diff --git a/cmd/wuffs-c/internal/cgen/base-header.h b/cmd/wuffs-c/internal/cgen/base-header.h index 852af1b2..0bb3e403 100644 --- a/cmd/wuffs-c/internal/cgen/base-header.h +++ b/cmd/wuffs-c/internal/cgen/base-header.h @@ -43,11 +43,40 @@ // For now, the API and ABI are simply unstable and can change at any time. // // TODO: don't hard code this in base-header.h. -#define WUFFS_VERSION (0x00001) +#define WUFFS_VERSION 0x00001 -// ---------------- I/O +// ---------------- Fundamentals -// wuffs_base__slice_u8 is a 1-dimensional buffer (a pointer and length). +// Flicks are a unit of time. One flick (frame-tick) is 1 / 705_600_000 of a +// second. See https://github.com/OculusVR/Flicks +typedef uint64_t wuffs_base__flicks; + +#define WUFFS_BASE__FLICKS_PER_SECOND 705600000ULL + +// wuffs_base__rectangle is a rectangle on the integer grid. It contains all +// points (x, y) such that ((min_x <= x) && (x < max_x)) and likewise for y. It +// is therefore empty if min_x >= max_x. There are multiple representations of +// an empty rectangle. +// +// A value with all fields zero is a valid, empty rectangle. +// +// The X and Y axes increase right and down. +typedef struct { + uint32_t min_x; + uint32_t min_y; + uint32_t max_x; + uint32_t max_y; +} wuffs_base__rectangle; + +static inline uint32_t wuffs_base__rectangle__width(wuffs_base__rectangle r) { + return r.max_x > r.min_x ? r.max_x - r.min_x : 0; +} + +static inline uint32_t wuffs_base__rectangle__height(wuffs_base__rectangle r) { + return r.max_y > r.min_y ? r.max_y - r.min_y : 0; +} + +// wuffs_base__slice_u8 is a 1-dimensional buffer. // // A value with all fields NULL or zero is a valid, empty slice. typedef struct { @@ -55,6 +84,20 @@ typedef struct { size_t len; } wuffs_base__slice_u8; +// wuffs_base__table_u8 is a 2-dimensional buffer. +// +// A value with all fields NULL or zero is a valid, empty table. +typedef struct { + uint8_t* ptr; + size_t width; + size_t height; + size_t stride; +} wuffs_base__table_u8; + +// ---------------- I/O + +// TODO: rename buf1 to io_buffer, writer1 to io_writer, etc. + // wuffs_base__buf1 is a 1-dimensional buffer (a pointer and length), plus // additional indexes into that buffer, plus an opened / closed flag. // @@ -106,18 +149,286 @@ typedef struct { // ---------------- Images +// wuffs_base__pixel_format encodes the format of the bytes that constitute an +// image frame's pixel data. Its bits: +// - bit 31 is reserved. +// - bits 30 .. 28 encodes color (and channel order, in terms of memory). +// - bits 27 .. 26 are reserved. +// - bits 25 .. 24 encodes transparency. +// - bit 23 indicates big-endian/MSB-first (as opposed to little/LSB). +// - bit 22 indicates floating point (as opposed to integer). +// - bits 21 .. 20 are the number of planes, minus 1. Zero means packed. +// - bits 19 .. 16 encodes the number of bits (depth) in an index value. +// Zero means direct, not palette-indexed. +// - bits 15 .. 12 encodes the number of bits (depth) in the 3rd channel. +// - bits 11 .. 8 encodes the number of bits (depth) in the 2nd channel. +// - bits 7 .. 4 encodes the number of bits (depth) in the 1st channel. +// - bits 3 .. 0 encodes the number of bits (depth) in the 0th channel. +// +// The bit fields of a wuffs_base__pixel_format are not independent. For +// example, the number of planes should not be greater than the number of +// channels. Similarly, bits 15..4 are unused (and should be zero) if bits +// 31..24 (color and transparency) together imply only 1 channel (gray, no +// alpha) and floating point samples should mean a bit depth of 16, 32 or 64. +// +// Formats hold between 1 and 4 channels. For example: Y (1 channel: gray), YA +// (2 channels: gray and alpha), BGR (3 channels: blue, green, red) or CMYK (4 +// channels: cyan, magenta, yellow, black). +// +// For direct formats with N > 1 channels, those channels can be laid out in +// either 1 (packed) or N (planar) planes. For example, RGBA data is usually +// packed, but YUV data is usually planar, due to chroma subsampling (for +// details, see the wuffs_base__pixel_subsampling type). For indexed formats, +// the palette (always 256 × 4 bytes) holds up to 4 packed bytes of color data +// per index value, and there is only 1 plane (for the index). The distance +// between successive palette elements is always 4 bytes. +// +// The color field is encoded in 3 bits: +// - 0 means A (Alpha). +// - 1 means Y or YA (Gray, Alpha). +// - 2 means BGR, BGRX or BGRA (Blue, Green, Red, X-padding or Alpha). +// - 3 means RGB, RGBX or RGBA (Red, Green, Blue, X-padding or Alpha). +// - 4 means YUV or YUVA (Luma, Chroma-blue, Chroma-red, Alpha). +// - 5 means CMY or CMYK (Cyan, Magenta, Yellow, Black). +// - all other values are reserved. +// +// In Wuffs, channels are given in memory order, regardless of endianness, +// since the C type for the pixel data is an array of bytes, not an array of +// uint32_t. For example, packed BGRA with 8 bits per channel means that the +// bytes in memory are always Blue, Green, Red then Alpha. On big-endian +// systems, that is the uint32_t 0xBBGGRRAA. On little-endian, 0xAARRGGBB. +// +// When the color field (3 bits) encodes multiple options, the transparency +// field (2 bits) distinguishes them: +// - 0 means fully opaque, no extra channels +// - 1 means fully opaque, one extra channel (X or K, padding or black). +// - 2 means one extra alpha channel, other channels are non-premultiplied. +// - 3 means one extra alpha channel, other channels are premultiplied. +// +// The zero wuffs_base__pixel_format value is an invalid pixel format, as it is +// invalid to combine the zero color (alpha only) with the zero transparency. +// +// Bit depth is encoded in 4 bits: +// - 0 means the channel or index is unused. +// - x means a bit depth of x, for x in the range 1..8. +// - 9 means a bit depth of 10. +// - 10 means a bit depth of 12. +// - 11 means a bit depth of 16. +// - 12 means a bit depth of 24. +// - 13 means a bit depth of 32. +// - 14 means a bit depth of 48. +// - 15 means a bit depth of 64. +// +// For example, wuffs_base__pixel_format 0x3280BBBB is a natural format for +// decoding a PNG image - network byte order (also known as big-endian), +// packed, non-premultiplied alpha - that happens to be 16-bit-depth truecolor +// with alpha (RGBA). In memory order: +// +// ptr+0 ptr+1 ptr+2 ptr+3 ptr+4 ptr+5 ptr+6 ptr+7 +// Rhi Rlo Ghi Glo Bhi Blo Ahi Alo +// +// For example, the value wuffs_base__pixel_format 0x20000565 means BGR with no +// alpha or padding, 5/6/5 bits for blue/green/red, packed 2 bytes per pixel, +// laid out LSB-first in memory order: +// +// ptr+0........... ptr+1........... +// MSB LSB MSB LSB +// G₂G₁G₀B₄B₃B₂B₁B₀ R₄R₃R₂R₁R₀G₅G₄G₃ +// +// On little-endian systems (but not big-endian), this Wuffs pixel format value +// (0x20000565) corresponds to the Cairo library's CAIRO_FORMAT_RGB16_565, the +// SDL2 (Simple DirectMedia Layer 2) library's SDL_PIXELFORMAT_RGB565 and the +// Skia library's kRGB_565_SkColorType. Note BGR in Wuffs versus RGB in the +// other libraries. +// +// Regardless of endianness, this Wuffs pixel format value (0x20000565) +// corresponds to the V4L2 (Video For Linux 2) library's V4L2_PIX_FMT_RGB565 +// and the Wayland-DRM library's WL_DRM_FORMAT_RGB565. +// +// Different software libraries name their pixel formats (and especially their +// channel order) either according to memory layout or as bits of a native +// integer type like uint32_t. The two conventions differ because of a system's +// endianness. As mentioned earlier, Wuffs pixel formats are always in memory +// order. More detail of other software libraries' naming conventions is in the +// Pixel Format Guide at https://afrantzis.github.io/pixel-format-guide/ +// +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_base__pixel_format__num_planes instead. +typedef uint32_t wuffs_base__pixel_format; + +// Common 8-bit-depth pixel formats. This list is not exhaustive; not all valid +// wuffs_base__pixel_format values are present. + +#define WUFFS_BASE__PIXEL_FORMAT__INVALID ((wuffs_base__pixel_format)0x00000000) + +#define WUFFS_BASE__PIXEL_FORMAT__A ((wuffs_base__pixel_format)0x02000008) + +#define WUFFS_BASE__PIXEL_FORMAT__Y ((wuffs_base__pixel_format)0x10000008) +#define WUFFS_BASE__PIXEL_FORMAT__YA_NONPREMUL \ + ((wuffs_base__pixel_format)0x12000008) +#define WUFFS_BASE__PIXEL_FORMAT__YA_PREMUL \ + ((wuffs_base__pixel_format)0x13000008) + +#define WUFFS_BASE__PIXEL_FORMAT__BGR ((wuffs_base__pixel_format)0x20000888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRX ((wuffs_base__pixel_format)0x21008888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRX_INDEXED \ + ((wuffs_base__pixel_format)0x21088888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_NONPREMUL \ + ((wuffs_base__pixel_format)0x22008888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_NONPREMUL_INDEXED \ + ((wuffs_base__pixel_format)0x22088888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_PREMUL \ + ((wuffs_base__pixel_format)0x23008888) + +#define WUFFS_BASE__PIXEL_FORMAT__RGB ((wuffs_base__pixel_format)0x30000888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBX ((wuffs_base__pixel_format)0x31008888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBX_INDEXED \ + ((wuffs_base__pixel_format)0x31088888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_NONPREMUL \ + ((wuffs_base__pixel_format)0x32008888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_NONPREMUL_INDEXED \ + ((wuffs_base__pixel_format)0x32088888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_PREMUL \ + ((wuffs_base__pixel_format)0x33008888) + +#define WUFFS_BASE__PIXEL_FORMAT__YUV ((wuffs_base__pixel_format)0x40200888) +#define WUFFS_BASE__PIXEL_FORMAT__YUVK ((wuffs_base__pixel_format)0x41308888) +#define WUFFS_BASE__PIXEL_FORMAT__YUVA_NONPREMUL \ + ((wuffs_base__pixel_format)0x42308888) + +#define WUFFS_BASE__PIXEL_FORMAT__CMY ((wuffs_base__pixel_format)0x50200888) +#define WUFFS_BASE__PIXEL_FORMAT__CMYK ((wuffs_base__pixel_format)0x51308888) + +static inline bool wuffs_base__pixel_format__is_valid( + wuffs_base__pixel_format f) { + return f != 0; +} + +static inline bool wuffs_base__pixel_format__is_indexed( + wuffs_base__pixel_format f) { + return ((f >> 16) & 0x0F) != 0; +} + +#define WUFFS_BASE__PIXEL_FORMAT__NUM_PLANES_MAX 4 + +static inline uint32_t wuffs_base__pixel_format__num_planes( + wuffs_base__pixel_format f) { + return f ? (((f >> 20) & 0x03) + 1) : 0; +} + +typedef struct { + wuffs_base__table_u8 planes[WUFFS_BASE__PIXEL_FORMAT__NUM_PLANES_MAX]; +} wuffs_base__pixel_buffer; + +// wuffs_base__pixel_subsampling encodes the mapping of pixel space coordinates +// (x, y) to pixel buffer indices (i, j). That mapping can differ for each +// plane p. For a depth of 8 bits (1 byte), the p'th plane's sample starts at +// (planes[p].ptr + (j * planes[p].stride) + i). +// +// For packed pixel formats, the mapping is trivial: i = x and j = y. For +// planar pixel formats, the mapping can differ due to chroma subsampling. For +// example, consider a three plane YUV pixel format with 4:2:2 subsampling. For +// the luma (Y) channel, there is one sample for every pixel, but for the +// chroma (U, V) channels, there is one sample for every two pixels: pairs of +// horizontally adjacent pixels form one macropixel, i = x / 2 and j == y. In +// general, for a given p: +// - i = (x + bias_x) >> shift_x. +// - j = (y + bias_y) >> shift_y. +// where biases and shifts are in the range 0..3 and 0..2 respectively. +// +// In general, the biases will be zero after decoding an image. However, making +// a sub-image may change the bias, since the (x, y) coordinates are relative +// to the sub-image's top-left origin, but the backing pixel buffers were +// created relative to the original image's origin. +// +// For each plane p, each of those four numbers (biases and shifts) are encoded +// in two bits, which combine to form an 8 bit unsigned integer: +// +// e_p = (bias_x << 6) | (shift_x << 4) | (bias_y << 2) | (shift_y << 0) +// +// Those e_p values (e_0 for the first plane, e_1 for the second plane, etc) +// combine to form a wuffs_base__pixel_subsampling value: +// +// pixsub = (e_3 << 24) | (e_2 << 16) | (e_1 << 8) | (e_0 << 0) +// +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_base__pixel_subsampling__bias_x instead. +typedef uint32_t wuffs_base__pixel_subsampling; + +#define WUFFS_BASE__PIXEL_SUBSAMPLING__NONE ((wuffs_base__pixel_subsampling)0); + +#define WUFFS_BASE__PIXEL_SUBSAMPLING__444 \ + ((wuffs_base__pixel_subsampling)0x000000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__440 \ + ((wuffs_base__pixel_subsampling)0x010100); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__422 \ + ((wuffs_base__pixel_subsampling)0x101000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__420 \ + ((wuffs_base__pixel_subsampling)0x111100); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__411 \ + ((wuffs_base__pixel_subsampling)0x202000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__410 \ + ((wuffs_base__pixel_subsampling)0x212100); + +static inline uint32_t wuffs_base__pixel_subsampling__bias_x( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 6; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__shift_x( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 4; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__bias_y( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 2; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__shift_y( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 0; + return (s >> shift) & 0x03; +} + typedef struct { // Do not access the private_impl's fields directly. There is no API/ABI // compatibility or safety guarantee if you do so. struct { - uint32_t flags; - uint32_t w; - uint32_t h; - // TODO: color model, including both packed RGBA and planar, - // chroma-subsampled YCbCr. + wuffs_base__pixel_format pixfmt; + wuffs_base__pixel_subsampling pixsub; + uint32_t width; + uint32_t height; + uint32_t num_loops; } private_impl; } wuffs_base__image_config; +static inline void wuffs_base__image_config__initialize( + wuffs_base__image_config* c, + wuffs_base__pixel_format pixfmt, + wuffs_base__pixel_subsampling pixsub, + uint32_t width, + uint32_t height, + uint32_t num_loops) { + if (!c) { + return; + } + // TODO: move the check from wuffs_base__image_config__is_valid here. Should + // this function return bool? An error type? + c->private_impl.pixfmt = pixfmt; + c->private_impl.pixsub = pixsub; + c->private_impl.width = width; + c->private_impl.height = height; + c->private_impl.num_loops = num_loops; +} + static inline void wuffs_base__image_config__invalidate( wuffs_base__image_config* c) { if (c) { @@ -125,24 +436,42 @@ static inline void wuffs_base__image_config__invalidate( } } -static inline bool wuffs_base__image_config__valid( +static inline bool wuffs_base__image_config__is_valid( wuffs_base__image_config* c) { - if (!c || !(c->private_impl.flags & 1)) { + if (!c || !c->private_impl.pixfmt) { return false; } - uint64_t wh = ((uint64_t)c->private_impl.w) * ((uint64_t)c->private_impl.h); + uint64_t wh = + ((uint64_t)c->private_impl.width) * ((uint64_t)c->private_impl.height); // TODO: handle things other than 1 byte per pixel. + // + // TODO: move the check to wuffs_base__image_config__initialize. return wh <= ((uint64_t)SIZE_MAX); } +static inline wuffs_base__pixel_format wuffs_base__image_config__pixel_format( + wuffs_base__image_config* c) { + return c ? c->private_impl.pixfmt : 0; +} + +static inline wuffs_base__pixel_subsampling +wuffs_base__image_config__pixel_subsampling(wuffs_base__image_config* c) { + return wuffs_base__image_config__is_valid(c) ? c->private_impl.pixsub : 0; +} + static inline uint32_t wuffs_base__image_config__width( wuffs_base__image_config* c) { - return wuffs_base__image_config__valid(c) ? c->private_impl.w : 0; + return wuffs_base__image_config__is_valid(c) ? c->private_impl.width : 0; } static inline uint32_t wuffs_base__image_config__height( wuffs_base__image_config* c) { - return wuffs_base__image_config__valid(c) ? c->private_impl.h : 0; + return wuffs_base__image_config__is_valid(c) ? c->private_impl.height : 0; +} + +static inline uint32_t wuffs_base__image_config__num_loops( + wuffs_base__image_config* c) { + return wuffs_base__image_config__is_valid(c) ? c->private_impl.num_loops : 0; } // TODO: this is the right API for planar (not packed) pixbufs? Should it allow @@ -150,26 +479,96 @@ static inline uint32_t wuffs_base__image_config__height( // example, decoding a JPEG image straight to RGBA instead of to YCbCr? static inline size_t wuffs_base__image_config__pixbuf_size( wuffs_base__image_config* c) { - if (wuffs_base__image_config__valid(c)) { - uint64_t wh = ((uint64_t)c->private_impl.w) * ((uint64_t)c->private_impl.h); + if (wuffs_base__image_config__is_valid(c)) { + uint64_t wh = + ((uint64_t)c->private_impl.width) * ((uint64_t)c->private_impl.height); // TODO: handle things other than 1 byte per pixel. return (size_t)wh; } return 0; } -static inline void wuffs_base__image_config__initialize( - wuffs_base__image_config* c, - uint32_t width, - uint32_t height, - uint32_t TODO_color_model) { - if (!c) { +typedef struct { + // Do not access the private_impl's fields directly. There is no API/ABI + // compatibility or safety guarantee if you do so. + struct { + wuffs_base__image_config config; + uint32_t loop_count; // 0-based count of the current loop. + wuffs_base__pixel_buffer pixbuf; + // TODO: color spaces. + wuffs_base__rectangle dirty_rect; + wuffs_base__flicks duration; + uint8_t palette[1024]; + } private_impl; +} wuffs_base__image_buffer; + +static inline void wuffs_base__image_buffer__initialize( + wuffs_base__image_buffer* f, + wuffs_base__image_config config, + wuffs_base__pixel_buffer pixbuf) { + if (!f) { return; } - c->private_impl.flags = 1; - c->private_impl.w = width; - c->private_impl.h = height; - // TODO: color model. + *f = ((wuffs_base__image_buffer){}); + f->private_impl.config = config; + f->private_impl.pixbuf = pixbuf; +} + +static inline void wuffs_base__image_buffer__update( + wuffs_base__image_buffer* f, + wuffs_base__rectangle dirty_rect, + wuffs_base__flicks duration, + uint8_t* palette_ptr, + size_t palette_len) { + if (!f) { + return; + } + f->private_impl.dirty_rect = dirty_rect; + f->private_impl.duration = duration; + if (palette_ptr) { + memmove(f->private_impl.palette, palette_ptr, + palette_len <= 1024 ? palette_len : 1024); + } +} + +// wuffs_base__image_buffer__loop returns whether the image decoder should loop +// back to the beginning of the animation, assuming that we've reached the end +// of the encoded stream. If so, it increments f's count of the animation loops +// played so far. +static inline bool wuffs_base__image_buffer__loop(wuffs_base__image_buffer* f) { + if (!f) { + return false; + } + uint32_t n = f->private_impl.config.private_impl.num_loops; + if (n == 0) { + return true; + } + if (f->private_impl.loop_count < n - 1) { + f->private_impl.loop_count++; + return true; + } + return false; +} + +// wuffs_base__image_buffer__dirty_rect returns an upper bound for what part of +// this frame's pixels differs from the previous frame. +static inline wuffs_base__rectangle wuffs_base__image_buffer__dirty_rect( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.dirty_rect : ((wuffs_base__rectangle){0}); +} + +// wuffs_base__image_buffer__duration returns the amount of time to display +// this frame. Zero means to display forever - a still (non-animated) image. +static inline wuffs_base__flicks wuffs_base__image_buffer__duration( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.duration : 0; +} + +// wuffs_base__image_buffer__palette returns the palette that the pixel data +// can index. The backing array is inside f and has length 1024. +static inline uint8_t* wuffs_base__image_buffer__palette( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.palette : NULL; } #endif // WUFFS_BASE_HEADER_H diff --git a/cmd/wuffs-c/internal/cgen/base-impl.h b/cmd/wuffs-c/internal/cgen/base-impl.h index fd2a7249..a87a0004 100644 --- a/cmd/wuffs-c/internal/cgen/base-impl.h +++ b/cmd/wuffs-c/internal/cgen/base-impl.h @@ -27,14 +27,14 @@ typedef struct { // but it should catch 99.99% of cases. // // Its (non-zero) value is arbitrary, based on md5sum("wuffs"). -#define WUFFS_BASE__MAGIC (0x3CCB6C71U) +#define WUFFS_BASE__MAGIC 0x3CCB6C71U // WUFFS_BASE__ALREADY_ZEROED is passed from a container struct's initializer // to a containee struct's initializer when the container has already zeroed // the containee's memory. // // Its (non-zero) value is arbitrary, based on md5sum("zeroed"). -#define WUFFS_BASE__ALREADY_ZEROED (0x68602EF1U) +#define WUFFS_BASE__ALREADY_ZEROED 0x68602EF1U // Denote intentional fallthroughs for -Wimplicit-fallthrough. // diff --git a/cmd/wuffs-c/internal/cgen/cgen.go b/cmd/wuffs-c/internal/cgen/cgen.go index 0fee3aaf..bb69bd75 100644 --- a/cmd/wuffs-c/internal/cgen/cgen.go +++ b/cmd/wuffs-c/internal/cgen/cgen.go @@ -91,11 +91,11 @@ const ( statusCodeNamespaceShift = 10 statusCodeCodeBits = 8 statusCodeDescription = "" + - "// Status codes are int32_t values:\n" + - "// - the sign bit indicates a non-recoverable status code: an error\n" + - "// - bits 10-30 hold the packageid: a namespace\n" + - "// - bits 8-9 are reserved\n" + - "// - bits 0-7 are a package-namespaced numeric code\n" + "// Status codes are int32_t values. Its bits:\n" + + "// - bit 31 (the sign bit) indicates unrecoverable-ness: an error.\n" + + "// - bits 30 .. 10 are the packageid: a namespace.\n" + + "// - bits 9 .. 8 are reserved.\n" + + "// - bits 7 .. 0 are a package-namespaced numeric code.\n" ) func init() { @@ -235,8 +235,9 @@ func (g *gen) genHeader(b *buffer) error { b.writes("// ---------------- Status Codes\n\n") b.writes(statusCodeDescription) - b.printf("//\n// Do not manipulate these bits directly. Use the API functions such as\n"+ - "// %sstatus__is_error instead.\n", g.pkgPrefix) + b.printf("//\n// Do not manipulate these bits directly; they are private implementation\n"+ + "// details. Use methods such as %sstatus__is_error instead.\n", g.pkgPrefix) + b.printf("typedef int32_t %sstatus;\n\n", g.pkgPrefix) pkgID := g.checker.PackageID() b.printf("#define %spackageid %d // 0x%08X\n\n", g.pkgPrefix, pkgID, pkgID) diff --git a/cmd/wuffs-c/internal/cgen/data.go b/cmd/wuffs-c/internal/cgen/data.go index 8c673089..794e7939 100644 --- a/cmd/wuffs-c/internal/cgen/data.go +++ b/cmd/wuffs-c/internal/cgen/data.go @@ -18,26 +18,42 @@ package cgen const baseHeader = "" + "#ifndef WUFFS_BASE_HEADER_H\n#define WUFFS_BASE_HEADER_H\n\n// Copyright 2017 The Wuffs Authors.\n//\n// Licensed under the Apache License, Version 2.0 (the \"License\");\n// you may not use this file except in compliance with the License.\n// You may obtain a copy of the License at\n//\n// https://www.apache.org/licenses/LICENSE-2.0\n//\n// Unless required by applicable law or agreed to in writing, software\n// distributed under the License is distributed on an \"AS IS\" BASIS,\n// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n// See the License for the specific language governing permissions and\n// limitations under the License.\n\n#include \n#include \n#include \n\n// Wuffs requires a word size of at least 32 bits because it assumes that\n// converting a u32 to usize will never overflow. For example, the size of a\n// decoded image is often represented, explicitly or implicitly in an image\n// file, as a u32, and it is convenient to compare that to a buffer size.\n//\n// Si" + - "milarly, the word size is at most 64 bits because it assumes that\n// converting a usize to u64 will never overflow.\n#if __WORDSIZE < 32\n#error \"Wuffs requires a word size of at least 32 bits\"\n#elif __WORDSIZE > 64\n#error \"Wuffs requires a word size of at most 64 bits\"\n#endif\n\n// WUFFS_VERSION is the major.minor version number as a uint32. The major\n// number is the high 16 bits. The minor number is the low 16 bits.\n//\n// The intention is to bump the version number at least on every API / ABI\n// backwards incompatible change.\n//\n// For now, the API and ABI are simply unstable and can change at any time.\n//\n// TODO: don't hard code this in base-header.h.\n#define WUFFS_VERSION (0x00001)\n\n// ---------------- I/O\n\n// wuffs_base__slice_u8 is a 1-dimensional buffer (a pointer and length).\n//\n// A value with all fields NULL or zero is a valid, empty slice.\ntypedef struct {\n uint8_t* ptr;\n size_t len;\n} wuffs_base__slice_u8;\n\n// wuffs_base__buf1 is a 1-dimensional buffer (a pointer and length), plus\n// additional in" + - "dexes into that buffer, plus an opened / closed flag.\n//\n// A value with all fields NULL or zero is a valid, empty buffer.\ntypedef struct {\n uint8_t* ptr; // Pointer.\n size_t len; // Length.\n size_t wi; // Write index. Invariant: wi <= len.\n size_t ri; // Read index. Invariant: ri <= wi.\n bool closed; // No further writes are expected.\n} wuffs_base__buf1;\n\n// wuffs_base__limit1 provides a limited view of a 1-dimensional byte stream:\n// its first N bytes. That N can be greater than a buffer's current read or\n// write capacity. N decreases naturally over time as bytes are read from or\n// written to the stream.\n//\n// A value with all fields NULL or zero is a valid, unlimited view.\ntypedef struct wuffs_base__limit1 {\n uint64_t* ptr_to_len; // Pointer to N.\n struct wuffs_base__limit1* next; // Linked list of limits.\n} wuffs_base__limit1;\n\ntypedef struct {\n // TODO: move buf into private_impl? As it is, it looks like users can modify\n // the buf field to point to a different buf" + - "fer, which can turn the limit and\n // mark fields into dangling pointers.\n wuffs_base__buf1* buf;\n // Do not access the private_impl's fields directly. There is no API/ABI\n // compatibility or safety guarantee if you do so.\n struct {\n wuffs_base__limit1 limit;\n uint8_t* mark;\n } private_impl;\n} wuffs_base__reader1;\n\ntypedef struct {\n // TODO: move buf into private_impl? As it is, it looks like users can modify\n // the buf field to point to a different buffer, which can turn the limit and\n // mark fields into dangling pointers.\n wuffs_base__buf1* buf;\n // Do not access the private_impl's fields directly. There is no API/ABI\n // compatibility or safety guarantee if you do so.\n struct {\n wuffs_base__limit1 limit;\n uint8_t* mark;\n } private_impl;\n} wuffs_base__writer1;\n\n// ---------------- Images\n\ntypedef struct {\n // Do not access the private_impl's fields directly. There is no API/ABI\n // compatibility or safety guarantee if you do so.\n struct {\n uint32_t flags;\n uint32_t w;\n " + - " uint32_t h;\n // TODO: color model, including both packed RGBA and planar,\n // chroma-subsampled YCbCr.\n } private_impl;\n} wuffs_base__image_config;\n\nstatic inline void wuffs_base__image_config__invalidate(\n wuffs_base__image_config* c) {\n if (c) {\n *c = ((wuffs_base__image_config){});\n }\n}\n\nstatic inline bool wuffs_base__image_config__valid(\n wuffs_base__image_config* c) {\n if (!c || !(c->private_impl.flags & 1)) {\n return false;\n }\n uint64_t wh = ((uint64_t)c->private_impl.w) * ((uint64_t)c->private_impl.h);\n // TODO: handle things other than 1 byte per pixel.\n return wh <= ((uint64_t)SIZE_MAX);\n}\n\nstatic inline uint32_t wuffs_base__image_config__width(\n wuffs_base__image_config* c) {\n return wuffs_base__image_config__valid(c) ? c->private_impl.w : 0;\n}\n\nstatic inline uint32_t wuffs_base__image_config__height(\n wuffs_base__image_config* c) {\n return wuffs_base__image_config__valid(c) ? c->private_impl.h : 0;\n}\n\n// TODO: this is the right API for planar (not packed) pixbuf" + - "s? Should it allow\n// decoding into a color model different from the format's intrinsic one? For\n// example, decoding a JPEG image straight to RGBA instead of to YCbCr?\nstatic inline size_t wuffs_base__image_config__pixbuf_size(\n wuffs_base__image_config* c) {\n if (wuffs_base__image_config__valid(c)) {\n uint64_t wh = ((uint64_t)c->private_impl.w) * ((uint64_t)c->private_impl.h);\n // TODO: handle things other than 1 byte per pixel.\n return (size_t)wh;\n }\n return 0;\n}\n\nstatic inline void wuffs_base__image_config__initialize(\n wuffs_base__image_config* c,\n uint32_t width,\n uint32_t height,\n uint32_t TODO_color_model) {\n if (!c) {\n return;\n }\n c->private_impl.flags = 1;\n c->private_impl.w = width;\n c->private_impl.h = height;\n // TODO: color model.\n}\n\n#endif // WUFFS_BASE_HEADER_H\n" + + "milarly, the word size is at most 64 bits because it assumes that\n// converting a usize to u64 will never overflow.\n#if __WORDSIZE < 32\n#error \"Wuffs requires a word size of at least 32 bits\"\n#elif __WORDSIZE > 64\n#error \"Wuffs requires a word size of at most 64 bits\"\n#endif\n\n// WUFFS_VERSION is the major.minor version number as a uint32. The major\n// number is the high 16 bits. The minor number is the low 16 bits.\n//\n// The intention is to bump the version number at least on every API / ABI\n// backwards incompatible change.\n//\n// For now, the API and ABI are simply unstable and can change at any time.\n//\n// TODO: don't hard code this in base-header.h.\n#define WUFFS_VERSION 0x00001\n\n// ---------------- Fundamentals\n\n// Flicks are a unit of time. One flick (frame-tick) is 1 / 705_600_000 of a\n// second. See https://github.com/OculusVR/Flicks\ntypedef uint64_t wuffs_base__flicks;\n\n#define WUFFS_BASE__FLICKS_PER_SECOND 705600000ULL\n\n// wuffs_base__rectangle is a rectangle on the integer grid. It contains all\n// p" + + "oints (x, y) such that ((min_x <= x) && (x < max_x)) and likewise for y. It\n// is therefore empty if min_x >= max_x. There are multiple representations of\n// an empty rectangle.\n//\n// A value with all fields zero is a valid, empty rectangle.\n//\n// The X and Y axes increase right and down.\ntypedef struct {\n uint32_t min_x;\n uint32_t min_y;\n uint32_t max_x;\n uint32_t max_y;\n} wuffs_base__rectangle;\n\nstatic inline uint32_t wuffs_base__rectangle__width(wuffs_base__rectangle r) {\n return r.max_x > r.min_x ? r.max_x - r.min_x : 0;\n}\n\nstatic inline uint32_t wuffs_base__rectangle__height(wuffs_base__rectangle r) {\n return r.max_y > r.min_y ? r.max_y - r.min_y : 0;\n}\n\n// wuffs_base__slice_u8 is a 1-dimensional buffer.\n//\n// A value with all fields NULL or zero is a valid, empty slice.\ntypedef struct {\n uint8_t* ptr;\n size_t len;\n} wuffs_base__slice_u8;\n\n// wuffs_base__table_u8 is a 2-dimensional buffer.\n//\n// A value with all fields NULL or zero is a valid, empty table.\ntypedef struct {\n uint8_t* ptr;\n size_" + + "t width;\n size_t height;\n size_t stride;\n} wuffs_base__table_u8;\n\n// ---------------- I/O\n\n// TODO: rename buf1 to io_buffer, writer1 to io_writer, etc.\n\n// wuffs_base__buf1 is a 1-dimensional buffer (a pointer and length), plus\n// additional indexes into that buffer, plus an opened / closed flag.\n//\n// A value with all fields NULL or zero is a valid, empty buffer.\ntypedef struct {\n uint8_t* ptr; // Pointer.\n size_t len; // Length.\n size_t wi; // Write index. Invariant: wi <= len.\n size_t ri; // Read index. Invariant: ri <= wi.\n bool closed; // No further writes are expected.\n} wuffs_base__buf1;\n\n// wuffs_base__limit1 provides a limited view of a 1-dimensional byte stream:\n// its first N bytes. That N can be greater than a buffer's current read or\n// write capacity. N decreases naturally over time as bytes are read from or\n// written to the stream.\n//\n// A value with all fields NULL or zero is a valid, unlimited view.\ntypedef struct wuffs_base__limit1 {\n uint64_t* ptr_to_len; " + + " // Pointer to N.\n struct wuffs_base__limit1* next; // Linked list of limits.\n} wuffs_base__limit1;\n\ntypedef struct {\n // TODO: move buf into private_impl? As it is, it looks like users can modify\n // the buf field to point to a different buffer, which can turn the limit and\n // mark fields into dangling pointers.\n wuffs_base__buf1* buf;\n // Do not access the private_impl's fields directly. There is no API/ABI\n // compatibility or safety guarantee if you do so.\n struct {\n wuffs_base__limit1 limit;\n uint8_t* mark;\n } private_impl;\n} wuffs_base__reader1;\n\ntypedef struct {\n // TODO: move buf into private_impl? As it is, it looks like users can modify\n // the buf field to point to a different buffer, which can turn the limit and\n // mark fields into dangling pointers.\n wuffs_base__buf1* buf;\n // Do not access the private_impl's fields directly. There is no API/ABI\n // compatibility or safety guarantee if you do so.\n struct {\n wuffs_base__limit1 limit;\n uint8_t* mark;\n } private_impl" + + ";\n} wuffs_base__writer1;\n\n// ---------------- Images\n\n// wuffs_base__pixel_format encodes the format of the bytes that constitute an\n// image frame's pixel data. Its bits:\n// - bit 31 is reserved.\n// - bits 30 .. 28 encodes color (and channel order, in terms of memory).\n// - bits 27 .. 26 are reserved.\n// - bits 25 .. 24 encodes transparency.\n// - bit 23 indicates big-endian/MSB-first (as opposed to little/LSB).\n// - bit 22 indicates floating point (as opposed to integer).\n// - bits 21 .. 20 are the number of planes, minus 1. Zero means packed.\n// - bits 19 .. 16 encodes the number of bits (depth) in an index value.\n// Zero means direct, not palette-indexed.\n// - bits 15 .. 12 encodes the number of bits (depth) in the 3rd channel.\n// - bits 11 .. 8 encodes the number of bits (depth) in the 2nd channel.\n// - bits 7 .. 4 encodes the number of bits (depth) in the 1st channel.\n// - bits 3 .. 0 encodes the number of bits (depth) in the 0th channel.\n//\n// The " + + "bit fields of a wuffs_base__pixel_format are not independent. For\n// example, the number of planes should not be greater than the number of\n// channels. Similarly, bits 15..4 are unused (and should be zero) if bits\n// 31..24 (color and transparency) together imply only 1 channel (gray, no\n// alpha) and floating point samples should mean a bit depth of 16, 32 or 64.\n//\n// Formats hold between 1 and 4 channels. For example: Y (1 channel: gray), YA\n// (2 channels: gray and alpha), BGR (3 channels: blue, green, red) or CMYK (4\n// channels: cyan, magenta, yellow, black).\n//\n// For direct formats with N > 1 channels, those channels can be laid out in\n// either 1 (packed) or N (planar) planes. For example, RGBA data is usually\n// packed, but YUV data is usually planar, due to chroma subsampling (for\n// details, see the wuffs_base__pixel_subsampling type). For indexed formats,\n// the palette (always 256 × 4 bytes) holds up to 4 packed bytes of color data\n// per index value, and there is only 1 plane (for the index)." + + " The distance\n// between successive palette elements is always 4 bytes.\n//\n// The color field is encoded in 3 bits:\n// - 0 means A (Alpha).\n// - 1 means Y or YA (Gray, Alpha).\n// - 2 means BGR, BGRX or BGRA (Blue, Green, Red, X-padding or Alpha).\n// - 3 means RGB, RGBX or RGBA (Red, Green, Blue, X-padding or Alpha).\n// - 4 means YUV or YUVA (Luma, Chroma-blue, Chroma-red, Alpha).\n// - 5 means CMY or CMYK (Cyan, Magenta, Yellow, Black).\n// - all other values are reserved.\n//\n// In Wuffs, channels are given in memory order, regardless of endianness,\n// since the C type for the pixel data is an array of bytes, not an array of\n// uint32_t. For example, packed BGRA with 8 bits per channel means that the\n// bytes in memory are always Blue, Green, Red then Alpha. On big-endian\n// systems, that is the uint32_t 0xBBGGRRAA. On little-endian, 0xAARRGGBB.\n//\n// When the color field (3 bits) encodes multiple options, the transparency\n// field (2 bits) distinguishes them:\n// - " + + "0 means fully opaque, no extra channels\n// - 1 means fully opaque, one extra channel (X or K, padding or black).\n// - 2 means one extra alpha channel, other channels are non-premultiplied.\n// - 3 means one extra alpha channel, other channels are premultiplied.\n//\n// The zero wuffs_base__pixel_format value is an invalid pixel format, as it is\n// invalid to combine the zero color (alpha only) with the zero transparency.\n//\n// Bit depth is encoded in 4 bits:\n// - 0 means the channel or index is unused.\n// - x means a bit depth of x, for x in the range 1..8.\n// - 9 means a bit depth of 10.\n// - 10 means a bit depth of 12.\n// - 11 means a bit depth of 16.\n// - 12 means a bit depth of 24.\n// - 13 means a bit depth of 32.\n// - 14 means a bit depth of 48.\n// - 15 means a bit depth of 64.