col) (unsafeCoerce $ z col) + (unsafeCoerce $ w col) + else + glColor3f (unsafeCoerce $ x col) (unsafeCoerce $ y col) (unsafeCoerce $ z col) + + +-- | Renders a sphere. +-- Center is the sphere's center. +-- Radius is the sphere's radius. +-- Colour is a Vector representing the sphere's colour. Colour may hold an alpha channel. +render :: Center -> Radius -> Colour -> IO () +render c radius col = do + glPushMatrix + glTranslatef (unsafeCoerce $ x c) (unsafeCoerce $ y c) (unsafeCoerce $ z c) + applyColour col + + let r = unsafeCoerce $ (realToFrac radius :: Double) + let style = GLU.QuadricStyle (Just Smooth) GLU.NoTextureCoordinates GLU.Outside GLU.FillStyle + GLU.renderQuadric style $ GLU.Sphere r 16 16 + + glPopMatrix + \ No newline at end of file diff --git a/Spear/old/Spear/Render/StaticModel.hs b/Spear/old/Spear/Render/StaticModel.hs new file mode 100644 index 0000000..5168cf2 --- /dev/null +++ b/Spear/old/Spear/Render/StaticModel.hs @@ -0,0 +1,152 @@ +module Spear.Render.StaticModel + ( -- * Data types + StaticModelResource, + StaticModelRenderer, + + -- * Construction and destruction + staticModelResource, + staticModelRenderer, + + -- * Manipulation + box, + modelRes, + + -- * Rendering + bind, + render, + + -- * Collision + mkColsFromStatic, + ) +where + +import Spear.Assets.Model +import Spear.Game +import Spear.GL +import Spear.Math.AABB +import Spear.Math.Algebra +import Spear.Math.Collision +import Spear.Math.Matrix4 (Matrix4) +import Spear.Math.Vector +import Spear.Render.Material +import Spear.Render.Model +import Spear.Render.Program + +import qualified Data.Vector as V +import Foreign.C.Types +import Prelude hiding ((*)) +import Unsafe.Coerce (unsafeCoerce) + + +data StaticModelResource = StaticModelResource + { vao :: VAO, + nVertices :: Int, + material :: Material, + texture :: Texture, + boxes :: V.Vector Box, + rkey :: ReleaseKey + } + +instance Eq StaticModelResource where + m1 == m2 = vao m1 == vao m2 + +instance Ord StaticModelResource where + m1 < m2 = vao m1 < vao m2 + m1 <= m2 = vao m1 <= vao m2 + +instance ResourceClass StaticModelResource where + getResource = rkey + +data StaticModelRenderer = StaticModelRenderer {model :: StaticModelResource} + +instance Eq StaticModelRenderer where + m1 == m2 = model m1 == model m2 + +instance Ord StaticModelRenderer where + m1 < m2 = model m1 < model m2 + m1 <= m2 = model m1 <= model m2 + +-- | Create a model resource from the given model. +staticModelResource :: + StaticProgramChannels -> + Material -> + Texture -> + Model -> + Game s StaticModelResource +staticModelResource (StaticProgramChannels vertChan normChan texChan) material texture model = do + RenderModel elements _ numVertices <- liftIO . renderModelFromModel $ model + elementBuf <- newBuffer + vao <- newVAO + boxes <- liftIO $ modelBoxes model + + liftIO $ do + let elemSize = 32::CUInt + elemSize' = fromIntegral elemSize + n = numVertices + + bindVAO vao + + bindBuffer ArrayBuffer elementBuf + bufferData' ArrayBuffer (fromIntegral $ elemSize * n) elements StaticDraw + + attribVAOPointer vertChan 3 GL_FLOAT False elemSize' 0 + attribVAOPointer normChan 3 GL_FLOAT False elemSize' 12 + attribVAOPointer texChan 2 GL_FLOAT False elemSize' 24 + + enableVAOAttrib vertChan + enableVAOAttrib normChan + enableVAOAttrib texChan + + rkey <- register $ do + release $ vaoKey vao + release $ bufferKey elementBuf + + return $ + StaticModelResource + vao + (unsafeCoerce numVertices) + material + texture + boxes + rkey + +-- | Create a renderer from the given model resource. +staticModelRenderer :: StaticModelResource -> StaticModelRenderer +staticModelRenderer = StaticModelRenderer + +-- | Get the model's ith bounding box. +box :: Int -> StaticModelResource -> Box +box i model = boxes model V.! i + +-- | Get the renderer's model resource. +modelRes :: StaticModelRenderer -> StaticModelResource +modelRes = model + +-- | Bind the given renderer to prepare it