diff options
| -rw-r--r-- | TODO | 3 | ||||
| -rw-r--r-- | external/stb_image_write.h | 1724 | ||||
| -rw-r--r-- | include/capture/capture.h | 20 | ||||
| -rw-r--r-- | include/encoder/video/image.h | 15 | ||||
| -rw-r--r-- | include/image_writer.h | 31 | ||||
| -rw-r--r-- | meson.build | 2 | ||||
| -rw-r--r-- | src/capture/capture.c | 12 | ||||
| -rw-r--r-- | src/capture/kms.c | 31 | ||||
| -rw-r--r-- | src/capture/nvfbc.c | 56 | ||||
| -rw-r--r-- | src/capture/portal.c | 32 | ||||
| -rw-r--r-- | src/capture/xcomposite.c | 29 | ||||
| -rw-r--r-- | src/egl.c | 39 | ||||
| -rw-r--r-- | src/encoder/video/image.c | 130 | ||||
| -rw-r--r-- | src/image_writer.c | 85 | ||||
| -rw-r--r-- | src/main.cpp | 255 | ||||
| -rw-r--r-- | src/utils.c | 5 |
16 files changed, 2097 insertions, 372 deletions
@@ -230,3 +230,6 @@ Add an option to pass http headers when streaming. Some streaming services requi When adding vulkan video support add VK_VIDEO_ENCODE_TUNING_MODE_LOW_LATENCY_KHR. Implement screenshot without invoking opengl (which is slow to start on some systems). + +Automatically use desktop portal on wayland when hdr is enabled (or night light) by checking if kms hdr metadata exists, if hdr video codec is not used. + Or maybe do this in the ui?
\ No newline at end of file diff --git a/external/stb_image_write.h b/external/stb_image_write.h new file mode 100644 index 0000000..e4b32ed --- /dev/null +++ b/external/stb_image_write.h @@ -0,0 +1,1724 @@ +/* stb_image_write - v1.16 - public domain - http://nothings.org/stb + writes out PNG/BMP/TGA/JPEG/HDR images to C stdio - Sean Barrett 2010-2015 + no warranty implied; use at your own risk + + Before #including, + + #define STB_IMAGE_WRITE_IMPLEMENTATION + + in the file that you want to have the implementation. + + Will probably not work correctly with strict-aliasing optimizations. + +ABOUT: + + This header file is a library for writing images to C stdio or a callback. + + The PNG output is not optimal; it is 20-50% larger than the file + written by a decent optimizing implementation; though providing a custom + zlib compress function (see STBIW_ZLIB_COMPRESS) can mitigate that. + This library is designed for source code compactness and simplicity, + not optimal image file size or run-time performance. + +BUILDING: + + You can #define STBIW_ASSERT(x) before the #include to avoid using assert.h. + You can #define STBIW_MALLOC(), STBIW_REALLOC(), and STBIW_FREE() to replace + malloc,realloc,free. + You can #define STBIW_MEMMOVE() to replace memmove() + You can #define STBIW_ZLIB_COMPRESS to use a custom zlib-style compress function + for PNG compression (instead of the builtin one), it must have the following signature: + unsigned char * my_compress(unsigned char *data, int data_len, int *out_len, int quality); + The returned data will be freed with STBIW_FREE() (free() by default), + so it must be heap allocated with STBIW_MALLOC() (malloc() by default), + +UNICODE: + + If compiling for Windows and you wish to use Unicode filenames, compile + with + #define STBIW_WINDOWS_UTF8 + and pass utf8-encoded filenames. Call stbiw_convert_wchar_to_utf8 to convert + Windows wchar_t filenames to utf8. + +USAGE: + + There are five functions, one for each image file format: + + int stbi_write_png(char const *filename, int w, int h, int comp, const void *data, int stride_in_bytes); + int stbi_write_bmp(char const *filename, int w, int h, int comp, const void *data); + int stbi_write_tga(char const *filename, int w, int h, int comp, const void *data); + int stbi_write_jpg(char const *filename, int w, int h, int comp, const void *data, int quality); + int stbi_write_hdr(char const *filename, int w, int h, int comp, const float *data); + + void stbi_flip_vertically_on_write(int flag); // flag is non-zero to flip data vertically + + There are also five equivalent functions that use an arbitrary write function. You are + expected to open/close your file-equivalent before and after calling these: + + int stbi_write_png_to_func(stbi_write_func *func, void *context, int w, int h, int comp, const void *data, int stride_in_bytes); + int stbi_write_bmp_to_func(stbi_write_func *func, void *context, int w, int h, int comp, const void *data); + int stbi_write_tga_to_func(stbi_write_func *func, void *context, int w, int h, int comp, const void *data); + int stbi_write_hdr_to_func(stbi_write_func *func, void *context, int w, int h, int comp, const float *data); + int stbi_write_jpg_to_func(stbi_write_func *func, void *context, int x, int y, int comp, const void *data, int quality); + + where the callback is: + void stbi_write_func(void *context, void *data, int size); + + You can configure it with these global variables: + int stbi_write_tga_with_rle; // defaults to true; set to 0 to disable RLE + int stbi_write_png_compression_level; // defaults to 8; set to higher for more compression + int stbi_write_force_png_filter; // defaults to -1; set to 0..5 to force a filter mode + + + You can define STBI_WRITE_NO_STDIO to disable the file variant of these + functions, so the library will not use stdio.h at all. However, this will + also disable HDR writing, because it requires stdio for formatted output. + + Each function returns 0 on failure and non-0 on success. + + The functions create an image file defined by the parameters. The image + is a rectangle of pixels stored from left-to-right, top-to-bottom. + Each pixel contains 'comp' channels of data stored interleaved with 8-bits + per channel, in the following order: 1=Y, 2=YA, 3=RGB, 4=RGBA. (Y is + monochrome color.) The rectangle is 'w' pixels wide and 'h' pixels tall. + The *data pointer points to the first byte of the top-left-most pixel. + For PNG, "stride_in_bytes" is the distance in bytes from the first byte of + a row of pixels to the first byte of the next row of pixels. + + PNG creates output files with the same number of components as the input. + The BMP format expands Y to RGB in the file format and does not + output alpha. + + PNG supports writing rectangles of data even when the bytes storing rows of + data are not consecutive in memory (e.g. sub-rectangles of a larger image), + by supplying the stride between the beginning of adjacent rows. The other + formats do not. (Thus you cannot write a native-format BMP through the BMP + writer, both because it is in BGR order and because it may have padding + at the end of the line.) + + PNG allows you to set the deflate compression level by setting the global + variable 'stbi_write_png_compression_level' (it defaults to 8). + + HDR expects linear float data. Since the format is always 32-bit rgb(e) + data, alpha (if provided) is discarded, and for monochrome data it is + replicated across all three channels. + + TGA supports RLE or non-RLE compressed data. To use non-RLE-compressed + data, set the global variable 'stbi_write_tga_with_rle' to 0. + + JPEG does ignore alpha channels in input data; quality is between 1 and 100. + Higher quality looks better but results in a bigger image. + JPEG baseline (no JPEG progressive). + +CREDITS: + + + Sean Barrett - PNG/BMP/TGA + Baldur Karlsson - HDR + Jean-Sebastien Guay - TGA monochrome + Tim Kelsey - misc enhancements + Alan Hickman - TGA RLE + Emmanuel Julien - initial file IO callback implementation + Jon Olick - original jo_jpeg.cpp code + Daniel Gibson - integrate JPEG, allow external zlib + Aarni Koskela - allow choosing PNG filter + + bugfixes: + github:Chribba + Guillaume Chereau + github:jry2 + github:romigrou + Sergio Gonzalez + Jonas Karlsson + Filip Wasil + Thatcher Ulrich + github:poppolopoppo + Patrick Boettcher + github:xeekworx + Cap Petschulat + Simon Rodriguez + Ivan Tikhonov + github:ignotion + Adam Schackart + Andrew Kensler + +LICENSE + + See end of file for license information. + +*/ + +#ifndef INCLUDE_STB_IMAGE_WRITE_H +#define INCLUDE_STB_IMAGE_WRITE_H + +#include <stdlib.h> + +// if STB_IMAGE_WRITE_STATIC causes problems, try defining STBIWDEF to 'inline' or 'static inline' +#ifndef STBIWDEF +#ifdef STB_IMAGE_WRITE_STATIC +#define STBIWDEF static +#else +#ifdef __cplusplus +#define STBIWDEF extern "C" +#else +#define STBIWDEF extern +#endif +#endif +#endif + +#ifndef STB_IMAGE_WRITE_STATIC // C++ forbids static forward declarations +STBIWDEF int stbi_write_tga_with_rle; +STBIWDEF int stbi_write_png_compression_level; +STBIWDEF int stbi_write_force_png_filter; +#endif + +#ifndef STBI_WRITE_NO_STDIO +STBIWDEF int stbi_write_png(char const *filename, int w, int h, int comp, const void *data, int stride_in_bytes); +STBIWDEF int stbi_write_bmp(char const *filename, int w, int h, int comp, const void *data); +STBIWDEF int stbi_write_tga(char const *filename, int w, int h, int comp, const void *data); +STBIWDEF int stbi_write_hdr(char const *filename, int