\n//\n// For example, wuffs_base__pixel_format 0x3280BBBB is a natural format for\n// decoding a PNG image - network byte order (also known as big-endian),\n// packed, non-premultiplied alpha - that happens to be 16-bi" + + "t-depth truecolor\n// with alpha (RGBA). In memory order:\n//\n// ptr+0 ptr+1 ptr+2 ptr+3 ptr+4 ptr+5 ptr+6 ptr+7\n// Rhi Rlo Ghi Glo Bhi Blo Ahi Alo\n//\n// For example, the value wuffs_base__pixel_format 0x20000565 means BGR with no\n// alpha or padding, 5/6/5 bits for blue/green/red, packed 2 bytes per pixel,\n// laid out LSB-first in memory order:\n//\n// ptr+0........... ptr+1...........\n// MSB LSB MSB LSB\n// G₂G₁G₀B₄B₃B₂B₁B₀ R₄R₃R₂R₁R₀G₅G₄G₃\n//\n// On little-endian systems (but not big-endian), this Wuffs pixel format value\n// (0x20000565) corresponds to the Cairo library's CAIRO_FORMAT_RGB16_565, the\n// SDL2 (Simple DirectMedia Layer 2) library's SDL_PIXELFORMAT_RGB565 and the\n// Skia library's kRGB_565_SkColorType. Note BGR in Wuffs versus RGB in the\n// other libraries.\n//\n// Regardless of endianness, this Wuffs pixel format value (0x20000565)\n// corresponds to the V4L2 (Video For Linux 2) library's V4L2_PIX_FMT_RGB565\n// and t" + + "he Wayland-DRM library's WL_DRM_FORMAT_RGB565.\n//\n// Different software libraries name their pixel formats (and especially their\n// channel order) either according to memory layout or as bits of a native\n// integer type like uint32_t. The two conventions differ because of a system's\n// endianness. As mentioned earlier, Wuffs pixel formats are always in memory\n// order. More detail of other software libraries' naming conventions is in the\n// Pixel Format Guide at https://afrantzis.github.io/pixel-format-guide/\n//\n// Do not manipulate these bits directly; they are private implementation\n// details. Use methods such as wuffs_base__pixel_format__num_planes instead.\ntypedef uint32_t wuffs_base__pixel_format;\n\n// Common 8-bit-depth pixel formats. This list is not exhaustive; not all valid\n// wuffs_base__pixel_format values are present.\n\n#define WUFFS_BASE__PIXEL_FORMAT__INVALID ((wuffs_base__pixel_format)0x00000000)\n\n#define WUFFS_BASE__PIXEL_FORMAT__A ((wuffs_base__pixel_format)0x02000008)\n\n#define WUFFS_BASE__PIX" + + "EL_FORMAT__Y ((wuffs_base__pixel_format)0x10000008)\n#define WUFFS_BASE__PIXEL_FORMAT__YA_NONPREMUL \\\n ((wuffs_base__pixel_format)0x12000008)\n#define WUFFS_BASE__PIXEL_FORMAT__YA_PREMUL \\\n ((wuffs_base__pixel_format)0x13000008)\n\n#define WUFFS_BASE__PIXEL_FORMAT__BGR ((wuffs_base__pixel_format)0x20000888)\n#define WUFFS_BASE__PIXEL_FORMAT__BGRX ((wuffs_base__pixel_format)0x21008888)\n#define WUFFS_BASE__PIXEL_FORMAT__BGRX_INDEXED \\\n ((wuffs_base__pixel_format)0x21088888)\n#define WUFFS_BASE__PIXEL_FORMAT__BGRA_NONPREMUL \\\n ((wuffs_base__pixel_format)0x22008888)\n#define WUFFS_BASE__PIXEL_FORMAT__BGRA_NONPREMUL_INDEXED \\\n ((wuffs_base__pixel_format)0x22088888)\n#define WUFFS_BASE__PIXEL_FORMAT__BGRA_PREMUL \\\n ((wuffs_base__pixel_format)0x23008888)\n\n#define WUFFS_BASE__PIXEL_FORMAT__RGB ((wuffs_base__pixel_format)0x30000888)\n#define WUFFS_BASE__PIXEL_FORMAT__RGBX ((wuffs_base__pixel_format)0x31008888)\n#define WUFFS_BASE__PIXEL_FORMAT__RGBX_INDEXED \\\n ((wuffs_base__pixel_format)0x31088888)\n#define WUFFS_BASE__PI" + + "XEL_FORMAT__RGBA_NONPREMUL \\\n ((wuffs_base__pixel_format)0x32008888)\n#define WUFFS_BASE__PIXEL_FORMAT__RGBA_NONPREMUL_INDEXED \\\n ((wuffs_base__pixel_format)0x32088888)\n#define WUFFS_BASE__PIXEL_FORMAT__RGBA_PREMUL \\\n ((wuffs_base__pixel_format)0x33008888)\n\n#define WUFFS_BASE__PIXEL_FORMAT__YUV ((wuffs_base__pixel_format)0x40200888)\n#define WUFFS_BASE__PIXEL_FORMAT__YUVK ((wuffs_base__pixel_format)0x41308888)\n#define WUFFS_BASE__PIXEL_FORMAT__YUVA_NONPREMUL \\\n ((wuffs_base__pixel_format)0x42308888)\n\n#define WUFFS_BASE__PIXEL_FORMAT__CMY ((wuffs_base__pixel_format)0x50200888)\n#define WUFFS_BASE__PIXEL_FORMAT__CMYK ((wuffs_base__pixel_format)0x51308888)\n\nstatic inline bool wuffs_base__pixel_format__is_valid(\n wuffs_base__pixel_format f) {\n return f != 0;\n}\n\nstatic inline bool wuffs_base__pixel_format__is_indexed(\n wuffs_base__pixel_format f) {\n return ((f >> 16) & 0x0F) != 0;\n}\n\n#define WUFFS_BASE__PIXEL_FORMAT__NUM_PLANES_MAX 4\n\nstatic inline uint32_t wuffs_base__pixel_format__num_planes(\n wuffs_" + + "base__pixel_format f) {\n return f ? (((f >> 20) & 0x03) + 1) : 0;\n}\n\ntypedef struct {\n wuffs_base__table_u8 planes[WUFFS_BASE__PIXEL_FORMAT__NUM_PLANES_MAX];\n} wuffs_base__pixel_buffer;\n\n// wuffs_base__pixel_subsampling encodes the mapping of pixel space coordinates\n// (x, y) to pixel buffer indices (i, j). That mapping can differ for each\n// plane p. For a depth of 8 bits (1 byte), the p'th plane's sample starts at\n// (planes[p].ptr + (j * planes[p].stride) + i).\n//\n// For packed pixel formats, the mapping is trivial: i = x and j = y. For\n// planar pixel formats, the mapping can differ due to chroma subsampling. For\n// example, consider a three plane YUV pixel format with 4:2:2 subsampling. For\n// the luma (Y) channel, there is one sample for every pixel, but for the\n// chroma (U, V) channels, there is one sample for every two pixels: pairs of\n// horizontally adjacent pixels form one macropixel, i = x / 2 and j == y. In\n// general, for a given p:\n// - i = (x + bias_x) >> shift_x.\n// - j = (y + bias_y) >>" + + " shift_y.\n// where biases and shifts are in the range 0..3 and 0..2 respectively.\n//\n// In general, the biases will be zero after decoding an image. However, making\n// a sub-image may change the bias, since the (x, y) coordinates are relative\n// to the sub-image's top-left origin, but the backing pixel buffers were\n// created relative to the original image's origin.\n//\n// For each plane p, each of those four numbers (biases and shifts) are encoded\n// in two bits, which combine to form an 8 bit unsigned integer:\n//\n// e_p = (bias_x << 6) | (shift_x << 4) | (bias_y << 2) | (shift_y << 0)\n//\n// Those e_p values (e_0 for the first plane, e_1 for the second plane, etc)\n// combine to form a wuffs_base__pixel_subsampling value:\n//\n// pixsub = (e_3 << 24) | (e_2 << 16) | (e_1 << 8) | (e_0 << 0)\n//\n// Do not manipulate these bits directly; they are private implementation\n// details. Use methods such as wuffs_base__pixel_subsampling__bias_x instead.\ntypedef uint32_t wuffs_base__pixel_subsampling;\n\n#define WUFFS_BASE_" + + "_PIXEL_SUBSAMPLING__NONE ((wuffs_base__pixel_subsampling)0);\n\n#define WUFFS_BASE__PIXEL_SUBSAMPLING__444 \\\n ((wuffs_base__pixel_subsampling)0x000000);\n#define WUFFS_BASE__PIXEL_SUBSAMPLING__440 \\\n ((wuffs_base__pixel_subsampling)0x010100);\n#define WUFFS_BASE__PIXEL_SUBSAMPLING__422 \\\n ((wuffs_base__pixel_subsampling)0x101000);\n#define WUFFS_BASE__PIXEL_SUBSAMPLING__420 \\\n ((wuffs_base__pixel_subsampling)0x111100);\n#define WUFFS_BASE__PIXEL_SUBSAMPLING__411 \\\n ((wuffs_base__pixel_subsampling)0x202000);\n#define WUFFS_BASE__PIXEL_SUBSAMPLING__410 \\\n ((wuffs_base__pixel_subsampling)0x212100);\n\nstatic inline uint32_t wuffs_base__pixel_subsampling__bias_x(\n wuffs_base__pixel_subsampling s,\n uint32_t plane) {\n uint32_t shift = ((plane & 0x03) * 8) + 6;\n return (s >> shift) & 0x03;\n}\n\nstatic inline uint32_t wuffs_base__pixel_subsampling__shift_x(\n wuffs_base__pixel_subsampling s,\n uint32_t plane) {\n uint32_t shift = ((plane & 0x03) * 8) + 4;\n return (s >> shift) & 0x03;\n}\n\nstatic inline uint32_t" + + " wuffs_base__pixel_subsampling__bias_y(\n wuffs_base__pixel_subsampling s,\n uint32_t plane) {\n uint32_t shift = ((plane & 0x03) * 8) + 2;\n return (s >> shift) & 0x03;\n}\n\nstatic inline uint32_t wuffs_base__pixel_subsampling__shift_y(\n wuffs_base__pixel_subsampling s,\n uint32_t plane) {\n uint32_t shift = ((plane & 0x03) * 8) + 0;\n return (s >> shift) & 0x03;\n}\n\ntypedef struct {\n // Do not access the private_impl's fields directly. There is no API/ABI\n // compatibility or safety guarantee if you do so.\n struct {\n wuffs_base__pixel_format pixfmt;\n wuffs_base__pixel_subsampling pixsub;\n uint32_t width;\n uint32_t height;\n uint32_t num_loops;\n } private_impl;\n} wuffs_base__image_config;\n\nstatic inline void wuffs_base__image_config__initialize(\n wuffs_base__image_config* c,\n wuffs_base__pixel_format pixfmt,\n wuffs_base__pixel_subsampling pixsub,\n uint32_t width,\n uint32_t height,\n uint32_t num_loops) {\n if (!c) {\n return;\n }\n // TODO: move the check from wuffs" + + "_base__image_config__is_valid here. Should\n // this function return bool? An error type?\n c->private_impl.pixfmt = pixfmt;\n c->private_impl.pixsub = pixsub;\n c->private_impl.width = width;\n c->private_impl.height = height;\n c->private_impl.num_loops = num_loops;\n}\n\nstatic inline void wuffs_base__image_config__invalidate(\n wuffs_base__image_config* c) {\n if (c) {\n *c = ((wuffs_base__image_config){});\n }\n}\n\nstatic inline bool wuffs_base__image_config__is_valid(\n wuffs_base__image_config* c) {\n if (!c || !c->private_impl.pixfmt) {\n return false;\n }\n uint64_t wh =\n ((uint64_t)c->private_impl.width) * ((uint64_t)c->private_impl.height);\n // TODO: handle things other than 1 byte per pixel.\n //\n // TODO: move the check to wuffs_base__image_config__initialize.\n return wh <= ((uint64_t)SIZE_MAX);\n}\n\nstatic inline wuffs_base__pixel_format wuffs_base__image_config__pixel_format(\n wuffs_base__image_config* c) {\n return c ? c->private_impl.pixfmt : 0;\n}\n\nstatic inline wuffs_base__pixel_" + + "subsampling\nwuffs_base__image_config__pixel_subsampling(wuffs_base__image_config* c) {\n return wuffs_base__image_config__is_valid(c) ? c->private_impl.pixsub : 0;\n}\n\nstatic inline uint32_t wuffs_base__image_config__width(\n wuffs_base__image_config* c) {\n return wuffs_base__image_config__is_valid(c) ? c->private_impl.width : 0;\n}\n\nstatic inline uint32_t wuffs_base__image_config__height(\n wuffs_base__image_config* c) {\n return wuffs_base__image_config__is_valid(c) ? c->private_impl.height : 0;\n}\n\nstatic inline uint32_t wuffs_base__image_config__num_loops(\n wuffs_base__image_config* c) {\n return wuffs_base__image_config__is_valid(c) ? c->private_impl.num_loops : 0;\n}\n\n// TODO: this is the right API for planar (not packed) pixbufs? Should it allow\n// decoding into a color model different from the format's intrinsic one? For\n// example, decoding a JPEG image straight to RGBA instead of to YCbCr?\nstatic inline size_t wuffs_base__image_config__pixbuf_size(\n wuffs_base__image_config* c) {\n if (wuffs_" + + "base__image_config__is_valid(c)) {\n uint64_t wh =\n ((uint64_t)c->private_impl.width) * ((uint64_t)c->private_impl.height);\n // TODO: handle things other than 1 byte per pixel.\n return (size_t)wh;\n }\n return 0;\n}\n\ntypedef struct {\n // Do not access the private_impl's fields directly. There is no API/ABI\n // compatibility or safety guarantee if you do so.\n struct {\n wuffs_base__image_config config;\n uint32_t loop_count; // 0-based count of the current loop.\n wuffs_base__pixel_buffer pixbuf;\n // TODO: color spaces.\n wuffs_base__rectangle dirty_rect;\n wuffs_base__flicks duration;\n uint8_t palette[1024];\n } private_impl;\n} wuffs_base__image_buffer;\n\nstatic inline void wuffs_base__image_buffer__initialize(\n wuffs_base__image_buffer* f,\n wuffs_base__image_config config,\n wuffs_base__pixel_buffer pixbuf) {\n if (!f) {\n return;\n }\n *f = ((wuffs_base__image_buffer){});\n f->private_impl.config = config;\n f->private_impl.pixbuf = pixbuf;\n}\n\nstatic inline void wu" + + "ffs_base__image_buffer__update(\n wuffs_base__image_buffer* f,\n wuffs_base__rectangle dirty_rect,\n wuffs_base__flicks duration,\n uint8_t* palette_ptr,\n size_t palette_len) {\n if (!f) {\n return;\n }\n f->private_impl.dirty_rect = dirty_rect;\n f->private_impl.duration = duration;\n if (palette_ptr) {\n memmove(f->private_impl.palette, palette_ptr,\n palette_len <= 1024 ? palette_len : 1024);\n }\n}\n\n// wuffs_base__image_buffer__loop returns whether the image decoder should loop\n// back to the beginning of the animation, assuming that we've reached the end\n// of the encoded stream. If so, it increments f's count of the animation loops\n// played so far.\nstatic inline bool wuffs_base__image_buffer__loop(wuffs_base__image_buffer* f) {\n if (!f) {\n return false;\n }\n uint32_t n = f->private_impl.config.private_impl.num_loops;\n if (n == 0) {\n return true;\n }\n if (f->private_impl.loop_count < n - 1) {\n f->private_impl.loop_count++;\n return true;\n }\n return false;\n}\n\n// w" + + "uffs_base__image_buffer__dirty_rect returns an upper bound for what part of\n// this frame's pixels differs from the previous frame.\nstatic inline wuffs_base__rectangle wuffs_base__image_buffer__dirty_rect(\n wuffs_base__image_buffer* f) {\n return f ? f->private_impl.dirty_rect : ((wuffs_base__rectangle){0});\n}\n\n// wuffs_base__image_buffer__duration returns the amount of time to display\n// this frame. Zero means to display forever - a still (non-animated) image.\nstatic inline wuffs_base__flicks wuffs_base__image_buffer__duration(\n wuffs_base__image_buffer* f) {\n return f ? f->private_impl.duration : 0;\n}\n\n// wuffs_base__image_buffer__palette returns the palette that the pixel data\n// can index. The backing array is inside f and has length 1024.\nstatic inline uint8_t* wuffs_base__image_buffer__palette(\n wuffs_base__image_buffer* f) {\n return f ? f->private_impl.palette : NULL;\n}\n\n#endif // WUFFS_BASE_HEADER_H\n" + "" const baseImpl = "" + "// Copyright 2017 The Wuffs Authors.\n//\n// Licensed under the Apache License, Version 2.0 (the \"License\");\n// you may not use this file except in compliance with the License.\n// You may obtain a copy of the License at\n//\n// https://www.apache.org/licenses/LICENSE-2.0\n//\n// Unless required by applicable law or agreed to in writing, software\n// distributed under the License is distributed on an \"AS IS\" BASIS,\n// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n// See the License for the specific language governing permissions and\n// limitations under the License.\n\n// wuffs_base__empty_struct is used when a Wuffs function returns an empty\n// struct. In C, if a function f returns void, you can't say \"x = f()\", but in\n// Wuffs, if a function g returns empty, you can say \"y = g()\".\ntypedef struct {\n} wuffs_base__empty_struct;\n\n#define WUFFS_BASE__IGNORE_POTENTIALLY_UNUSED_VARIABLE(x) (void)(x)\n\n// WUFFS_BASE__MAGIC is a magic number to check that initializers are called.\n// It's not foolp" + - "roof, given C doesn't automatically zero memory before use,\n// but it should catch 99.99% of cases.\n//\n// Its (non-zero) value is arbitrary, based on md5sum(\"wuffs\").\n#define WUFFS_BASE__MAGIC (0x3CCB6C71U)\n\n// WUFFS_BASE__ALREADY_ZEROED is passed from a container struct's initializer\n// to a containee struct's initializer when the container has already zeroed\n// the containee's memory.\n//\n// Its (non-zero) value is arbitrary, based on md5sum(\"zeroed\").\n#define WUFFS_BASE__ALREADY_ZEROED (0x68602EF1U)\n\n// Denote intentional fallthroughs for -Wimplicit-fallthrough.\n//\n// The order matters here. Clang also defines \"__GNUC__\".\n#if defined(__clang__) && __cplusplus >= 201103L\n#define WUFFS_BASE__FALLTHROUGH [[clang::fallthrough]]\n#elif !defined(__clang__) && defined(__GNUC__) && (__GNUC__ >= 7)\n#define WUFFS_BASE__FALLTHROUGH __attribute__((fallthrough))\n#else\n#define WUFFS_BASE__FALLTHROUGH\n#endif\n\n// Use switch cases for coroutine suspension points, similar to the technique\n// in https://www.chiark.greenend.org" + - ".uk/~sgtatham/coroutines.html\n//\n// We use trivial macros instead of an explicit assignment and case statement\n// so that clang-format doesn't get confused by the unusual \"case\"s.\n#define WUFFS_BASE__COROUTINE_SUSPENSION_POINT_0 case 0:;\n#define WUFFS_BASE__COROUTINE_SUSPENSION_POINT(n) \\\n coro_susp_point = n; \\\n WUFFS_BASE__FALLTHROUGH; \\\n case n:;\n\n#define WUFFS_BASE__COROUTINE_SUSPENSION_POINT_MAYBE_SUSPEND(n) \\\n if (status < 0) { \\\n goto exit; \\\n } else if (status == 0) { \\\n goto ok; \\\n } \\\n coro_susp_point = n; \\\n goto suspend; \\\n case n:;\n\n// Clang also defines \"__GNUC__\".\n#if defined(__GNUC__)\n#define WUFFS_BASE__LI" + - "KELY(expr) (__builtin_expect(!!(expr), 1))\n#define WUFFS_BASE__UNLIKELY(expr) (__builtin_expect(!!(expr), 0))\n#else\n#define WUFFS_BASE__LIKELY(expr) (expr)\n#define WUFFS_BASE__UNLIKELY(expr) (expr)\n#endif\n\n// Uncomment this #include for printf-debugging.\n// #include \n\n// ---------------- Static Inline Functions\n//\n// The helpers below are functions, instead of macros, because their arguments\n// can be an expression that we shouldn't evaluate more than once.\n//\n// They are in base-impl.h and hence copy/pasted into every generated C file,\n// instead of being in some \"base.c\" file, since a design goal is that users of\n// the generated C code can often just #include a single .c file, such as\n// \"gif.c\", without having to additionally include or otherwise build and link\n// a \"base.c\" file.\n//\n// They are static, so that linking multiple wuffs .o files won't complain about\n// duplicate function definitions.\n//\n// They are explicitly marked inline, even if modern compilers don't use the\n// inline attribute " + - "to guide optimizations such as inlining, to avoid the\n// -Wunused-function warning, and we like to compile with -Wall -Werror.\n\nstatic inline uint16_t wuffs_base__load_u16be(uint8_t* p) {\n return ((uint16_t)(p[0]) << 8) | ((uint16_t)(p[1]) << 0);\n}\n\nstatic inline uint16_t wuffs_base__load_u16le(uint8_t* p) {\n return ((uint16_t)(p[0]) << 0) | ((uint16_t)(p[1]) << 8);\n}\n\nstatic inline uint32_t wuffs_base__load_u32be(uint8_t* p) {\n return ((uint32_t)(p[0]) << 24) | ((uint32_t)(p[1]) << 16) |\n ((uint32_t)(p[2]) << 8) | ((uint32_t)(p[3]) << 0);\n}\n\nstatic inline uint32_t wuffs_base__load_u32le(uint8_t* p) {\n return ((uint32_t)(p[0]) << 0) | ((uint32_t)(p[1]) << 8) |\n ((uint32_t)(p[2]) << 16) | ((uint32_t)(p[3]) << 24);\n}\n\nstatic inline wuffs_base__slice_u8 wuffs_base__slice_u8__subslice_i(\n wuffs_base__slice_u8 s,\n uint64_t i) {\n if ((i <= SIZE_MAX) && (i <= s.len)) {\n return ((wuffs_base__slice_u8){\n .ptr = s.ptr + i,\n .len = s.len - i,\n });\n }\n return ((wuffs_base_" + - "_slice_u8){});\n}\n\nstatic inline wuffs_base__slice_u8 wuffs_base__slice_u8__subslice_j(\n wuffs_base__slice_u8 s,\n uint64_t j) {\n if ((j <= SIZE_MAX) && (j <= s.len)) {\n return ((wuffs_base__slice_u8){.ptr = s.ptr, .len = j});\n }\n return ((wuffs_base__slice_u8){});\n}\n\nstatic inline wuffs_base__slice_u8 wuffs_base__slice_u8__subslice_ij(\n wuffs_base__slice_u8 s,\n uint64_t i,\n uint64_t j) {\n if ((i <= j) && (j <= SIZE_MAX) && (j <= s.len)) {\n return ((wuffs_base__slice_u8){\n .ptr = s.ptr + i,\n .len = j - i,\n });\n }\n return ((wuffs_base__slice_u8){});\n}\n\n// wuffs_base__slice_u8__prefix returns up to the first up_to bytes of s.\nstatic inline wuffs_base__slice_u8 wuffs_base__slice_u8__prefix(\n wuffs_base__slice_u8 s,\n uint64_t up_to) {\n if ((uint64_t)(s.len) > up_to) {\n s.len = up_to;\n }\n return s;\n}\n\n// wuffs_base__slice_u8__suffix returns up to the last up_to bytes of s.\nstatic inline wuffs_base__slice_u8 wuffs_base__slice_u8_suffix(\n wuffs_base__slice_u8 " + - "s,\n uint64_t up_to) {\n if ((uint64_t)(s.len) > up_to) {\n s.ptr += (uint64_t)(s.len) - up_to;\n s.len = up_to;\n }\n return s;\n}\n\n// wuffs_base__slice_u8__copy_from_slice calls memmove(dst.ptr, src.ptr,\n// length) where length is the minimum of dst.len and src.len.\n//\n// Passing a wuffs_base__slice_u8 with all fields NULL or zero (a valid, empty\n// slice) is valid and results in a no-op.\nstatic inline uint64_t wuffs_base__slice_u8__copy_from_slice(\n wuffs_base__slice_u8 dst,\n wuffs_base__slice_u8 src) {\n size_t length = dst.len < src.len ? dst.len : src.len;\n if (length > 0) {\n memmove(dst.ptr, src.ptr, length);\n }\n return length;\n}\n\nstatic inline uint32_t wuffs_base__writer1__copy_from_history32(\n uint8_t** ptr_ptr,\n uint8_t* start, // May be NULL, meaning an unmarked writer1.\n uint8_t* end,\n uint32_t distance,\n uint32_t length) {\n if (!start || !distance) {\n return 0;\n }\n uint8_t* ptr = *ptr_ptr;\n if ((size_t)(ptr - start) < (size_t)(distance)) {\n return 0;\n }" + - "\n start = ptr - distance;\n size_t n = end - ptr;\n if ((size_t)(length) > n) {\n length = n;\n } else {\n n = length;\n }\n // TODO: unrolling by 3 seems best for the std/deflate benchmarks, but that\n // is mostly because 3 is the minimum length for the deflate format. This\n // function implementation shouldn't overfit to that one format. Perhaps the\n // copy_from_history32 Wuffs method should also take an unroll hint argument,\n // and the cgen can look if that argument is the constant expression '3'.\n //\n // See also wuffs_base__writer1__copy_from_history32__bco below.\n //\n // Alternatively, or additionally, have a sloppy_copy_from_history32 method\n // that copies 8 bytes at a time, possibly writing more than length bytes?\n for (; n >= 3; n -= 3) {\n *ptr++ = *start++;\n *ptr++ = *start++;\n *ptr++ = *start++;\n }\n for (; n; n--) {\n *ptr++ = *start++;\n }\n *ptr_ptr = ptr;\n return length;\n}\n\n// wuffs_base__writer1__copy_from_history32__bco is a Bounds Check Optimized\n// version of " + - "the wuffs_base__writer1__copy_from_history32 function above. The\n// caller needs to prove that:\n// - start != NULL\n// - distance > 0\n// - distance <= (*ptr_ptr - start)\n// - length <= (end - *ptr_ptr)\nstatic inline uint32_t wuffs_base__writer1__copy_from_history32__bco(\n uint8_t** ptr_ptr,\n uint8_t* start,\n uint8_t* end,\n uint32_t distance,\n uint32_t length) {\n uint8_t* ptr = *ptr_ptr;\n start = ptr - distance;\n uint32_t n = length;\n for (; n >= 3; n -= 3) {\n *ptr++ = *start++;\n *ptr++ = *start++;\n *ptr++ = *start++;\n }\n for (; n; n--) {\n *ptr++ = *start++;\n }\n *ptr_ptr = ptr;\n return length;\n}\n\nstatic inline uint32_t wuffs_base__writer1__copy_from_reader32(\n uint8_t** ptr_wptr,\n uint8_t* wend,\n uint8_t** ptr_rptr,\n uint8_t* rend,\n uint32_t length) {\n uint8_t* wptr = *ptr_wptr;\n size_t n = length;\n if (n > wend - wptr) {\n n = wend - wptr;\n }\n uint8_t* rptr = *ptr_rptr;\n if (n > rend - rptr) {\n n = rend - rptr;\n }\n if (n > 0) {\n " + - "memmove(wptr, rptr, n);\n *ptr_wptr += n;\n *ptr_rptr += n;\n }\n return n;\n}\n\nstatic inline uint64_t wuffs_base__writer1__copy_from_slice(\n uint8_t** ptr_wptr,\n uint8_t* wend,\n wuffs_base__slice_u8 src) {\n uint8_t* wptr = *ptr_wptr;\n size_t n = src.len;\n if (n > wend - wptr) {\n n = wend - wptr;\n }\n if (n > 0) {\n memmove(wptr, src.ptr, n);\n *ptr_wptr += n;\n }\n return n;\n}\n\nstatic inline uint32_t wuffs_base__writer1__copy_from_slice32(\n uint8_t** ptr_wptr,\n uint8_t* wend,\n wuffs_base__slice_u8 src,\n uint32_t length) {\n uint8_t* wptr = *ptr_wptr;\n size_t n = src.len;\n if (n > length) {\n n = length;\n }\n if (n > wend - wptr) {\n n = wend - wptr;\n }\n if (n > 0) {\n memmove(wptr, src.ptr, n);\n *ptr_wptr += n;\n }\n return n;\n}\n\n// Note that the *__limit and *__mark methods are private (in base-impl.h) not\n// public (in base-header.h). We assume that, at the boundary between user code\n// and Wuffs code, the reader1 and writer1's private_impl fields (including" + - "\n// limit and mark) are NULL. Otherwise, some internal assumptions break down.\n// For example, limits could be represented as pointers, even though\n// conceptually they are counts, but that pointer-to-count correspondence\n// becomes invalid if a buffer is re-used (e.g. on resuming a coroutine).\n//\n// Admittedly, some of the Wuffs test code calls these methods, but that test\n// code is still Wuffs code, not user code. Other Wuffs test code modifies\n// private_impl fields directly.\n\nstatic inline wuffs_base__reader1 wuffs_base__reader1__limit(\n wuffs_base__reader1* o,\n uint64_t* ptr_to_len) {\n wuffs_base__reader1 ret = *o;\n ret.private_impl.limit.ptr_to_len = ptr_to_len;\n ret.private_impl.limit.next = &o->private_impl.limit;\n return ret;\n}\n\nstatic inline wuffs_base__empty_struct wuffs_base__reader1__mark(\n wuffs_base__reader1* o,\n uint8_t* mark) {\n o->private_impl.mark = mark;\n return ((wuffs_base__empty_struct){});\n}\n\n// TODO: static inline wuffs_base__writer1 wuffs_base__writer1__limit()\n\nst" + - "atic inline wuffs_base__empty_struct wuffs_base__writer1__mark(\n wuffs_base__writer1* o,\n uint8_t* mark) {\n o->private_impl.mark = mark;\n return ((wuffs_base__empty_struct){});\n}\n" + + "roof, given C doesn't automatically zero memory before use,\n// but it should catch 99.99% of cases.\n//\n// Its (non-zero) value is arbitrary, based on md5sum(\"wuffs\").\n#define WUFFS_BASE__MAGIC 0x3CCB6C71U\n\n// WUFFS_BASE__ALREADY_ZEROED is passed from a container struct's initializer\n// to a containee struct's initializer when the container has already zeroed\n// the containee's memory.\n//\n// Its (non-zero) value is arbitrary, based on md5sum(\"zeroed\").\n#define WUFFS_BASE__ALREADY_ZEROED 0x68602EF1U\n\n// Denote intentional fallthroughs for -Wimplicit-fallthrough.\n//\n// The order matters here. Clang also defines \"__GNUC__\".\n#if defined(__clang__) && __cplusplus >= 201103L\n#define WUFFS_BASE__FALLTHROUGH [[clang::fallthrough]]\n#elif !defined(__clang__) && defined(__GNUC__) && (__GNUC__ >= 7)\n#define WUFFS_BASE__FALLTHROUGH __attribute__((fallthrough))\n#else\n#define WUFFS_BASE__FALLTHROUGH\n#endif\n\n// Use switch cases for coroutine suspension points, similar to the technique\n// in https://www.chiark.greenend.org.uk/" + + "~sgtatham/coroutines.html\n//\n// We use trivial macros instead of an explicit assignment and case statement\n// so that clang-format doesn't get confused by the unusual \"case\"s.\n#define WUFFS_BASE__COROUTINE_SUSPENSION_POINT_0 case 0:;\n#define WUFFS_BASE__COROUTINE_SUSPENSION_POINT(n) \\\n coro_susp_point = n; \\\n WUFFS_BASE__FALLTHROUGH; \\\n case n:;\n\n#define WUFFS_BASE__COROUTINE_SUSPENSION_POINT_MAYBE_SUSPEND(n) \\\n if (status < 0) { \\\n goto exit; \\\n } else if (status == 0) { \\\n goto ok; \\\n } \\\n coro_susp_point = n; \\\n goto suspend; \\\n case n:;\n\n// Clang also defines \"__GNUC__\".\n#if defined(__GNUC__)\n#define WUFFS_BASE__LIKELY" + + "(expr) (__builtin_expect(!!(expr), 1))\n#define WUFFS_BASE__UNLIKELY(expr) (__builtin_expect(!!(expr), 0))\n#else\n#define WUFFS_BASE__LIKELY(expr) (expr)\n#define WUFFS_BASE__UNLIKELY(expr) (expr)\n#endif\n\n// Uncomment this #include for printf-debugging.\n// #include \n\n// ---------------- Static Inline Functions\n//\n// The helpers below are functions, instead of macros, because their arguments\n// can be an expression that we shouldn't evaluate more than once.\n//\n// They are in base-impl.h and hence copy/pasted into every generated C file,\n// instead of being in some \"base.c\" file, since a design goal is that users of\n// the generated C code can often just #include a single .c file, such as\n// \"gif.c\", without having to additionally include or otherwise build and link\n// a \"base.c\" file.\n//\n// They are static, so that linking multiple wuffs .o files won't complain about\n// duplicate function definitions.\n//\n// They are explicitly marked inline, even if modern compilers don't use the\n// inline attribute to g" + + "uide optimizations such as inlining, to avoid the\n// -Wunused-function warning, and we like to compile with -Wall -Werror.\n\nstatic inline uint16_t wuffs_base__load_u16be(uint8_t* p) {\n return ((uint16_t)(p[0]) << 8) | ((uint16_t)(p[1]) << 0);\n}\n\nstatic inline uint16_t wuffs_base__load_u16le(uint8_t* p) {\n return ((uint16_t)(p[0]) << 0) | ((uint16_t)(p[1]) << 8);\n}\n\nstatic inline uint32_t wuffs_base__load_u32be(uint8_t* p) {\n return ((uint32_t)(p[0]) << 24) | ((uint32_t)(p[1]) << 16) |\n ((uint32_t)(p[2]) << 8) | ((uint32_t)(p[3]) << 0);\n}\n\nstatic inline uint32_t wuffs_base__load_u32le(uint8_t* p) {\n return ((uint32_t)(p[0]) << 0) | ((uint32_t)(p[1]) << 8) |\n ((uint32_t)(p[2]) << 16) | ((uint32_t)(p[3]) << 24);\n}\n\nstatic inline wuffs_base__slice_u8 wuffs_base__slice_u8__subslice_i(\n wuffs_base__slice_u8 s,\n uint64_t i) {\n if ((i <= SIZE_MAX) && (i <= s.len)) {\n return ((wuffs_base__slice_u8){\n .ptr = s.ptr + i,\n .len = s.len - i,\n });\n }\n return ((wuffs_base__sli" + + "ce_u8){});\n}\n\nstatic inline wuffs_base__slice_u8 wuffs_base__slice_u8__subslice_j(\n wuffs_base__slice_u8 s,\n uint64_t j) {\n if ((j <= SIZE_MAX) && (j <= s.len)) {\n return ((wuffs_base__slice_u8){.ptr = s.ptr, .len = j});\n }\n return ((wuffs_base__slice_u8){});\n}\n\nstatic inline wuffs_base__slice_u8 wuffs_base__slice_u8__subslice_ij(\n wuffs_base__slice_u8 s,\n uint64_t i,\n uint64_t j) {\n if ((i <= j) && (j <= SIZE_MAX) && (j <= s.len)) {\n return ((wuffs_base__slice_u8){\n .ptr = s.ptr + i,\n .len = j - i,\n });\n }\n return ((wuffs_base__slice_u8){});\n}\n\n// wuffs_base__slice_u8__prefix returns up to the first up_to bytes of s.\nstatic inline wuffs_base__slice_u8 wuffs_base__slice_u8__prefix(\n wuffs_base__slice_u8 s,\n uint64_t up_to) {\n if ((uint64_t)(s.len) > up_to) {\n s.len = up_to;\n }\n return s;\n}\n\n// wuffs_base__slice_u8__suffix returns up to the last up_to bytes of s.\nstatic inline wuffs_base__slice_u8 wuffs_base__slice_u8_suffix(\n wuffs_base__slice_u8 s,\n " + + " uint64_t up_to) {\n if ((uint64_t)(s.len) > up_to) {\n s.ptr += (uint64_t)(s.len) - up_to;\n s.len = up_to;\n }\n return s;\n}\n\n// wuffs_base__slice_u8__copy_from_slice calls memmove(dst.ptr, src.ptr,\n// length) where length is the minimum of dst.len and src.len.\n//\n// Passing a wuffs_base__slice_u8 with all fields NULL or zero (a valid, empty\n// slice) is valid and results in a no-op.\nstatic inline uint64_t wuffs_base__slice_u8__copy_from_slice(\n wuffs_base__slice_u8 dst,\n wuffs_base__slice_u8 src) {\n size_t length = dst.len < src.len ? dst.len : src.len;\n if (length > 0) {\n memmove(dst.ptr, src.ptr, length);\n }\n return length;\n}\n\nstatic inline uint32_t wuffs_base__writer1__copy_from_history32(\n uint8_t** ptr_ptr,\n uint8_t* start, // May be NULL, meaning an unmarked writer1.\n uint8_t* end,\n uint32_t distance,\n uint32_t length) {\n if (!start || !distance) {\n return 0;\n }\n uint8_t* ptr = *ptr_ptr;\n if ((size_t)(ptr - start) < (size_t)(distance)) {\n return 0;\n }\n s" + + "tart = ptr - distance;\n size_t n = end - ptr;\n if ((size_t)(length) > n) {\n length = n;\n } else {\n n = length;\n }\n // TODO: unrolling by 3 seems best for the std/deflate benchmarks, but that\n // is mostly because 3 is the minimum length for the deflate format. This\n // function implementation shouldn't overfit to that one format. Perhaps the\n // copy_from_history32 Wuffs method should also take an unroll hint argument,\n // and the cgen can look if that argument is the constant expression '3'.\n //\n // See also wuffs_base__writer1__copy_from_history32__bco below.\n //\n // Alternatively, or additionally, have a sloppy_copy_from_history32 method\n // that copies 8 bytes at a time, possibly writing more than length bytes?\n for (; n >= 3; n -= 3) {\n *ptr++ = *start++;\n *ptr++ = *start++;\n *ptr++ = *start++;\n }\n for (; n; n--) {\n *ptr++ = *start++;\n }\n *ptr_ptr = ptr;\n return length;\n}\n\n// wuffs_base__writer1__copy_from_history32__bco is a Bounds Check Optimized\n// version of the " + + "wuffs_base__writer1__copy_from_history32 function above. The\n// caller needs to prove that:\n// - start != NULL\n// - distance > 0\n// - distance <= (*ptr_ptr - start)\n// - length <= (end - *ptr_ptr)\nstatic inline uint32_t wuffs_base__writer1__copy_from_history32__bco(\n uint8_t** ptr_ptr,\n uint8_t* start,\n uint8_t* end,\n uint32_t distance,\n uint32_t length) {\n uint8_t* ptr = *ptr_ptr;\n start = ptr - distance;\n uint32_t n = length;\n for (; n >= 3; n -= 3) {\n *ptr++ = *start++;\n *ptr++ = *start++;\n *ptr++ = *start++;\n }\n for (; n; n--) {\n *ptr++ = *start++;\n }\n *ptr_ptr = ptr;\n return length;\n}\n\nstatic inline uint32_t wuffs_base__writer1__copy_from_reader32(\n uint8_t** ptr_wptr,\n uint8_t* wend,\n uint8_t** ptr_rptr,\n uint8_t* rend,\n uint32_t length) {\n uint8_t* wptr = *ptr_wptr;\n size_t n = length;\n if (n > wend - wptr) {\n n = wend - wptr;\n }\n uint8_t* rptr = *ptr_rptr;\n if (n > rend - rptr) {\n n = rend - rptr;\n }\n if (n > 0) {\n memm" + + "ove(wptr, rptr, n);\n *ptr_wptr += n;\n *ptr_rptr += n;\n }\n return n;\n}\n\nstatic inline uint64_t wuffs_base__writer1__copy_from_slice(\n uint8_t** ptr_wptr,\n uint8_t* wend,\n wuffs_base__slice_u8 src) {\n uint8_t* wptr = *ptr_wptr;\n size_t n = src.len;\n if (n > wend - wptr) {\n n = wend - wptr;\n }\n if (n > 0) {\n memmove(wptr, src.ptr, n);\n *ptr_wptr += n;\n }\n return n;\n}\n\nstatic inline uint32_t wuffs_base__writer1__copy_from_slice32(\n uint8_t** ptr_wptr,\n uint8_t* wend,\n wuffs_base__slice_u8 src,\n uint32_t length) {\n uint8_t* wptr = *ptr_wptr;\n size_t n = src.len;\n if (n > length) {\n n = length;\n }\n if (n > wend - wptr) {\n n = wend - wptr;\n }\n if (n > 0) {\n memmove(wptr, src.ptr, n);\n *ptr_wptr += n;\n }\n return n;\n}\n\n// Note that the *__limit and *__mark methods are private (in base-impl.h) not\n// public (in base-header.h). We assume that, at the boundary between user code\n// and Wuffs code, the reader1 and writer1's private_impl fields (including\n// " + + "limit and mark) are NULL. Otherwise, some internal assumptions break down.\n// For example, limits could be represented as pointers, even though\n// conceptually they are counts, but that pointer-to-count correspondence\n// becomes invalid if a buffer is re-used (e.g. on resuming a coroutine).\n//\n// Admittedly, some of the Wuffs test code calls these methods, but that test\n// code is still Wuffs code, not user code. Other Wuffs test code modifies\n// private_impl fields directly.\n\nstatic inline wuffs_base__reader1 wuffs_base__reader1__limit(\n wuffs_base__reader1* o,\n uint64_t* ptr_to_len) {\n wuffs_base__reader1 ret = *o;\n ret.private_impl.limit.ptr_to_len = ptr_to_len;\n ret.private_impl.limit.next = &o->private_impl.limit;\n return ret;\n}\n\nstatic inline wuffs_base__empty_struct wuffs_base__reader1__mark(\n wuffs_base__reader1* o,\n uint8_t* mark) {\n o->private_impl.mark = mark;\n return ((wuffs_base__empty_struct){});\n}\n\n// TODO: static inline wuffs_base__writer1 wuffs_base__writer1__limit()\n\nstatic" + + " inline wuffs_base__empty_struct wuffs_base__writer1__mark(\n wuffs_base__writer1* o,\n uint8_t* mark) {\n o->private_impl.mark = mark;\n return ((wuffs_base__empty_struct){});\n}\n" + "" type template_args_short_read struct { diff --git a/cmd/wuffs-c/internal/cgen/expr.go b/cmd/wuffs-c/internal/cgen/expr.go index d70ca1d4..c418bfc7 100644 --- a/cmd/wuffs-c/internal/cgen/expr.go +++ b/cmd/wuffs-c/internal/cgen/expr.go @@ -362,7 +362,7 @@ func (g *gen) writeExprOther(b *buffer, n *a.Expr, rp replacementPolicy, pp pare ".len = v_r.private_impl.mark ? (size_t)(b_rptr_src - v_r.private_impl.mark) : 0, })") return nil } - if isThatMethod(g.tm, n, g.tm.ByName("initialize").Key(), 3) { + if isThatMethod(g.tm, n, g.tm.ByName("initialize").Key(), 5) { // TODO: don't hard-code a_dst. b.printf("wuffs_base__image_config__initialize(a_dst") for _, o := range n.Args() { diff --git a/fuzz/c/std/gif_fuzzer.cc b/fuzz/c/std/gif_fuzzer.cc index 284225b3..06c98313 100644 --- a/fuzz/c/std/gif_fuzzer.cc +++ b/fuzz/c/std/gif_fuzzer.cc @@ -50,7 +50,7 @@ void fuzz(wuffs_base__reader1 src_reader, uint32_t hash) { wuffs_base__image_config ic = {{0}}; s = wuffs_gif__decoder__decode_config(&dec, &ic, src_reader); - if (s || !wuffs_base__image_config__valid(&ic)) { + if (s || !wuffs_base__image_config__is_valid(&ic)) { goto exit; } diff --git a/gen/c/std/crc32.c b/gen/c/std/crc32.c index 8cbfe6e0..8b432bdd 100644 --- a/gen/c/std/crc32.c +++ b/gen/c/std/crc32.c @@ -46,11 +46,40 @@ // For now, the API and ABI are simply unstable and can change at any time. // // TODO: don't hard code this in base-header.h. -#define WUFFS_VERSION (0x00001) +#define WUFFS_VERSION 0x00001 -// ---------------- I/O +// ---------------- Fundamentals -// wuffs_base__slice_u8 is a 1-dimensional buffer (a pointer and length). +// Flicks are a unit of time. One flick (frame-tick) is 1 / 705_600_000 of a +// second. See https://github.com/OculusVR/Flicks +typedef uint64_t wuffs_base__flicks; + +#define WUFFS_BASE__FLICKS_PER_SECOND 705600000ULL + +// wuffs_base__rectangle is a rectangle on the integer grid. It contains all +// points (x, y) such that ((min_x <= x) && (x < max_x)) and likewise for y. It +// is therefore empty if min_x >= max_x. There are multiple representations of +// an empty rectangle. +// +// A value with all fields zero is a valid, empty rectangle. +// +// The X and Y axes increase right and down. +typedef struct { + uint32_t min_x; + uint32_t min_y; + uint32_t max_x; + uint32_t max_y; +} wuffs_base__rectangle; + +static inline uint32_t wuffs_base__rectangle__width(wuffs_base__rectangle r) { + return r.max_x > r.min_x ? r.max_x - r.min_x : 0; +} + +static inline uint32_t wuffs_base__rectangle__height(wuffs_base__rectangle r) { + return r.max_y > r.min_y ? r.max_y - r.min_y : 0; +} + +// wuffs_base__slice_u8 is a 1-dimensional buffer. // // A value with all fields NULL or zero is a valid, empty slice. typedef struct { @@ -58,6 +87,20 @@ typedef struct { size_t len; } wuffs_base__slice_u8; +// wuffs_base__table_u8 is a 2-dimensional buffer. +// +// A value with all fields NULL or zero is a valid, empty table. +typedef struct { + uint8_t* ptr; + size_t width; + size_t height; + size_t stride; +} wuffs_base__table_u8; + +// ---------------- I/O + +// TODO: rename buf1 to io_buffer, writer1 to io_writer, etc. + // wuffs_base__buf1 is a 1-dimensional buffer (a pointer and length), plus // additional indexes into that buffer, plus an opened / closed flag. // @@ -109,18 +152,286 @@ typedef struct { // ---------------- Images +// wuffs_base__pixel_format encodes the format of the bytes that constitute an +// image frame's pixel data. Its bits: +// - bit 31 is reserved. +// - bits 30 .. 