for rendering. +bind :: StaticProgramUniforms -> StaticModelRenderer -> IO () +bind (StaticProgramUniforms kaLoc kdLoc ksLoc shiLoc texLoc _ _ _) (StaticModelRenderer model) = + let (Material _ ka kd ks shi) = material model + in do + bindVAO . vao $ model + bindTexture $ texture model + activeTexture $= GL_TEXTURE0 + glUniform1i texLoc 0 + +-- | Render the given renderer. +render :: StaticProgramUniforms -> StaticModelRenderer -> IO () +render uniforms (StaticModelRenderer model) = + let (Material _ ka kd ks shi) = material model + in do + uniform (kaLoc uniforms) ka + uniform (kdLoc uniforms) kd + uniform (ksLoc uniforms) ks + glUniform1f (shiLoc uniforms) $ unsafeCoerce shi + drawArrays GL_TRIANGLES 0 $ nVertices model + +-- | Compute AABB collisioners in view space from the given model. +mkColsFromStatic :: + -- | Modelview matrix + Matrix4 -> + StaticModelResource -> + [Collisioner2] +mkColsFromStatic modelview modelRes = mkCols modelview (box 0 modelRes) diff --git a/Spear/old/Spear/Scene/Graph.hs b/Spear/old/Spear/Scene/Graph.hs new file mode 100644 index 0000000..8f8b5f9 --- /dev/null +++ b/Spear/old/Spear/Scene/Graph.hs @@ -0,0 +1,143 @@ +module Spear.Scene.Graph +( + Property +, SceneGraph(..) +, ParseError +, loadSceneGraph +, loadSceneGraphFromFile +, node +) +where + + +import qualified Data.ByteString.Char8 as B +import Data.List (find, intersperse) +import Data.Maybe (isJust) +import Text.Parsec.Char +import Text.Parsec.Combinator +import Text.Parsec.Error +import Text.Parsec.Prim +import qualified Text.Parsec.ByteString as P +import qualified Text.Parsec.Token as PT + + +type Property = (String, [String]) + + +data SceneGraph + = SceneNode + { nodeID :: String + , properties :: [Property] + , children :: [SceneGraph] + } + | SceneLeaf + { nodeID :: String + , properties :: [Property] + } + + +instance Show SceneGraph where + show sceneGraph = show' "" sceneGraph + where + show' tab (SceneNode nid props children) = + tab ++ nid ++ "\n" ++ tab ++ "{\n" ++ (printProps tab props) ++ + (concat . fmap (show' $ " " ++ tab) $ children) ++ '\n':tab ++ "}\n" + + show' tab (SceneLeaf nid props) = + tab ++ nid ++ '\n':tab ++ "{\n" ++ tab ++ (printProps tab props) ++ '\n':tab ++ "}\n" + + +printProp :: Property -> String +printProp (key, vals) = key ++ " = " ++ (concat $ intersperse ", " vals) + + +printProps :: String -> [Property] -> String +printProps tab props = + let + tab' = '\n':(tab ++ tab) + longestKeyLen = maximum . fmap (length . fst) $ props + + align :: Int -> String -> String + align len str = + let (key, vals) = break ((==) '=') str + thisLen = length key + padLen = len - thisLen + 1 + pad = replicate padLen ' ' + in + key ++ pad ++ vals + in + case concat . intersperse tab' . fmap (align longestKeyLen . printProp) $ props of + [] -> [] + xs -> tab ++ xs + + +-- | Load the scene graph from the given string. +loadSceneGraph :: String -> Either ParseError SceneGraph +loadSceneGraph str = parse sceneGraph "(unknown)" $ B.pack str + + +-- | Load the scene graph specified by the given file. +loadSceneGraphFromFile :: FilePath -> IO (Either ParseError SceneGraph) +loadSceneGraphFromFile = P.parseFromFile sceneGraph + + +-- | Get the node identified by the given string from the given scene graph. +node :: String -> SceneGraph -> Maybe SceneGraph +node str SceneLeaf {} = Nothing +node str n@(SceneNode nid _ children) + | str == nid = Just n + | otherwise = case find isJust $ fmap (node str) children of + Nothing -> Nothing + Just x -> x + + +sceneGraph :: P.Parser SceneGraph +sceneGraph = do + g <- graph + whitespace + eof + return g + + +graph :: P.Parser SceneGraph +graph = do + nid <- name + whitespace + char '{' + props <- many . try $ whitespace >> property + children <- many . try $ whitespace >> graph + whitespace + char '}' + + return $ case null children of + True -> SceneLeaf nid props + False -> SceneNode nid props children + + +property :: P.Parser Property +property = do + key <- name + spaces >> char '=' >> spaces + vals <- cells name + return (key, vals) + + +cells :: P.Parser String -> P.Parser [String] +cells p = do + val <- p + vals <- remainingCells p + return $ val:vals + + +remainingCells :: P.Parser String -> P.Parser [String] +remainingCells p = + try (whitespace >> char ',' >> whitespace >> cells p) + <|> (return []) + + +name :: P.Parser String +name = many1 $ choice [oneOf "-/.