28 encodes color (and channel order, in terms of memory). +// - bits 27 .. 26 are reserved. +// - bits 25 .. 24 encodes transparency. +// - bit 23 indicates big-endian/MSB-first (as opposed to little/LSB). +// - bit 22 indicates floating point (as opposed to integer). +// - bits 21 .. 20 are the number of planes, minus 1. Zero means packed. +// - bits 19 .. 16 encodes the number of bits (depth) in an index value. +// Zero means direct, not palette-indexed. +// - bits 15 .. 12 encodes the number of bits (depth) in the 3rd channel. +// - bits 11 .. 8 encodes the number of bits (depth) in the 2nd channel. +// - bits 7 .. 4 encodes the number of bits (depth) in the 1st channel. +// - bits 3 .. 0 encodes the number of bits (depth) in the 0th channel. +// +// The bit fields of a wuffs_base__pixel_format are not independent. For +// example, the number of planes should not be greater than the number of +// channels. Similarly, bits 15..4 are unused (and should be zero) if bits +// 31..24 (color and transparency) together imply only 1 channel (gray, no +// alpha) and floating point samples should mean a bit depth of 16, 32 or 64. +// +// Formats hold between 1 and 4 channels. For example: Y (1 channel: gray), YA +// (2 channels: gray and alpha), BGR (3 channels: blue, green, red) or CMYK (4 +// channels: cyan, magenta, yellow, black). +// +// For direct formats with N > 1 channels, those channels can be laid out in +// either 1 (packed) or N (planar) planes. For example, RGBA data is usually +// packed, but YUV data is usually planar, due to chroma subsampling (for +// details, see the wuffs_base__pixel_subsampling type). For indexed formats, +// the palette (always 256 × 4 bytes) holds up to 4 packed bytes of color data +// per index value, and there is only 1 plane (for the index). The distance +// between successive palette elements is always 4 bytes. +// +// The color field is encoded in 3 bits: +// - 0 means A (Alpha). +// - 1 means Y or YA (Gray, Alpha). +// - 2 means BGR, BGRX or BGRA (Blue, Green, Red, X-padding or Alpha). +// - 3 means RGB, RGBX or RGBA (Red, Green, Blue, X-padding or Alpha). +// - 4 means YUV or YUVA (Luma, Chroma-blue, Chroma-red, Alpha). +// - 5 means CMY or CMYK (Cyan, Magenta, Yellow, Black). +// - all other values are reserved. +// +// In Wuffs, channels are given in memory order, regardless of endianness, +// since the C type for the pixel data is an array of bytes, not an array of +// uint32_t. For example, packed BGRA with 8 bits per channel means that the +// bytes in memory are always Blue, Green, Red then Alpha. On big-endian +// systems, that is the uint32_t 0xBBGGRRAA. On little-endian, 0xAARRGGBB. +// +// When the color field (3 bits) encodes multiple options, the transparency +// field (2 bits) distinguishes them: +// - 0 means fully opaque, no extra channels +// - 1 means fully opaque, one extra channel (X or K, padding or black). +// - 2 means one extra alpha channel, other channels are non-premultiplied. +// - 3 means one extra alpha channel, other channels are premultiplied. +// +// The zero wuffs_base__pixel_format value is an invalid pixel format, as it is +// invalid to combine the zero color (alpha only) with the zero transparency. +// +// Bit depth is encoded in 4 bits: +// - 0 means the channel or index is unused. +// - x means a bit depth of x, for x in the range 1..8. +// - 9 means a bit depth of 10. +// - 10 means a bit depth of 12. +// - 11 means a bit depth of 16. +// - 12 means a bit depth of 24. +// - 13 means a bit depth of 32. +// - 14 means a bit depth of 48. +// - 15 means a bit depth of 64. +// +// For example, wuffs_base__pixel_format 0x3280BBBB is a natural format for +// decoding a PNG image - network byte order (also known as big-endian), +// packed, non-premultiplied alpha - that happens to be 16-bit-depth truecolor +// with alpha (RGBA). In memory order: +// +// ptr+0 ptr+1 ptr+2 ptr+3 ptr+4 ptr+5 ptr+6 ptr+7 +// Rhi Rlo Ghi Glo Bhi Blo Ahi Alo +// +// For example, the value wuffs_base__pixel_format 0x20000565 means BGR with no +// alpha or padding, 5/6/5 bits for blue/green/red, packed 2 bytes per pixel, +// laid out LSB-first in memory order: +// +// ptr+0........... ptr+1........... +// MSB LSB MSB LSB +// G₂G₁G₀B₄B₃B₂B₁B₀ R₄R₃R₂R₁R₀G₅G₄G₃ +// +// On little-endian systems (but not big-endian), this Wuffs pixel format value +// (0x20000565) corresponds to the Cairo library's CAIRO_FORMAT_RGB16_565, the +// SDL2 (Simple DirectMedia Layer 2) library's SDL_PIXELFORMAT_RGB565 and the +// Skia library's kRGB_565_SkColorType. Note BGR in Wuffs versus RGB in the +// other libraries. +// +// Regardless of endianness, this Wuffs pixel format value (0x20000565) +// corresponds to the V4L2 (Video For Linux 2) library's V4L2_PIX_FMT_RGB565 +// and the Wayland-DRM library's WL_DRM_FORMAT_RGB565. +// +// Different software libraries name their pixel formats (and especially their +// channel order) either according to memory layout or as bits of a native +// integer type like uint32_t. The two conventions differ because of a system's +// endianness. As mentioned earlier, Wuffs pixel formats are always in memory +// order. More detail of other software libraries' naming conventions is in the +// Pixel Format Guide at https://afrantzis.github.io/pixel-format-guide/ +// +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_base__pixel_format__num_planes instead. +typedef uint32_t wuffs_base__pixel_format; + +// Common 8-bit-depth pixel formats. This list is not exhaustive; not all valid +// wuffs_base__pixel_format values are present. + +#define WUFFS_BASE__PIXEL_FORMAT__INVALID ((wuffs_base__pixel_format)0x00000000) + +#define WUFFS_BASE__PIXEL_FORMAT__A ((wuffs_base__pixel_format)0x02000008) + +#define WUFFS_BASE__PIXEL_FORMAT__Y ((wuffs_base__pixel_format)0x10000008) +#define WUFFS_BASE__PIXEL_FORMAT__YA_NONPREMUL \ + ((wuffs_base__pixel_format)0x12000008) +#define WUFFS_BASE__PIXEL_FORMAT__YA_PREMUL \ + ((wuffs_base__pixel_format)0x13000008) + +#define WUFFS_BASE__PIXEL_FORMAT__BGR ((wuffs_base__pixel_format)0x20000888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRX ((wuffs_base__pixel_format)0x21008888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRX_INDEXED \ + ((wuffs_base__pixel_format)0x21088888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_NONPREMUL \ + ((wuffs_base__pixel_format)0x22008888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_NONPREMUL_INDEXED \ + ((wuffs_base__pixel_format)0x22088888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_PREMUL \ + ((wuffs_base__pixel_format)0x23008888) + +#define WUFFS_BASE__PIXEL_FORMAT__RGB ((wuffs_base__pixel_format)0x30000888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBX ((wuffs_base__pixel_format)0x31008888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBX_INDEXED \ + ((wuffs_base__pixel_format)0x31088888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_NONPREMUL \ + ((wuffs_base__pixel_format)0x32008888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_NONPREMUL_INDEXED \ + ((wuffs_base__pixel_format)0x32088888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_PREMUL \ + ((wuffs_base__pixel_format)0x33008888) + +#define WUFFS_BASE__PIXEL_FORMAT__YUV ((wuffs_base__pixel_format)0x40200888) +#define WUFFS_BASE__PIXEL_FORMAT__YUVK ((wuffs_base__pixel_format)0x41308888) +#define WUFFS_BASE__PIXEL_FORMAT__YUVA_NONPREMUL \ + ((wuffs_base__pixel_format)0x42308888) + +#define WUFFS_BASE__PIXEL_FORMAT__CMY ((wuffs_base__pixel_format)0x50200888) +#define WUFFS_BASE__PIXEL_FORMAT__CMYK ((wuffs_base__pixel_format)0x51308888) + +static inline bool wuffs_base__pixel_format__is_valid( + wuffs_base__pixel_format f) { + return f != 0; +} + +static inline bool wuffs_base__pixel_format__is_indexed( + wuffs_base__pixel_format f) { + return ((f >> 16) & 0x0F) != 0; +} + +#define WUFFS_BASE__PIXEL_FORMAT__NUM_PLANES_MAX 4 + +static inline uint32_t wuffs_base__pixel_format__num_planes( + wuffs_base__pixel_format f) { + return f ? (((f >> 20) & 0x03) + 1) : 0; +} + +typedef struct { + wuffs_base__table_u8 planes[WUFFS_BASE__PIXEL_FORMAT__NUM_PLANES_MAX]; +} wuffs_base__pixel_buffer; + +// wuffs_base__pixel_subsampling encodes the mapping of pixel space coordinates +// (x, y) to pixel buffer indices (i, j). That mapping can differ for each +// plane p. For a depth of 8 bits (1 byte), the p'th plane's sample starts at +// (planes[p].ptr + (j * planes[p].stride) + i). +// +// For packed pixel formats, the mapping is trivial: i = x and j = y. For +// planar pixel formats, the mapping can differ due to chroma subsampling. For +// example, consider a three plane YUV pixel format with 4:2:2 subsampling. For +// the luma (Y) channel, there is one sample for every pixel, but for the +// chroma (U, V) channels, there is one sample for every two pixels: pairs of +// horizontally adjacent pixels form one macropixel, i = x / 2 and j == y. In +// general, for a given p: +// - i = (x + bias_x) >> shift_x. +// - j = (y + bias_y) >> shift_y. +// where biases and shifts are in the range 0..3 and 0..2 respectively. +// +// In general, the biases will be zero after decoding an image. However, making +// a sub-image may change the bias, since the (x, y) coordinates are relative +// to the sub-image's top-left origin, but the backing pixel buffers were +// created relative to the original image's origin. +// +// For each plane p, each of those four numbers (biases and shifts) are encoded +// in two bits, which combine to form an 8 bit unsigned integer: +// +// e_p = (bias_x << 6) | (shift_x << 4) | (bias_y << 2) | (shift_y << 0) +// +// Those e_p values (e_0 for the first plane, e_1 for the second plane, etc) +// combine to form a wuffs_base__pixel_subsampling value: +// +// pixsub = (e_3 << 24) | (e_2 << 16) | (e_1 << 8) | (e_0 << 0) +// +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_base__pixel_subsampling__bias_x instead. +typedef uint32_t wuffs_base__pixel_subsampling; + +#define WUFFS_BASE__PIXEL_SUBSAMPLING__NONE ((wuffs_base__pixel_subsampling)0); + +#define WUFFS_BASE__PIXEL_SUBSAMPLING__444 \ + ((wuffs_base__pixel_subsampling)0x000000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__440 \ + ((wuffs_base__pixel_subsampling)0x010100); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__422 \ + ((wuffs_base__pixel_subsampling)0x101000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__420 \ + ((wuffs_base__pixel_subsampling)0x111100); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__411 \ + ((wuffs_base__pixel_subsampling)0x202000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__410 \ + ((wuffs_base__pixel_subsampling)0x212100); + +static inline uint32_t wuffs_base__pixel_subsampling__bias_x( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 6; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__shift_x( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 4; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__bias_y( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 2; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__shift_y( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 0; + return (s >> shift) & 0x03; +} + typedef struct { // Do not access the private_impl's fields directly. There is no API/ABI // compatibility or safety guarantee if you do so. struct { - uint32_t flags; - uint32_t w; - uint32_t h; - // TODO: color model, including both packed RGBA and planar, - // chroma-subsampled YCbCr. + wuffs_base__pixel_format pixfmt; + wuffs_base__pixel_subsampling pixsub; + uint32_t width; + uint32_t height; + uint32_t num_loops; } private_impl; } wuffs_base__image_config; +static inline void wuffs_base__image_config__initialize( + wuffs_base__image_config* c, + wuffs_base__pixel_format pixfmt, + wuffs_base__pixel_subsampling pixsub, + uint32_t width, + uint32_t height, + uint32_t num_loops) { + if (!c) { + return; + } + // TODO: move the check from wuffs_base__image_config__is_valid here. Should + // this function return bool? An error type? + c->private_impl.pixfmt = pixfmt; + c->private_impl.pixsub = pixsub; + c->private_impl.width = width; + c->private_impl.height = height; + c->private_impl.num_loops = num_loops; +} + static inline void wuffs_base__image_config__invalidate( wuffs_base__image_config* c) { if (c) { @@ -128,24 +439,42 @@ static inline void wuffs_base__image_config__invalidate( } } -static inline bool wuffs_base__image_config__valid( +static inline bool wuffs_base__image_config__is_valid( wuffs_base__image_config* c) { - if (!c || !(c->private_impl.flags & 1)) { + if (!c || !c->private_impl.pixfmt) { return false; } - uint64_t wh = ((uint64_t)c->private_impl.w) * ((uint64_t)c->private_impl.h); + uint64_t wh = + ((uint64_t)c->private_impl.width) * ((uint64_t)c->private_impl.height); // TODO: handle things other than 1 byte per pixel. + // + // TODO: move the check to wuffs_base__image_config__initialize. return wh <= ((uint64_t)SIZE_MAX); } +static inline wuffs_base__pixel_format wuffs_base__image_config__pixel_format( + wuffs_base__image_config* c) { + return c ? c->private_impl.pixfmt : 0; +} + +static inline wuffs_base__pixel_subsampling +wuffs_base__image_config__pixel_subsampling(wuffs_base__image_config* c) { + return wuffs_base__image_config__is_valid(c) ? c->private_impl.pixsub : 0; +} + static inline uint32_t wuffs_base__image_config__width( wuffs_base__image_config* c) { - return wuffs_base__image_config__valid(c) ? c->private_impl.w : 0; + return wuffs_base__image_config__is_valid(c) ? c->private_impl.width : 0; } static inline uint32_t wuffs_base__image_config__height( wuffs_base__image_config* c) { - return wuffs_base__image_config__valid(c) ? c->private_impl.h : 0; + return wuffs_base__image_config__is_valid(c) ? c->private_impl.height : 0; +} + +static inline uint32_t wuffs_base__image_config__num_loops( + wuffs_base__image_config* c) { + return wuffs_base__image_config__is_valid(c) ? c->private_impl.num_loops : 0; } // TODO: this is the right API for planar (not packed) pixbufs? Should it allow @@ -153,26 +482,96 @@ static inline uint32_t wuffs_base__image_config__height( // example, decoding a JPEG image straight to RGBA instead of to YCbCr? static inline size_t wuffs_base__image_config__pixbuf_size( wuffs_base__image_config* c) { - if (wuffs_base__image_config__valid(c)) { - uint64_t wh = ((uint64_t)c->private_impl.w) * ((uint64_t)c->private_impl.h); + if (wuffs_base__image_config__is_valid(c)) { + uint64_t wh = + ((uint64_t)c->private_impl.width) * ((uint64_t)c->private_impl.height); // TODO: handle things other than 1 byte per pixel. return (size_t)wh; } return 0; } -static inline void wuffs_base__image_config__initialize( - wuffs_base__image_config* c, - uint32_t width, - uint32_t height, - uint32_t TODO_color_model) { - if (!c) { +typedef struct { + // Do not access the private_impl's fields directly. There is no API/ABI + // compatibility or safety guarantee if you do so. + struct { + wuffs_base__image_config config; + uint32_t loop_count; // 0-based count of the current loop. + wuffs_base__pixel_buffer pixbuf; + // TODO: color spaces. + wuffs_base__rectangle dirty_rect; + wuffs_base__flicks duration; + uint8_t palette[1024]; + } private_impl; +} wuffs_base__image_buffer; + +static inline void wuffs_base__image_buffer__initialize( + wuffs_base__image_buffer* f, + wuffs_base__image_config config, + wuffs_base__pixel_buffer pixbuf) { + if (!f) { return; } - c->private_impl.flags = 1; - c->private_impl.w = width; - c->private_impl.h = height; - // TODO: color model. + *f = ((wuffs_base__image_buffer){}); + f->private_impl.config = config; + f->private_impl.pixbuf = pixbuf; +} + +static inline void wuffs_base__image_buffer__update( + wuffs_base__image_buffer* f, + wuffs_base__rectangle dirty_rect, + wuffs_base__flicks duration, + uint8_t* palette_ptr, + size_t palette_len) { + if (!f) { + return; + } + f->private_impl.dirty_rect = dirty_rect; + f->private_impl.duration = duration; + if (palette_ptr) { + memmove(f->private_impl.palette, palette_ptr, + palette_len <= 1024 ? palette_len : 1024); + } +} + +// wuffs_base__image_buffer__loop returns whether the image decoder should loop +// back to the beginning of the animation, assuming that we've reached the end +// of the encoded stream. If so, it increments f's count of the animation loops +// played so far. +static inline bool wuffs_base__image_buffer__loop(wuffs_base__image_buffer* f) { + if (!f) { + return false; + } + uint32_t n = f->private_impl.config.private_impl.num_loops; + if (n == 0) { + return true; + } + if (f->private_impl.loop_count < n - 1) { + f->private_impl.loop_count++; + return true; + } + return false; +} + +// wuffs_base__image_buffer__dirty_rect returns an upper bound for what part of +// this frame's pixels differs from the previous frame. +static inline wuffs_base__rectangle wuffs_base__image_buffer__dirty_rect( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.dirty_rect : ((wuffs_base__rectangle){0}); +} + +// wuffs_base__image_buffer__duration returns the amount of time to display +// this frame. Zero means to display forever - a still (non-animated) image. +static inline wuffs_base__flicks wuffs_base__image_buffer__duration( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.duration : 0; +} + +// wuffs_base__image_buffer__palette returns the palette that the pixel data +// can index. The backing array is inside f and has length 1024. +static inline uint8_t* wuffs_base__image_buffer__palette( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.palette : NULL; } #endif // WUFFS_BASE_HEADER_H @@ -185,14 +584,14 @@ extern "C" { // ---------------- Status Codes -// Status codes are int32_t values: -// - the sign bit indicates a non-recoverable status code: an error -// - bits 10-30 hold the packageid: a namespace -// - bits 8-9 are reserved -// - bits 0-7 are a package-namespaced numeric code +// Status codes are int32_t values. Its bits: +// - bit 31 (the sign bit) indicates unrecoverable-ness: an error. +// - bits 30 .. 10 are the packageid: a namespace. +// - bits 9 .. 8 are reserved. +// - bits 7 .. 0 are a package-namespaced numeric code. // -// Do not manipulate these bits directly. Use the API functions such as -// wuffs_crc32__status__is_error instead. +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_crc32__status__is_error instead. typedef int32_t wuffs_crc32__status; #define wuffs_crc32__packageid 810620 // 0x000C5E7C @@ -290,14 +689,14 @@ typedef struct { // but it should catch 99.99% of cases. // // Its (non-zero) value is arbitrary, based on md5sum("wuffs"). -#define WUFFS_BASE__MAGIC (0x3CCB6C71U) +#define WUFFS_BASE__MAGIC 0x3CCB6C71U // WUFFS_BASE__ALREADY_ZEROED is passed from a container struct's initializer // to a containee struct's initializer when the container has already zeroed // the containee's memory. // // Its (non-zero) value is arbitrary, based on md5sum("zeroed"). -#define WUFFS_BASE__ALREADY_ZEROED (0x68602EF1U) +#define WUFFS_BASE__ALREADY_ZEROED 0x68602EF1U // Denote intentional fallthroughs for -Wimplicit-fallthrough. // diff --git a/gen/c/std/deflate.c b/gen/c/std/deflate.c index a772698c..61af6a27 100644 --- a/gen/c/std/deflate.c +++ b/gen/c/std/deflate.c @@ -46,11 +46,40 @@ // For now, the API and ABI are simply unstable and can change at any time. // // TODO: don't hard code this in base-header.h. -#define WUFFS_VERSION (0x00001) +#define WUFFS_VERSION 0x00001 -// ---------------- I/O +// ---------------- Fundamentals -// wuffs_base__slice_u8 is a 1-dimensional buffer (a pointer and length). +// Flicks are a unit of time. One flick (frame-tick) is 1 / 705_600_000 of a +// second. See https://github.com/OculusVR/Flicks +typedef uint64_t wuffs_base__flicks; + +#define WUFFS_BASE__FLICKS_PER_SECOND 705600000ULL + +// wuffs_base__rectangle is a rectangle on the integer grid. It contains all +// points (x, y) such that ((min_x <= x) && (x < max_x)) and likewise for y. It +// is therefore empty if min_x >= max_x. There are multiple representations of +// an empty rectangle. +// +// A value with all fields zero is a valid, empty rectangle. +// +// The X and Y axes increase right and down. +typedef struct { + uint32_t min_x; + uint32_t min_y; + uint32_t max_x; + uint32_t max_y; +} wuffs_base__rectangle; + +static inline uint32_t wuffs_base__rectangle__width(wuffs_base__rectangle r) { + return r.max_x > r.min_x ? r.max_x - r.min_x : 0; +} + +static inline uint32_t wuffs_base__rectangle__height(wuffs_base__rectangle r) { + return r.max_y > r.min_y ? r.max_y - r.min_y : 0; +} + +// wuffs_base__slice_u8 is a 1-dimensional buffer. // // A value with all fields NULL or zero is a valid, empty slice. typedef struct { @@ -58,6 +87,20 @@ typedef struct { size_t len; } wuffs_base__slice_u8; +// wuffs_base__table_u8 is a 2-dimensional buffer. +// +// A value with all fields NULL or zero is a valid, empty table. +typedef struct { + uint8_t* ptr; + size_t width; + size_t height; + size_t stride; +} wuffs_base__table_u8; + +// ---------------- I/O + +// TODO: rename buf1 to io_buffer, writer1 to io_writer, etc. + // wuffs_base__buf1 is a 1-dimensional buffer (a pointer and length), plus // additional indexes into that buffer, plus an opened / closed flag. // @@ -109,18 +152,286 @@ typedef struct { // ---------------- Images +// wuffs_base__pixel_format encodes the format of the bytes that constitute an +// image frame's pixel data. Its bits: +// - bit 31 is reserved. +// - bits 30 .. 28 encodes color (and channel order, in terms of memory). +// - bits 27 .. 26 are reserved. +// - bits 25 .. 24 encodes transparency. +// - bit 23 indicates big-endian/MSB-first (as opposed to little/LSB). +// - bit 22 indicates floating point (as opposed to integer). +// - bits 21 .. 20 are the number of planes, minus 1. Zero means packed. +// - bits 19 .. 16 encodes the number of bits (depth) in an index value. +// Zero means direct, not palette-indexed. +// - bits 15 .. 12 encodes the number of bits (depth) in the 3rd channel. +// - bits 11 .. 8 encodes the number of bits (depth) in the 2nd channel. +// - bits 7 .. 4 encodes the number of bits (depth) in the 1st channel. +// - bits 3 .. 0 encodes the number of bits (depth) in the 0th channel. +// +// The bit fields of a wuffs_base__pixel_format are not independent. For +// example, the number of planes should not be greater than the number of +// channels. Similarly, bits 15..4 are unused (and should be zero) if bits +// 31..24 (color and transparency) together imply only 1 channel (gray, no +// alpha) and floating point samples should mean a bit depth of 16, 32 or 64. +// +// Formats hold between 1 and 4 channels. For example: Y (1 channel: gray), YA +// (2 channels: gray and alpha), BGR (3 channels: blue, green, red) or CMYK (4 +// channels: cyan, magenta, yellow, black). +// +// For direct formats with N > 1 channels, those channels can be laid out in +// either 1 (packed) or N (planar) planes. For example, RGBA data is usually +// packed, but YUV data is usually planar, due to chroma subsampling (for +// details, see the wuffs_base__pixel_subsampling type). For indexed formats, +// the palette (always 256 × 4 bytes) holds up to 4 packed bytes of color data +// per index value, and there is only 1 plane (for the index). The distance +// between successive palette elements is always 4 bytes. +// +// The color field is encoded in 3 bits: +// - 0 means A (Alpha). +// - 1 means Y or YA (Gray, Alpha). +// - 2 means BGR, BGRX or BGRA (Blue, Green, Red, X-padding or Alpha). +// - 3 means RGB, RGBX or RGBA (Red, Green, Blue, X-padding or Alpha). +// - 4 means YUV or YUVA (Luma, Chroma-blue, Chroma-red, Alpha). +// - 5 means CMY or CMYK (Cyan, Magenta, Yellow, Black). +// - all other values are reserved. +// +// In Wuffs, channels are given in memory order, regardless of endianness, +// since the C type for the pixel data is an array of bytes, not an array of +// uint32_t. For example, packed BGRA with 8 bits per channel means that the +// bytes in memory are always Blue, Green, Red then Alpha. On big-endian +// systems, that is the uint32_t 0xBBGGRRAA. On little-endian, 0xAARRGGBB. +// +// When the color field (3 bits) encodes multiple options, the transparency +// field (2 bits) distinguishes them: +// - 0 means fully opaque, no extra channels +// - 1 means fully opaque, one extra channel (X or K, padding or black). +// - 2 means one extra alpha channel, other channels are non-premultiplied. +// - 3 means one extra alpha channel, other channels are premultiplied. +// +// The zero wuffs_base__pixel_format value is an invalid pixel format, as it is +// invalid to combine the zero color (alpha only) with the zero transparency. +// +// Bit depth is encoded in 4 bits: +// - 0 means the channel or index is unused. +// - x means a bit depth of x, for x in the range 1..8. +// - 9 means a bit depth of 10. +// - 10 means a bit depth of 12. +// - 11 means a bit depth of 16. +// - 12 means a bit depth of 24. +// - 13 means a bit depth of 32. +// - 14 means a bit depth of 48. +// - 15 means a bit depth of 64. +// +// For example, wuffs_base__pixel_format 0x3280BBBB is a natural format for +// decoding a PNG image - network byte order (also known as big-endian), +// packed, non-premultiplied alpha - that happens to be 16-bit-depth truecolor +// with alpha (RGBA). In memory order: +// +// ptr+0 ptr+1 ptr+2 ptr+3 ptr+4 ptr+5 ptr+6 ptr+7 +// Rhi Rlo Ghi Glo Bhi Blo Ahi Alo +// +// For example, the value wuffs_base__pixel_format 0x20000565 means BGR with no +// alpha or padding, 5/6/5 bits for blue/green/red, packed 2 bytes per pixel, +// laid out LSB-first in memory order: +// +// ptr+0........... ptr+1........... +// MSB LSB MSB LSB +// G₂G₁G₀B₄B₃B₂B₁B₀ R₄R₃R₂R₁R₀G₅G₄G₃ +// +// On little-endian systems (but not big-endian), this Wuffs pixel format value +// (0x20000565) corresponds to the Cairo library's CAIRO_FORMAT_RGB16_565, the +// SDL2 (Simple DirectMedia Layer 2) library's SDL_PIXELFORMAT_RGB565 and the +// Skia library's kRGB_565_SkColorType. Note BGR in Wuffs versus RGB in the +// other libraries. +// +// Regardless of endianness, this Wuffs pixel format value (0x20000565) +// corresponds to the V4L2 (Video For Linux 2) library's V4L2_PIX_FMT_RGB565 +// and the Wayland-DRM library's WL_DRM_FORMAT_RGB565. +// +// Different software libraries name their pixel formats (and especially their +// channel order) either according to memory layout or as bits of a native +// integer type like uint32_t. The two conventions differ because of a system's +// endianness. As mentioned earlier, Wuffs pixel formats are always in memory +// order. More detail of other software libraries' naming conventions is in the +// Pixel Format Guide at https://afrantzis.github.io/pixel-format-guide/ +// +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_base__pixel_format__num_planes instead. +typedef uint32_t wuffs_base__pixel_format; + +// Common 8-bit-depth pixel formats. This list is not exhaustive; not all valid +// wuffs_base__pixel_format values are present. + +#define WUFFS_BASE__PIXEL_FORMAT__INVALID ((wuffs_base__pixel_format)0x00000000) + +#define WUFFS_BASE__PIXEL_FORMAT__A ((wuffs_base__pixel_format)0x02000008) + +#define WUFFS_BASE__PIXEL_FORMAT__Y ((wuffs_base__pixel_format)0x10000008) +#define WUFFS_BASE__PIXEL_FORMAT__YA_NONPREMUL \ + ((wuffs_base__pixel_format)0x12000008) +#define WUFFS_BASE__PIXEL_FORMAT__YA_PREMUL \ + ((wuffs_base__pixel_format)0x13000008) + +#define WUFFS_BASE__PIXEL_FORMAT__BGR ((wuffs_base__pixel_format)0x20000888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRX ((wuffs_base__pixel_format)0x21008888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRX_INDEXED \ + ((wuffs_base__pixel_format)0x21088888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_NONPREMUL \ + ((wuffs_base__pixel_format)0x22008888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_NONPREMUL_INDEXED \ + ((wuffs_base__pixel_format)0x22088888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_PREMUL \ + ((wuffs_base__pixel_format)0x23008888) + +#define WUFFS_BASE__PIXEL_FORMAT__RGB ((wuffs_base__pixel_format)0x30000888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBX ((wuffs_base__pixel_format)0x31008888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBX_INDEXED \ + ((wuffs_base__pixel_format)0x31088888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_NONPREMUL \ + ((wuffs_base__pixel_format)0x32008888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_NONPREMUL_INDEXED \ + ((wuffs_base__pixel_format)0x32088888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_PREMUL \ + ((wuffs_base__pixel_format)0x33008888) + +#define WUFFS_BASE__PIXEL_FORMAT__YUV ((wuffs_base__pixel_format)0x40200888) +#define WUFFS_BASE__PIXEL_FORMAT__YUVK ((wuffs_base__pixel_format)0x41308888) +#define WUFFS_BASE__PIXEL_FORMAT__YUVA_NONPREMUL \ + ((wuffs_base__pixel_format)0x42308888) + +#define WUFFS_BASE__PIXEL_FORMAT__CMY ((wuffs_base__pixel_format)0x50200888) +#define WUFFS_BASE__PIXEL_FORMAT__CMYK ((wuffs_base__pixel_format)0x51308888) + +static inline bool wuffs_base__pixel_format__is_valid( + wuffs_base__pixel_format f) { + return f != 0; +} + +static inline bool wuffs_base__pixel_format__is_indexed( + wuffs_base__pixel_format f) { + return ((f >> 16) & 0x0F) != 0; +} + +#define WUFFS_BASE__PIXEL_FORMAT__NUM_PLANES_MAX 4 + +static inline uint32_t wuffs_base__pixel_format__num_planes( + wuffs_base__pixel_format f) { + return f ? (((f >> 20) & 0x03) + 1) : 0; +} + +typedef struct { + wuffs_base__table_u8 planes[WUFFS_BASE__PIXEL_FORMAT__NUM_PLANES_MAX]; +} wuffs_base__pixel_buffer; + +// wuffs_base__pixel_subsampling encodes the mapping of pixel space coordinates +// (x, y) to pixel buffer indices (i, j). That mapping can differ for each +// plane p. For a depth of 8 bits (1 byte), the p'th plane's sample starts at +// (planes[p].ptr + (j * planes[p].stride) + i). +// +// For packed pixel formats, the mapping is trivial: i = x and j = y. For +// planar pixel formats, the mapping can differ due to chroma subsampling. For +// example, consider a three plane YUV pixel format with 4:2:2 subsampling. For +// the luma (Y) channel, there is one sample for every pixel, but for the +// chroma (U, V) channels, there is one sample for every two pixels: pairs of +// horizontally adjacent pixels form one macropixel, i = x / 2 and j == y. In +// general, for a given p: +// - i = (x + bias_x) >> shift_x. +// - j = (y + bias_y) >> shift_y. +// where biases and shifts are in the range 0..3 and 0..2 respectively. +// +// In general, the biases will be zero after decoding an image. However, making +// a sub-image may change the bias, since the (x, y) coordinates are relative +// to the sub-image's top-left origin, but the backing pixel buffers were +// created relative to the original image's origin. +// +// For each plane p, each of those four numbers (biases and shifts) are encoded +// in two bits, which combine to form an 8 bit unsigned integer: +// +// e_p = (bias_x << 6) | (shift_x << 4) | (bias_y << 2) | (shift_y << 0) +// +// Those e_p values (e_0 for the first plane, e_1 for the second plane, etc) +// combine to form a wuffs_base__pixel_subsampling value: +// +// pixsub = (e_3 << 24) | (e_2 << 16) | (e_1 << 8) | (e_0 << 0) +// +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_base__pixel_subsampling__bias_x instead. +typedef uint32_t wuffs_base__pixel_subsampling; + +#define WUFFS_BASE__PIXEL_SUBSAMPLING__NONE ((wuffs_base__pixel_subsampling)0); + +#define WUFFS_BASE__PIXEL_SUBSAMPLING__444 \ + ((wuffs_base__pixel_subsampling)0x000000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__440 \ + ((wuffs_base__pixel_subsampling)0x010100); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__422 \ + ((wuffs_base__pixel_subsampling)0x101000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__420 \ + ((wuffs_base__pixel_subsampling)0x111100); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__411 \ + ((wuffs_base__pixel_subsampling)0x202000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__410 \ + ((wuffs_base__pixel_subsampling)0x212100); + +static inline uint32_t wuffs_base__pixel_subsampling__bias_x( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 6; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__shift_x( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 4; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__bias_y( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 2; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__shift_y( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 0; + return (s >> shift) & 0x03; +} + typedef struct { // Do not access the private_impl's fields directly. There is no API/ABI // compatibility or safety guarantee if you do so. struct { - uint32_t flags; - uint32_t w; - uint32_t h; - // TODO: color model, including both packed RGBA and planar, - // chroma-subsampled YCbCr. + wuffs_base__pixel_format pixfmt; + wuffs_base__pixel_subsampling pixsub; + uint32_t width; + uint32_t height; + uint32_t num_loops; } private_impl; } wuffs_base__image_config; +static inline void wuffs_base__image_config__initialize( + wuffs_base__image_config* c, + wuffs_base__pixel_format pixfmt, + wuffs_base__pixel_subsampling pixsub, + uint32_t width, + uint32_t height, + uint32_t num_loops) { + if (!c) { + return; + } + // TODO: move the check from wuffs_base__image_config__is_valid here. Should + // this function return bool? An error type? + c->private_impl.pixfmt = pixfmt; + c->private_impl.pixsub = pixsub; + c->private_impl.width = width; + c->private_impl.height = height; + c->private_impl.num_loops = num_loops; +} + static inline void wuffs_base__image_config__invalidate( wuffs_base__image_config* c) { if (c) { @@ -128,24 +439,42 @@ static inline void wuffs_base__image_config__invalidate( } } -static inline bool wuffs_base__image_config__valid( +static inline bool wuffs_base__image_config__is_valid( wuffs_base__image_config* c) { - if (!c || !(c->private_impl.flags & 1)) { + if (!c || !c->private_impl.pixfmt) { return false; } - uint64_t wh = ((uint64_t)c->private_impl.w) * ((uint64_t)c->private_impl.h); + uint64_t wh = + ((uint64_t)c->private_impl.width) * ((uint64_t)c->private_impl.height); // TODO: handle things other than 1 byte per pixel. + // + // TODO: move the check to wuffs_base__image_config__initialize. return wh <= ((uint64_t)SIZE_MAX); } +static inline wuffs_base__pixel_format wuffs_base__image_config__pixel_format( + wuffs_base__image_config* c) { + return c ? c->private_impl.pixfmt : 0; +} + +static inline wuffs_base__pixel_subsampling +wuffs_base__image_config__pixel_subsampling(wuffs_base__image_config* c) { + return wuffs_base__image_config__is_valid(c) ? c->private_impl.pixsub : 0; +} + static inline uint32_t wuffs_base__image_config__width( wuffs_base__image_config* c) { - return wuffs_base__image_config__valid(c) ? c->private_impl.w : 0; + return wuffs_base__image_config__is_valid(c) ? c->private_impl.width : 0; } static inline uint32_t wuffs_base__image_config__height( wuffs_base__image_config* c) { - return wuffs_base__image_config__valid(c) ? c->private_impl.h : 0; + return wuffs_base__image_config__is_valid(c) ? c->private_impl.height : 0; +} + +static inline uint32_t wuffs_base__image_config__num_loops( + wuffs_base__image_config* c) { + return wuffs_base__image_config__is_valid(c) ? c->private_impl.num_loops : 0; } // TODO: this is the right API for planar (not packed) pixbufs? Should it allow @@ -153,26 +482,96 @@ static inline uint32_t wuffs_base__image_config__height( // example, decoding a JPEG image straight to RGBA instead of to YCbCr? static inline size_t wuffs_base__image_config__pixbuf_size( wuffs_base__image_config* c) { - if (wuffs_base__image_config__valid(c)) { - uint64_t wh = ((uint64_t)c->private_impl.w) * ((uint64_t)c->private_impl.h); + if (wuffs_base__image_config__is_valid(c)) { + uint64_t wh = + ((uint64_t)c->private_impl.width) * ((uint64_t)c->private_impl.height); // TODO: handle things other than 1 byte per pixel. return (size_t)wh; } return 0; } -static inline void wuffs_base__image_config__initialize( - wuffs_base__image_config* c, - uint32_t width, - uint32_t height, - uint32_t TODO_color_model) { - if (!c) { +typedef struct { + // Do not access the private_impl's fields directly. There is no API/ABI + // compatibility or safety guarantee if you do so. + struct { + wuffs_base__image_config config; + uint32_t loop_count; // 0-based count of the current loop. + wuffs_base__pixel_buffer pixbuf; + // TODO: color spaces. + wuffs_base__rectangle dirty_rect; + wuffs_base__flicks duration; + uint8_t palette[1024]; + } private_impl; +} wuffs_base__image_buffer; + +static inline void wuffs_base__image_buffer__initialize( + wuffs_base__image_buffer* f, + wuffs_base__image_config config, + wuffs_base__pixel_buffer pixbuf) { + if (!f) { return; } - c->private_impl.flags = 1; - c->private_impl.w = width; - c->private_impl.h = height; - // TODO: color model. + *f = ((wuffs_base__image_buffer){}); + f->private_impl.config = config; + f->private_impl.pixbuf = pixbuf; +} + +static inline void wuffs_base__image_buffer__update( + wuffs_base__image_buffer* f, + wuffs_base__rectangle dirty_rect, + wuffs_base__flicks duration, + uint8_t* palette_ptr, + size_t palette_len) { + if (!f) { + return; + } + f->private_impl.dirty_rect = dirty_rect; + f->private_impl.duration = duration; + if (palette_ptr) { + memmove(f->private_impl.palette, palette_ptr, + palette_len <= 1024 ? palette_len : 1024); + } +} + +// wuffs_base__image_buffer__loop returns whether the image decoder should loop +// back to the beginning of the animation, assuming that we've reached the end +// of the encoded stream. If so, it increments f's count of the animation loops +// played so far. +static inline bool wuffs_base__image_buffer__loop(wuffs_base__image_buffer* f) { + if (!f) { + return false; + } + uint32_t n = f->private_impl.config.private_impl.num_loops; + if (n == 0) { + return true; + } + if (f->private_impl.loop_count < n - 1) { + f->private_impl.loop_count++; + return true; + } + return false; +} + +// wuffs_base__image_buffer__dirty_rect returns an upper bound for what part of +// this frame's pixels differs from the previous frame. +static inline wuffs_base__rectangle wuffs_base__image_buffer__dirty_rect( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.dirty_rect : ((wuffs_base__rectangle){0}); +} + +// wuffs_base__image_buffer__duration returns the amount of time to display +// this frame. Zero means to display forever - a still (non-animated) image. +static inline wuffs_base__flicks wuffs_base__image_buffer__duration( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.duration : 0; +} + +// wuffs_base__image_buffer__palette returns the palette that the pixel data +// can index. The backing array is inside f and has length 1024. +static inline uint8_t* wuffs_base__image_buffer__palette( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.palette : NULL; } #endif // WUFFS_BASE_HEADER_H @@ -185,14 +584,14 @@ extern "C" { // ---------------- Status Codes -// Status codes are int32_t values: -// - the sign bit indicates a non-recoverable status code: an error -// - bits 10-30 hold the packageid: a namespace -// - bits 8-9 are reserved -// - bits 0-7 are a package-namespaced numeric code +// Status codes are int32_t values. Its bits: +// - bit 31 (the sign bit) indicates unrecoverable-ness: an error. +// - bits 30 .. 10 are the packageid: a namespace. +// - bits 9 .. 8 are reserved. +// - bits 7 .. 0 are a package-namespaced numeric code. // -// Do not manipulate these bits directly. Use the API functions such as -// wuffs_deflate__status__is_error instead. +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_deflate__status__is_error instead. typedef int32_t wuffs_deflate__status; #define wuffs_deflate__packageid 848533 // 0x000CF295 @@ -382,14 +781,14 @@ typedef struct { // but it should catch 99.99% of cases. // // Its (non-zero) value is arbitrary, based on md5sum("wuffs"). -#define WUFFS_BASE__MAGIC (0x3CCB6C71U) +#define WUFFS_BASE__MAGIC 0x3CCB6C71U // WUFFS_BASE__ALREADY_ZEROED is passed from a container struct's initializer // to a containee struct's initializer when the container has already zeroed // the containee's memory. // // Its (non-zero) value is arbitrary, based on md5sum("zeroed"). -#define WUFFS_BASE__ALREADY_ZEROED (0x68602EF1U) +#define WUFFS_BASE__ALREADY_ZEROED 0x68602EF1U // Denote intentional fallthroughs for -Wimplicit-fallthrough. // diff --git a/gen/c/std/gif.c b/gen/c/std/gif.c index 9c64c814..d9320835 100644 --- a/gen/c/std/gif.c +++ b/gen/c/std/gif.c @@ -46,11 +46,40 @@ // For now, the API and ABI are simply unstable and can change at any time. // // TODO: don't hard code this in base-header.h. -#define WUFFS_VERSION (0x00001) +#define WUFFS_VERSION 0x00001 -// ---------------- I/O +// ---------------- Fundamentals -// wuffs_base__slice_u8 is a 1-dimensional buffer (a pointer and length). +// Flicks are a unit of time. One flick (frame-tick) is 1 / 705_600_000 of a +// second. See https://github.com/OculusVR/Flicks +typedef uint64_t wuffs_base__flicks; + +#define WUFFS_BASE__FLICKS_PER_SECOND 705600000ULL + +// wuffs_base__rectangle is a rectangle on the integer grid. It contains all +// points (x, y) such that ((min_x <= x) && (x < max_x)) and likewise for y. It +// is therefore empty if min_x >= max_x. There are multiple representations of +// an empty rectangle. +// +// A value with all fields zero is a valid, empty rectangle. +// +// The X and Y axes increase right and down. +typedef struct { + uint32_t min_x; + uint32_t min_y; + uint32_t max_x; + uint32_t max_y; +} wuffs_base__rectangle; + +static inline uint32_t wuffs_base__rectangle__width(wuffs_base__rectangle r) { + return r.max_x > r.min_x ? r.max_x - r.min_x : 0; +} + +static inline uint32_t wuffs_base__rectangle__height(wuffs_base__rectangle r) { + return r.max_y > r.min_y ? r.max_y - r.min_y : 0; +} + +// wuffs_base__slice_u8 is a 1-dimensional buffer. // // A value with all fields NULL or zero is a valid, empty slice. typedef struct { @@ -58,6 +87,20 @@ typedef struct { size_t len; } wuffs_base__slice_u8; +// wuffs_base__table_u8 is a 2-dimensional buffer. +// +// A value with all fields NULL or zero is a valid, empty table. +typedef struct { + uint8_t* ptr; + size_t width; + size_t height; + size_t stride; +} wuffs_base__table_u8; + +// ---------------- I/O + +// TODO: rename buf1 to io_buffer, writer1 to io_writer, etc. + // wuffs_base__buf1 is a 1-dimensional buffer (a pointer and length), plus // additional indexes into that buffer, plus an opened / closed flag. // @@ -109,18 +152,286 @@ typedef struct { // ---------------- Images +// wuffs_base__pixel_format encodes the format of the bytes that constitute an +// image frame's pixel data. Its bits: +// - bit 31 is reserved. +// - bits 30 .. 28 encodes color (and channel order, in terms of memory). +// - bits 27 .. 26 are reserved. +// - bits 25 .. 24 encodes transparency. +// - bit 23 indicates big-endian/MSB-first (as opposed to little/LSB). +// - bit 22 indicates floating point (as opposed to integer). +// - bits 21 .. 20 are the number of planes, minus 1. Zero means packed. +// - bits 19 .. 16 encodes the number of bits (depth) in an index value. +// Zero means direct, not palette-indexed. +// - bits 15 .. 12 encodes the number of bits (depth) in the 3rd channel. +// - bits 11 .. 8 encodes the number of bits (depth) in the 2nd channel. +// - bits 7 .. 4 encodes the number of bits (depth) in the 1st channel. +// - bits 3 .. 0 encodes the number of bits (depth) in the 0th channel. +// +// The bit fields of a wuffs_base__pixel_format are not independent. For +// example, the number of planes should not be greater than the number of +// channels. Similarly, bits 15..4 are unused (and should be zero) if bits +// 31..24 (color and transparency) together imply only 1 channel (gray, no +// alpha) and floating point samples should mean a bit depth of 16, 32 or 64. +// +// Formats hold between 1 and 4 channels. For example: Y (1 channel: gray), YA +// (2 channels: gray and alpha), BGR (3 channels: blue, green, red) or CMYK (4 +// channels: cyan, magenta, yellow, black). +// +// For direct formats with N > 1 channels, those channels can be laid out in +// either 1 (packed) or N (planar) planes. For example, RGBA data is usually +// packed, but YUV data is usually planar, due to chroma subsampling (for +// details, see the wuffs_base__pixel_subsampling type). For indexed formats, +// the palette (always 256 × 4 bytes) holds up to 4 packed bytes of color data +// per index value, and there is only 1 plane (for the index). The distance +// between successive palette elements is always 4 bytes. +// +// The color field is encoded in 3 bits: +// - 0 means A (Alpha). +// - 1 means Y or YA (Gray, Alpha). +// - 2 means BGR, BGRX or BGRA (Blue, Green, Red, X-padding or Alpha). +// - 3 means RGB, RGBX or RGBA (Red, Green, Blue, X-padding or Alpha). +// - 4 means YUV or YUVA (Luma, Chroma-blue, Chroma-red, Alpha). +// - 5 means CMY or CMYK (Cyan, Magenta, Yellow, Black). +// - all other values are reserved. +// +// In Wuffs, channels are given in memory order, regardless of endianness, +// since the C type for the pixel data is an array of bytes, not an array of +// uint32_t. For example, packed BGRA with 8 bits per channel means that the +// bytes in memory are always Blue, Green, Red then Alpha. On big-endian +// systems, that is the uint32_t 0xBBGGRRAA. On little-endian, 0xAARRGGBB. +// +// When the color field (3 bits) encodes multiple options, the transparency +// field (2 bits) distinguishes them: +// - 0 means fully opaque, no extra channels +// - 1 means fully opaque, one extra channel (X or K, padding or black). +// - 2 means one extra alpha channel, other channels are non-premultiplied. +// - 3 means one extra alpha channel, other channels are premultiplied. +// +// The zero wuffs_base__pixel_format value is an invalid pixel format, as it is +// invalid to combine the zero color (alpha only) with the zero transparency. +// +// Bit depth is encoded in 4 bits: +// - 0 means the channel or index is unused. +// - x means a bit depth of x, for x in the range 1..8. +// - 9 means a bit depth of 10. +// - 10 means a bit depth of 12. +// - 11 means a bit depth of 16. +// - 12 means a bit depth of 24. +// - 13 means a bit depth of 32. +// - 14 means a bit depth of 48. +// - 15 means a bit depth of 64. +// +// For example, wuffs_base__pixel_format 0x3280BBBB is a natural format for +// decoding a PNG image - network byte order (also known as big-endian), +// packed, non-premultiplied alpha - that happens to be 16-bit-depth truecolor +// with alpha (RGBA). In memory order: +// +// ptr+0 ptr+1 ptr+2 ptr+3 ptr+4 ptr+5 ptr+6 ptr+7 +// Rhi Rlo Ghi Glo Bhi Blo Ahi Alo +// +// For example, the value wuffs_base__pixel_format 0x20000565 means BGR with no +// alpha or padding, 5/6/5 bits for blue/green/red, packed 2 bytes per pixel, +// laid out LSB-first in memory order: +// +// ptr+0........... ptr+1........... +// MSB LSB MSB LSB +// G₂G₁G₀B₄B₃B₂B₁B₀ R₄R₃R₂R₁R₀G₅G₄G₃ +// +// On little-endian systems (but not big-endian), this Wuffs pixel format value +// (0x20000565) corresponds to the Cairo library's CAIRO_FORMAT_RGB16_565, the +// SDL2 (Simple DirectMedia Layer 2) library's SDL_PIXELFORMAT_RGB565 and the +// Skia library's kRGB_565_SkColorType. Note BGR in Wuffs versus RGB in the +// other libraries. +// +// Regardless of endianness, this Wuffs pixel format value (0x20000565) +// corresponds to the V4L2 (Video For Linux 2) library's V4L2_PIX_FMT_RGB565 +// and the Wayland-DRM library's WL_DRM_FORMAT_RGB565. +// +// Different software libraries name their pixel formats (and especially their +// channel order) either according to memory layout or as bits of a native +// integer type like uint32_t. The two conventions differ because of a system's +// endianness. As mentioned earlier, Wuffs pixel formats are always in memory +// order. More detail of other software libraries' naming conventions is in the +// Pixel Format Guide at https://afrantzis.github.io/pixel-format-guide/ +// +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_base__pixel_format__num_planes instead. +typedef uint32_t wuffs_base__pixel_format; + +// Common 8-bit-depth pixel formats. This list is not exhaustive; not all valid +// wuffs_base__pixel_format values are present. + +#define WUFFS_BASE__PIXEL_FORMAT__INVALID ((wuffs_base__pixel_format)0x00000000) + +#define WUFFS_BASE__PIXEL_FORMAT__A ((wuffs_base__pixel_format)0x02000008) + +#define WUFFS_BASE__PIXEL_FORMAT__Y ((wuffs_base__pixel_format)0x10000008) +#define WUFFS_BASE__PIXEL_FORMAT__YA_NONPREMUL \ + ((wuffs_base__pixel_format)0x12000008) +#define WUFFS_BASE__PIXEL_FORMAT__YA_PREMUL \ + ((wuffs_base__pixel_format)0x13000008) + +#define WUFFS_BASE__PIXEL_FORMAT__BGR ((wuffs_base__pixel_format)0x20000888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRX ((wuffs_base__pixel_format)0x21008888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRX_INDEXED \ + ((wuffs_base__pixel_format)0x21088888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_NONPREMUL \ + ((wuffs_base__pixel_format)0x22008888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_NONPREMUL_INDEXED \ + ((wuffs_base__pixel_format)0x22088888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_PREMUL \ + ((wuffs_base__pixel_format)0x23008888) + +#define WUFFS_BASE__PIXEL_FORMAT__RGB ((wuffs_base__pixel_format)0x30000888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBX ((wuffs_base__pixel_format)0x31008888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBX_INDEXED \ + ((wuffs_base__pixel_format)0x31088888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_NONPREMUL \ + ((wuffs_base__pixel_format)0x32008888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_NONPREMUL_INDEXED \ + ((wuffs_base__pixel_format)0x32088888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_PREMUL \ + ((wuffs_base__pixel_format)0x33008888) + +#define WUFFS_BASE__PIXEL_FORMAT__YUV ((wuffs_base__pixel_format)0x40200888) +#define WUFFS_BASE__PIXEL_FORMAT__YUVK ((wuffs_base__pixel_format)0x41308888) +#define WUFFS_BASE__PIXEL_FORMAT__YUVA_NONPREMUL \ + ((wuffs_base__pixel_format)0x42308888) + +#define WUFFS_BASE__PIXEL_FORMAT__CMY ((wuffs_base__pixel_format)0x50200888) +#define WUFFS_BASE__PIXEL_FORMAT__CMYK ((wuffs_base__pixel_format)0x51308888) + +static inline bool wuffs_base__pixel_format__is_valid( + wuffs_base__pixel_format f) { + return f != 0; +} + +static inline bool wuffs_base__pixel_format__is_indexed( + wuffs_base__pixel_format f) { + return ((f >> 16) & 0x0F) != 0; +} + +#define WUFFS_BASE__PIXEL_FORMAT__NUM_PLANES_MAX 4 + +static inline uint32_t wuffs_base__pixel_format__num_planes( + wuffs_base__pixel_format f) { + return f ? (((f >> 20) & 0x03) + 1) : 0; +} + +typedef struct { + wuffs_base__table_u8 planes[WUFFS_BASE__PIXEL_FORMAT__NUM_PLANES_MAX]; +} wuffs_base__pixel_buffer; + +// wuffs_base__pixel_subsampling encodes the mapping of pixel space coordinates +// (x, y) to pixel buffer indices (i, j). That mapping can differ for each +// plane p. For a depth of 8 bits (1 byte), the p'th plane's sample starts at +// (planes[p].ptr + (j * planes[p].stride) + i). +// +// For packed pixel formats, the mapping is trivial: i = x and j = y. For +// planar pixel formats, the mapping can differ due to chroma subsampling. For +// example, consider a three plane YUV pixel format with 4:2:2 subsampling. For +// the luma (Y) channel, there is one sample for every pixel, but for the +// chroma (U, V) channels, there is one sample for every two pixels: pairs of +// horizontally adjacent pixels form one macropixel, i = x / 2 and j == y. In +// general, for a given p: +// - i = (x + bias_x) >> shift_x. +// - j = (y + bias_y) >> shift_y. +// where biases and shifts are in the range 0..3 and 0..2 respectively. +// +// In general, the biases will be zero after decoding an image. However, making +// a sub-image may change the bias, since the (x, y) coordinates are relative +// to the sub-image's top-left origin, but the backing pixel buffers were +// created relative to the original image's origin. +// +// For each plane p, each of those four numbers (biases and shifts) are encoded +// in two bits, which combine to form an 8 bit unsigned integer: +// +// e_p = (bias_x << 6) | (shift_x << 4) | (bias_y << 2) | (shift_y << 0) +// +// Those e_p values (e_0 for the first plane, e_1 for the second plane, etc) +// combine to form a wuffs_base__pixel_subsampling value: +// +// pixsub = (e_3 << 24) | (e_2 << 16) | (e_1 << 8) | (e_0 << 0) +// +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_base__pixel_subsampling__bias_x instead. +typedef uint32_t wuffs_base__pixel_subsampling; + +#define WUFFS_BASE__PIXEL_SUBSAMPLING__NONE ((wuffs_base__pixel_subsampling)0); + +#define WUFFS_BASE__PIXEL_SUBSAMPLING__444 \ + ((wuffs_base__pixel_subsampling)0x000000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__440 \ + ((wuffs_base__pixel_subsampling)0x010100); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__422 \ + ((wuffs_base__pixel_subsampling)0x101000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__420 \ + ((wuffs_base__pixel_subsampling)0x111100); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__411 \ + ((wuffs_base__pixel_subsampling)0x202000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__410 \ + ((wuffs_base__pixel_subsampling)0x212100); + +static inline uint32_t wuffs_base__pixel_subsampling__bias_x( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 6; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__shift_x( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 4; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__bias_y( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 2; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__shift_y( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 0; + return (s >> shift) & 0x03; +} + typedef struct { // Do not access the private_impl's fields directly. There is no API/ABI // compatibility or safety guarantee if you do so. struct { - uint32_t flags; - uint32_t w; - uint32_t h; - // TODO: color model, including both packed RGBA and planar, - // chroma-subsampled YCbCr. + wuffs_base__pixel_format pixfmt; + wuffs_base__pixel_subsampling pixsub; + uint32_t width; + uint32_t height; + uint32_t num_loops; } private_impl; } wuffs_base__image_config; +static inline void wuffs_base__image_config__initialize( + wuffs_base__image_config* c, + wuffs_base__pixel_format pixfmt, + wuffs_base__pixel_subsampling pixsub, + uint32_t width, + uint32_t height, + uint32_t num_loops) { + if (!c) { + return; + } + // TODO: move the check from wuffs_base__image_config__is_valid here. Should + // this function return bool? An error type? + c->private_impl.pixfmt = pixfmt; + c->private_impl.pixsub = pixsub; + c->private_impl.width = width; + c->private_impl.height = height; + c->private_impl.num_loops = num_loops; +} + static inline void wuffs_base__image_config__invalidate( wuffs_base__image_config* c) { if (c) { @@ -128,24 +439,42 @@ static inline void wuffs_base__image_config__invalidate( } } -static inline bool wuffs_base__image_config__valid( +static inline bool wuffs_base__image_config__is_valid( wuffs_base__image_config* c) { - if (!c || !(c->private_impl.flags & 1)) { + if (!c || !c->private_impl.pixfmt) { return false; } - uint64_t wh = ((uint64_t)c->private_impl.w) * ((uint64_t)c->private_impl.h); + uint64_t wh = + ((uint64_t)c->private_impl.width) * ((uint64_t)c->private_impl.height); // TODO: handle things other than 1 byte per pixel. + // + // TODO: move the check to wuffs_base__image_config__initialize. return wh <= ((uint64_t)SIZE_MAX); } +static inline wuffs_base__pixel_format wuffs_base__image_config__pixel_format( + wuffs_base__image_config* c) { + return c ? c->private_impl.pixfmt : 0; +} + +static inline wuffs_base__pixel_subsampling +wuffs_base__image_config__pixel_subsampling(wuffs_base__image_config* c) { + return wuffs_base__image_config__is_valid(c) ? c->private_impl.pixsub : 0; +} + static inline uint32_t wuffs_base__image_config__width( wuffs_base__image_config* c) { - return wuffs_base__image_config__valid(c) ? c->private_impl.w : 0; + return wuffs_base__image_config__is_valid(c) ? c->private_impl.width : 0; } static inline uint32_t wuffs_base__image_config__height( wuffs_base__image_config* c) { - return wuffs_base__image_config__valid(c) ? c->private_impl.h : 0; + return wuffs_base__image_config__is_valid(c) ? c->private_impl.height : 0; +} + +static inline uint32_t wuffs_base__image_config__num_loops( + wuffs_base__image_config* c) { + return wuffs_base__image_config__is_valid(c) ? c->private_impl.num_loops : 0; } // TODO: this is the right API for planar (not packed) pixbufs? Should it allow @@ -153,26 +482,96 @@ static inline uint32_t wuffs_base__image_config__height( // example, decoding a JPEG image straight to RGBA instead of to YCbCr? static inline size_t wuffs_base__image_config__pixbuf_size( wuffs_base__image_config* c) { - if (wuffs_base__image_config__valid(c)) { - uint64_t wh = ((uint64_t)c->private_impl.w) * ((uint64_t)c->private_impl.h); + if (wuffs_base__image_config__is_valid(c)) { + uint64_t wh = + ((uint64_t)c->private_impl.width) * ((uint64_t)c->private_impl.height); // TODO: handle things other than 1 byte per pixel. return (size_t)wh; } return 0; } -static inline void wuffs_base__image_config__initialize( - wuffs_base__image_config* c, - uint32_t width, - uint32_t height, - uint32_t TODO_color_model) { - if (!c) { +typedef struct { + // Do not access the private_impl's fields directly. There is no API/ABI + // compatibility or safety guarantee if you do so. + struct { + wuffs_base__image_config config; + uint32_t loop_count; // 0-based count of the current loop. + wuffs_base__pixel_buffer pixbuf; + // TODO: color spaces. + wuffs_base__rectangle dirty_rect; + wuffs_base__flicks duration; + uint8_t palette[1024]; + } private_impl; +} wuffs_base__image_buffer; + +static inline void wuffs_base__image_buffer__initialize( + wuffs_base__image_buffer* f, + wuffs_base__image_config config, + wuffs_base__pixel_buffer pixbuf) { + if (!f) { return; } - c->private_impl.flags = 1; - c->private_impl.w = width; - c->private_impl.h = height; - // TODO: color model. + *f = ((wuffs_base__image_buffer){}); + f->private_impl.config = config; + f->private_impl.pixbuf = pixbuf; +} + +static inline void wuffs_base__image_buffer__update( + wuffs_base__image_buffer* f, + wuffs_base__rectangle dirty_rect, + wuffs_base__flicks duration, + uint8_t* palette_ptr, + size_t palette_len) { + if (!f) { + return; + } + f->private_impl.dirty_rect = dirty_rect; + f->private_impl.duration = duration; + if (palette_ptr) { + memmove(f->private_impl.palette, palette_ptr, + palette_len <= 1024 ? palette_len : 1024); + } +} + +// wuffs_base__image_buffer__loop returns whether the image decoder should loop +// back to the beginning of the animation, assuming that we've reached the end +// of the encoded stream. If so, it increments f's count of the animation loops +// played so far. +static inline bool wuffs_base__image_buffer__loop(wuffs_base__image_buffer* f) { + if (!f) { + return false; + } + uint32_t n = f->private_impl.config.private_impl.num_loops; + if (n == 0) { + return true; + } + if (f->private_impl.loop_count < n - 1) { + f->private_impl.loop_count++; + return true; + } + return false; +} + +// wuffs_base__image_buffer__dirty_rect returns an upper bound for what part of +// this frame's pixels differs from the previous frame. +static inline wuffs_base__rectangle wuffs_base__image_buffer__dirty_rect( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.dirty_rect : ((wuffs_base__rectangle){0}); +} + +// wuffs_base__image_buffer__duration returns the amount of time to display +// this frame. Zero means to display forever - a still (non-animated) image. +static inline wuffs_base__flicks wuffs_base__image_buffer__duration( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.duration : 0; +} + +// wuffs_base__image_buffer__palette returns the palette that the pixel data +// can index. The backing array is inside f and has length 1024. +static inline uint8_t* wuffs_base__image_buffer__palette( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.palette : NULL; } #endif // WUFFS_BASE_HEADER_H @@ -185,14 +584,14 @@ extern "C" { // ---------------- Status Codes -// Status codes are int32_t values: -// - the sign bit indicates a non-recoverable status code: an error -// - bits 10-30 hold the packageid: a namespace -// - bits 8-9 are reserved -// - bits 0-7 are a package-namespaced numeric code +// Status codes are int32_t values. Its bits: +// - bit 31 (the sign bit) indicates unrecoverable-ness: an error. +// - bits 30 .. 10 are the packageid: a namespace. +// - bits 9 .. 8 are reserved. +// - bits 7 .. 0 are a package-namespaced numeric code. // -// Do not manipulate these bits directly. Use the API functions such as -// wuffs_gif__status__is_error instead. +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_gif__status__is_error instead. typedef int32_t wuffs_gif__status; #define wuffs_gif__packageid 1017222 // 0x000F8586 @@ -400,14 +799,14 @@ typedef struct { // but it should catch 99.99% of cases. // // Its (non-zero) value is arbitrary, based on md5sum("wuffs"). -#define WUFFS_BASE__MAGIC (0x3CCB6C71U) +#define WUFFS_BASE__MAGIC 0x3CCB6C71U // WUFFS_BASE__ALREADY_ZEROED is passed from a container struct's initializer // to a containee struct's initializer when the container has already zeroed // the containee's memory. // // Its (non-zero) value is arbitrary, based on md5sum("zeroed"). -#define WUFFS_BASE__ALREADY_ZEROED (0x68602EF1U) +#define WUFFS_BASE__ALREADY_ZEROED 0x68602EF1U // Denote intentional fallthroughs for -Wimplicit-fallthrough. // @@ -985,8 +1384,9 @@ wuffs_gif__status wuffs_gif__decoder__decode_config( } } label_0_break:; - wuffs_base__image_config__initialize(a_dst, self->private_impl.f_width, - self->private_impl.f_height, 0); + wuffs_base__image_config__initialize( + a_dst, 570984584, 0, self->private_impl.f_width, + self->private_impl.f_height, self->private_impl.f_num_loops); self->private_impl.f_call_sequence = 1; goto ok; diff --git a/gen/c/std/gzip.c b/gen/c/std/gzip.c index 9889780d..0a814d42 100644 --- a/gen/c/std/gzip.c +++ b/gen/c/std/gzip.c @@ -46,11 +46,40 @@ // For now, the API and ABI are simply unstable and can change at any time. // // TODO: don't hard code this in base-header.h. -#define WUFFS_VERSION (0x00001) +#define WUFFS_VERSION 0x00001 -// ---------------- I/O +// ---------------- Fundamentals -// wuffs_base__slice_u8 is a 1-dimensional buffer (a pointer and length). +// Flicks are a unit of time. One flick (frame-tick) is 1 / 705_600_000 of a +// second. See https://github.com/OculusVR/Flicks +typedef uint64_t wuffs_base__flicks; + +#define WUFFS_BASE__FLICKS_PER_SECOND 705600000ULL + +// wuffs_base__rectangle is a rectangle on the integer grid. It contains all +// points (x, y) such that ((min_x <= x) && (x < max_x)) and likewise for y. It +// is therefore empty if min_x >= max_x. There are multiple representations of +// an empty rectangle. +// +// A value with all fields zero is a valid, empty rectangle. +// +// The X and Y axes increase right and down. +typedef struct { + uint32_t min_x; + uint32_t min_y; + uint32_t max_x; + uint32_t max_y; +} wuffs_base__rectangle; + +static inline uint32_t wuffs_base__rectangle__width(wuffs_base__rectangle r) { + return r.max_x > r.min_x ? r.max_x - r.min_x : 0; +} + +static inline uint32_t wuffs_base__rectangle__height(wuffs_base__rectangle r) { + return r.max_y > r.min_y ? r.max_y - r.min_y : 0; +} + +// wuffs_base__slice_u8 is a 1-dimensional buffer. // // A value with all fields NULL or zero is a valid, empty slice. typedef struct { @@ -58,6 +87,20 @@ typedef struct { size_t len; } wuffs_base__slice_u8; +// wuffs_base__table_u8 is a 2-dimensional buffer. +// +// A value with all fields NULL or zero is a valid, empty table. +typedef struct { + uint8_t* ptr; + size_t width; + size_t height; + size_t stride; +} wuffs_base__table_u8; + +// ---------------- I/O + +// TODO: rename buf1 to io_buffer, writer1 to io_writer, etc. + // wuffs_base__buf1 is a 1-dimensional buffer (a pointer and length), plus // additional indexes into that buffer, plus an opened / closed flag. // @@ -109,18 +152,286 @@ typedef struct { // ---------------- Images +// wuffs_base__pixel_format encodes the format of the bytes that constitute an +// image frame's pixel data. Its bits: +// - bit 31 is reserved. +// - bits 30 .. 28 encodes color (and channel order, in terms of memory). +// - bits 27 .. 26 are reserved. +// - bits 25 .. 24 encodes transparency. +// - bit 23 indicates big-endian/MSB-first (as opposed to little/LSB). +// - bit 22 indicates floating point (as opposed to integer). +// - bits 21 .. 20 are the number of planes, minus 1. Zero means packed. +// - bits 19 .. 16 encodes the number of bits (depth) in an index value. +// Zero means direct, not palette-indexed. +// - bits 15 .. 12 encodes the number of bits (depth) in the 3rd channel. +// - bits 11 .. 8 encodes the number of bits (depth) in the 2nd channel. +// - bits 7 .. 4 encodes the number of bits (depth) in the 1st channel. +// - bits 3 .. 0 encodes the number of bits (depth) in the 0th channel. +// +// The bit fields of a wuffs_base__pixel_format are not independent. For +// example, the number of planes should not be greater than the number of +// channels. Similarly, bits 15..4 are unused (and should be zero) if bits +// 31..24 (color and transparency) together imply only 1 channel (gray, no +// alpha) and floating point samples should mean a bit depth of 16, 32 or 64. +// +// Formats hold between 1 and 4 channels. For example: Y (1 channel: gray), YA +// (2 channels: gray and alpha), BGR (3 channels: blue, green, red) or CMYK (4 +// channels: cyan, magenta, yellow, black). +// +// For direct formats with N > 1 channels, those channels can be laid out in +// either 1 (packed) or N (planar) planes. For example, RGBA data is usually +// packed, but YUV data is usually planar, due to chroma subsampling (for +// details, see the wuffs_base__pixel_subsampling type). For indexed formats, +// the palette (always 256 × 4 bytes) holds up to 4 packed bytes of color data +// per index value, and there is only 1 plane (for the index). The distance +// between successive palette elements is always 4 bytes. +// +// The color field is encoded in 3 bits: +// - 0 means A (Alpha). +// - 1 means Y or YA (Gray, Alpha). +// - 2 means BGR, BGRX or BGRA (Blue, Green, Red, X-padding or Alpha). +// - 3 means RGB, RGBX or RGBA (Red, Green, Blue, X-padding or Alpha). +// - 4 means YUV or YUVA (Luma, Chroma-blue, Chroma-red, Alpha). +// - 5 means CMY or CMYK (Cyan, Magenta, Yellow, Black). +// - all other values are reserved. +// +// In Wuffs, channels are given in memory order, regardless of endianness, +// since the C type for the pixel data is an array of bytes, not an array of +// uint32_t. For example, packed BGRA with 8 bits per channel means that the +// bytes in memory are always Blue, Green, Red then Alpha. On big-endian +// systems, that is the uint32_t 0xBBGGRRAA. On little-endian, 0xAARRGGBB. +// +// When the color field (3 bits) encodes multiple options, the transparency +// field (2 bits) distinguishes them: +// - 0 means fully opaque, no extra channels +// - 1 means fully opaque, one extra channel (X or K, padding or black). +// - 2 means one extra alpha channel, other channels are non-premultiplied. +// - 3 means one extra alpha channel, other channels are premultiplied. +// +// The zero wuffs_base__pixel_format value is an invalid pixel format, as it is +// invalid to combine the zero color (alpha only) with the zero transparency. +// +// Bit depth is encoded in 4 bits: +// - 0 means the channel or index is unused. +// - x means a bit depth of x, for x in the range 1..8. +// - 9 means a bit depth of 10. +// - 10 means a bit depth of 12. +// - 11 means a bit depth of 16. +// - 12 means a bit depth of 24. +// - 13 means a bit depth of 32. +// - 14 means a bit depth of 48. +// - 15 means a bit depth of 64. +// +// For example, wuffs_base__pixel_format 0x3280BBBB is a natural format for +// decoding a PNG image - network byte order (also known as big-endian), +// packed, non-premultiplied alpha - that happens to be 16-bit-depth truecolor +// with alpha (RGBA). In memory order: +// +// ptr+0 ptr+1 ptr+2 ptr+3 ptr+4 ptr+5 ptr+6 ptr+7 +// Rhi Rlo Ghi Glo Bhi Blo Ahi Alo +// +// For example, the value wuffs_base__pixel_format 0x20000565 means BGR with no +// alpha or padding, 5/6/5 bits for blue/green/red, packed 2 bytes per pixel, +// laid out LSB-first in memory order: +// +// ptr+0........... ptr+1........... +// MSB LSB MSB LSB +// G₂G₁G₀B₄B₃B₂B₁B₀ R₄R₃R₂R₁R₀G₅G₄G₃ +// +// On little-endian systems (but not big-endian), this Wuffs pixel format value +// (0x20000565) corresponds to the Cairo library's CAIRO_FORMAT_RGB16_565, the +// SDL2 (Simple DirectMedia Layer 2) library's SDL_PIXELFORMAT_RGB565 and the +// Skia library's kRGB_565_SkColorType. Note BGR in Wuffs versus RGB in the +// other libraries. +// +// Regardless of endianness, this Wuffs pixel format value (0x20000565) +// corresponds to the V4L2 (Video For Linux 2) library's V4L2_PIX_FMT_RGB565 +// and the Wayland-DRM library's WL_DRM_FORMAT_RGB565. +// +// Different software libraries name their pixel formats (and especially their +// channel order) either according to memory layout or as bits of a native +// integer type like uint32_t. The two conventions differ because of a system's +// endianness. As mentioned earlier, Wuffs pixel formats are always in memory +// order. More detail of other software libraries' naming conventions is in the +// Pixel Format Guide at https://afrantzis.github.io/pixel-format-guide/ +// +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_base__pixel_format__num_planes instead. +typedef uint32_t wuffs_base__pixel_format; + +// Common 8-bit-depth pixel formats. This list is not exhaustive; not all valid +// wuffs_base__pixel_format values are present. + +#define WUFFS_BASE__PIXEL_FORMAT__INVALID ((wuffs_base__pixel_format)0x00000000) + +#define WUFFS_BASE__PIXEL_FORMAT__A ((wuffs_base__pixel_format)0x02000008) + +#define WUFFS_BASE__PIXEL_FORMAT__Y ((wuffs_base__pixel_format)0x10000008) +#define WUFFS_BASE__PIXEL_FORMAT__YA_NONPREMUL \ + ((wuffs_base__pixel_format)0x12000008) +#define WUFFS_BASE__PIXEL_FORMAT__YA_PREMUL \ + ((wuffs_base__pixel_format)0x13000008) + +#define WUFFS_BASE__PIXEL_FORMAT__BGR ((wuffs_base__pixel_format)0x20000888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRX ((wuffs_base__pixel_format)0x21008888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRX_INDEXED \ + ((wuffs_base__pixel_format)0x21088888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_NONPREMUL \ + ((wuffs_base__pixel_format)0x22008888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_NONPREMUL_INDEXED \ + ((wuffs_base__pixel_format)0x22088888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_PREMUL \ + ((wuffs_base__pixel_format)0x23008888) + +#define WUFFS_BASE__PIXEL_FORMAT__RGB ((wuffs_base__pixel_format)0x30000888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBX ((wuffs_base__pixel_format)0x31008888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBX_INDEXED \ + ((wuffs_base__pixel_format)0x31088888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_NONPREMUL \ + ((wuffs_base__pixel_format)0x32008888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_NONPREMUL_INDEXED \ + ((wuffs_base__pixel_format)0x32088888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_PREMUL \ + ((wuffs_base__pixel_format)0x33008888) + +#define WUFFS_BASE__PIXEL_FORMAT__YUV ((wuffs_base__pixel_format)0x40200888) +#define WUFFS_BASE__PIXEL_FORMAT__YUVK ((wuffs_base__pixel_format)0x41308888) +#define WUFFS_BASE__PIXEL_FORMAT__YUVA_NONPREMUL \ + ((wuffs_base__pixel_format)0x42308888) + +#define WUFFS_BASE__PIXEL_FORMAT__CMY ((wuffs_base__pixel_format)0x50200888) +#define WUFFS_BASE__PIXEL_FORMAT__CMYK ((wuffs_base__pixel_format)0x51308888) + +static inline bool wuffs_base__pixel_format__is_valid( + wuffs_base__pixel_format f) { + return f != 0; +} + +static inline bool wuffs_base__pixel_format__is_indexed( + wuffs_base__pixel_format f) { + return ((f >> 16) & 0x0F) != 0; +} + +#define WUFFS_BASE__PIXEL_FORMAT__NUM_PLANES_MAX 4 + +static inline uint32_t wuffs_base__pixel_format__num_planes( + wuffs_base__pixel_format f) { + return f ? (((f >> 20) & 0x03) + 1) : 0; +} + +typedef struct { + wuffs_base__table_u8 planes[WUFFS_BASE__PIXEL_FORMAT__NUM_PLANES_MAX]; +} wuffs_base__pixel_buffer; + +// wuffs_base__pixel_subsampling encodes the mapping of pixel space coordinates +// (x, y) to pixel buffer indices (i, j). That mapping can differ for each +// plane p. For a depth of 8 bits (1 byte), the p'th plane's sample starts at +// (planes[p].ptr + (j * planes[p].stride) + i). +// +// For packed pixel formats, the mapping is trivial: i = x and j = y. For +// planar pixel formats, the mapping can differ due to chroma subsampling. For +// example, consider a three plane YUV pixel format with 4:2:2 subsampling. For +// the luma (Y) channel, there is one sample for every pixel, but for the +// chroma (U, V) channels, there is one sample for every two pixels: pairs of +// horizontally adjacent pixels form one macropixel, i = x / 2 and j == y. In +// general, for a given p: +// - i = (x + bias_x) >> shift_x. +// - j = (y + bias_y) >> shift_y. +// where biases and shifts are in the range 0..3 and 0..2 respectively. +// +// In general, the biases will be zero after decoding an image. However, making +// a sub-image may change the bias, since the (x, y) coordinates are relative +// to the sub-image's top-left origin, but the backing pixel buffers were +// created relative to the original image's origin. +// +// For each plane p, each of those four numbers (biases and shifts) are encoded +// in two bits, which combine to form an 8 bit unsigned integer: +// +// e_p = (bias_x << 6) | (shift_x << 4) | (bias_y << 2) | (shift_y << 0) +// +// Those e_p values (e_0 for the first plane, e_1 for the second plane, etc) +// combine to form a wuffs_base__pixel_subsampling value: +// +// pixsub = (e_3 << 24) | (e_2 << 16) | (e_1 << 8) | (e_0 << 0) +// +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_base__pixel_subsampling__bias_x instead. +typedef uint32_t wuffs_base__pixel_subsampling; + +#define WUFFS_BASE__PIXEL_SUBSAMPLING__NONE ((wuffs_base__pixel_subsampling)0); + +#define WUFFS_BASE__PIXEL_SUBSAMPLING__444 \ + ((wuffs_base__pixel_subsampling)0x000000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__440 \ + ((wuffs_base__pixel_subsampling)0x010100); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__422 \ + ((wuffs_base__pixel_subsampling)0x101000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__420 \ + ((wuffs_base__pixel_subsampling)0x111100); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__411 \ + ((wuffs_base__pixel_subsampling)0x202000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__410 \ + ((wuffs_base__pixel_subsampling)0x212100); + +static inline uint32_t wuffs_base__pixel_subsampling__bias_x( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 6; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__shift_x( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 4; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__bias_y( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 2; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__shift_y( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 0; + return (s >> shift) & 0x03; +} + typedef struct { // Do not access the private_impl's fields directly. There is no API/ABI // compatibility or safety guarantee if you do so. struct { - uint32_t flags; - uint32_t w; - uint32_t h; - // TODO: color model, including both packed RGBA and planar, - // chroma-subsampled YCbCr. + wuffs_base__pixel_format pixfmt; + wuffs_base__pixel_subsampling pixsub; + uint32_t width; + uint32_t height; + uint32_t num_loops; } private_impl; } wuffs_base__image_config; +static inline void wuffs_base__image_config__initialize( + wuffs_base__image_config* c, + wuffs_base__pixel_format pixfmt, + wuffs_base__pixel_subsampling pixsub, + uint32_t width, + uint32_t height, + uint32_t num_loops) { + if (!c) { + return; + } + // TODO: move the check from wuffs_base__image_config__is_valid here. Should + // this function return bool? An error type? + c->private_impl.pixfmt = pixfmt; + c->private_impl.pixsub = pixsub; + c->private_impl.width = width; + c->private_impl.height = height; + c->private_impl.num_loops = num_loops; +} + static inline void wuffs_base__image_config__invalidate( wuffs_base__image_config* c) { if (c) { @@ -128,24 +439,42 @@ static inline void wuffs_base__image_config__invalidate( } } -static inline bool wuffs_base__image_config__valid( +static inline bool wuffs_base__image_config__is_valid( wuffs_base__image_config* c) { - if (!c || !