()?_", alphaNum] + + +whitespace :: P.Parser () +whitespace = skipMany $ choice [space, newline] diff --git a/Spear/old/Spear/Scene/Loader.hs b/Spear/old/Spear/Scene/Loader.hs new file mode 100644 index 0000000..0593c77 --- /dev/null +++ b/Spear/old/Spear/Scene/Loader.hs @@ -0,0 +1,364 @@ +{-# LANGUAGE FlexibleContexts #-} + +module Spear.Scene.Loader + ( SceneResources (..), + loadScene, + validate, + resourceMap, + value, + unspecified, + mandatory, + asString, + asFloat, + asVec3, + asVec4, + ) +where + +import Control.Monad.State.Strict +import Control.Monad.Trans (lift) +import Data.List as L (find) +import Data.Map as M +import qualified Data.StateVar as SV (get) +import Prelude hiding ((*)) +import Spear.Assets.Model as Model +import Spear.Game +import qualified Spear.GL as GL +import Spear.Math.Algebra +import Spear.Math.Collision +import Spear.Math.Matrix3 as M3 +import Spear.Math.Matrix4 as M4 +import Spear.Math.MatrixUtils (fastNormalMatrix) +import Spear.Math.Vector +import Spear.Render.AnimatedModel as AM +import Spear.Render.Material +import Spear.Render.Program +import Spear.Render.StaticModel as SM +import Spear.Scene.Graph +import Spear.Scene.SceneResources +import Text.Printf (printf) + +type Loader = Game SceneResources + +-- | Load the scene specified by the given file. +loadScene :: FilePath -> Game s (SceneResources, SceneGraph) +loadScene file = do + result <- liftIO $ loadSceneGraphFromFile file + case result of + Left err -> gameError $ show err + Right g -> case validate g of + Nothing -> do + sceneRes <- resourceMap g + return (sceneRes, g) + Just err -> gameError err + +-- | Validate the given SceneGraph. +validate :: SceneGraph -> Maybe String +validate _ = Nothing + +-- | Load the scene described by the given 'SceneGraph'. +resourceMap :: SceneGraph -> Game s SceneResources +resourceMap g = exec runChildGame emptySceneResources (resourceMap' g) + +resourceMap' :: SceneGraph -> Loader () +resourceMap' node@(SceneLeaf nid props) = do + case nid of + "shader-program" -> newShaderProgram node + "model" -> newModel node + x -> return () +resourceMap' node@(SceneNode nid props children) = do + mapM_ resourceMap' children + +-- Lookup the given resource in the data pool. Load it if it is not present, otherwise return it. +loadResource :: + -- | Resource name. + String -> + -- | Map getter. + (SceneResources -> Map String a) -> + -- | Function to modify resources. + (String -> a -> Loader ()) -> + -- | Resource loader. + Loader a -> + Loader a +loadResource key field modifyResources load = do + sceneData <- get + case M.lookup key $ field sceneData of + Just val -> return val + Nothing -> do + liftIO $ printf "Loading %s..." key + resource <- load + liftIO $ printf "done\n" + modifyResources key resource + return resource + +addShader name shader = modify $ \sceneData -> + sceneData {shaders = M.insert name shader $ shaders sceneData} + +addCustomProgram name prog = modify $ \sceneData -> + sceneData {customPrograms = M.insert name prog $ customPrograms sceneData} + +addStaticProgram name prog = modify $ \sceneData -> + sceneData {staticPrograms = M.insert name prog $ staticPrograms sceneData} + +addAnimatedProgram name prog = modify $ \sceneData -> + sceneData {animatedPrograms = M.insert name prog $ animatedPrograms sceneData} + +addTexture name tex = modify $ \sceneData -> + sceneData {textures = M.insert name tex $ textures sceneData} + +addStaticModel name model = modify $ + \sceneData -> sceneData {staticModels = M.insert name model $ staticModels sceneData} + +addAnimatedModel name model = modify $ + \sceneData -> sceneData {animatedModels = M.insert name model $ animatedModels sceneData} + +-- Get the given resource from the data pool. +getResource :: (SceneResources -> Map String a) -> String -> Loader a +getResource field key = do + sceneData <- get + case M.lookup key $ field sceneData of + Just val -> return val + Nothing -> gameError $ "Oops, the given resource has not been loaded: " ++ key + +---------------------- +-- Resource Loading -- +---------------------- + +newModel :: SceneGraph -> Loader () +newModel (SceneLeaf _ props) = do + name <- asString $ mandatory' "name" props + file <- asString $ mandatory' "file" props + tex <- asString $ mandatory' "texture" props + prog <- asString $ mandatory' "shader-program" props + ke <- asVec4 $ mandatory' "ke" props + ka <- asVec4 $ mandatory' "ka" props + kd <- asVec4 $ mandatory' "kd" props + ks <- asVec4 $ mandatory' "ks" props + shi <- asFloat $ mandatory' "shi" props + + let rotation = asRotation $ value "rotation" props + scale = asVec3 $ value "scale" props + + liftIO $ printf "Loading model %s..." name + model <- loadModel' file rotation scale + liftIO . putStrLn $ "done" + texture <- loadTexture tex + sceneRes <- get + + let material = Material ke ka kd ks shi + + case animated model of + False -> + case M.lookup prog $ staticPrograms sceneRes of + Nothing -> (gameError $ "Static shader program " ++ prog ++ " does not exist") >> return () + Just p -> + let StaticProgram _ channels _ = p + in do + model' <- staticModelResource channels material texture model + loadResource name staticModels addStaticModel (return model') + return () + True -> + case M.lookup prog $ animatedPrograms sceneRes of + Nothing -> (gameError $ "Animated shader program " ++ prog ++ " does not exist") >> return () + Just p -> + let AnimatedProgram _ channels _ = p + in do + model' <- animatedModelResource channels material texture model + loadResource name animatedModels addAnimatedModel (return model') + return () + +loadModel' :: FilePath -> Maybe Rotation -> Maybe Vector3 -> Game s Model +loadModel' file rotation scale = do + let transform = + ( case rotation of + Nothing -> Prelude.id + Just rot -> rotateModel rot + ) + . ( case scale of + Nothing -> Prelude.id + Just s -> flip Model.transformVerts $ + \(Vec3 x' y' z') -> Vec3 (x s * x') (y s * y') (z s * z') + ) + + (fmap transform $ Model.loadModel file) >>= liftIO . toGround + +rotateModel :: Rotation -> Model -> Model +rotateModel (Rotation ax ay az order) model = + let mat = case order of + XYZ -> rotateZ az * rotateY ay * rotateX ax + XZY -> rotateY ay * rotateZ az * rotateX ax + YXZ -> rotateZ az * rotateX ax * rotateY ay + YZX -> rotateX ax * rotateZ az * rotateY ay + ZXY -> rotateY ay * rotateX ax * rotateZ az + ZYX -> rotateX ax * rotateY ay * rotateZ az + normalMat = fastNormalMatrix mat + + vTransform (Vec3 x' y' z') = + let v = mat `M4.mulp` (vec3 x' y' z') in Vec3 (x v) (y v) (z v) + + nTransform (Vec3 x' y' z') = + let v = normalMat `M3.mul` (vec3 x' y' z') in Vec3 (x v) (y v) (z v) + in flip Model.transformVerts vTransform . flip Model.transformNormals nTransform $ model + +loadTexture :: FilePath -> Loader GL.Texture +loadTexture file = + loadResource file textures addTexture $ + GL.loadTextureImage file GL.GL_LINEAR GL.GL_LINEAR + +newShaderProgram :: SceneGraph -> Loader () +newShaderProgram (SceneLeaf _ props) = do + (vsName, vertShader) <- Spear.Scene.Loader.loadShader GL.VertexShader props + (fsName, fragShader) <- Spear.Scene.Loader.loadShader GL.FragmentShader props + name <- asString $ mandatory' "name" props + stype <- asString $ mandatory' "type" props + prog <- GL.newProgram [vertShader, fragShader] + + let getUniformLoc name = (liftIO . SV.get $ GL.uniformLocation prog name) `GL.assertGL` name + + case stype of + "static" -> do + ambient <- asString $ mandatory' "ambient" props + diffuse <- asString $ mandatory' "diffuse" props + specular <- asString $ mandatory' "specular" props + shininess <- asString $ mandatory' "shininess" props + texture <- asString $ mandatory' "texture" props + modelview <- asString $ mandatory' "modelview" props + normalmat <- asString $ mandatory' "normalmat" props + projection <- asString $ mandatory' "projection" props + + ka <- getUniformLoc ambient + kd <- getUniformLoc diffuse + ks <- getUniformLoc specular + shi <- getUniformLoc shininess + tex <- getUniformLoc texture + mview <- getUniformLoc modelview + nmat <- getUniformLoc normalmat + proj <- getUniformLoc projection + + vertChan <- fmap read $ asString $ mandatory' "vertex-channel" props + normChan <- fmap read $ asString $ mandatory' "normal-channel" props + texChan <- fmap read $ asString $ mandatory' "texture-channel" props + + let channels = StaticProgramChannels vertChan normChan texChan + uniforms = StaticProgramUniforms ka kd ks shi tex mview nmat proj + + loadResource name staticPrograms addStaticProgram $ + return $ StaticProgram prog channels uniforms + return () + "animated" -> do + ambient <- asString $ mandatory' "ambient" props + diffuse <- asString $ mandatory' "diffuse" props + specular <- asString $ mandatory' "specular" props + shininess <- asString $ mandatory' "shininess" props + texture <- asString $ mandatory' "texture" props + modelview <- asString $ mandatory' "modelview" props + normalmat <- asString $ mandatory' "normalmat" props + projection <- asString $ mandatory' "projection" props + + ka <- getUniformLoc ambient + kd <- getUniformLoc diffuse + ks <- getUniformLoc specular + shi <- getUniformLoc shininess + tex <- getUniformLoc texture + mview <- getUniformLoc modelview + nmat <- getUniformLoc normalmat + proj <- getUniformLoc projection + + vertChan1 <- fmap read $ asString $ mandatory' "vertex-channel1" props + vertChan2 <- fmap read $ asString $ mandatory' "vertex-channel2" props + normChan1 <- fmap read $ asString $ mandatory' "normal-channel1" props + normChan2 <- fmap read $ asString $ mandatory' "normal-channel2" props + texChan <- fmap read $ asString $ mandatory' "texture-channel" props + fp <- asString $ mandatory' "fp" props + p <- getUniformLoc fp + + let channels = AnimatedProgramChannels vertChan1 vertChan2 normChan1 normChan2 texChan + uniforms = AnimatedProgramUniforms ka kd ks shi tex p mview nmat proj + + loadResource name animatedPrograms addAnimatedProgram $ + return $ AnimatedProgram prog channels uniforms + return () + _ -> do + loadResource name customPrograms addCustomProgram $ return prog + return () + +loadShader :: GL.ShaderType -> [Property] -> Loader (String, GL.GLSLShader) +loadShader _ [] = gameError $ "Loader::vertexShader: empty list" +loadShader shaderType ((stype, file) : xs) = + if shaderType == GL.VertexShader && stype == "vertex-shader" + || shaderType == GL.FragmentShader && stype == "fragment-shader" + then + let f = concat file + in loadShader' f shaderType >>= \shader -> return (f, shader) + else Spear.Scene.Loader.loadShader shaderType xs + +loadShader' :: String -> GL.ShaderType -> Loader GL.GLSLShader +loadShader' file shaderType = loadResource file shaders addShader $ GL.loadShader shaderType file + +---------------------- +-- Helper functions -- +---------------------- + +-- Get the value of the given key. +value :: String -> [Property] -> Maybe [String] +value name props = case L.find ((==) name . fst) props of + Nothing -> Nothing + Just prop -> Just . snd $ prop + +unspecified :: Maybe a -> a -> a +unspecified (Just x) _ = x +unspecified Nothing x = x + +mandatory :: String -> [Property] -> Game s [String] +mandatory name props = case value name props of + Nothing -> gameError $ "Loader::mandatory: key not found: " ++ name + Just x -> return x + +mandatory' :: String -> [Property] -> Loader [String] +mandatory' name props = mandatory name