(c->private_impl.flags & 1)) { + if (!c || !c->private_impl.pixfmt) { return false; } - uint64_t wh = ((uint64_t)c->private_impl.w) * ((uint64_t)c->private_impl.h); + uint64_t wh = + ((uint64_t)c->private_impl.width) * ((uint64_t)c->private_impl.height); // TODO: handle things other than 1 byte per pixel. + // + // TODO: move the check to wuffs_base__image_config__initialize. return wh <= ((uint64_t)SIZE_MAX); } +static inline wuffs_base__pixel_format wuffs_base__image_config__pixel_format( + wuffs_base__image_config* c) { + return c ? c->private_impl.pixfmt : 0; +} + +static inline wuffs_base__pixel_subsampling +wuffs_base__image_config__pixel_subsampling(wuffs_base__image_config* c) { + return wuffs_base__image_config__is_valid(c) ? c->private_impl.pixsub : 0; +} + static inline uint32_t wuffs_base__image_config__width( wuffs_base__image_config* c) { - return wuffs_base__image_config__valid(c) ? c->private_impl.w : 0; + return wuffs_base__image_config__is_valid(c) ? c->private_impl.width : 0; } static inline uint32_t wuffs_base__image_config__height( wuffs_base__image_config* c) { - return wuffs_base__image_config__valid(c) ? c->private_impl.h : 0; + return wuffs_base__image_config__is_valid(c) ? c->private_impl.height : 0; +} + +static inline uint32_t wuffs_base__image_config__num_loops( + wuffs_base__image_config* c) { + return wuffs_base__image_config__is_valid(c) ? c->private_impl.num_loops : 0; } // TODO: this is the right API for planar (not packed) pixbufs? Should it allow @@ -153,26 +482,96 @@ static inline uint32_t wuffs_base__image_config__height( // example, decoding a JPEG image straight to RGBA instead of to YCbCr? static inline size_t wuffs_base__image_config__pixbuf_size( wuffs_base__image_config* c) { - if (wuffs_base__image_config__valid(c)) { - uint64_t wh = ((uint64_t)c->private_impl.w) * ((uint64_t)c->private_impl.h); + if (wuffs_base__image_config__is_valid(c)) { + uint64_t wh = + ((uint64_t)c->private_impl.width) * ((uint64_t)c->private_impl.height); // TODO: handle things other than 1 byte per pixel. return (size_t)wh; } return 0; } -static inline void wuffs_base__image_config__initialize( - wuffs_base__image_config* c, - uint32_t width, - uint32_t height, - uint32_t TODO_color_model) { - if (!c) { +typedef struct { + // Do not access the private_impl's fields directly. There is no API/ABI + // compatibility or safety guarantee if you do so. + struct { + wuffs_base__image_config config; + uint32_t loop_count; // 0-based count of the current loop. + wuffs_base__pixel_buffer pixbuf; + // TODO: color spaces. + wuffs_base__rectangle dirty_rect; + wuffs_base__flicks duration; + uint8_t palette[1024]; + } private_impl; +} wuffs_base__image_buffer; + +static inline void wuffs_base__image_buffer__initialize( + wuffs_base__image_buffer* f, + wuffs_base__image_config config, + wuffs_base__pixel_buffer pixbuf) { + if (!f) { return; } - c->private_impl.flags = 1; - c->private_impl.w = width; - c->private_impl.h = height; - // TODO: color model. + *f = ((wuffs_base__image_buffer){}); + f->private_impl.config = config; + f->private_impl.pixbuf = pixbuf; +} + +static inline void wuffs_base__image_buffer__update( + wuffs_base__image_buffer* f, + wuffs_base__rectangle dirty_rect, + wuffs_base__flicks duration, + uint8_t* palette_ptr, + size_t palette_len) { + if (!f) { + return; + } + f->private_impl.dirty_rect = dirty_rect; + f->private_impl.duration = duration; + if (palette_ptr) { + memmove(f->private_impl.palette, palette_ptr, + palette_len <= 1024 ? palette_len : 1024); + } +} + +// wuffs_base__image_buffer__loop returns whether the image decoder should loop +// back to the beginning of the animation, assuming that we've reached the end +// of the encoded stream. If so, it increments f's count of the animation loops +// played so far. +static inline bool wuffs_base__image_buffer__loop(wuffs_base__image_buffer* f) { + if (!f) { + return false; + } + uint32_t n = f->private_impl.config.private_impl.num_loops; + if (n == 0) { + return true; + } + if (f->private_impl.loop_count < n - 1) { + f->private_impl.loop_count++; + return true; + } + return false; +} + +// wuffs_base__image_buffer__dirty_rect returns an upper bound for what part of +// this frame's pixels differs from the previous frame. +static inline wuffs_base__rectangle wuffs_base__image_buffer__dirty_rect( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.dirty_rect : ((wuffs_base__rectangle){0}); +} + +// wuffs_base__image_buffer__duration returns the amount of time to display +// this frame. Zero means to display forever - a still (non-animated) image. +static inline wuffs_base__flicks wuffs_base__image_buffer__duration( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.duration : 0; +} + +// wuffs_base__image_buffer__palette returns the palette that the pixel data +// can index. The backing array is inside f and has length 1024. +static inline uint8_t* wuffs_base__image_buffer__palette( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.palette : NULL; } #endif // WUFFS_BASE_HEADER_H @@ -194,14 +593,14 @@ extern "C" { // ---------------- Status Codes -// Status codes are int32_t values: -// - the sign bit indicates a non-recoverable status code: an error -// - bits 10-30 hold the packageid: a namespace -// - bits 8-9 are reserved -// - bits 0-7 are a package-namespaced numeric code +// Status codes are int32_t values. Its bits: +// - bit 31 (the sign bit) indicates unrecoverable-ness: an error. +// - bits 30 .. 10 are the packageid: a namespace. +// - bits 9 .. 8 are reserved. +// - bits 7 .. 0 are a package-namespaced numeric code. // -// Do not manipulate these bits directly. Use the API functions such as -// wuffs_crc32__status__is_error instead. +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_crc32__status__is_error instead. typedef int32_t wuffs_crc32__status; #define wuffs_crc32__packageid 810620 // 0x000C5E7C @@ -284,14 +683,14 @@ extern "C" { // ---------------- Status Codes -// Status codes are int32_t values: -// - the sign bit indicates a non-recoverable status code: an error -// - bits 10-30 hold the packageid: a namespace -// - bits 8-9 are reserved -// - bits 0-7 are a package-namespaced numeric code +// Status codes are int32_t values. Its bits: +// - bit 31 (the sign bit) indicates unrecoverable-ness: an error. +// - bits 30 .. 10 are the packageid: a namespace. +// - bits 9 .. 8 are reserved. +// - bits 7 .. 0 are a package-namespaced numeric code. // -// Do not manipulate these bits directly. Use the API functions such as -// wuffs_deflate__status__is_error instead. +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_deflate__status__is_error instead. typedef int32_t wuffs_deflate__status; #define wuffs_deflate__packageid 848533 // 0x000CF295 @@ -457,14 +856,14 @@ extern "C" { // ---------------- Status Codes -// Status codes are int32_t values: -// - the sign bit indicates a non-recoverable status code: an error -// - bits 10-30 hold the packageid: a namespace -// - bits 8-9 are reserved -// - bits 0-7 are a package-namespaced numeric code +// Status codes are int32_t values. Its bits: +// - bit 31 (the sign bit) indicates unrecoverable-ness: an error. +// - bits 30 .. 10 are the packageid: a namespace. +// - bits 9 .. 8 are reserved. +// - bits 7 .. 0 are a package-namespaced numeric code. // -// Do not manipulate these bits directly. Use the API functions such as -// wuffs_gzip__status__is_error instead. +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_gzip__status__is_error instead. typedef int32_t wuffs_gzip__status; #define wuffs_gzip__packageid 1041911 // 0x000FE5F7 @@ -586,14 +985,14 @@ typedef struct { // but it should catch 99.99% of cases. // // Its (non-zero) value is arbitrary, based on md5sum("wuffs"). -#define WUFFS_BASE__MAGIC (0x3CCB6C71U) +#define WUFFS_BASE__MAGIC 0x3CCB6C71U // WUFFS_BASE__ALREADY_ZEROED is passed from a container struct's initializer // to a containee struct's initializer when the container has already zeroed // the containee's memory. // // Its (non-zero) value is arbitrary, based on md5sum("zeroed"). -#define WUFFS_BASE__ALREADY_ZEROED (0x68602EF1U) +#define WUFFS_BASE__ALREADY_ZEROED 0x68602EF1U // Denote intentional fallthroughs for -Wimplicit-fallthrough. // diff --git a/gen/c/std/zlib.c b/gen/c/std/zlib.c index 70e0a783..183c1ba9 100644 --- a/gen/c/std/zlib.c +++ b/gen/c/std/zlib.c @@ -46,11 +46,40 @@ // For now, the API and ABI are simply unstable and can change at any time. // // TODO: don't hard code this in base-header.h. -#define WUFFS_VERSION (0x00001) +#define WUFFS_VERSION 0x00001 -// ---------------- I/O +// ---------------- Fundamentals -// wuffs_base__slice_u8 is a 1-dimensional buffer (a pointer and length). +// Flicks are a unit of time. One flick (frame-tick) is 1 / 705_600_000 of a +// second. See https://github.com/OculusVR/Flicks +typedef uint64_t wuffs_base__flicks; + +#define WUFFS_BASE__FLICKS_PER_SECOND 705600000ULL + +// wuffs_base__rectangle is a rectangle on the integer grid. It contains all +// points (x, y) such that ((min_x <= x) && (x < max_x)) and likewise for y. It +// is therefore empty if min_x >= max_x. There are multiple representations of +// an empty rectangle. +// +// A value with all fields zero is a valid, empty rectangle. +// +// The X and Y axes increase right and down. +typedef struct { + uint32_t min_x; + uint32_t min_y; + uint32_t max_x; + uint32_t max_y; +} wuffs_base__rectangle; + +static inline uint32_t wuffs_base__rectangle__width(wuffs_base__rectangle r) { + return r.max_x > r.min_x ? r.max_x - r.min_x : 0; +} + +static inline uint32_t wuffs_base__rectangle__height(wuffs_base__rectangle r) { + return r.max_y > r.min_y ? r.max_y - r.min_y : 0; +} + +// wuffs_base__slice_u8 is a 1-dimensional buffer. // // A value with all fields NULL or zero is a valid, empty slice. typedef struct { @@ -58,6 +87,20 @@ typedef struct { size_t len; } wuffs_base__slice_u8; +// wuffs_base__table_u8 is a 2-dimensional buffer. +// +// A value with all fields NULL or zero is a valid, empty table. +typedef struct { + uint8_t* ptr; + size_t width; + size_t height; + size_t stride; +} wuffs_base__table_u8; + +// ---------------- I/O + +// TODO: rename buf1 to io_buffer, writer1 to io_writer, etc. + // wuffs_base__buf1 is a 1-dimensional buffer (a pointer and length), plus // additional indexes into that buffer, plus an opened / closed flag. // @@ -109,18 +152,286 @@ typedef struct { // ---------------- Images +// wuffs_base__pixel_format encodes the format of the bytes that constitute an +// image frame's pixel data. Its bits: +// - bit 31 is reserved. +// - bits 30 .. 28 encodes color (and channel order, in terms of memory). +// - bits 27 .. 26 are reserved. +// - bits 25 .. 24 encodes transparency. +// - bit 23 indicates big-endian/MSB-first (as opposed to little/LSB). +// - bit 22 indicates floating point (as opposed to integer). +// - bits 21 .. 20 are the number of planes, minus 1. Zero means packed. +// - bits 19 .. 16 encodes the number of bits (depth) in an index value. +// Zero means direct, not palette-indexed. +// - bits 15 .. 12 encodes the number of bits (depth) in the 3rd channel. +// - bits 11 .. 8 encodes the number of bits (depth) in the 2nd channel. +// - bits 7 .. 4 encodes the number of bits (depth) in the 1st channel. +// - bits 3 .. 0 encodes the number of bits (depth) in the 0th channel. +// +// The bit fields of a wuffs_base__pixel_format are not independent. For +// example, the number of planes should not be greater than the number of +// channels. Similarly, bits 15..4 are unused (and should be zero) if bits +// 31..24 (color and transparency) together imply only 1 channel (gray, no +// alpha) and floating point samples should mean a bit depth of 16, 32 or 64. +// +// Formats hold between 1 and 4 channels. For example: Y (1 channel: gray), YA +// (2 channels: gray and alpha), BGR (3 channels: blue, green, red) or CMYK (4 +// channels: cyan, magenta, yellow, black). +// +// For direct formats with N > 1 channels, those channels can be laid out in +// either 1 (packed) or N (planar) planes. For example, RGBA data is usually +// packed, but YUV data is usually planar, due to chroma subsampling (for +// details, see the wuffs_base__pixel_subsampling type). For indexed formats, +// the palette (always 256 × 4 bytes) holds up to 4 packed bytes of color data +// per index value, and there is only 1 plane (for the index). The distance +// between successive palette elements is always 4 bytes. +// +// The color field is encoded in 3 bits: +// - 0 means A (Alpha). +// - 1 means Y or YA (Gray, Alpha). +// - 2 means BGR, BGRX or BGRA (Blue, Green, Red, X-padding or Alpha). +// - 3 means RGB, RGBX or RGBA (Red, Green, Blue, X-padding or Alpha). +// - 4 means YUV or YUVA (Luma, Chroma-blue, Chroma-red, Alpha). +// - 5 means CMY or CMYK (Cyan, Magenta, Yellow, Black). +// - all other values are reserved. +// +// In Wuffs, channels are given in memory order, regardless of endianness, +// since the C type for the pixel data is an array of bytes, not an array of +// uint32_t. For example, packed BGRA with 8 bits per channel means that the +// bytes in memory are always Blue, Green, Red then Alpha. On big-endian +// systems, that is the uint32_t 0xBBGGRRAA. On little-endian, 0xAARRGGBB. +// +// When the color field (3 bits) encodes multiple options, the transparency +// field (2 bits) distinguishes them: +// - 0 means fully opaque, no extra channels +// - 1 means fully opaque, one extra channel (X or K, padding or black). +// - 2 means one extra alpha channel, other channels are non-premultiplied. +// - 3 means one extra alpha channel, other channels are premultiplied. +// +// The zero wuffs_base__pixel_format value is an invalid pixel format, as it is +// invalid to combine the zero color (alpha only) with the zero transparency. +// +// Bit depth is encoded in 4 bits: +// - 0 means the channel or index is unused. +// - x means a bit depth of x, for x in the range 1..8. +// - 9 means a bit depth of 10. +// - 10 means a bit depth of 12. +// - 11 means a bit depth of 16. +// - 12 means a bit depth of 24. +// - 13 means a bit depth of 32. +// - 14 means a bit depth of 48. +// - 15 means a bit depth of 64. +// +// For example, wuffs_base__pixel_format 0x3280BBBB is a natural format for +// decoding a PNG image - network byte order (also known as big-endian), +// packed, non-premultiplied alpha - that happens to be 16-bit-depth truecolor +// with alpha (RGBA). In memory order: +// +// ptr+0 ptr+1 ptr+2 ptr+3 ptr+4 ptr+5 ptr+6 ptr+7 +// Rhi Rlo Ghi Glo Bhi Blo Ahi Alo +// +// For example, the value wuffs_base__pixel_format 0x20000565 means BGR with no +// alpha or padding, 5/6/5 bits for blue/green/red, packed 2 bytes per pixel, +// laid out LSB-first in memory order: +// +// ptr+0........... ptr+1........... +// MSB LSB MSB LSB +// G₂G₁G₀B₄B₃B₂B₁B₀ R₄R₃R₂R₁R₀G₅G₄G₃ +// +// On little-endian systems (but not big-endian), this Wuffs pixel format value +// (0x20000565) corresponds to the Cairo library's CAIRO_FORMAT_RGB16_565, the +// SDL2 (Simple DirectMedia Layer 2) library's SDL_PIXELFORMAT_RGB565 and the +// Skia library's kRGB_565_SkColorType. Note BGR in Wuffs versus RGB in the +// other libraries. +// +// Regardless of endianness, this Wuffs pixel format value (0x20000565) +// corresponds to the V4L2 (Video For Linux 2) library's V4L2_PIX_FMT_RGB565 +// and the Wayland-DRM library's WL_DRM_FORMAT_RGB565. +// +// Different software libraries name their pixel formats (and especially their +// channel order) either according to memory layout or as bits of a native +// integer type like uint32_t. The two conventions differ because of a system's +// endianness. As mentioned earlier, Wuffs pixel formats are always in memory +// order. More detail of other software libraries' naming conventions is in the +// Pixel Format Guide at https://afrantzis.github.io/pixel-format-guide/ +// +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_base__pixel_format__num_planes instead. +typedef uint32_t wuffs_base__pixel_format; + +// Common 8-bit-depth pixel formats. This list is not exhaustive; not all valid +// wuffs_base__pixel_format values are present. + +#define WUFFS_BASE__PIXEL_FORMAT__INVALID ((wuffs_base__pixel_format)0x00000000) + +#define WUFFS_BASE__PIXEL_FORMAT__A ((wuffs_base__pixel_format)0x02000008) + +#define WUFFS_BASE__PIXEL_FORMAT__Y ((wuffs_base__pixel_format)0x10000008) +#define WUFFS_BASE__PIXEL_FORMAT__YA_NONPREMUL \ + ((wuffs_base__pixel_format)0x12000008) +#define WUFFS_BASE__PIXEL_FORMAT__YA_PREMUL \ + ((wuffs_base__pixel_format)0x13000008) + +#define WUFFS_BASE__PIXEL_FORMAT__BGR ((wuffs_base__pixel_format)0x20000888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRX ((wuffs_base__pixel_format)0x21008888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRX_INDEXED \ + ((wuffs_base__pixel_format)0x21088888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_NONPREMUL \ + ((wuffs_base__pixel_format)0x22008888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_NONPREMUL_INDEXED \ + ((wuffs_base__pixel_format)0x22088888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_PREMUL \ + ((wuffs_base__pixel_format)0x23008888) + +#define WUFFS_BASE__PIXEL_FORMAT__RGB ((wuffs_base__pixel_format)0x30000888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBX ((wuffs_base__pixel_format)0x31008888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBX_INDEXED \ + ((wuffs_base__pixel_format)0x31088888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_NONPREMUL \ + ((wuffs_base__pixel_format)0x32008888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_NONPREMUL_INDEXED \ + ((wuffs_base__pixel_format)0x32088888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_PREMUL \ + ((wuffs_base__pixel_format)0x33008888) + +#define WUFFS_BASE__PIXEL_FORMAT__YUV ((wuffs_base__pixel_format)0x40200888) +#define WUFFS_BASE__PIXEL_FORMAT__YUVK ((wuffs_base__pixel_format)0x41308888) +#define WUFFS_BASE__PIXEL_FORMAT__YUVA_NONPREMUL \ + ((wuffs_base__pixel_format)0x42308888) + +#define WUFFS_BASE__PIXEL_FORMAT__CMY ((wuffs_base__pixel_format)0x50200888) +#define WUFFS_BASE__PIXEL_FORMAT__CMYK ((wuffs_base__pixel_format)0x51308888) + +static inline bool wuffs_base__pixel_format__is_valid( + wuffs_base__pixel_format f) { + return f != 0; +} + +static inline bool wuffs_base__pixel_format__is_indexed( + wuffs_base__pixel_format f) { + return ((f >> 16) & 0x0F) != 0; +} + +#define WUFFS_BASE__PIXEL_FORMAT__NUM_PLANES_MAX 4 + +static inline uint32_t wuffs_base__pixel_format__num_planes( + wuffs_base__pixel_format f) { + return f ? (((f >> 20) & 0x03) + 1) : 0; +} + +typedef struct { + wuffs_base__table_u8 planes[WUFFS_BASE__PIXEL_FORMAT__NUM_PLANES_MAX]; +} wuffs_base__pixel_buffer; + +// wuffs_base__pixel_subsampling encodes the mapping of pixel space coordinates +// (x, y) to pixel buffer indices (i, j). That mapping can differ for each +// plane p. For a depth of 8 bits (1 byte), the p'th plane's sample starts at +// (planes[p].ptr + (j * planes[p].stride) + i). +// +// For packed pixel formats, the mapping is trivial: i = x and j = y. For +// planar pixel formats, the mapping can differ due to chroma subsampling. For +// example, consider a three plane YUV pixel format with 4:2:2 subsampling. For +// the luma (Y) channel, there is one sample for every pixel, but for the +// chroma (U, V) channels, there is one sample for every two pixels: pairs of +// horizontally adjacent pixels form one macropixel, i = x / 2 and j == y. In +// general, for a given p: +// - i = (x + bias_x) >> shift_x. +// - j = (y + bias_y) >> shift_y. +// where biases and shifts are in the range 0..3 and 0..2 respectively. +// +// In general, the biases will be zero after decoding an image. However, making +// a sub-image may change the bias, since the (x, y) coordinates are relative +// to the sub-image's top-left origin, but the backing pixel buffers were +// created relative to the original image's origin. +// +// For each plane p, each of those four numbers (biases and shifts) are encoded +// in two bits, which combine to form an 8 bit unsigned integer: +// +// e_p = (bias_x << 6) | (shift_x << 4) | (bias_y << 2) | (shift_y << 0) +// +// Those e_p values (e_0 for the first plane, e_1 for the second plane, etc) +// combine to form a wuffs_base__pixel_subsampling value: +// +// pixsub = (e_3 << 24) | (e_2 << 16) | (e_1 << 8) | (e_0 << 0) +// +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_base__pixel_subsampling__bias_x instead. +typedef uint32_t wuffs_base__pixel_subsampling; + +#define WUFFS_BASE__PIXEL_SUBSAMPLING__NONE ((wuffs_base__pixel_subsampling)0); + +#define WUFFS_BASE__PIXEL_SUBSAMPLING__444 \ + ((wuffs_base__pixel_subsampling)0x000000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__440 \ + ((wuffs_base__pixel_subsampling)0x010100); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__422 \ + ((wuffs_base__pixel_subsampling)0x101000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__420 \ + ((wuffs_base__pixel_subsampling)0x111100); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__411 \ + ((wuffs_base__pixel_subsampling)0x202000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__410 \ + ((wuffs_base__pixel_subsampling)0x212100); + +static inline uint32_t wuffs_base__pixel_subsampling__bias_x( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 6; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__shift_x( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 4; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__bias_y( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 2; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__shift_y( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 0; + return (s >> shift) & 0x03; +} + typedef struct { // Do not access the private_impl's fields directly. There is no API/ABI // compatibility or safety guarantee if you do so. struct { - uint32_t flags; - uint32_t w; - uint32_t h; - // TODO: color model, including both packed RGBA and planar, - // chroma-subsampled YCbCr. + wuffs_base__pixel_format pixfmt; + wuffs_base__pixel_subsampling pixsub; + uint32_t width; + uint32_t height; + uint32_t num_loops; } private_impl; } wuffs_base__image_config; +static inline void wuffs_base__image_config__initialize( + wuffs_base__image_config* c, + wuffs_base__pixel_format pixfmt, + wuffs_base__pixel_subsampling pixsub, + uint32_t width, + uint32_t height, + uint32_t num_loops) { + if (!c) { + return; + } + // TODO: move the check from wuffs_base__image_config__is_valid here. Should + // this function return bool? An error type? + c->private_impl.pixfmt = pixfmt; + c->private_impl.pixsub = pixsub; + c->private_impl.width = width; + c->private_impl.height = height; + c->private_impl.num_loops = num_loops; +} + static inline void wuffs_base__image_config__invalidate( wuffs_base__image_config* c) { if (c) { @@ -128,24 +439,42 @@ static inline void wuffs_base__image_config__invalidate( } } -static inline bool wuffs_base__image_config__valid( +static inline bool wuffs_base__image_config__is_valid( wuffs_base__image_config* c) { - if (!c || !(c->private_impl.flags & 1)) { + if (!c || !c->private_impl.pixfmt) { return false; } - uint64_t wh = ((uint64_t)c->private_impl.w) * ((uint64_t)c->private_impl.h); + uint64_t wh = + ((uint64_t)c->private_impl.width) * ((uint64_t)c->private_impl.height); // TODO: handle things other than 1 byte per pixel. + // + // TODO: move the check to wuffs_base__image_config__initialize. return wh <= ((uint64_t)SIZE_MAX); } +static inline wuffs_base__pixel_format wuffs_base__image_config__pixel_format( + wuffs_base__image_config* c) { + return c ? c->private_impl.pixfmt : 0; +} + +static inline wuffs_base__pixel_subsampling +wuffs_base__image_config__pixel_subsampling(wuffs_base__image_config* c) { + return wuffs_base__image_config__is_valid(c) ? c->private_impl.pixsub : 0; +} + static inline uint32_t wuffs_base__image_config__width( wuffs_base__image_config* c) { - return wuffs_base__image_config__valid(c) ? c->private_impl.w : 0; + return wuffs_base__image_config__is_valid(c) ? c->private_impl.width : 0; } static inline uint32_t wuffs_base__image_config__height( wuffs_base__image_config* c) { - return wuffs_base__image_config__valid(c) ? c->private_impl.h : 0; + return wuffs_base__image_config__is_valid(c) ? c->private_impl.height : 0; +} + +static inline uint32_t wuffs_base__image_config__num_loops( + wuffs_base__image_config* c) { + return wuffs_base__image_config__is_valid(c) ? c->private_impl.num_loops : 0; } // TODO: this is the right API for planar (not packed) pixbufs? Should it allow @@ -153,26 +482,96 @@ static inline uint32_t wuffs_base__image_config__height( // example, decoding a JPEG image straight to RGBA instead of to YCbCr? static inline size_t wuffs_base__image_config__pixbuf_size( wuffs_base__image_config* c) { - if (wuffs_base__image_config__valid(c)) { - uint64_t wh = ((uint64_t)c->private_impl.w) * ((uint64_t)c->private_impl.h); + if (wuffs_base__image_config__is_valid(c)) { + uint64_t wh = + ((uint64_t)c->private_impl.width) * ((uint64_t)c->private_impl.height); // TODO: handle things other than 1 byte per pixel. return (size_t)wh; } return 0; } -static inline void wuffs_base__image_config__initialize( - wuffs_base__image_config* c, - uint32_t width, - uint32_t height, - uint32_t TODO_color_model) { - if (!c) { +typedef struct { + // Do not access the private_impl's fields directly. There is no API/ABI + // compatibility or safety guarantee if you do so. + struct { + wuffs_base__image_config config; + uint32_t loop_count; // 0-based count of the current loop. + wuffs_base__pixel_buffer pixbuf; + // TODO: color spaces. + wuffs_base__rectangle dirty_rect; + wuffs_base__flicks duration; + uint8_t palette[1024]; + } private_impl; +} wuffs_base__image_buffer; + +static inline void wuffs_base__image_buffer__initialize( + wuffs_base__image_buffer* f, + wuffs_base__image_config config, + wuffs_base__pixel_buffer pixbuf) { + if (!f) { return; } - c->private_impl.flags = 1; - c->private_impl.w = width; - c->private_impl.h = height; - // TODO: color model. + *f = ((wuffs_base__image_buffer){}); + f->private_impl.config = config; + f->private_impl.pixbuf = pixbuf; +} + +static inline void wuffs_base__image_buffer__update( + wuffs_base__image_buffer* f, + wuffs_base__rectangle dirty_rect, + wuffs_base__flicks duration, + uint8_t* palette_ptr, + size_t palette_len) { + if (!f) { + return; + } + f->private_impl.dirty_rect = dirty_rect; + f->private_impl.duration = duration; + if (palette_ptr) { + memmove(f->private_impl.palette, palette_ptr, + palette_len <= 1024 ? palette_len : 1024); + } +} + +// wuffs_base__image_buffer__loop returns whether the image decoder should loop +// back to the beginning of the animation, assuming that we've reached the end +// of the encoded stream. If so, it increments f's count of the animation loops +// played so far. +static inline bool wuffs_base__image_buffer__loop(wuffs_base__image_buffer* f) { + if (!f) { + return false; + } + uint32_t n = f->private_impl.config.private_impl.num_loops; + if (n == 0) { + return true; + } + if (f->private_impl.loop_count < n - 1) { + f->private_impl.loop_count++; + return true; + } + return false; +} + +// wuffs_base__image_buffer__dirty_rect returns an upper bound for what part of +// this frame's pixels differs from the previous frame. +static inline wuffs_base__rectangle wuffs_base__image_buffer__dirty_rect( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.dirty_rect : ((wuffs_base__rectangle){0}); +} + +// wuffs_base__image_buffer__duration returns the amount of time to display +// this frame. Zero means to display forever - a still (non-animated) image. +static inline wuffs_base__flicks wuffs_base__image_buffer__duration( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.duration : 0; +} + +// wuffs_base__image_buffer__palette returns the palette that the pixel data +// can index. The backing array is inside f and has length 1024. +static inline uint8_t* wuffs_base__image_buffer__palette( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.palette : NULL; } #endif // WUFFS_BASE_HEADER_H @@ -194,14 +593,14 @@ extern "C" { // ---------------- Status Codes -// Status codes are int32_t values: -// - the sign bit indicates a non-recoverable status code: an error -// - bits 10-30 hold the packageid: a namespace -// - bits 8-9 are reserved -// - bits 0-7 are a package-namespaced numeric code +// Status codes are int32_t values. Its bits: +// - bit 31 (the sign bit) indicates unrecoverable-ness: an error. +// - bits 30 .. 10 are the packageid: a namespace. +// - bits 9 .. 8 are reserved. +// - bits 7 .. 0 are a package-namespaced numeric code. // -// Do not manipulate these bits directly. Use the API functions such as -// wuffs_deflate__status__is_error instead. +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_deflate__status__is_error instead. typedef int32_t wuffs_deflate__status; #define wuffs_deflate__packageid 848533 // 0x000CF295 @@ -367,14 +766,14 @@ extern "C" { // ---------------- Status Codes -// Status codes are int32_t values: -// - the sign bit indicates a non-recoverable status code: an error -// - bits 10-30 hold the packageid: a namespace -// - bits 8-9 are reserved -// - bits 0-7 are a package-namespaced numeric code +// Status codes are int32_t values. Its bits: +// - bit 31 (the sign bit) indicates unrecoverable-ness: an error. +// - bits 30 .. 10 are the packageid: a namespace. +// - bits 9 .. 8 are reserved. +// - bits 7 .. 0 are a package-namespaced numeric code. // -// Do not manipulate these bits directly. Use the API functions such as -// wuffs_zlib__status__is_error instead. +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_zlib__status__is_error instead. typedef int32_t wuffs_zlib__status; #define wuffs_zlib__packageid 2064249 // 0x001F7F79 @@ -512,14 +911,14 @@ typedef struct { // but it should catch 99.99% of cases. // // Its (non-zero) value is arbitrary, based on md5sum("wuffs"). -#define WUFFS_BASE__MAGIC (0x3CCB6C71U) +#define WUFFS_BASE__MAGIC 0x3CCB6C71U // WUFFS_BASE__ALREADY_ZEROED is passed from a container struct's initializer // to a containee struct's initializer when the container has already zeroed // the containee's memory. // // Its (non-zero) value is arbitrary, based on md5sum("zeroed"). -#define WUFFS_BASE__ALREADY_ZEROED (0x68602EF1U) +#define WUFFS_BASE__ALREADY_ZEROED 0x68602EF1U // Denote intentional fallthroughs for -Wimplicit-fallthrough. // diff --git a/gen/h/std/crc32.h b/gen/h/std/crc32.h index 7805e9f3..6a4359ae 100644 --- a/gen/h/std/crc32.h +++ b/gen/h/std/crc32.h @@ -46,11 +46,40 @@ // For now, the API and ABI are simply unstable and can change at any time. // // TODO: don't hard code this in base-header.h. -#define WUFFS_VERSION (0x00001) +#define WUFFS_VERSION 0x00001 -// ---------------- I/O +// ---------------- Fundamentals -// wuffs_base__slice_u8 is a 1-dimensional buffer (a pointer and length). +// Flicks are a unit of time. One flick (frame-tick) is 1 / 705_600_000 of a +// second. See https://github.com/OculusVR/Flicks +typedef uint64_t wuffs_base__flicks; + +#define WUFFS_BASE__FLICKS_PER_SECOND 705600000ULL + +// wuffs_base__rectangle is a rectangle on the integer grid. It contains all +// points (x, y) such that ((min_x <= x) && (x < max_x)) and likewise for y. It +// is therefore empty if min_x >= max_x. There are multiple representations of +// an empty rectangle. +// +// A value with all fields zero is a valid, empty rectangle. +// +// The X and Y axes increase right and down. +typedef struct { + uint32_t min_x; + uint32_t min_y; + uint32_t max_x; + uint32_t max_y; +} wuffs_base__rectangle; + +static inline uint32_t wuffs_base__rectangle__width(wuffs_base__rectangle r) { + return r.max_x > r.min_x ? r.max_x - r.min_x : 0; +} + +static inline uint32_t wuffs_base__rectangle__height(wuffs_base__rectangle r) { + return r.max_y > r.min_y ? r.max_y - r.min_y : 0; +} + +// wuffs_base__slice_u8 is a 1-dimensional buffer. // // A value with all fields NULL or zero is a valid, empty slice. typedef struct { @@ -58,6 +87,20 @@ typedef struct { size_t len; } wuffs_base__slice_u8; +// wuffs_base__table_u8 is a 2-dimensional buffer. +// +// A value with all fields NULL or zero is a valid, empty table. +typedef struct { + uint8_t* ptr; + size_t width; + size_t height; + size_t stride; +} wuffs_base__table_u8; + +// ---------------- I/O + +// TODO: rename buf1 to io_buffer, writer1 to io_writer, etc. + // wuffs_base__buf1 is a 1-dimensional buffer (a pointer and length), plus // additional indexes into that buffer, plus an opened / closed flag. // @@ -109,18 +152,286 @@ typedef struct { // ---------------- Images +// wuffs_base__pixel_format encodes the format of the bytes that constitute an +// image frame's pixel data. Its bits: +// - bit 31 is reserved. +// - bits 30 .. 28 encodes color (and channel order, in terms of memory). +// - bits 27 .. 26 are reserved. +// - bits 25 .. 24 encodes transparency. +// - bit 23 indicates big-endian/MSB-first (as opposed to little/LSB). +// - bit 22 indicates floating point (as opposed to integer). +// - bits 21 .. 20 are the number of planes, minus 1. Zero means packed. +// - bits 19 .. 16 encodes the number of bits (depth) in an index value. +// Zero means direct, not palette-indexed. +// - bits 15 .. 12 encodes the number of bits (depth) in the 3rd channel. +// - bits 11 .. 8 encodes the number of bits (depth) in the 2nd channel. +// - bits 7 .. 4 encodes the number of bits (depth) in the 1st channel. +// - bits 3 .. 0 encodes the number of bits (depth) in the 0th channel. +// +// The bit fields of a wuffs_base__pixel_format are not independent. For +// example, the number of planes should not be greater than the number of +// channels. Similarly, bits 15..4 are unused (and should be zero) if bits +// 31..24 (color and transparency) together imply only 1 channel (gray, no +// alpha) and floating point samples should mean a bit depth of 16, 32 or 64. +// +// Formats hold between 1 and 4 channels. For example: Y (1 channel: gray), YA +// (2 channels: gray and alpha), BGR (3 channels: blue, green, red) or CMYK (4 +// channels: cyan, magenta, yellow, black). +// +// For direct formats with N > 1 channels, those channels can be laid out in +// either 1 (packed) or N (planar) planes. For example, RGBA data is usually +// packed, but YUV data is usually planar, due to chroma subsampling (for +// details, see the wuffs_base__pixel_subsampling type). For indexed formats, +// the palette (always 256 × 4 bytes) holds up to 4 packed bytes of color data +// per index value, and there is only 1 plane (for the index). The distance +// between successive palette elements is always 4 bytes. +// +// The color field is encoded in 3 bits: +// - 0 means A (Alpha). +// - 1 means Y or YA (Gray, Alpha). +// - 2 means BGR, BGRX or BGRA (Blue, Green, Red, X-padding or Alpha). +// - 3 means RGB, RGBX or RGBA (Red, Green, Blue, X-padding or Alpha). +// - 4 means YUV or YUVA (Luma, Chroma-blue, Chroma-red, Alpha). +// - 5 means CMY or CMYK (Cyan, Magenta, Yellow, Black). +// - all other values are reserved. +// +// In Wuffs, channels are given in memory order, regardless of endianness, +// since the C type for the pixel data is an array of bytes, not an array of +// uint32_t. For example, packed BGRA with 8 bits per channel means that the +// bytes in memory are always Blue, Green, Red then Alpha. On big-endian +// systems, that is the uint32_t 0xBBGGRRAA. On little-endian, 0xAARRGGBB. +// +// When the color field (3 bits) encodes multiple options, the transparency +// field (2 bits) distinguishes them: +// - 0 means fully opaque, no extra channels +// - 1 means fully opaque, one extra channel (X or K, padding or black). +// - 2 means one extra alpha channel, other channels are non-premultiplied. +// - 3 means one extra alpha channel, other channels are premultiplied. +// +// The zero wuffs_base__pixel_format value is an invalid pixel format, as it is +// invalid to combine the zero color (alpha only) with the zero transparency. +// +// Bit depth is encoded in 4 bits: +// - 0 means the channel or index is unused. +// - x means a bit depth of x, for x in the range 1..8. +// - 9 means a bit depth of 10. +// - 10 means a bit depth of 12. +// - 11 means a bit depth of 16. +// - 12 means a bit depth of 24. +// - 13 means a bit depth of 32. +// - 14 means a bit depth of 48. +// - 15 means a bit depth of 64. +// +// For example, wuffs_base__pixel_format 0x3280BBBB is a natural format for +// decoding a PNG image - network byte order (also known as big-endian), +// packed, non-premultiplied alpha - that happens to be 16-bit-depth truecolor +// with alpha (RGBA). In memory order: +// +// ptr+0 ptr+1 ptr+2 ptr+3 ptr+4 ptr+5 ptr+6 ptr+7 +// Rhi Rlo Ghi Glo Bhi Blo Ahi Alo +// +// For example, the value wuffs_base__pixel_format 0x20000565 means BGR with no +// alpha or padding, 5/6/5 bits for blue/green/red, packed 2 bytes per pixel, +// laid out LSB-first in memory order: +// +// ptr+0........... ptr+1........... +// MSB LSB MSB LSB +// G₂G₁G₀B₄B₃B₂B₁B₀ R₄R₃R₂R₁R₀G₅G₄G₃ +// +// On little-endian systems (but not big-endian), this Wuffs pixel format value +// (0x20000565) corresponds to the Cairo library's CAIRO_FORMAT_RGB16_565, the +// SDL2 (Simple DirectMedia Layer 2) library's SDL_PIXELFORMAT_RGB565 and the +// Skia library's kRGB_565_SkColorType. Note BGR in Wuffs versus RGB in the +// other libraries. +// +// Regardless of endianness, this Wuffs pixel format value (0x20000565) +// corresponds to the V4L2 (Video For Linux 2) library's V4L2_PIX_FMT_RGB565 +// and the Wayland-DRM library's WL_DRM_FORMAT_RGB565. +// +// Different software libraries name their pixel formats (and especially their +// channel order) either according to memory layout or as bits of a native +// integer type like uint32_t. The two conventions differ because of a system's +// endianness. As mentioned earlier, Wuffs pixel formats are always in memory +// order. More detail of other software libraries' naming conventions is in the +// Pixel Format Guide at https://afrantzis.github.io/pixel-format-guide/ +// +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_base__pixel_format__num_planes instead. +typedef uint32_t wuffs_base__pixel_format; + +// Common 8-bit-depth pixel formats. This list is not exhaustive; not all valid +// wuffs_base__pixel_format values are present. + +#define WUFFS_BASE__PIXEL_FORMAT__INVALID ((wuffs_base__pixel_format)0x00000000) + +#define WUFFS_BASE__PIXEL_FORMAT__A ((wuffs_base__pixel_format)0x02000008) + +#define WUFFS_BASE__PIXEL_FORMAT__Y ((wuffs_base__pixel_format)0x10000008) +#define WUFFS_BASE__PIXEL_FORMAT__YA_NONPREMUL \ + ((wuffs_base__pixel_format)0x12000008) +#define WUFFS_BASE__PIXEL_FORMAT__YA_PREMUL \ + ((wuffs_base__pixel_format)0x13000008) + +#define WUFFS_BASE__PIXEL_FORMAT__BGR ((wuffs_base__pixel_format)0x20000888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRX ((wuffs_base__pixel_format)0x21008888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRX_INDEXED \ + ((wuffs_base__pixel_format)0x21088888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_NONPREMUL \ + ((wuffs_base__pixel_format)0x22008888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_NONPREMUL_INDEXED \ + ((wuffs_base__pixel_format)0x22088888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_PREMUL \ + ((wuffs_base__pixel_format)0x23008888) + +#define WUFFS_BASE__PIXEL_FORMAT__RGB ((wuffs_base__pixel_format)0x30000888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBX ((wuffs_base__pixel_format)0x31008888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBX_INDEXED \ + ((wuffs_base__pixel_format)0x31088888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_NONPREMUL \ + ((wuffs_base__pixel_format)0x32008888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_NONPREMUL_INDEXED \ + ((wuffs_base__pixel_format)0x32088888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_PREMUL \ + ((wuffs_base__pixel_format)0x33008888) + +#define WUFFS_BASE__PIXEL_FORMAT__YUV ((wuffs_base__pixel_format)0x40200888) +#define WUFFS_BASE__PIXEL_FORMAT__YUVK ((wuffs_base__pixel_format)0x41308888) +#define WUFFS_BASE__PIXEL_FORMAT__YUVA_NONPREMUL \ + ((wuffs_base__pixel_format)0x42308888) + +#define WUFFS_BASE__PIXEL_FORMAT__CMY ((wuffs_base__pixel_format)0x50200888) +#define WUFFS_BASE__PIXEL_FORMAT__CMYK ((wuffs_base__pixel_format)0x51308888) + +static inline bool wuffs_base__pixel_format__is_valid( + wuffs_base__pixel_format f) { + return f != 0; +} + +static inline bool wuffs_base__pixel_format__is_indexed( + wuffs_base__pixel_format f) { + return ((f >> 16) & 0x0F) != 0; +} + +#define WUFFS_BASE__PIXEL_FORMAT__NUM_PLANES_MAX 4 + +static inline uint32_t wuffs_base__pixel_format__num_planes( + wuffs_base__pixel_format f) { + return f ? (((f >> 20) & 0x03) + 1) : 0; +} + +typedef struct { + wuffs_base__table_u8 planes[WUFFS_BASE__PIXEL_FORMAT__NUM_PLANES_MAX]; +} wuffs_base__pixel_buffer; + +// wuffs_base__pixel_subsampling encodes the mapping of pixel space coordinates +// (x, y) to pixel buffer indices (i, j). That mapping can differ for each +// plane p. For a depth of 8 bits (1 byte), the p'th plane's sample starts at +// (planes[p].ptr + (j * planes[p].stride) + i). +// +// For packed pixel formats, the mapping is trivial: i = x and j = y. For +// planar pixel formats, the mapping can differ due to chroma subsampling. For +// example, consider a three plane YUV pixel format with 4:2:2 subsampling. For +// the luma (Y) channel, there is one sample for every pixel, but for the +// chroma (U, V) channels, there is one sample for every two pixels: pairs of +// horizontally adjacent pixels form one macropixel, i = x / 2 and j == y. In +// general, for a given p: +// - i = (x + bias_x) >> shift_x. +// - j = (y + bias_y) >> shift_y. +// where biases and shifts are in the range 0..3 and 0..2 respectively. +// +// In general, the biases will be zero after decoding an image. However, making +// a sub-image may change the bias, since the (x, y) coordinates are relative +// to the sub-image's top-left origin, but the backing pixel buffers were +// created relative to the original image's origin. +// +// For each plane p, each of those four numbers (biases and shifts) are encoded +// in two bits, which combine to form an 8 bit unsigned integer: +// +// e_p = (bias_x << 6) | (shift_x << 4) | (bias_y << 2) | (shift_y << 0) +// +// Those e_p values (e_0 for the first plane, e_1 for the second plane, etc) +// combine to form a wuffs_base__pixel_subsampling value: +// +// pixsub = (e_3 << 24) | (e_2 << 16) | (e_1 << 8) | (e_0 << 0) +// +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_base__pixel_subsampling__bias_x instead. +typedef uint32_t wuffs_base__pixel_subsampling; + +#define WUFFS_BASE__PIXEL_SUBSAMPLING__NONE ((wuffs_base__pixel_subsampling)0); + +#define WUFFS_BASE__PIXEL_SUBSAMPLING__444 \ + ((wuffs_base__pixel_subsampling)0x000000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__440 \ + ((wuffs_base__pixel_subsampling)0x010100); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__422 \ + ((wuffs_base__pixel_subsampling)0x101000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__420 \ + ((wuffs_base__pixel_subsampling)0x111100); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__411 \ + ((wuffs_base__pixel_subsampling)0x202000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__410 \ + ((wuffs_base__pixel_subsampling)0x212100); + +static inline uint32_t wuffs_base__pixel_subsampling__bias_x( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 6; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__shift_x( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 4; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__bias_y( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 2; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__shift_y( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 0; + return (s >> shift) & 0x03; +} + typedef struct { // Do not access the private_impl's fields directly. There is no API/ABI // compatibility or safety guarantee if you do so. struct { - uint32_t flags; - uint32_t w; - uint32_t h; - // TODO: color model, including both packed RGBA and planar, - // chroma-subsampled YCbCr. + wuffs_base__pixel_format pixfmt; + wuffs_base__pixel_subsampling pixsub; + uint32_t width; + uint32_t height; + uint32_t num_loops; } private_impl; } wuffs_base__image_config; +static inline void wuffs_base__image_config__initialize( + wuffs_base__image_config* c, + wuffs_base__pixel_format pixfmt, + wuffs_base__pixel_subsampling pixsub, + uint32_t width, + uint32_t height, + uint32_t num_loops) { + if (!c) { + return; + } + // TODO: move the check from wuffs_base__image_config__is_valid here. Should + // this function return bool? An error type? + c->private_impl.pixfmt = pixfmt; + c->private_impl.pixsub = pixsub; + c->private_impl.width = width; + c->private_impl.height = height; + c->private_impl.num_loops = num_loops; +} + static inline void wuffs_base__image_config__invalidate( wuffs_base__image_config* c) { if (c) { @@ -128,24 +439,42 @@ static inline void wuffs_base__image_config__invalidate( } } -static inline bool wuffs_base__image_config__valid( +static inline bool wuffs_base__image_config__is_valid( wuffs_base__image_config* c) { - if (!c || !(c->private_impl.flags & 1)) { + if (!c || !c->private_impl.pixfmt) { return false; } - uint64_t wh = ((uint64_t)c->private_impl.w) * ((uint64_t)c->private_impl.h); + uint64_t wh = + ((uint64_t)c->private_impl.width) * ((uint64_t)c->private_impl.height); // TODO: handle things other than 1 byte per pixel. + // + // TODO: move the check to wuffs_base__image_config__initialize. return wh <= ((uint64_t)SIZE_MAX); } +static inline wuffs_base__pixel_format wuffs_base__image_config__pixel_format( + wuffs_base__image_config* c) { + return c ? c->private_impl.pixfmt : 0; +} + +static inline wuffs_base__pixel_subsampling +wuffs_base__image_config__pixel_subsampling(wuffs_base__image_config* c) { + return wuffs_base__image_config__is_valid(c) ? c->private_impl.pixsub : 0; +} + static inline uint32_t wuffs_base__image_config__width( wuffs_base__image_config* c) { - return wuffs_base__image_config__valid(c) ? c->private_impl.w : 0; + return wuffs_base__image_config__is_valid(c) ? c->private_impl.width : 0; } static inline uint32_t wuffs_base__image_config__height( wuffs_base__image_config* c) { - return wuffs_base__image_config__valid(c) ? c->private_impl.h : 0; + return wuffs_base__image_config__is_valid(c) ? c->private_impl.height : 0; +} + +static inline uint32_t wuffs_base__image_config__num_loops( + wuffs_base__image_config* c) { + return wuffs_base__image_config__is_valid(c) ? c->private_impl.num_loops : 0; } // TODO: this is the right API for planar (not packed) pixbufs? Should it allow @@ -153,26 +482,96 @@ static inline uint32_t wuffs_base__image_config__height( // example, decoding a JPEG image straight to RGBA instead of to YCbCr? static inline size_t wuffs_base__image_config__pixbuf_size( wuffs_base__image_config* c) { - if (wuffs_base__image_config__valid(c)) { - uint64_t wh = ((uint64_t)c->private_impl.w) * ((uint64_t)c->private_impl.h); + if (wuffs_base__image_config__is_valid(c)) { + uint64_t wh = + ((uint64_t)c->private_impl.width) * ((uint64_t)c->private_impl.height); // TODO: handle things other than 1 byte per pixel. return (size_t)wh; } return 0; } -static inline void wuffs_base__image_config__initialize( - wuffs_base__image_config* c, - uint32_t width, - uint32_t height, - uint32_t TODO_color_model) { - if (!c) { +typedef struct { + // Do not access the private_impl's fields directly. There is no API/ABI + // compatibility or safety guarantee if you do so. + struct { + wuffs_base__image_config config; + uint32_t loop_count; // 0-based count of the current loop. + wuffs_base__pixel_buffer pixbuf; + // TODO: color spaces. + wuffs_base__rectangle dirty_rect; + wuffs_base__flicks duration; + uint8_t palette[1024]; + } private_impl; +} wuffs_base__image_buffer; + +static inline void wuffs_base__image_buffer__initialize( + wuffs_base__image_buffer* f, + wuffs_base__image_config config, + wuffs_base__pixel_buffer pixbuf) { + if (!f) { return; } - c->private_impl.flags = 1; - c->private_impl.w = width; - c->private_impl.h = height; - // TODO: color model. + *f = ((wuffs_base__image_buffer){}); + f->private_impl.config = config; + f->private_impl.pixbuf = pixbuf; +} + +static inline void wuffs_base__image_buffer__update( + wuffs_base__image_buffer* f, + wuffs_base__rectangle dirty_rect, + wuffs_base__flicks duration, + uint8_t* palette_ptr, + size_t palette_len) { + if (!f) { + return; + } + f->private_impl.dirty_rect = dirty_rect; + f->private_impl.duration = duration; + if (palette_ptr) { + memmove(f->private_impl.palette, palette_ptr, + palette_len <= 1024 ? palette_len : 1024); + } +} + +// wuffs_base__image_buffer__loop returns whether the image decoder should loop +// back to the beginning of the animation, assuming that we've reached the end +// of the encoded stream. If so, it increments f's count of the animation loops +// played so far. +static inline bool wuffs_base__image_buffer__loop(wuffs_base__image_buffer* f) { + if (!f) { + return false; + } + uint32_t n = f->private_impl.config.private_impl.num_loops; + if (n == 0) { + return true; + } + if (f->private_impl.loop_count < n - 1) { + f->private_impl.loop_count++; + return true; + } + return false; +} + +// wuffs_base__image_buffer__dirty_rect returns an upper bound for what part of +// this frame's pixels differs from the previous frame. +static inline wuffs_base__rectangle wuffs_base__image_buffer__dirty_rect( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.dirty_rect : ((wuffs_base__rectangle){0}); +} + +// wuffs_base__image_buffer__duration returns the amount of time to display +// this frame. Zero means to display forever - a still (non-animated) image. +static inline wuffs_base__flicks wuffs_base__image_buffer__duration( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.duration : 0; +} + +// wuffs_base__image_buffer__palette returns the palette that the pixel data +// can index. The backing array is inside f and has length 1024. +static inline uint8_t* wuffs_base__image_buffer__palette( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.palette : NULL; } #endif // WUFFS_BASE_HEADER_H @@ -185,14 +584,14 @@ extern "C" { // ---------------- Status Codes -// Status codes are int32_t values: -// - the sign bit indicates a non-recoverable status code: an error -// - bits 10-30 hold the packageid: a namespace -// - bits 8-9 are reserved -// - bits 0-7 are a package-namespaced numeric code +// Status codes are int32_t values. Its bits: +// - bit 31 (the sign bit) indicates unrecoverable-ness: an error. +// - bits 30 .. 10 are the packageid: a namespace. +// - bits 9 .. 8 are reserved. +// - bits 7 .. 0 are a package-namespaced numeric code. // -// Do not manipulate these bits directly. Use the API functions such as -// wuffs_crc32__status__is_error instead. +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_crc32__status__is_error instead. typedef int32_t wuffs_crc32__status; #define wuffs_crc32__packageid 810620 // 0x000C5E7C diff --git a/gen/h/std/deflate.h b/gen/h/std/deflate.h index bd05270a..a4e5fc98 100644 --- a/gen/h/std/deflate.h +++ b/gen/h/std/deflate.h @@ -46,11 +46,40 @@ // For now, the API and ABI are simply unstable and can change at any time. // // TODO: don't hard code this in base-header.h. -#define WUFFS_VERSION (0x00001) +#define WUFFS_VERSION 0x00001 -// ---------------- I/O +// ---------------- Fundamentals -// wuffs_base__slice_u8 is a 1-dimensional buffer (a pointer and length). +// Flicks are a unit of time. One flick (frame-tick) is 1 / 705_600_000 of a +// second. See https://github.com/OculusVR/Flicks +typedef uint64_t wuffs_base__flicks; + +#define WUFFS_BASE__FLICKS_PER_SECOND 705600000ULL + +// wuffs_base__rectangle is a rectangle on the integer grid. It contains all +// points (x, y) such that ((min_x <= x) && (x < max_x)) and likewise for y. It +// is therefore empty if min_x >= max_x. There are multiple representations of +// an empty rectangle. +// +// A value with all fields zero is a valid, empty rectangle. +// +// The X and Y axes increase right and down. +typedef struct { + uint32_t min_x; + uint32_t min_y; + uint32_t max_x; + uint32_t max_y; +} wuffs_base__rectangle; + +static inline uint32_t wuffs_base__rectangle__width(wuffs_base__rectangle r) { + return r.max_x > r.min_x ? r.max_x - r.min_x : 0; +} + +static inline uint32_t wuffs_base__rectangle__height(wuffs_base__rectangle r) { + return r.max_y > r.min_y ? r.max_y - r.min_y : 0; +} + +// wuffs_base__slice_u8 is a 1-dimensional buffer. // // A value with all fields NULL or zero is a valid, empty slice. typedef struct { @@ -58,6 +87,20 @@ typedef struct { size_t len; } wuffs_base__slice_u8; +// wuffs_base__table_u8 is a 2-dimensional buffer. +// +// A value with all fields NULL or zero is a valid, empty table. +typedef struct { + uint8_t* ptr; + size_t width; + size_t height; + size_t stride; +} wuffs_base__table_u8; + +// ---------------- I/O + +// TODO: rename buf1 to io_buffer, writer1 to io_writer, etc. + // wuffs_base__buf1 is a 1-dimensional buffer (a pointer and length), plus // additional indexes into that buffer, plus an opened / closed flag. // @@ -109,18 +152,286 @@ typedef struct { // ---------------- Images +// wuffs_base__pixel_format encodes the format of the bytes that constitute an +// image frame's pixel data. Its bits: +// - bit 31 is reserved. +// - bits 30 .. 28 encodes color (and channel order, in terms of memory). +// - bits 27 .. 26 are reserved. +// - bits 25 .. 24 encodes transparency. +// - bit 23 indicates big-endian/MSB-first (as opposed to little/LSB). +// - bit 22 indicates floating point (as opposed to integer). +// - bits 21 .. 20 are the number of planes, minus 1. Zero means packed. +// - bits 19 .. 16 encodes the number of bits (depth) in an index value. +// Zero means direct, not palette-indexed. +// - bits 15 .. 12 encodes the number of bits (depth) in the 3rd channel. +// - bits 11 .. 8 encodes the number of bits (depth) in the 2nd channel. +// - bits 7 .. 4 encodes the number of bits (depth) in the 1st channel. +// - bits 3 .. 0 encodes the number of bits (depth) in the 0th channel. +// +// The bit fields of a wuffs_base__pixel_format are not independent. For +// example, the number of planes should not be greater than the number of +// channels. Similarly, bits 15..4 are unused (and should be zero) if bits +// 31..24 (color and transparency) together imply only 1 channel (gray, no +// alpha) and floating point samples should mean a bit depth of 16, 32 or 64. +// +// Formats hold between 1 and 4 channels. For example: Y (1 channel: gray), YA +// (2 channels: gray and alpha), BGR (3 channels: blue, green, red) or CMYK (4 +// channels: cyan, magenta, yellow, black). +// +// For direct formats with N > 1 channels, those channels can be laid out in +// either 1 (packed) or N (planar) planes. For example, RGBA data is usually +// packed, but YUV data is usually planar, due to chroma subsampling (for +// details, see the wuffs_base__pixel_subsampling type). For indexed formats, +// the palette (always 256 × 4 bytes) holds up to 4 packed bytes of color data +// per index value, and there is only 1 plane (for the index). The distance +// between successive palette elements is always 4 bytes. +// +// The color field is encoded in 3 bits: +// - 0 means A (Alpha). +// - 1 means Y or YA (Gray, Alpha). +// - 2 means BGR, BGRX or BGRA (Blue, Green, Red, X-padding or Alpha). +// - 3 means RGB, RGBX or RGBA (Red, Green, Blue, X-padding or Alpha). +// - 4 means YUV or YUVA (Luma, Chroma-blue, Chroma-red, Alpha). +// - 5 means CMY or CMYK (Cyan, Magenta, Yellow, Black). +// - all other values are reserved. +// +// In Wuffs, channels are given in memory order, regardless of endianness, +// since the C type for the pixel data is an array of bytes, not an array of +// uint32_t. For example, packed BGRA with 8 bits per channel means that the +// bytes in memory are always Blue, Green, Red then Alpha. On big-endian +// systems, that is the uint32_t 0xBBGGRRAA. On little-endian, 0xAARRGGBB. +// +// When the color field (3 bits) encodes multiple options, the transparency +// field (2 bits) distinguishes them: +// - 0 means fully opaque, no extra channels +// - 1 means fully opaque, one extra channel (X or K, padding or black). +// - 2 means one extra alpha channel, other channels are non-premultiplied. +// - 3 means one extra alpha channel, other channels are premultiplied. +// +// The zero wuffs_base__pixel_format value is an invalid pixel format, as it is +// invalid to combine the zero color (alpha only) with the zero transparency. +// +// Bit depth is encoded in 4 bits: +// - 0 means the channel or index is unused. +// - x means a bit depth of x, for x in the range 1..8. +// - 9 means a bit depth of 10. +// - 10 means a bit depth of 12. +// - 11 means a bit depth of 16. +// - 12 means a bit depth of 24. +// - 13 means a bit depth of 32. +// - 14 means a bit depth of 48. +// - 15 means a bit depth of 64. +// +// For example, wuffs_base__pixel_format 0x3280BBBB is a natural format for +// decoding a PNG image - network byte order (also known as big-endian), +// packed, non-premultiplied alpha - that happens to be 16-bit-depth truecolor +// with alpha (RGBA). In memory order: +// +// ptr+0 ptr+1 ptr+2 ptr+3 ptr+4 ptr+5 ptr+6 ptr+7 +// Rhi Rlo Ghi Glo Bhi Blo Ahi Alo +// +// For example, the value wuffs_base__pixel_format 0x20000565 means BGR with no +// alpha or padding, 5/6/5 bits for blue/green/red, packed 2 bytes per pixel, +// laid out LSB-first in memory order: +// +// ptr+0........... ptr+1........... +// MSB LSB MSB LSB +// G₂G₁G₀B₄B₃B₂B₁B₀ R₄R₃R₂R₁R₀G₅G₄G₃ +// +// On little-endian systems (but not big-endian), this Wuffs pixel format value +// (0x20000565) corresponds to the Cairo library's CAIRO_FORMAT_RGB16_565, the +// SDL2 (Simple DirectMedia Layer 2) library's SDL_PIXELFORMAT_RGB565 and the +// Skia library's kRGB_565_SkColorType. Note BGR in Wuffs versus RGB in the +// other libraries. +// +// Regardless of endianness, this Wuffs pixel format value (0x20000565) +// corresponds to the V4L2 (Video For Linux 2) library's V4L2_PIX_FMT_RGB565 +// and the Wayland-DRM library's WL_DRM_FORMAT_RGB565. +// +// Different software libraries name their pixel formats (and especially their +// channel order) either according to memory layout or as bits of a native +// integer type like uint32_t. The two conventions differ because of a system's +// endianness. As mentioned earlier, Wuffs pixel formats are always in memory +// order. More detail of other software libraries' naming conventions is in the +// Pixel Format Guide at https://afrantzis.github.io/pixel-format-guide/ +// +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_base__pixel_format__num_planes instead. +typedef uint32_t wuffs_base__pixel_format; + +// Common 8-bit-depth pixel formats. This list is not exhaustive; not all valid +// wuffs_base__pixel_format values are present. + +#define WUFFS_BASE__PIXEL_FORMAT__INVALID ((wuffs_base__pixel_format)0x00000000) + +#define WUFFS_BASE__PIXEL_FORMAT__A ((wuffs_base__pixel_format)0x02000008) + +#define WUFFS_BASE__PIXEL_FORMAT__Y ((wuffs_base__pixel_format)0x10000008) +#define WUFFS_BASE__PIXEL_FORMAT__YA_NONPREMUL \ + ((wuffs_base__pixel_format)0x12000008) +#define WUFFS_BASE__PIXEL_FORMAT__YA_PREMUL \ + ((wuffs_base__pixel_format)0x13000008) + +#define WUFFS_BASE__PIXEL_FORMAT__BGR ((wuffs_base__pixel_format)0x20000888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRX ((wuffs_base__pixel_format)0x21008888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRX_INDEXED \ + ((wuffs_base__pixel_format)0x21088888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_NONPREMUL \ + ((wuffs_base__pixel_format)0x22008888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_NONPREMUL_INDEXED \ + ((wuffs_base__pixel_format)0x22088888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_PREMUL \ + ((wuffs_base__pixel_format)0x23008888) + +#define WUFFS_BASE__PIXEL_FORMAT__RGB ((wuffs_base__pixel_format)0x30000888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBX ((wuffs_base__pixel_format)0x31008888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBX_INDEXED \ + ((wuffs_base__pixel_format)0x31088888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_NONPREMUL \ + ((wuffs_base__pixel_format)0x32008888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_NONPREMUL_INDEXED \ + ((wuffs_base__pixel_format)0x32088888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_PREMUL \ + ((wuffs_base__pixel_format)0x33008888) + +#define WUFFS_BASE__PIXEL_FORMAT__YUV ((wuffs_base__pixel_format)0x40200888) +#define WUFFS_BASE__PIXEL_FORMAT__YUVK ((wuffs_base__pixel_format)0x41308888) +#define WUFFS_BASE__PIXEL_FORMAT__YUVA_NONPREMUL \ + ((wuffs_base__pixel_format)0x42308888) + +#define WUFFS_BASE__PIXEL_FORMAT__CMY ((wuffs_base__pixel_format)0x50200888) +#define WUFFS_BASE__PIXEL_FORMAT__CMYK ((wuffs_base__pixel_format)0x51308888) + +static inline bool wuffs_base__pixel_format__is_valid( + wuffs_base__pixel_format f) { + return f != 0; +} + +static inline bool wuffs_base__pixel_format__is_indexed( + wuffs_base__pixel_format f) { + return ((f >> 16) & 0x0F) != 0; +} + +#define WUFFS_BASE__PIXEL_FORMAT__NUM_PLANES_MAX 4 + +static inline uint32_t wuffs_base__pixel_format__num_planes( + wuffs_base__pixel_format f) { + return f ? (((f >> 20) & 0x03) + 1) : 0; +} + +typedef struct { + wuffs_base__table_u8 planes[WUFFS_BASE__PIXEL_FORMAT__NUM_PLANES_MAX]; +} wuffs_base__pixel_buffer; + +// wuffs_base__pixel_subsampling encodes the mapping of pixel space coordinates +// (x, y) to pixel buffer indices (i, j). That mapping can differ for each +// plane p. For a depth of 8 bits (1 byte), the p'th plane's sample starts at +// (planes[p].ptr + (j * planes[p].stride) + i). +// +// For packed pixel formats, the mapping is trivial: i = x and j = y. For +// planar pixel formats, the mapping can differ due to chroma subsampling. For +// example, consider a three plane YUV pixel format with 4:2:2 subsampling. For +// the luma (Y) channel, there is one sample for every pixel, but for the +// chroma (U, V) channels, there is one sample for every two pixels: pairs of +// horizontally adjacent pixels form one macropixel, i = x / 2 and j == y. In +// general, for a given p: +// - i = (x + bias_x) >> shift_x. +// - j = (y + bias_y) >> shift_y. +// where biases and shifts are in the range 0..3 and 0..2 respectively. +// +// In general, the biases will be zero after decoding an image. However, making +// a sub-image may change the bias, since the (x, y) coordinates are relative +// to the sub-image's top-left origin, but the backing pixel buffers were +// created relative to the original image's origin. +// +// For each plane p, each of those four numbers (biases and shifts) are encoded +// in two bits, which combine to form an 8 bit unsigned integer: +// +// e_p = (bias_x << 6) | (shift_x << 4) | (bias_y << 2) | (shift_y << 0) +// +// Those e_p values (e_0 for the first plane, e_1 for the second plane, etc) +// combine to form a wuffs_base__pixel_subsampling value: +// +// pixsub = (e_3 << 24) | (e_2 << 16) | (e_1 << 8) | (e_0 << 0) +// +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_base__pixel_subsampling__bias_x instead. +typedef uint32_t wuffs_base__pixel_subsampling; + +#define WUFFS_BASE__PIXEL_SUBSAMPLING__NONE ((wuffs_base__pixel_subsampling)0); + +#define WUFFS_BASE__PIXEL_SUBSAMPLING__444 \ + ((wuffs_base__pixel_subsampling)0x000000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__440 \ + ((wuffs_base__pixel_subsampling)0x010100); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__422 \ + ((wuffs_base__pixel_subsampling)0x101000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__420 \ + ((wuffs_base__pixel_subsampling)0x111100); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__411 \ + ((wuffs_base__pixel_subsampling)0x202000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__410 \ + ((wuffs_base__pixel_subsampling)0x212100); + +static inline uint32_t wuffs_base__pixel_subsampling__bias_x( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 6; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__shift_x( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 4; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__bias_y( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 2; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__shift_y( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 0; + return (s >> shift) & 0x03; +} + typedef struct { // Do not access the private_impl's fields directly. There is no API/ABI // compatibility or safety guarantee if you do so. struct { - uint32_t flags; - uint32_t w; - uint32_t h; - // TODO: color model, including both packed RGBA and planar, - // chroma-subsampled YCbCr. + wuffs_base__pixel_format pixfmt; + wuffs_base__pixel_subsampling pixsub; + uint32_t width; + uint32_t height; + uint32_t num_loops; } private_impl; } wuffs_base__image_config; +static inline void wuffs_base__image_config__initialize( + wuffs_base__image_config* c, + wuffs_base__pixel_format pixfmt, + wuffs_base__pixel_subsampling pixsub, + uint32_t width, + uint32_t height, + uint32_t num_loops) { + if (!c) { + return; + } + // TODO: move the check from wuffs_base__image_config__is_valid here. Should + // this function return bool? An error type? + c->private_impl.pixfmt = pixfmt; + c->private_impl.pixsub = pixsub; + c->private_impl.width = width; + c->private_impl.height = height; + c->private_impl.num_loops = num_loops; +} + static inline void wuffs_base__image_config__invalidate( wuffs_base__image_config* c) { if (c) { @@ -128,24 +439,42 @@ static inline void wuffs_base__image_config__invalidate( } } -static inline bool wuffs_base__image_config__valid( +static inline bool wuffs_base__image_config__is_valid( wuffs_base__image_config* c) { - if (!c || !