props + +asString :: Functor f => f [String] -> f String +asString = fmap concat + +asFloat :: Functor f => f [String] -> f Float +asFloat = fmap (read . concat) + +asVec2 :: Functor f => f [String] -> f Vector2 +asVec2 val = fmap toVec2 val + where + toVec2 (x : y : _) = vec2 (read x) (read y) + toVec2 (x : []) = let x' = read x in vec2 x' x' + +asVec3 :: Functor f => f [String] -> f Vector3 +asVec3 val = fmap toVec3 val + where + toVec3 (x : y : z : _) = vec3 (read x) (read y) (read z) + toVec3 (x : []) = let x' = read x in vec3 x' x' x' + +asVec4 :: Functor f => f [String] -> f Vector4 +asVec4 val = fmap toVec4 val + where + toVec4 (x : y : z : w : _) = vec4 (read x) (read y) (read z) (read w) + toVec4 (x : []) = let x' = read x in vec4 x' x' x' x' + +asRotation :: Functor f => f [String] -> f Rotation +asRotation val = fmap parseRotation val + where + parseRotation (ax : ay : az : order : _) = Rotation (read ax) (read ay) (read az) (readOrder order) + +data Rotation = Rotation + { ax :: Float, + ay :: Float, + az :: Float, + order :: RotationOrder + } + +data RotationOrder = XYZ | XZY | YXZ | YZX | ZXY | ZYX deriving (Eq) + +readOrder :: String -> RotationOrder +readOrder "xyz" = XYZ +readOrder "xzy" = XZY +readOrder "yxz" = YXZ +readOrder "yzx" = YZX +readOrder "zxy" = ZXY +readOrder "zyx" = ZYX diff --git a/Spear/old/Spear/Scene/SceneResources.hs b/Spear/old/Spear/Scene/SceneResources.hs new file mode 100644 index 0000000..de2fc80 --- /dev/null +++ b/Spear/old/Spear/Scene/SceneResources.hs @@ -0,0 +1,70 @@ +module Spear.Scene.SceneResources +( + -- * Data types + SceneResources(..) +, StaticProgram(..) +, AnimatedProgram(..) + -- * Construction +, emptySceneResources + -- * Accessors +, getShader +, getCustomProgram +, getStaticProgram +, getAnimatedProgram +, getTexture +, getStaticModel +, getAnimatedModel +) +where + +import Spear.Assets.Model as Model +import Spear.GL as GL +import Spear.Math.Vector +import Spear.Render.AnimatedModel +import Spear.Render.Material +import Spear.Render.Program +import Spear.Render.StaticModel + +import Data.Map as M + +data SceneResources = SceneResources + { shaders :: Map String GLSLShader + , customPrograms :: Map String GLSLProgram + , staticPrograms :: Map String StaticProgram + , animatedPrograms :: Map String AnimatedProgram + , textures :: Map String Texture + , staticModels :: Map String StaticModelResource + , animatedModels :: Map String AnimatedModelResource + } + +-- | Build an empty instance of 'SceneResources'. +emptySceneResources = + SceneResources M.empty M.empty M.empty M.empty M.empty M.empty M.empty + +-- | Get the shader specified by the given string. +getShader :: SceneResources -> String -> Maybe GLSLShader +getShader res key = M.lookup key $ shaders res + +-- | Get the custom program specified by the given string. +getCustomProgram :: SceneResources -> String -> Maybe GLSLProgram +getCustomProgram res key = M.lookup key $ customPrograms res + +-- | Get the static program specified by the given string. +getStaticProgram :: SceneResources -> String -> Maybe StaticProgram +getStaticProgram res key = M.lookup key $ staticPrograms res + +-- | Get the animated program specified by the given string. +getAnimatedProgram :: SceneResources -> String -> Maybe AnimatedProgram +getAnimatedProgram res key = M.lookup key $ animatedPrograms res + +-- | Get the texture specified by the given string. +getTexture :: SceneResources -> String -> Maybe Texture +getTexture res key = M.lookup key $ textures res + +-- | Get the static model resource specified by the given string. +getStaticModel :: SceneResources -> String -> Maybe StaticModelResource +getStaticModel res key = M.lookup key $ staticModels res + +-- | Get the animated model resource specified by the given string. +getAnimatedModel :: SceneResources -> String -> Maybe AnimatedModelResource +getAnimatedModel res key = M.lookup key $ animatedModels res -- cgit v1.2.3