(c->private_impl.flags & 1)) { + if (!c || !c->private_impl.pixfmt) { return false; } - uint64_t wh = ((uint64_t)c->private_impl.w) * ((uint64_t)c->private_impl.h); + uint64_t wh = + ((uint64_t)c->private_impl.width) * ((uint64_t)c->private_impl.height); // TODO: handle things other than 1 byte per pixel. + // + // TODO: move the check to wuffs_base__image_config__initialize. return wh <= ((uint64_t)SIZE_MAX); } +static inline wuffs_base__pixel_format wuffs_base__image_config__pixel_format( + wuffs_base__image_config* c) { + return c ? c->private_impl.pixfmt : 0; +} + +static inline wuffs_base__pixel_subsampling +wuffs_base__image_config__pixel_subsampling(wuffs_base__image_config* c) { + return wuffs_base__image_config__is_valid(c) ? c->private_impl.pixsub : 0; +} + static inline uint32_t wuffs_base__image_config__width( wuffs_base__image_config* c) { - return wuffs_base__image_config__valid(c) ? c->private_impl.w : 0; + return wuffs_base__image_config__is_valid(c) ? c->private_impl.width : 0; } static inline uint32_t wuffs_base__image_config__height( wuffs_base__image_config* c) { - return wuffs_base__image_config__valid(c) ? c->private_impl.h : 0; + return wuffs_base__image_config__is_valid(c) ? c->private_impl.height : 0; +} + +static inline uint32_t wuffs_base__image_config__num_loops( + wuffs_base__image_config* c) { + return wuffs_base__image_config__is_valid(c) ? c->private_impl.num_loops : 0; } // TODO: this is the right API for planar (not packed) pixbufs? Should it allow @@ -153,26 +482,96 @@ static inline uint32_t wuffs_base__image_config__height( // example, decoding a JPEG image straight to RGBA instead of to YCbCr? static inline size_t wuffs_base__image_config__pixbuf_size( wuffs_base__image_config* c) { - if (wuffs_base__image_config__valid(c)) { - uint64_t wh = ((uint64_t)c->private_impl.w) * ((uint64_t)c->private_impl.h); + if (wuffs_base__image_config__is_valid(c)) { + uint64_t wh = + ((uint64_t)c->private_impl.width) * ((uint64_t)c->private_impl.height); // TODO: handle things other than 1 byte per pixel. return (size_t)wh; } return 0; } -static inline void wuffs_base__image_config__initialize( - wuffs_base__image_config* c, - uint32_t width, - uint32_t height, - uint32_t TODO_color_model) { - if (!c) { +typedef struct { + // Do not access the private_impl's fields directly. There is no API/ABI + // compatibility or safety guarantee if you do so. + struct { + wuffs_base__image_config config; + uint32_t loop_count; // 0-based count of the current loop. + wuffs_base__pixel_buffer pixbuf; + // TODO: color spaces. + wuffs_base__rectangle dirty_rect; + wuffs_base__flicks duration; + uint8_t palette[1024]; + } private_impl; +} wuffs_base__image_buffer; + +static inline void wuffs_base__image_buffer__initialize( + wuffs_base__image_buffer* f, + wuffs_base__image_config config, + wuffs_base__pixel_buffer pixbuf) { + if (!f) { return; } - c->private_impl.flags = 1; - c->private_impl.w = width; - c->private_impl.h = height; - // TODO: color model. + *f = ((wuffs_base__image_buffer){}); + f->private_impl.config = config; + f->private_impl.pixbuf = pixbuf; +} + +static inline void wuffs_base__image_buffer__update( + wuffs_base__image_buffer* f, + wuffs_base__rectangle dirty_rect, + wuffs_base__flicks duration, + uint8_t* palette_ptr, + size_t palette_len) { + if (!f) { + return; + } + f->private_impl.dirty_rect = dirty_rect; + f->private_impl.duration = duration; + if (palette_ptr) { + memmove(f->private_impl.palette, palette_ptr, + palette_len <= 1024 ? palette_len : 1024); + } +} + +// wuffs_base__image_buffer__loop returns whether the image decoder should loop +// back to the beginning of the animation, assuming that we've reached the end +// of the encoded stream. If so, it increments f's count of the animation loops +// played so far. +static inline bool wuffs_base__image_buffer__loop(wuffs_base__image_buffer* f) { + if (!f) { + return false; + } + uint32_t n = f->private_impl.config.private_impl.num_loops; + if (n == 0) { + return true; + } + if (f->private_impl.loop_count < n - 1) { + f->private_impl.loop_count++; + return true; + } + return false; +} + +// wuffs_base__image_buffer__dirty_rect returns an upper bound for what part of +// this frame's pixels differs from the previous frame. +static inline wuffs_base__rectangle wuffs_base__image_buffer__dirty_rect( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.dirty_rect : ((wuffs_base__rectangle){0}); +} + +// wuffs_base__image_buffer__duration returns the amount of time to display +// this frame. Zero means to display forever - a still (non-animated) image. +static inline wuffs_base__flicks wuffs_base__image_buffer__duration( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.duration : 0; +} + +// wuffs_base__image_buffer__palette returns the palette that the pixel data +// can index. The backing array is inside f and has length 1024. +static inline uint8_t* wuffs_base__image_buffer__palette( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.palette : NULL; } #endif // WUFFS_BASE_HEADER_H @@ -185,14 +584,14 @@ extern "C" { // ---------------- Status Codes -// Status codes are int32_t values: -// - the sign bit indicates a non-recoverable status code: an error -// - bits 10-30 hold the packageid: a namespace -// - bits 8-9 are reserved -// - bits 0-7 are a package-namespaced numeric code +// Status codes are int32_t values. Its bits: +// - bit 31 (the sign bit) indicates unrecoverable-ness: an error. +// - bits 30 .. 10 are the packageid: a namespace. +// - bits 9 .. 8 are reserved. +// - bits 7 .. 0 are a package-namespaced numeric code. // -// Do not manipulate these bits directly. Use the API functions such as -// wuffs_deflate__status__is_error instead. +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_deflate__status__is_error instead. typedef int32_t wuffs_deflate__status; #define wuffs_deflate__packageid 848533 // 0x000CF295 diff --git a/gen/h/std/gif.h b/gen/h/std/gif.h index 8585ad2d..4a6d89af 100644 --- a/gen/h/std/gif.h +++ b/gen/h/std/gif.h @@ -46,11 +46,40 @@ // For now, the API and ABI are simply unstable and can change at any time. // // TODO: don't hard code this in base-header.h. -#define WUFFS_VERSION (0x00001) +#define WUFFS_VERSION 0x00001 -// ---------------- I/O +// ---------------- Fundamentals -// wuffs_base__slice_u8 is a 1-dimensional buffer (a pointer and length). +// Flicks are a unit of time. One flick (frame-tick) is 1 / 705_600_000 of a +// second. See https://github.com/OculusVR/Flicks +typedef uint64_t wuffs_base__flicks; + +#define WUFFS_BASE__FLICKS_PER_SECOND 705600000ULL + +// wuffs_base__rectangle is a rectangle on the integer grid. It contains all +// points (x, y) such that ((min_x <= x) && (x < max_x)) and likewise for y. It +// is therefore empty if min_x >= max_x. There are multiple representations of +// an empty rectangle. +// +// A value with all fields zero is a valid, empty rectangle. +// +// The X and Y axes increase right and down. +typedef struct { + uint32_t min_x; + uint32_t min_y; + uint32_t max_x; + uint32_t max_y; +} wuffs_base__rectangle; + +static inline uint32_t wuffs_base__rectangle__width(wuffs_base__rectangle r) { + return r.max_x > r.min_x ? r.max_x - r.min_x : 0; +} + +static inline uint32_t wuffs_base__rectangle__height(wuffs_base__rectangle r) { + return r.max_y > r.min_y ? r.max_y - r.min_y : 0; +} + +// wuffs_base__slice_u8 is a 1-dimensional buffer. // // A value with all fields NULL or zero is a valid, empty slice. typedef struct { @@ -58,6 +87,20 @@ typedef struct { size_t len; } wuffs_base__slice_u8; +// wuffs_base__table_u8 is a 2-dimensional buffer. +// +// A value with all fields NULL or zero is a valid, empty table. +typedef struct { + uint8_t* ptr; + size_t width; + size_t height; + size_t stride; +} wuffs_base__table_u8; + +// ---------------- I/O + +// TODO: rename buf1 to io_buffer, writer1 to io_writer, etc. + // wuffs_base__buf1 is a 1-dimensional buffer (a pointer and length), plus // additional indexes into that buffer, plus an opened / closed flag. // @@ -109,18 +152,286 @@ typedef struct { // ---------------- Images +// wuffs_base__pixel_format encodes the format of the bytes that constitute an +// image frame's pixel data. Its bits: +// - bit 31 is reserved. +// - bits 30 .. 28 encodes color (and channel order, in terms of memory). +// - bits 27 .. 26 are reserved. +// - bits 25 .. 24 encodes transparency. +// - bit 23 indicates big-endian/MSB-first (as opposed to little/LSB). +// - bit 22 indicates floating point (as opposed to integer). +// - bits 21 .. 20 are the number of planes, minus 1. Zero means packed. +// - bits 19 .. 16 encodes the number of bits (depth) in an index value. +// Zero means direct, not palette-indexed. +// - bits 15 .. 12 encodes the number of bits (depth) in the 3rd channel. +// - bits 11 .. 8 encodes the number of bits (depth) in the 2nd channel. +// - bits 7 .. 4 encodes the number of bits (depth) in the 1st channel. +// - bits 3 .. 0 encodes the number of bits (depth) in the 0th channel. +// +// The bit fields of a wuffs_base__pixel_format are not independent. For +// example, the number of planes should not be greater than the number of +// channels. Similarly, bits 15..4 are unused (and should be zero) if bits +// 31..24 (color and transparency) together imply only 1 channel (gray, no +// alpha) and floating point samples should mean a bit depth of 16, 32 or 64. +// +// Formats hold between 1 and 4 channels. For example: Y (1 channel: gray), YA +// (2 channels: gray and alpha), BGR (3 channels: blue, green, red) or CMYK (4 +// channels: cyan, magenta, yellow, black). +// +// For direct formats with N > 1 channels, those channels can be laid out in +// either 1 (packed) or N (planar) planes. For example, RGBA data is usually +// packed, but YUV data is usually planar, due to chroma subsampling (for +// details, see the wuffs_base__pixel_subsampling type). For indexed formats, +// the palette (always 256 × 4 bytes) holds up to 4 packed bytes of color data +// per index value, and there is only 1 plane (for the index). The distance +// between successive palette elements is always 4 bytes. +// +// The color field is encoded in 3 bits: +// - 0 means A (Alpha). +// - 1 means Y or YA (Gray, Alpha). +// - 2 means BGR, BGRX or BGRA (Blue, Green, Red, X-padding or Alpha). +// - 3 means RGB, RGBX or RGBA (Red, Green, Blue, X-padding or Alpha). +// - 4 means YUV or YUVA (Luma, Chroma-blue, Chroma-red, Alpha). +// - 5 means CMY or CMYK (Cyan, Magenta, Yellow, Black). +// - all other values are reserved. +// +// In Wuffs, channels are given in memory order, regardless of endianness, +// since the C type for the pixel data is an array of bytes, not an array of +// uint32_t. For example, packed BGRA with 8 bits per channel means that the +// bytes in memory are always Blue, Green, Red then Alpha. On big-endian +// systems, that is the uint32_t 0xBBGGRRAA. On little-endian, 0xAARRGGBB. +// +// When the color field (3 bits) encodes multiple options, the transparency +// field (2 bits) distinguishes them: +// - 0 means fully opaque, no extra channels +// - 1 means fully opaque, one extra channel (X or K, padding or black). +// - 2 means one extra alpha channel, other channels are non-premultiplied. +// - 3 means one extra alpha channel, other channels are premultiplied. +// +// The zero wuffs_base__pixel_format value is an invalid pixel format, as it is +// invalid to combine the zero color (alpha only) with the zero transparency. +// +// Bit depth is encoded in 4 bits: +// - 0 means the channel or index is unused. +// - x means a bit depth of x, for x in the range 1..8. +// - 9 means a bit depth of 10. +// - 10 means a bit depth of 12. +// - 11 means a bit depth of 16. +// - 12 means a bit depth of 24. +// - 13 means a bit depth of 32. +// - 14 means a bit depth of 48. +// - 15 means a bit depth of 64. +// +// For example, wuffs_base__pixel_format 0x3280BBBB is a natural format for +// decoding a PNG image - network byte order (also known as big-endian), +// packed, non-premultiplied alpha - that happens to be 16-bit-depth truecolor +// with alpha (RGBA). In memory order: +// +// ptr+0 ptr+1 ptr+2 ptr+3 ptr+4 ptr+5 ptr+6 ptr+7 +// Rhi Rlo Ghi Glo Bhi Blo Ahi Alo +// +// For example, the value wuffs_base__pixel_format 0x20000565 means BGR with no +// alpha or padding, 5/6/5 bits for blue/green/red, packed 2 bytes per pixel, +// laid out LSB-first in memory order: +// +// ptr+0........... ptr+1........... +// MSB LSB MSB LSB +// G₂G₁G₀B₄B₃B₂B₁B₀ R₄R₃R₂R₁R₀G₅G₄G₃ +// +// On little-endian systems (but not big-endian), this Wuffs pixel format value +// (0x20000565) corresponds to the Cairo library's CAIRO_FORMAT_RGB16_565, the +// SDL2 (Simple DirectMedia Layer 2) library's SDL_PIXELFORMAT_RGB565 and the +// Skia library's kRGB_565_SkColorType. Note BGR in Wuffs versus RGB in the +// other libraries. +// +// Regardless of endianness, this Wuffs pixel format value (0x20000565) +// corresponds to the V4L2 (Video For Linux 2) library's V4L2_PIX_FMT_RGB565 +// and the Wayland-DRM library's WL_DRM_FORMAT_RGB565. +// +// Different software libraries name their pixel formats (and especially their +// channel order) either according to memory layout or as bits of a native +// integer type like uint32_t. The two conventions differ because of a system's +// endianness. As mentioned earlier, Wuffs pixel formats are always in memory +// order. More detail of other software libraries' naming conventions is in the +// Pixel Format Guide at https://afrantzis.github.io/pixel-format-guide/ +// +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_base__pixel_format__num_planes instead. +typedef uint32_t wuffs_base__pixel_format; + +// Common 8-bit-depth pixel formats. This list is not exhaustive; not all valid +// wuffs_base__pixel_format values are present. + +#define WUFFS_BASE__PIXEL_FORMAT__INVALID ((wuffs_base__pixel_format)0x00000000) + +#define WUFFS_BASE__PIXEL_FORMAT__A ((wuffs_base__pixel_format)0x02000008) + +#define WUFFS_BASE__PIXEL_FORMAT__Y ((wuffs_base__pixel_format)0x10000008) +#define WUFFS_BASE__PIXEL_FORMAT__YA_NONPREMUL \ + ((wuffs_base__pixel_format)0x12000008) +#define WUFFS_BASE__PIXEL_FORMAT__YA_PREMUL \ + ((wuffs_base__pixel_format)0x13000008) + +#define WUFFS_BASE__PIXEL_FORMAT__BGR ((wuffs_base__pixel_format)0x20000888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRX ((wuffs_base__pixel_format)0x21008888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRX_INDEXED \ + ((wuffs_base__pixel_format)0x21088888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_NONPREMUL \ + ((wuffs_base__pixel_format)0x22008888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_NONPREMUL_INDEXED \ + ((wuffs_base__pixel_format)0x22088888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_PREMUL \ + ((wuffs_base__pixel_format)0x23008888) + +#define WUFFS_BASE__PIXEL_FORMAT__RGB ((wuffs_base__pixel_format)0x30000888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBX ((wuffs_base__pixel_format)0x31008888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBX_INDEXED \ + ((wuffs_base__pixel_format)0x31088888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_NONPREMUL \ + ((wuffs_base__pixel_format)0x32008888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_NONPREMUL_INDEXED \ + ((wuffs_base__pixel_format)0x32088888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_PREMUL \ + ((wuffs_base__pixel_format)0x33008888) + +#define WUFFS_BASE__PIXEL_FORMAT__YUV ((wuffs_base__pixel_format)0x40200888) +#define WUFFS_BASE__PIXEL_FORMAT__YUVK ((wuffs_base__pixel_format)0x41308888) +#define WUFFS_BASE__PIXEL_FORMAT__YUVA_NONPREMUL \ + ((wuffs_base__pixel_format)0x42308888) + +#define WUFFS_BASE__PIXEL_FORMAT__CMY ((wuffs_base__pixel_format)0x50200888) +#define WUFFS_BASE__PIXEL_FORMAT__CMYK ((wuffs_base__pixel_format)0x51308888) + +static inline bool wuffs_base__pixel_format__is_valid( + wuffs_base__pixel_format f) { + return f != 0; +} + +static inline bool wuffs_base__pixel_format__is_indexed( + wuffs_base__pixel_format f) { + return ((f >> 16) & 0x0F) != 0; +} + +#define WUFFS_BASE__PIXEL_FORMAT__NUM_PLANES_MAX 4 + +static inline uint32_t wuffs_base__pixel_format__num_planes( + wuffs_base__pixel_format f) { + return f ? (((f >> 20) & 0x03) + 1) : 0; +} + +typedef struct { + wuffs_base__table_u8 planes[WUFFS_BASE__PIXEL_FORMAT__NUM_PLANES_MAX]; +} wuffs_base__pixel_buffer; + +// wuffs_base__pixel_subsampling encodes the mapping of pixel space coordinates +// (x, y) to pixel buffer indices (i, j). That mapping can differ for each +// plane p. For a depth of 8 bits (1 byte), the p'th plane's sample starts at +// (planes[p].ptr + (j * planes[p].stride) + i). +// +// For packed pixel formats, the mapping is trivial: i = x and j = y. For +// planar pixel formats, the mapping can differ due to chroma subsampling. For +// example, consider a three plane YUV pixel format with 4:2:2 subsampling. For +// the luma (Y) channel, there is one sample for every pixel, but for the +// chroma (U, V) channels, there is one sample for every two pixels: pairs of +// horizontally adjacent pixels form one macropixel, i = x / 2 and j == y. In +// general, for a given p: +// - i = (x + bias_x) >> shift_x. +// - j = (y + bias_y) >> shift_y. +// where biases and shifts are in the range 0..3 and 0..2 respectively. +// +// In general, the biases will be zero after decoding an image. However, making +// a sub-image may change the bias, since the (x, y) coordinates are relative +// to the sub-image's top-left origin, but the backing pixel buffers were +// created relative to the original image's origin. +// +// For each plane p, each of those four numbers (biases and shifts) are encoded +// in two bits, which combine to form an 8 bit unsigned integer: +// +// e_p = (bias_x << 6) | (shift_x << 4) | (bias_y << 2) | (shift_y << 0) +// +// Those e_p values (e_0 for the first plane, e_1 for the second plane, etc) +// combine to form a wuffs_base__pixel_subsampling value: +// +// pixsub = (e_3 << 24) | (e_2 << 16) | (e_1 << 8) | (e_0 << 0) +// +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_base__pixel_subsampling__bias_x instead. +typedef uint32_t wuffs_base__pixel_subsampling; + +#define WUFFS_BASE__PIXEL_SUBSAMPLING__NONE ((wuffs_base__pixel_subsampling)0); + +#define WUFFS_BASE__PIXEL_SUBSAMPLING__444 \ + ((wuffs_base__pixel_subsampling)0x000000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__440 \ + ((wuffs_base__pixel_subsampling)0x010100); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__422 \ + ((wuffs_base__pixel_subsampling)0x101000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__420 \ + ((wuffs_base__pixel_subsampling)0x111100); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__411 \ + ((wuffs_base__pixel_subsampling)0x202000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__410 \ + ((wuffs_base__pixel_subsampling)0x212100); + +static inline uint32_t wuffs_base__pixel_subsampling__bias_x( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 6; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__shift_x( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 4; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__bias_y( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 2; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__shift_y( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 0; + return (s >> shift) & 0x03; +} + typedef struct { // Do not access the private_impl's fields directly. There is no API/ABI // compatibility or safety guarantee if you do so. struct { - uint32_t flags; - uint32_t w; - uint32_t h; - // TODO: color model, including both packed RGBA and planar, - // chroma-subsampled YCbCr. + wuffs_base__pixel_format pixfmt; + wuffs_base__pixel_subsampling pixsub; + uint32_t width; + uint32_t height; + uint32_t num_loops; } private_impl; } wuffs_base__image_config; +static inline void wuffs_base__image_config__initialize( + wuffs_base__image_config* c, + wuffs_base__pixel_format pixfmt, + wuffs_base__pixel_subsampling pixsub, + uint32_t width, + uint32_t height, + uint32_t num_loops) { + if (!c) { + return; + } + // TODO: move the check from wuffs_base__image_config__is_valid here. Should + // this function return bool? An error type? + c->private_impl.pixfmt = pixfmt; + c->private_impl.pixsub = pixsub; + c->private_impl.width = width; + c->private_impl.height = height; + c->private_impl.num_loops = num_loops; +} + static inline void wuffs_base__image_config__invalidate( wuffs_base__image_config* c) { if (c) { @@ -128,24 +439,42 @@ static inline void wuffs_base__image_config__invalidate( } } -static inline bool wuffs_base__image_config__valid( +static inline bool wuffs_base__image_config__is_valid( wuffs_base__image_config* c) { - if (!c || !(c->private_impl.flags & 1)) { + if (!c || !c->private_impl.pixfmt) { return false; } - uint64_t wh = ((uint64_t)c->private_impl.w) * ((uint64_t)c->private_impl.h); + uint64_t wh = + ((uint64_t)c->private_impl.width) * ((uint64_t)c->private_impl.height); // TODO: handle things other than 1 byte per pixel. + // + // TODO: move the check to wuffs_base__image_config__initialize. return wh <= ((uint64_t)SIZE_MAX); } +static inline wuffs_base__pixel_format wuffs_base__image_config__pixel_format( + wuffs_base__image_config* c) { + return c ? c->private_impl.pixfmt : 0; +} + +static inline wuffs_base__pixel_subsampling +wuffs_base__image_config__pixel_subsampling(wuffs_base__image_config* c) { + return wuffs_base__image_config__is_valid(c) ? c->private_impl.pixsub : 0; +} + static inline uint32_t wuffs_base__image_config__width( wuffs_base__image_config* c) { - return wuffs_base__image_config__valid(c) ? c->private_impl.w : 0; + return wuffs_base__image_config__is_valid(c) ? c->private_impl.width : 0; } static inline uint32_t wuffs_base__image_config__height( wuffs_base__image_config* c) { - return wuffs_base__image_config__valid(c) ? c->private_impl.h : 0; + return wuffs_base__image_config__is_valid(c) ? c->private_impl.height : 0; +} + +static inline uint32_t wuffs_base__image_config__num_loops( + wuffs_base__image_config* c) { + return wuffs_base__image_config__is_valid(c) ? c->private_impl.num_loops : 0; } // TODO: this is the right API for planar (not packed) pixbufs? Should it allow @@ -153,26 +482,96 @@ static inline uint32_t wuffs_base__image_config__height( // example, decoding a JPEG image straight to RGBA instead of to YCbCr? static inline size_t wuffs_base__image_config__pixbuf_size( wuffs_base__image_config* c) { - if (wuffs_base__image_config__valid(c)) { - uint64_t wh = ((uint64_t)c->private_impl.w) * ((uint64_t)c->private_impl.h); + if (wuffs_base__image_config__is_valid(c)) { + uint64_t wh = + ((uint64_t)c->private_impl.width) * ((uint64_t)c->private_impl.height); // TODO: handle things other than 1 byte per pixel. return (size_t)wh; } return 0; } -static inline void wuffs_base__image_config__initialize( - wuffs_base__image_config* c, - uint32_t width, - uint32_t height, - uint32_t TODO_color_model) { - if (!c) { +typedef struct { + // Do not access the private_impl's fields directly. There is no API/ABI + // compatibility or safety guarantee if you do so. + struct { + wuffs_base__image_config config; + uint32_t loop_count; // 0-based count of the current loop. + wuffs_base__pixel_buffer pixbuf; + // TODO: color spaces. + wuffs_base__rectangle dirty_rect; + wuffs_base__flicks duration; + uint8_t palette[1024]; + } private_impl; +} wuffs_base__image_buffer; + +static inline void wuffs_base__image_buffer__initialize( + wuffs_base__image_buffer* f, + wuffs_base__image_config config, + wuffs_base__pixel_buffer pixbuf) { + if (!f) { return; } - c->private_impl.flags = 1; - c->private_impl.w = width; - c->private_impl.h = height; - // TODO: color model. + *f = ((wuffs_base__image_buffer){}); + f->private_impl.config = config; + f->private_impl.pixbuf = pixbuf; +} + +static inline void wuffs_base__image_buffer__update( + wuffs_base__image_buffer* f, + wuffs_base__rectangle dirty_rect, + wuffs_base__flicks duration, + uint8_t* palette_ptr, + size_t palette_len) { + if (!f) { + return; + } + f->private_impl.dirty_rect = dirty_rect; + f->private_impl.duration = duration; + if (palette_ptr) { + memmove(f->private_impl.palette, palette_ptr, + palette_len <= 1024 ? palette_len : 1024); + } +} + +// wuffs_base__image_buffer__loop returns whether the image decoder should loop +// back to the beginning of the animation, assuming that we've reached the end +// of the encoded stream. If so, it increments f's count of the animation loops +// played so far. +static inline bool wuffs_base__image_buffer__loop(wuffs_base__image_buffer* f) { + if (!f) { + return false; + } + uint32_t n = f->private_impl.config.private_impl.num_loops; + if (n == 0) { + return true; + } + if (f->private_impl.loop_count < n - 1) { + f->private_impl.loop_count++; + return true; + } + return false; +} + +// wuffs_base__image_buffer__dirty_rect returns an upper bound for what part of +// this frame's pixels differs from the previous frame. +static inline wuffs_base__rectangle wuffs_base__image_buffer__dirty_rect( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.dirty_rect : ((wuffs_base__rectangle){0}); +} + +// wuffs_base__image_buffer__duration returns the amount of time to display +// this frame. Zero means to display forever - a still (non-animated) image. +static inline wuffs_base__flicks wuffs_base__image_buffer__duration( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.duration : 0; +} + +// wuffs_base__image_buffer__palette returns the palette that the pixel data +// can index. The backing array is inside f and has length 1024. +static inline uint8_t* wuffs_base__image_buffer__palette( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.palette : NULL; } #endif // WUFFS_BASE_HEADER_H @@ -185,14 +584,14 @@ extern "C" { // ---------------- Status Codes -// Status codes are int32_t values: -// - the sign bit indicates a non-recoverable status code: an error -// - bits 10-30 hold the packageid: a namespace -// - bits 8-9 are reserved -// - bits 0-7 are a package-namespaced numeric code +// Status codes are int32_t values. Its bits: +// - bit 31 (the sign bit) indicates unrecoverable-ness: an error. +// - bits 30 .. 10 are the packageid: a namespace. +// - bits 9 .. 8 are reserved. +// - bits 7 .. 0 are a package-namespaced numeric code. // -// Do not manipulate these bits directly. Use the API functions such as -// wuffs_gif__status__is_error instead. +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_gif__status__is_error instead. typedef int32_t wuffs_gif__status; #define wuffs_gif__packageid 1017222 // 0x000F8586 diff --git a/gen/h/std/gzip.h b/gen/h/std/gzip.h index c3b0e6c5..9efbabbd 100644 --- a/gen/h/std/gzip.h +++ b/gen/h/std/gzip.h @@ -46,11 +46,40 @@ // For now, the API and ABI are simply unstable and can change at any time. // // TODO: don't hard code this in base-header.h. -#define WUFFS_VERSION (0x00001) +#define WUFFS_VERSION 0x00001 -// ---------------- I/O +// ---------------- Fundamentals -// wuffs_base__slice_u8 is a 1-dimensional buffer (a pointer and length). +// Flicks are a unit of time. One flick (frame-tick) is 1 / 705_600_000 of a +// second. See https://github.com/OculusVR/Flicks +typedef uint64_t wuffs_base__flicks; + +#define WUFFS_BASE__FLICKS_PER_SECOND 705600000ULL + +// wuffs_base__rectangle is a rectangle on the integer grid. It contains all +// points (x, y) such that ((min_x <= x) && (x < max_x)) and likewise for y. It +// is therefore empty if min_x >= max_x. There are multiple representations of +// an empty rectangle. +// +// A value with all fields zero is a valid, empty rectangle. +// +// The X and Y axes increase right and down. +typedef struct { + uint32_t min_x; + uint32_t min_y; + uint32_t max_x; + uint32_t max_y; +} wuffs_base__rectangle; + +static inline uint32_t wuffs_base__rectangle__width(wuffs_base__rectangle r) { + return r.max_x > r.min_x ? r.max_x - r.min_x : 0; +} + +static inline uint32_t wuffs_base__rectangle__height(wuffs_base__rectangle r) { + return r.max_y > r.min_y ? r.max_y - r.min_y : 0; +} + +// wuffs_base__slice_u8 is a 1-dimensional buffer. // // A value with all fields NULL or zero is a valid, empty slice. typedef struct { @@ -58,6 +87,20 @@ typedef struct { size_t len; } wuffs_base__slice_u8; +// wuffs_base__table_u8 is a 2-dimensional buffer. +// +// A value with all fields NULL or zero is a valid, empty table. +typedef struct { + uint8_t* ptr; + size_t width; + size_t height; + size_t stride; +} wuffs_base__table_u8; + +// ---------------- I/O + +// TODO: rename buf1 to io_buffer, writer1 to io_writer, etc. + // wuffs_base__buf1 is a 1-dimensional buffer (a pointer and length), plus // additional indexes into that buffer, plus an opened / closed flag. // @@ -109,18 +152,286 @@ typedef struct { // ---------------- Images +// wuffs_base__pixel_format encodes the format of the bytes that constitute an +// image frame's pixel data. Its bits: +// - bit 31 is reserved. +// - bits 30 .. 28 encodes color (and channel order, in terms of memory). +// - bits 27 .. 26 are reserved. +// - bits 25 .. 24 encodes transparency. +// - bit 23 indicates big-endian/MSB-first (as opposed to little/LSB). +// - bit 22 indicates floating point (as opposed to integer). +// - bits 21 .. 20 are the number of planes, minus 1. Zero means packed. +// - bits 19 .. 16 encodes the number of bits (depth) in an index value. +// Zero means direct, not palette-indexed. +// - bits 15 .. 12 encodes the number of bits (depth) in the 3rd channel. +// - bits 11 .. 8 encodes the number of bits (depth) in the 2nd channel. +// - bits 7 .. 4 encodes the number of bits (depth) in the 1st channel. +// - bits 3 .. 0 encodes the number of bits (depth) in the 0th channel. +// +// The bit fields of a wuffs_base__pixel_format are not independent. For +// example, the number of planes should not be greater than the number of +// channels. Similarly, bits 15..4 are unused (and should be zero) if bits +// 31..24 (color and transparency) together imply only 1 channel (gray, no +// alpha) and floating point samples should mean a bit depth of 16, 32 or 64. +// +// Formats hold between 1 and 4 channels. For example: Y (1 channel: gray), YA +// (2 channels: gray and alpha), BGR (3 channels: blue, green, red) or CMYK (4 +// channels: cyan, magenta, yellow, black). +// +// For direct formats with N > 1 channels, those channels can be laid out in +// either 1 (packed) or N (planar) planes. For example, RGBA data is usually +// packed, but YUV data is usually planar, due to chroma subsampling (for +// details, see the wuffs_base__pixel_subsampling type). For indexed formats, +// the palette (always 256 × 4 bytes) holds up to 4 packed bytes of color data +// per index value, and there is only 1 plane (for the index). The distance +// between successive palette elements is always 4 bytes. +// +// The color field is encoded in 3 bits: +// - 0 means A (Alpha). +// - 1 means Y or YA (Gray, Alpha). +// - 2 means BGR, BGRX or BGRA (Blue, Green, Red, X-padding or Alpha). +// - 3 means RGB, RGBX or RGBA (Red, Green, Blue, X-padding or Alpha). +// - 4 means YUV or YUVA (Luma, Chroma-blue, Chroma-red, Alpha). +// - 5 means CMY or CMYK (Cyan, Magenta, Yellow, Black). +// - all other values are reserved. +// +// In Wuffs, channels are given in memory order, regardless of endianness, +// since the C type for the pixel data is an array of bytes, not an array of +// uint32_t. For example, packed BGRA with 8 bits per channel means that the +// bytes in memory are always Blue, Green, Red then Alpha. On big-endian +// systems, that is the uint32_t 0xBBGGRRAA. On little-endian, 0xAARRGGBB. +// +// When the color field (3 bits) encodes multiple options, the transparency +// field (2 bits) distinguishes them: +// - 0 means fully opaque, no extra channels +// - 1 means fully opaque, one extra channel (X or K, padding or black). +// - 2 means one extra alpha channel, other channels are non-premultiplied. +// - 3 means one extra alpha channel, other channels are premultiplied. +// +// The zero wuffs_base__pixel_format value is an invalid pixel format, as it is +// invalid to combine the zero color (alpha only) with the zero transparency. +// +// Bit depth is encoded in 4 bits: +// - 0 means the channel or index is unused. +// - x means a bit depth of x, for x in the range 1..8. +// - 9 means a bit depth of 10. +// - 10 means a bit depth of 12. +// - 11 means a bit depth of 16. +// - 12 means a bit depth of 24. +// - 13 means a bit depth of 32. +// - 14 means a bit depth of 48. +// - 15 means a bit depth of 64. +// +// For example, wuffs_base__pixel_format 0x3280BBBB is a natural format for +// decoding a PNG image - network byte order (also known as big-endian), +// packed, non-premultiplied alpha - that happens to be 16-bit-depth truecolor +// with alpha (RGBA). In memory order: +// +// ptr+0 ptr+1 ptr+2 ptr+3 ptr+4 ptr+5 ptr+6 ptr+7 +// Rhi Rlo Ghi Glo Bhi Blo Ahi Alo +// +// For example, the value wuffs_base__pixel_format 0x20000565 means BGR with no +// alpha or padding, 5/6/5 bits for blue/green/red, packed 2 bytes per pixel, +// laid out LSB-first in memory order: +// +// ptr+0........... ptr+1........... +// MSB LSB MSB LSB +// G₂G₁G₀B₄B₃B₂B₁B₀ R₄R₃R₂R₁R₀G₅G₄G₃ +// +// On little-endian systems (but not big-endian), this Wuffs pixel format value +// (0x20000565) corresponds to the Cairo library's CAIRO_FORMAT_RGB16_565, the +// SDL2 (Simple DirectMedia Layer 2) library's SDL_PIXELFORMAT_RGB565 and the +// Skia library's kRGB_565_SkColorType. Note BGR in Wuffs versus RGB in the +// other libraries. +// +// Regardless of endianness, this Wuffs pixel format value (0x20000565) +// corresponds to the V4L2 (Video For Linux 2) library's V4L2_PIX_FMT_RGB565 +// and the Wayland-DRM library's WL_DRM_FORMAT_RGB565. +// +// Different software libraries name their pixel formats (and especially their +// channel order) either according to memory layout or as bits of a native +// integer type like uint32_t. The two conventions differ because of a system's +// endianness. As mentioned earlier, Wuffs pixel formats are always in memory +// order. More detail of other software libraries' naming conventions is in the +// Pixel Format Guide at https://afrantzis.github.io/pixel-format-guide/ +// +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_base__pixel_format__num_planes instead. +typedef uint32_t wuffs_base__pixel_format; + +// Common 8-bit-depth pixel formats. This list is not exhaustive; not all valid +// wuffs_base__pixel_format values are present. + +#define WUFFS_BASE__PIXEL_FORMAT__INVALID ((wuffs_base__pixel_format)0x00000000) + +#define WUFFS_BASE__PIXEL_FORMAT__A ((wuffs_base__pixel_format)0x02000008) + +#define WUFFS_BASE__PIXEL_FORMAT__Y ((wuffs_base__pixel_format)0x10000008) +#define WUFFS_BASE__PIXEL_FORMAT__YA_NONPREMUL \ + ((wuffs_base__pixel_format)0x12000008) +#define WUFFS_BASE__PIXEL_FORMAT__YA_PREMUL \ + ((wuffs_base__pixel_format)0x13000008) + +#define WUFFS_BASE__PIXEL_FORMAT__BGR ((wuffs_base__pixel_format)0x20000888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRX ((wuffs_base__pixel_format)0x21008888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRX_INDEXED \ + ((wuffs_base__pixel_format)0x21088888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_NONPREMUL \ + ((wuffs_base__pixel_format)0x22008888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_NONPREMUL_INDEXED \ + ((wuffs_base__pixel_format)0x22088888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_PREMUL \ + ((wuffs_base__pixel_format)0x23008888) + +#define WUFFS_BASE__PIXEL_FORMAT__RGB ((wuffs_base__pixel_format)0x30000888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBX ((wuffs_base__pixel_format)0x31008888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBX_INDEXED \ + ((wuffs_base__pixel_format)0x31088888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_NONPREMUL \ + ((wuffs_base__pixel_format)0x32008888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_NONPREMUL_INDEXED \ + ((wuffs_base__pixel_format)0x32088888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_PREMUL \ + ((wuffs_base__pixel_format)0x33008888) + +#define WUFFS_BASE__PIXEL_FORMAT__YUV ((wuffs_base__pixel_format)0x40200888) +#define WUFFS_BASE__PIXEL_FORMAT__YUVK ((wuffs_base__pixel_format)0x41308888) +#define WUFFS_BASE__PIXEL_FORMAT__YUVA_NONPREMUL \ + ((wuffs_base__pixel_format)0x42308888) + +#define WUFFS_BASE__PIXEL_FORMAT__CMY ((wuffs_base__pixel_format)0x50200888) +#define WUFFS_BASE__PIXEL_FORMAT__CMYK ((wuffs_base__pixel_format)0x51308888) + +static inline bool wuffs_base__pixel_format__is_valid( + wuffs_base__pixel_format f) { + return f != 0; +} + +static inline bool wuffs_base__pixel_format__is_indexed( + wuffs_base__pixel_format f) { + return ((f >> 16) & 0x0F) != 0; +} + +#define WUFFS_BASE__PIXEL_FORMAT__NUM_PLANES_MAX 4 + +static inline uint32_t wuffs_base__pixel_format__num_planes( + wuffs_base__pixel_format f) { + return f ? (((f >> 20) & 0x03) + 1) : 0; +} + +typedef struct { + wuffs_base__table_u8 planes[WUFFS_BASE__PIXEL_FORMAT__NUM_PLANES_MAX]; +} wuffs_base__pixel_buffer; + +// wuffs_base__pixel_subsampling encodes the mapping of pixel space coordinates +// (x, y) to pixel buffer indices (i, j). That mapping can differ for each +// plane p. For a depth of 8 bits (1 byte), the p'th plane's sample starts at +// (planes[p].ptr + (j * planes[p].stride) + i). +// +// For packed pixel formats, the mapping is trivial: i = x and j = y. For +// planar pixel formats, the mapping can differ due to chroma subsampling. For +// example, consider a three plane YUV pixel format with 4:2:2 subsampling. For +// the luma (Y) channel, there is one sample for every pixel, but for the +// chroma (U, V) channels, there is one sample for every two pixels: pairs of +// horizontally adjacent pixels form one macropixel, i = x / 2 and j == y. In +// general, for a given p: +// - i = (x + bias_x) >> shift_x. +// - j = (y + bias_y) >> shift_y. +// where biases and shifts are in the range 0..3 and 0..2 respectively. +// +// In general, the biases will be zero after decoding an image. However, making +// a sub-image may change the bias, since the (x, y) coordinates are relative +// to the sub-image's top-left origin, but the backing pixel buffers were +// created relative to the original image's origin. +// +// For each plane p, each of those four numbers (biases and shifts) are encoded +// in two bits, which combine to form an 8 bit unsigned integer: +// +// e_p = (bias_x << 6) | (shift_x << 4) | (bias_y << 2) | (shift_y << 0) +// +// Those e_p values (e_0 for the first plane, e_1 for the second plane, etc) +// combine to form a wuffs_base__pixel_subsampling value: +// +// pixsub = (e_3 << 24) | (e_2 << 16) | (e_1 << 8) | (e_0 << 0) +// +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_base__pixel_subsampling__bias_x instead. +typedef uint32_t wuffs_base__pixel_subsampling; + +#define WUFFS_BASE__PIXEL_SUBSAMPLING__NONE ((wuffs_base__pixel_subsampling)0); + +#define WUFFS_BASE__PIXEL_SUBSAMPLING__444 \ + ((wuffs_base__pixel_subsampling)0x000000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__440 \ + ((wuffs_base__pixel_subsampling)0x010100); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__422 \ + ((wuffs_base__pixel_subsampling)0x101000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__420 \ + ((wuffs_base__pixel_subsampling)0x111100); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__411 \ + ((wuffs_base__pixel_subsampling)0x202000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__410 \ + ((wuffs_base__pixel_subsampling)0x212100); + +static inline uint32_t wuffs_base__pixel_subsampling__bias_x( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 6; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__shift_x( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 4; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__bias_y( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 2; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__shift_y( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 0; + return (s >> shift) & 0x03; +} + typedef struct { // Do not access the private_impl's fields directly. There is no API/ABI // compatibility or safety guarantee if you do so. struct { - uint32_t flags; - uint32_t w; - uint32_t h; - // TODO: color model, including both packed RGBA and planar, - // chroma-subsampled YCbCr. + wuffs_base__pixel_format pixfmt; + wuffs_base__pixel_subsampling pixsub; + uint32_t width; + uint32_t height; + uint32_t num_loops; } private_impl; } wuffs_base__image_config; +static inline void wuffs_base__image_config__initialize( + wuffs_base__image_config* c, + wuffs_base__pixel_format pixfmt, + wuffs_base__pixel_subsampling pixsub, + uint32_t width, + uint32_t height, + uint32_t num_loops) { + if (!c) { + return; + } + // TODO: move the check from wuffs_base__image_config__is_valid here. Should + // this function return bool? An error type? + c->private_impl.pixfmt = pixfmt; + c->private_impl.pixsub = pixsub; + c->private_impl.width = width; + c->private_impl.height = height; + c->private_impl.num_loops = num_loops; +} + static inline void wuffs_base__image_config__invalidate( wuffs_base__image_config* c) { if (c) { @@ -128,24 +439,42 @@ static inline void wuffs_base__image_config__invalidate( } } -static inline bool wuffs_base__image_config__valid( +static inline bool wuffs_base__image_config__is_valid( wuffs_base__image_config* c) { - if (!c || !(c->private_impl.flags & 1)) { + if (!c || !c->private_impl.pixfmt) { return false; } - uint64_t wh = ((uint64_t)c->private_impl.w) * ((uint64_t)c->private_impl.h); + uint64_t wh = + ((uint64_t)c->private_impl.width) * ((uint64_t)c->private_impl.height); // TODO: handle things other than 1 byte per pixel. + // + // TODO: move the check to wuffs_base__image_config__initialize. return wh <= ((uint64_t)SIZE_MAX); } +static inline wuffs_base__pixel_format wuffs_base__image_config__pixel_format( + wuffs_base__image_config* c) { + return c ? c->private_impl.pixfmt : 0; +} + +static inline wuffs_base__pixel_subsampling +wuffs_base__image_config__pixel_subsampling(wuffs_base__image_config* c) { + return wuffs_base__image_config__is_valid(c) ? c->private_impl.pixsub : 0; +} + static inline uint32_t wuffs_base__image_config__width( wuffs_base__image_config* c) { - return wuffs_base__image_config__valid(c) ? c->private_impl.w : 0; + return wuffs_base__image_config__is_valid(c) ? c->private_impl.width : 0; } static inline uint32_t wuffs_base__image_config__height( wuffs_base__image_config* c) { - return wuffs_base__image_config__valid(c) ? c->private_impl.h : 0; + return wuffs_base__image_config__is_valid(c) ? c->private_impl.height : 0; +} + +static inline uint32_t wuffs_base__image_config__num_loops( + wuffs_base__image_config* c) { + return wuffs_base__image_config__is_valid(c) ? c->private_impl.num_loops : 0; } // TODO: this is the right API for planar (not packed) pixbufs? Should it allow @@ -153,26 +482,96 @@ static inline uint32_t wuffs_base__image_config__height( // example, decoding a JPEG image straight to RGBA instead of to YCbCr? static inline size_t wuffs_base__image_config__pixbuf_size( wuffs_base__image_config* c) { - if (wuffs_base__image_config__valid(c)) { - uint64_t wh = ((uint64_t)c->private_impl.w) * ((uint64_t)c->private_impl.h); + if (wuffs_base__image_config__is_valid(c)) { + uint64_t wh = + ((uint64_t)c->private_impl.width) * ((uint64_t)c->private_impl.height); // TODO: handle things other than 1 byte per pixel. return (size_t)wh; } return 0; } -static inline void wuffs_base__image_config__initialize( - wuffs_base__image_config* c, - uint32_t width, - uint32_t height, - uint32_t TODO_color_model) { - if (!c) { +typedef struct { + // Do not access the private_impl's fields directly. There is no API/ABI + // compatibility or safety guarantee if you do so. + struct { + wuffs_base__image_config config; + uint32_t loop_count; // 0-based count of the current loop. + wuffs_base__pixel_buffer pixbuf; + // TODO: color spaces. + wuffs_base__rectangle dirty_rect; + wuffs_base__flicks duration; + uint8_t palette[1024]; + } private_impl; +} wuffs_base__image_buffer; + +static inline void wuffs_base__image_buffer__initialize( + wuffs_base__image_buffer* f, + wuffs_base__image_config config, + wuffs_base__pixel_buffer pixbuf) { + if (!f) { return; } - c->private_impl.flags = 1; - c->private_impl.w = width; - c->private_impl.h = height; - // TODO: color model. + *f = ((wuffs_base__image_buffer){}); + f->private_impl.config = config; + f->private_impl.pixbuf = pixbuf; +} + +static inline void wuffs_base__image_buffer__update( + wuffs_base__image_buffer* f, + wuffs_base__rectangle dirty_rect, + wuffs_base__flicks duration, + uint8_t* palette_ptr, + size_t palette_len) { + if (!f) { + return; + } + f->private_impl.dirty_rect = dirty_rect; + f->private_impl.duration = duration; + if (palette_ptr) { + memmove(f->private_impl.palette, palette_ptr, + palette_len <= 1024 ? palette_len : 1024); + } +} + +// wuffs_base__image_buffer__loop returns whether the image decoder should loop +// back to the beginning of the animation, assuming that we've reached the end +// of the encoded stream. If so, it increments f's count of the animation loops +// played so far. +static inline bool wuffs_base__image_buffer__loop(wuffs_base__image_buffer* f) { + if (!f) { + return false; + } + uint32_t n = f->private_impl.config.private_impl.num_loops; + if (n == 0) { + return true; + } + if (f->private_impl.loop_count < n - 1) { + f->private_impl.loop_count++; + return true; + } + return false; +} + +// wuffs_base__image_buffer__dirty_rect returns an upper bound for what part of +// this frame's pixels differs from the previous frame. +static inline wuffs_base__rectangle wuffs_base__image_buffer__dirty_rect( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.dirty_rect : ((wuffs_base__rectangle){0}); +} + +// wuffs_base__image_buffer__duration returns the amount of time to display +// this frame. Zero means to display forever - a still (non-animated) image. +static inline wuffs_base__flicks wuffs_base__image_buffer__duration( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.duration : 0; +} + +// wuffs_base__image_buffer__palette returns the palette that the pixel data +// can index. The backing array is inside f and has length 1024. +static inline uint8_t* wuffs_base__image_buffer__palette( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.palette : NULL; } #endif // WUFFS_BASE_HEADER_H @@ -194,14 +593,14 @@ extern "C" { // ---------------- Status Codes -// Status codes are int32_t values: -// - the sign bit indicates a non-recoverable status code: an error -// - bits 10-30 hold the packageid: a namespace -// - bits 8-9 are reserved -// - bits 0-7 are a package-namespaced numeric code +// Status codes are int32_t values. Its bits: +// - bit 31 (the sign bit) indicates unrecoverable-ness: an error. +// - bits 30 .. 10 are the packageid: a namespace. +// - bits 9 .. 8 are reserved. +// - bits 7 .. 0 are a package-namespaced numeric code. // -// Do not manipulate these bits directly. Use the API functions such as -// wuffs_crc32__status__is_error instead. +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_crc32__status__is_error instead. typedef int32_t wuffs_crc32__status; #define wuffs_crc32__packageid 810620 // 0x000C5E7C @@ -284,14 +683,14 @@ extern "C" { // ---------------- Status Codes -// Status codes are int32_t values: -// - the sign bit indicates a non-recoverable status code: an error -// - bits 10-30 hold the packageid: a namespace -// - bits 8-9 are reserved -// - bits 0-7 are a package-namespaced numeric code +// Status codes are int32_t values. Its bits: +// - bit 31 (the sign bit) indicates unrecoverable-ness: an error. +// - bits 30 .. 10 are the packageid: a namespace. +// - bits 9 .. 8 are reserved. +// - bits 7 .. 0 are a package-namespaced numeric code. // -// Do not manipulate these bits directly. Use the API functions such as -// wuffs_deflate__status__is_error instead. +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_deflate__status__is_error instead. typedef int32_t wuffs_deflate__status; #define wuffs_deflate__packageid 848533 // 0x000CF295 @@ -457,14 +856,14 @@ extern "C" { // ---------------- Status Codes -// Status codes are int32_t values: -// - the sign bit indicates a non-recoverable status code: an error -// - bits 10-30 hold the packageid: a namespace -// - bits 8-9 are reserved -// - bits 0-7 are a package-namespaced numeric code +// Status codes are int32_t values. Its bits: +// - bit 31 (the sign bit) indicates unrecoverable-ness: an error. +// - bits 30 .. 10 are the packageid: a namespace. +// - bits 9 .. 8 are reserved. +// - bits 7 .. 0 are a package-namespaced numeric code. // -// Do not manipulate these bits directly. Use the API functions such as -// wuffs_gzip__status__is_error instead. +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_gzip__status__is_error instead. typedef int32_t wuffs_gzip__status; #define wuffs_gzip__packageid 1041911 // 0x000FE5F7 diff --git a/gen/h/std/zlib.h b/gen/h/std/zlib.h index 72797521..8ff097f4 100644 --- a/gen/h/std/zlib.h +++ b/gen/h/std/zlib.h @@ -46,11 +46,40 @@ // For now, the API and ABI are simply unstable and can change at any time. // // TODO: don't hard code this in base-header.h. -#define WUFFS_VERSION (0x00001) +#define WUFFS_VERSION 0x00001 -// ---------------- I/O +// ---------------- Fundamentals -// wuffs_base__slice_u8 is a 1-dimensional buffer (a pointer and length). +// Flicks are a unit of time. One flick (frame-tick) is 1 / 705_600_000 of a +// second. See https://github.com/OculusVR/Flicks +typedef uint64_t wuffs_base__flicks; + +#define WUFFS_BASE__FLICKS_PER_SECOND 705600000ULL + +// wuffs_base__rectangle is a rectangle on the integer grid. It contains all +// points (x, y) such that ((min_x <= x) && (x < max_x)) and likewise for y. It +// is therefore empty if min_x >= max_x. There are multiple representations of +// an empty rectangle. +// +// A value with all fields zero is a valid, empty rectangle. +// +// The X and Y axes increase right and down. +typedef struct { + uint32_t min_x; + uint32_t min_y; + uint32_t max_x; + uint32_t max_y; +} wuffs_base__rectangle; + +static inline uint32_t wuffs_base__rectangle__width(wuffs_base__rectangle r) { + return r.max_x > r.min_x ? r.max_x - r.min_x : 0; +} + +static inline uint32_t wuffs_base__rectangle__height(wuffs_base__rectangle r) { + return r.max_y > r.min_y ? r.max_y - r.min_y : 0; +} + +// wuffs_base__slice_u8 is a 1-dimensional buffer. // // A value with all fields NULL or zero is a valid, empty slice. typedef struct { @@ -58,6 +87,20 @@ typedef struct { size_t len; } wuffs_base__slice_u8; +// wuffs_base__table_u8 is a 2-dimensional buffer. +// +// A value with all fields NULL or zero is a valid, empty table. +typedef struct { + uint8_t* ptr; + size_t width; + size_t height; + size_t stride; +} wuffs_base__table_u8; + +// ---------------- I/O + +// TODO: rename buf1 to io_buffer, writer1 to io_writer, etc. + // wuffs_base__buf1 is a 1-dimensional buffer (a pointer and length), plus // additional indexes into that buffer, plus an opened / closed flag. // @@ -109,18 +152,286 @@ typedef struct { // ---------------- Images +// wuffs_base__pixel_format encodes the format of the bytes that constitute an +// image frame's pixel data. Its bits: +// - bit 31 is reserved. +// - bits 30 .. 28 encodes color (and channel order, in terms of memory). +// - bits 27 .. 26 are reserved. +// - bits 25 .. 24 encodes transparency. +// - bit 23 indicates big-endian/MSB-first (as opposed to little/LSB). +// - bit 22 indicates floating point (as opposed to integer). +// - bits 21 .. 20 are the number of planes, minus 1. Zero means packed. +// - bits 19 .. 16 encodes the number of bits (depth) in an index value. +// Zero means direct, not palette-indexed. +// - bits 15 .. 12 encodes the number of bits (depth) in the 3rd channel. +// - bits 11 .. 8 encodes the number of bits (depth) in the 2nd channel. +// - bits 7 .. 4 encodes the number of bits (depth) in the 1st channel. +// - bits 3 .. 0 encodes the number of bits (depth) in the 0th channel. +// +// The bit fields of a wuffs_base__pixel_format are not independent. For +// example, the number of planes should not be greater than the number of +// channels. Similarly, bits 15..4 are unused (and should be zero) if bits +// 31..24 (color and transparency) together imply only 1 channel (gray, no +// alpha) and floating point samples should mean a bit depth of 16, 32 or 64. +// +// Formats hold between 1 and 4 channels. For example: Y (1 channel: gray), YA +// (2 channels: gray and alpha), BGR (3 channels: blue, green, red) or CMYK (4 +// channels: cyan, magenta, yellow, black). +// +// For direct formats with N > 1 channels, those channels can be laid out in +// either 1 (packed) or N (planar) planes. For example, RGBA data is usually +// packed, but YUV data is usually planar, due to chroma subsampling (for +// details, see the wuffs_base__pixel_subsampling type). For indexed formats, +// the palette (always 256 × 4 bytes) holds up to 4 packed bytes of color data +// per index value, and there is only 1 plane (for the index). The distance +// between successive palette elements is always 4 bytes. +// +// The color field is encoded in 3 bits: +// - 0 means A (Alpha). +// - 1 means Y or YA (Gray, Alpha). +// - 2 means BGR, BGRX or BGRA (Blue, Green, Red, X-padding or Alpha). +// - 3 means RGB, RGBX or RGBA (Red, Green, Blue, X-padding or Alpha). +// - 4 means YUV or YUVA (Luma, Chroma-blue, Chroma-red, Alpha). +// - 5 means CMY or CMYK (Cyan, Magenta, Yellow, Black). +// - all other values are reserved. +// +// In Wuffs, channels are given in memory order, regardless of endianness, +// since the C type for the pixel data is an array of bytes, not an array of +// uint32_t. For example, packed BGRA with 8 bits per channel means that the +// bytes in memory are always Blue, Green, Red then Alpha. On big-endian +// systems, that is the uint32_t 0xBBGGRRAA. On little-endian, 0xAARRGGBB. +// +// When the color field (3 bits) encodes multiple options, the transparency +// field (2 bits) distinguishes them: +// - 0 means fully opaque, no extra channels +// - 1 means fully opaque, one extra channel (X or K, padding or black). +// - 2 means one extra alpha channel, other channels are non-premultiplied. +// - 3 means one extra alpha channel, other channels are premultiplied. +// +// The zero wuffs_base__pixel_format value is an invalid pixel format, as it is +// invalid to combine the zero color (alpha only) with the zero transparency. +// +// Bit depth is encoded in 4 bits: +// - 0 means the channel or index is unused. +// - x means a bit depth of x, for x in the range 1..8. +// - 9 means a bit depth of 10. +// - 10 means a bit depth of 12. +// - 11 means a bit depth of 16. +// - 12 means a bit depth of 24. +// - 13 means a bit depth of 32. +// - 14 means a bit depth of 48. +// - 15 means a bit depth of 64. +// +// For example, wuffs_base__pixel_format 0x3280BBBB is a natural format for +// decoding a PNG image - network byte order (also known as big-endian), +// packed, non-premultiplied alpha - that happens to be 16-bit-depth truecolor +// with alpha (RGBA). In memory order: +// +// ptr+0 ptr+1 ptr+2 ptr+3 ptr+4 ptr+5 ptr+6 ptr+7 +// Rhi Rlo Ghi Glo Bhi Blo Ahi Alo +// +// For example, the value wuffs_base__pixel_format 0x20000565 means BGR with no +// alpha or padding, 5/6/5 bits for blue/green/red, packed 2 bytes per pixel, +// laid out LSB-first in memory order: +// +// ptr+0........... ptr+1........... +// MSB LSB MSB LSB +// G₂G₁G₀B₄B₃B₂B₁B₀ R₄R₃R₂R₁R₀G₅G₄G₃ +// +// On little-endian systems (but not big-endian), this Wuffs pixel format value +// (0x20000565) corresponds to the Cairo library's CAIRO_FORMAT_RGB16_565, the +// SDL2 (Simple DirectMedia Layer 2) library's SDL_PIXELFORMAT_RGB565 and the +// Skia library's kRGB_565_SkColorType. Note BGR in Wuffs versus RGB in the +// other libraries. +// +// Regardless of endianness, this Wuffs pixel format value (0x20000565) +// corresponds to the V4L2 (Video For Linux 2) library's V4L2_PIX_FMT_RGB565 +// and the Wayland-DRM library's WL_DRM_FORMAT_RGB565. +// +// Different software libraries name their pixel formats (and especially their +// channel order) either according to memory layout or as bits of a native +// integer type like uint32_t. The two conventions differ because of a system's +// endianness. As mentioned earlier, Wuffs pixel formats are always in memory +// order. More detail of other software libraries' naming conventions is in the +// Pixel Format Guide at https://afrantzis.github.io/pixel-format-guide/ +// +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_base__pixel_format__num_planes instead. +typedef uint32_t wuffs_base__pixel_format; + +// Common 8-bit-depth pixel formats. This list is not exhaustive; not all valid +// wuffs_base__pixel_format values are present. + +#define WUFFS_BASE__PIXEL_FORMAT__INVALID ((wuffs_base__pixel_format)0x00000000) + +#define WUFFS_BASE__PIXEL_FORMAT__A ((wuffs_base__pixel_format)0x02000008) + +#define WUFFS_BASE__PIXEL_FORMAT__Y ((wuffs_base__pixel_format)0x10000008) +#define WUFFS_BASE__PIXEL_FORMAT__YA_NONPREMUL \ + ((wuffs_base__pixel_format)0x12000008) +#define WUFFS_BASE__PIXEL_FORMAT__YA_PREMUL \ + ((wuffs_base__pixel_format)0x13000008) + +#define WUFFS_BASE__PIXEL_FORMAT__BGR ((wuffs_base__pixel_format)0x20000888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRX ((wuffs_base__pixel_format)0x21008888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRX_INDEXED \ + ((wuffs_base__pixel_format)0x21088888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_NONPREMUL \ + ((wuffs_base__pixel_format)0x22008888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_NONPREMUL_INDEXED \ + ((wuffs_base__pixel_format)0x22088888) +#define WUFFS_BASE__PIXEL_FORMAT__BGRA_PREMUL \ + ((wuffs_base__pixel_format)0x23008888) + +#define WUFFS_BASE__PIXEL_FORMAT__RGB ((wuffs_base__pixel_format)0x30000888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBX ((wuffs_base__pixel_format)0x31008888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBX_INDEXED \ + ((wuffs_base__pixel_format)0x31088888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_NONPREMUL \ + ((wuffs_base__pixel_format)0x32008888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_NONPREMUL_INDEXED \ + ((wuffs_base__pixel_format)0x32088888) +#define WUFFS_BASE__PIXEL_FORMAT__RGBA_PREMUL \ + ((wuffs_base__pixel_format)0x33008888) + +#define WUFFS_BASE__PIXEL_FORMAT__YUV ((wuffs_base__pixel_format)0x40200888) +#define WUFFS_BASE__PIXEL_FORMAT__YUVK ((wuffs_base__pixel_format)0x41308888) +#define WUFFS_BASE__PIXEL_FORMAT__YUVA_NONPREMUL \ + ((wuffs_base__pixel_format)0x42308888) + +#define WUFFS_BASE__PIXEL_FORMAT__CMY ((wuffs_base__pixel_format)0x50200888) +#define WUFFS_BASE__PIXEL_FORMAT__CMYK ((wuffs_base__pixel_format)0x51308888) + +static inline bool wuffs_base__pixel_format__is_valid( + wuffs_base__pixel_format f) { + return f != 0; +} + +static inline bool wuffs_base__pixel_format__is_indexed( + wuffs_base__pixel_format f) { + return ((f >> 16) & 0x0F) != 0; +} + +#define WUFFS_BASE__PIXEL_FORMAT__NUM_PLANES_MAX 4 + +static inline uint32_t wuffs_base__pixel_format__num_planes( + wuffs_base__pixel_format f) { + return f ? (((f >> 20) & 0x03) + 1) : 0; +} + +typedef struct { + wuffs_base__table_u8 planes[WUFFS_BASE__PIXEL_FORMAT__NUM_PLANES_MAX]; +} wuffs_base__pixel_buffer; + +// wuffs_base__pixel_subsampling encodes the mapping of pixel space coordinates +// (x, y) to pixel buffer indices (i, j). That mapping can differ for each +// plane p. For a depth of 8 bits (1 byte), the p'th plane's sample starts at +// (planes[p].ptr + (j * planes[p].stride) + i). +// +// For packed pixel formats, the mapping is trivial: i = x and j = y. For +// planar pixel formats, the mapping can differ due to chroma subsampling. For +// example, consider a three plane YUV pixel format with 4:2:2 subsampling. For +// the luma (Y) channel, there is one sample for every pixel, but for the +// chroma (U, V) channels, there is one sample for every two pixels: pairs of +// horizontally adjacent pixels form one macropixel, i = x / 2 and j == y. In +// general, for a given p: +// - i = (x + bias_x) >> shift_x. +// - j = (y + bias_y) >> shift_y. +// where biases and shifts are in the range 0..3 and 0..2 respectively. +// +// In general, the biases will be zero after decoding an image. However, making +// a sub-image may change the bias, since the (x, y) coordinates are relative +// to the sub-image's top-left origin, but the backing pixel buffers were +// created relative to the original image's origin. +// +// For each plane p, each of those four numbers (biases and shifts) are encoded +// in two bits, which combine to form an 8 bit unsigned integer: +// +// e_p = (bias_x << 6) | (shift_x << 4) | (bias_y << 2) | (shift_y << 0) +// +// Those e_p values (e_0 for the first plane, e_1 for the second plane, etc) +// combine to form a wuffs_base__pixel_subsampling value: +// +// pixsub = (e_3 << 24) | (e_2 << 16) | (e_1 << 8) | (e_0 << 0) +// +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_base__pixel_subsampling__bias_x instead. +typedef uint32_t wuffs_base__pixel_subsampling; + +#define WUFFS_BASE__PIXEL_SUBSAMPLING__NONE ((wuffs_base__pixel_subsampling)0); + +#define WUFFS_BASE__PIXEL_SUBSAMPLING__444 \ + ((wuffs_base__pixel_subsampling)0x000000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__440 \ + ((wuffs_base__pixel_subsampling)0x010100); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__422 \ + ((wuffs_base__pixel_subsampling)0x101000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__420 \ + ((wuffs_base__pixel_subsampling)0x111100); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__411 \ + ((wuffs_base__pixel_subsampling)0x202000); +#define WUFFS_BASE__PIXEL_SUBSAMPLING__410 \ + ((wuffs_base__pixel_subsampling)0x212100); + +static inline uint32_t wuffs_base__pixel_subsampling__bias_x( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 6; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__shift_x( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 4; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__bias_y( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 2; + return (s >> shift) & 0x03; +} + +static inline uint32_t wuffs_base__pixel_subsampling__shift_y( + wuffs_base__pixel_subsampling s, + uint32_t plane) { + uint32_t shift = ((plane & 0x03) * 8) + 0; + return (s >> shift) & 0x03; +} + typedef struct { // Do not access the private_impl's fields directly. There is no API/ABI // compatibility or safety guarantee if you do so. struct { - uint32_t flags; - uint32_t w; - uint32_t h; - // TODO: color model, including both packed RGBA and planar, - // chroma-subsampled YCbCr. + wuffs_base__pixel_format pixfmt; + wuffs_base__pixel_subsampling pixsub; + uint32_t width; + uint32_t height; + uint32_t num_loops; } private_impl; } wuffs_base__image_config; +static inline void wuffs_base__image_config__initialize( + wuffs_base__image_config* c, + wuffs_base__pixel_format pixfmt, + wuffs_base__pixel_subsampling pixsub, + uint32_t width, + uint32_t height, + uint32_t num_loops) { + if (!c) { + return; + } + // TODO: move the check from wuffs_base__image_config__is_valid here. Should + // this function return bool? An error type? + c->private_impl.pixfmt = pixfmt; + c->private_impl.pixsub = pixsub; + c->private_impl.width = width; + c->private_impl.height = height; + c->private_impl.num_loops = num_loops; +} + static inline void wuffs_base__image_config__invalidate( wuffs_base__image_config* c) { if (c) { @@ -128,24 +439,42 @@ static inline void wuffs_base__image_config__invalidate( } } -static inline bool wuffs_base__image_config__valid( +static inline bool wuffs_base__image_config__is_valid( wuffs_base__image_config* c) { - if (!c || !(c->private_impl.flags & 1)) { + if (!c || !c->private_impl.pixfmt) { return false; } - uint64_t wh = ((uint64_t)c->private_impl.w) * ((uint64_t)c->private_impl.h); + uint64_t wh = + ((uint64_t)c->private_impl.width) * ((uint64_t)c->private_impl.height); // TODO: handle things other than 1 byte per pixel. + // + // TODO: move the check to wuffs_base__image_config__initialize. return wh <= ((uint64_t)SIZE_MAX); } +static inline wuffs_base__pixel_format wuffs_base__image_config__pixel_format( + wuffs_base__image_config* c) { + return c ? c->private_impl.pixfmt : 0; +} + +static inline wuffs_base__pixel_subsampling +wuffs_base__image_config__pixel_subsampling(wuffs_base__image_config* c) { + return wuffs_base__image_config__is_valid(c) ? c->private_impl.pixsub : 0; +} + static inline uint32_t wuffs_base__image_config__width( wuffs_base__image_config* c) { - return wuffs_base__image_config__valid(c) ? c->private_impl.w : 0; + return wuffs_base__image_config__is_valid(c) ? c->private_impl.width : 0; } static inline uint32_t wuffs_base__image_config__height( wuffs_base__image_config* c) { - return wuffs_base__image_config__valid(c) ? c->private_impl.h : 0; + return wuffs_base__image_config__is_valid(c) ? c->private_impl.height : 0; +} + +static inline uint32_t wuffs_base__image_config__num_loops( + wuffs_base__image_config* c) { + return wuffs_base__image_config__is_valid(c) ? c->private_impl.num_loops : 0; } // TODO: this is the right API for planar (not packed) pixbufs? Should it allow @@ -153,26 +482,96 @@ static inline uint32_t wuffs_base__image_config__height( // example, decoding a JPEG image straight to RGBA instead of to YCbCr? static inline size_t wuffs_base__image_config__pixbuf_size( wuffs_base__image_config* c) { - if (wuffs_base__image_config__valid(c)) { - uint64_t wh = ((uint64_t)c->private_impl.w) * ((uint64_t)c->private_impl.h); + if (wuffs_base__image_config__is_valid(c)) { + uint64_t wh = + ((uint64_t)c->private_impl.width) * ((uint64_t)c->private_impl.height); // TODO: handle things other than 1 byte per pixel. return (size_t)wh; } return 0; } -static inline void wuffs_base__image_config__initialize( - wuffs_base__image_config* c, - uint32_t width, - uint32_t height, - uint32_t TODO_color_model) { - if (!c) { +typedef struct { + // Do not access the private_impl's fields directly. There is no API/ABI + // compatibility or safety guarantee if you do so. + struct { + wuffs_base__image_config config; + uint32_t loop_count; // 0-based count of the current loop. + wuffs_base__pixel_buffer pixbuf; + // TODO: color spaces. + wuffs_base__rectangle dirty_rect; + wuffs_base__flicks duration; + uint8_t palette[1024]; + } private_impl; +} wuffs_base__image_buffer; + +static inline void wuffs_base__image_buffer__initialize( + wuffs_base__image_buffer* f, + wuffs_base__image_config config, + wuffs_base__pixel_buffer pixbuf) { + if (!f) { return; } - c->private_impl.flags = 1; - c->private_impl.w = width; - c->private_impl.h = height; - // TODO: color model. + *f = ((wuffs_base__image_buffer){}); + f->private_impl.config = config; + f->private_impl.pixbuf = pixbuf; +} + +static inline void wuffs_base__image_buffer__update( + wuffs_base__image_buffer* f, + wuffs_base__rectangle dirty_rect, + wuffs_base__flicks duration, + uint8_t* palette_ptr, + size_t palette_len) { + if (!f) { + return; + } + f->private_impl.dirty_rect = dirty_rect; + f->private_impl.duration = duration; + if (palette_ptr) { + memmove(f->private_impl.palette, palette_ptr, + palette_len <= 1024 ? palette_len : 1024); + } +} + +// wuffs_base__image_buffer__loop returns whether the image decoder should loop +// back to the beginning of the animation, assuming that we've reached the end +// of the encoded stream. If so, it increments f's count of the animation loops +// played so far. +static inline bool wuffs_base__image_buffer__loop(wuffs_base__image_buffer* f) { + if (!f) { + return false; + } + uint32_t n = f->private_impl.config.private_impl.num_loops; + if (n == 0) { + return true; + } + if (f->private_impl.loop_count < n - 1) { + f->private_impl.loop_count++; + return true; + } + return false; +} + +// wuffs_base__image_buffer__dirty_rect returns an upper bound for what part of +// this frame's pixels differs from the previous frame. +static inline wuffs_base__rectangle wuffs_base__image_buffer__dirty_rect( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.dirty_rect : ((wuffs_base__rectangle){0}); +} + +// wuffs_base__image_buffer__duration returns the amount of time to display +// this frame. Zero means to display forever - a still (non-animated) image. +static inline wuffs_base__flicks wuffs_base__image_buffer__duration( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.duration : 0; +} + +// wuffs_base__image_buffer__palette returns the palette that the pixel data +// can index. The backing array is inside f and has length 1024. +static inline uint8_t* wuffs_base__image_buffer__palette( + wuffs_base__image_buffer* f) { + return f ? f->private_impl.palette : NULL; } #endif // WUFFS_BASE_HEADER_H @@ -194,14 +593,14 @@ extern "C" { // ---------------- Status Codes -// Status codes are int32_t values: -// - the sign bit indicates a non-recoverable status code: an error -// - bits 10-30 hold the packageid: a namespace -// - bits 8-9 are reserved -// - bits 0-7 are a package-namespaced numeric code +// Status codes are int32_t values. Its bits: +// - bit 31 (the sign bit) indicates unrecoverable-ness: an error. +// - bits 30 .. 10 are the packageid: a namespace. +// - bits 9 .. 8 are reserved. +// - bits 7 .. 0 are a package-namespaced numeric code. // -// Do not manipulate these bits directly. Use the API functions such as -// wuffs_deflate__status__is_error instead. +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_deflate__status__is_error instead. typedef int32_t wuffs_deflate__status; #define wuffs_deflate__packageid 848533 // 0x000CF295 @@ -367,14 +766,14 @@ extern "C" { // ---------------- Status Codes -// Status codes are int32_t values: -// - the sign bit indicates a non-recoverable status code: an error -// - bits 10-30 hold the packageid: a namespace -// - bits 8-9 are reserved -// - bits 0-7 are a package-namespaced numeric code +// Status codes are int32_t values. Its bits: +// - bit 31 (the sign bit) indicates unrecoverable-ness: an error. +// - bits 30 .. 10 are the packageid: a namespace. +// - bits 9 .. 8 are reserved. +// - bits 7 .. 0 are a package-namespaced numeric code. // -// Do not manipulate these bits directly. Use the API functions such as -// wuffs_zlib__status__is_error instead. +// Do not manipulate these bits directly; they are private implementation +// details. Use methods such as wuffs_zlib__status__is_error instead. typedef int32_t wuffs_zlib__status; #define wuffs_zlib__packageid 2064249 // 0x001F7F79 diff --git a/lang/builtin/builtin.go b/lang/builtin/builtin.go index 77760cc1..5bf201a6 100644 --- a/lang/builtin/builtin.go +++ b/lang/builtin/builtin.go @@ -127,7 +127,7 @@ var Funcs = []string{ "writer1.mark()()", "writer1.since_mark()(ret[] u8)", - "image_config.initialize!(width u32, height u32, color_model u32)()", + "image_config.initialize!(pixfmt u32, pixsub u32, width u32, height u32, num_loops u32)()", } const ( diff --git a/std/deflate/decode_deflate.wuffs b/std/deflate/decode_deflate.wuffs index 2b85e1ef..28d1f3b4 100644 --- a/std/deflate/decode_deflate.wuffs +++ b/std/deflate/decode_deflate.wuffs @@ -78,16 +78,16 @@ pub struct decoder?( // the table value's number of decoder.bits to consume. If it compares // less, then more source bytes are read and the table lookup re-tried. // - // The table value is: - // - bit 31 indicates the symbol is a literal. - // - bit 30 indicates the symbol is a base number + extra bits. - // - bit 29 indicates the symbol is the end-of-block. - // - bit 28 indicates a redirect to another part of the table. - // - bit 27 indicates an invalid value. - // - bits 26-24 are zero. - // - bits 23- 8 is the redirect offset, literal (in bits 15-8) or base number. - // - bits 7- 4 is the redirected table's bits or number of extra bits. - // - bits 3- 0 are the number of decoder.bits to consume. + // The table value's bits: + // - bit 31 indicates a literal. + // - bit 30 indicates a base number + extra bits. + // - bit 29 indicates end-of-block. + // - bit 28 indicates a redirect to another part of the table. + // - bit 27 indicates an invalid value. + // - bits 26 .. 24 are zero. + // - bits 23 .. 8 are the redirect offset, literal (in bits 15-8) or base number. + // - bits 7 .. 4 are the redirected table's bits or number of extra bits. + // - bits 3 .. 0 are the number of decoder.bits to consume. // // Exactly one of the eight bits 31-24 should be set. // diff --git a/std/gif/decode_gif.wuffs b/std/gif/decode_gif.wuffs index 486e311c..a4fe519a 100644 --- a/std/gif/decode_gif.wuffs +++ b/std/gif/decode_gif.wuffs @@ -77,7 +77,11 @@ pub func decoder.decode_config?(dst ptr image_config, src reader1)() { } // TODO: rename initialize to set? - in.dst.initialize!(width:this.width, height:this.height, color_model:0) + // + // TODO: a Wuffs (not just C) name for the + // WUFFS_BASE__PIXEL_FORMAT__BGRA_NONPREMUL_INDEXED magic pixfmt constant. + in.dst.initialize!(pixfmt:0x22088888, pixsub:0, + width:this.width, height:this.height, num_loops:this.num_loops) this.call_sequence = 1 } diff --git a/test/c/std/gif.c b/test/c/std/gif.c index 5b0b1c71..f9a5ae71 100644 --- a/test/c/std/gif.c +++ b/test/c/std/gif.c @@ -401,8 +401,15 @@ bool do_test_wuffs_gif_decode(const char* filename, wuffs_gif__status__string(status)); return false; } + if (wuffs_base__image_config__pixel_format(&ic) != + WUFFS_BASE__PIXEL_FORMAT__BGRA_NONPREMUL_INDEXED) { + FAIL("pixel_format: got 0x%08" PRIX32 ", want 0x%08" PRIX32, + wuffs_base__image_config__pixel_format(&ic), + WUFFS_BASE__PIXEL_FORMAT__BGRA_NONPREMUL_INDEXED); + return false; + } - // bricks-dither.gif is a 160 × 120 static (not animated) GIF. + // bricks-dither.gif is a 160 × 120 still (not animated) GIF. if (wuffs_base__image_config__width(&ic) != 160) { FAIL("width: got %" PRIu32 ", want 160", wuffs_base__image_config__width(&ic)); @@ -413,9 +420,9 @@ bool do_test_wuffs_gif_decode(const char* filename, wuffs_base__image_config__height(&ic)); return false; } - // TODO: provide a public API for getting num_loops. - if (dec.private_impl.f_num_loops != 1) { - FAIL("num_loops: got %" PRIu32 ", want 1", dec.private_impl.f_num_loops); + if (wuffs_base__image_config__num_loops(&ic) != 1) { + FAIL("num_loops: got %" PRIu32 ", want 1", + wuffs_base__image_config__num_loops(&ic)); return false; } } @@ -579,10 +586,9 @@ void test_wuffs_gif_decode_animated() { // animated-red-blue.gif's num_loops should be 3. The value explicitly in the // wire format is 0x0002, but that value means "repeat 2 times after the // first play", so the total number of loops is 3. - // - // TODO: provide a public API for getting num_loops. - if (dec.private_impl.f_num_loops != 3) { - FAIL("num_loops: got %" PRIu32 ", want 3", dec.private_impl.f_num_loops); + if (wuffs_base__image_config__num_loops(&ic) != 3) { + FAIL("num_loops: got %" PRIu32 ", want 3", + wuffs_base__image_config__num_loops(&ic)); return; } @@ -612,6 +618,8 @@ void test_wuffs_gif_decode_animated() { wuffs_gif__status__string(WUFFS_GIF__SUSPENSION_END_OF_DATA)); return; } + + // TODO: test calling wuffs_base__image_buffer__loop. } void test_wuffs_gif_decode_input_is_a_gif() {