w, int h, int comp, const float *data); +STBIWDEF int stbi_write_jpg(char const *filename, int x, int y, int comp, const void *data, int quality); + +#ifdef STBIW_WINDOWS_UTF8 +STBIWDEF int stbiw_convert_wchar_to_utf8(char *buffer, size_t bufferlen, const wchar_t* input); +#endif +#endif + +typedef void stbi_write_func(void *context, void *data, int size); + +STBIWDEF int stbi_write_png_to_func(stbi_write_func *func, void *context, int w, int h, int comp, const void *data, int stride_in_bytes); +STBIWDEF int stbi_write_bmp_to_func(stbi_write_func *func, void *context, int w, int h, int comp, const void *data); +STBIWDEF int stbi_write_tga_to_func(stbi_write_func *func, void *context, int w, int h, int comp, const void *data); +STBIWDEF int stbi_write_hdr_to_func(stbi_write_func *func, void *context, int w, int h, int comp, const float *data); +STBIWDEF int stbi_write_jpg_to_func(stbi_write_func *func, void *context, int x, int y, int comp, const void *data, int quality); + +STBIWDEF void stbi_flip_vertically_on_write(int flip_boolean); + +#endif//INCLUDE_STB_IMAGE_WRITE_H + +#ifdef STB_IMAGE_WRITE_IMPLEMENTATION + +#ifdef _WIN32 + #ifndef _CRT_SECURE_NO_WARNINGS + #define _CRT_SECURE_NO_WARNINGS + #endif + #ifndef _CRT_NONSTDC_NO_DEPRECATE + #define _CRT_NONSTDC_NO_DEPRECATE + #endif +#endif + +#ifndef STBI_WRITE_NO_STDIO +#include <stdio.h> +#endif // STBI_WRITE_NO_STDIO + +#include <stdarg.h> +#include <stdlib.h> +#include <string.h> +#include <math.h> + +#if defined(STBIW_MALLOC) && defined(STBIW_FREE) && (defined(STBIW_REALLOC) || defined(STBIW_REALLOC_SIZED)) +// ok +#elif !defined(STBIW_MALLOC) && !defined(STBIW_FREE) && !defined(STBIW_REALLOC) && !defined(STBIW_REALLOC_SIZED) +// ok +#else +#error "Must define all or none of STBIW_MALLOC, STBIW_FREE, and STBIW_REALLOC (or STBIW_REALLOC_SIZED)." +#endif + +#ifndef STBIW_MALLOC +#define STBIW_MALLOC(sz) malloc(sz) +#define STBIW_REALLOC(p,newsz) realloc(p,newsz) +#define STBIW_FREE(p) free(p) +#endif + +#ifndef STBIW_REALLOC_SIZED +#define STBIW_REALLOC_SIZED(p,oldsz,newsz) STBIW_REALLOC(p,newsz) +#endif + + +#ifndef STBIW_MEMMOVE +#define STBIW_MEMMOVE(a,b,sz) memmove(a,b,sz) +#endif + + +#ifndef STBIW_ASSERT +#include <assert.h> +#define STBIW_ASSERT(x) assert(x) +#endif + +#define STBIW_UCHAR(x) (unsigned char) ((x) & 0xff) + +#ifdef STB_IMAGE_WRITE_STATIC +static int stbi_write_png_compression_level = 8; +static int stbi_write_tga_with_rle = 1; +static int stbi_write_force_png_filter = -1; +#else +int stbi_write_png_compression_level = 8; +int stbi_write_tga_with_rle = 1; +int stbi_write_force_png_filter = -1; +#endif + +static int stbi__flip_vertically_on_write = 0; + +STBIWDEF void stbi_flip_vertically_on_write(int flag) +{ + stbi__flip_vertically_on_write = flag; +} + +typedef struct +{ + stbi_write_func *func; + void *context; + unsigned char buffer[64]; + int buf_used; +} stbi__write_context; + +// initialize a callback-based context +static void stbi__start_write_callbacks(stbi__write_context *s, stbi_write_func *c, void *context) +{ + s->func = c; + s->context = context; +} + +#ifndef STBI_WRITE_NO_STDIO + +static void stbi__stdio_write(void *context, void *data, int size) +{ + fwrite(data,1,size,(FILE*) context); +} + +#if defined(_WIN32) && defined(STBIW_WINDOWS_UTF8) +#ifdef __cplusplus +#define STBIW_EXTERN extern "C" +#else +#define STBIW_EXTERN extern +#endif +STBIW_EXTERN __declspec(dllimport) int __stdcall MultiByteToWideChar(unsigned int cp, unsigned long flags, const char *str, int cbmb, wchar_t *widestr, int cchwide); +STBIW_EXTERN __declspec(dllimport) int __stdcall WideCharToMultiByte(unsigned int cp, unsigned long flags, const wchar_t *widestr, int cchwide, char *str, int cbmb, const char *defchar, int *used_default); + +STBIWDEF int stbiw_convert_wchar_to_utf8(char *buffer, size_t bufferlen, const wchar_t* input) +{ + return WideCharToMultiByte(65001 /* UTF8 */, 0, input, -1, buffer, (int) bufferlen, NULL, NULL); +} +#endif + +static FILE *stbiw__fopen(char const *filename, char const *mode) +{ + FILE *f; +#if defined(_WIN32) && defined(STBIW_WINDOWS_UTF8) + wchar_t wMode[64]; + wchar_t wFilename[1024]; + if (0 == MultiByteToWideChar(65001 /* UTF8 */, 0, filename, -1, wFilename, sizeof(wFilename)/sizeof(*wFilename))) + return 0; + + if (0 == MultiByteToWideChar(65001 /* UTF8 */, 0, mode, -1, wMode, sizeof(wMode)/sizeof(*wMode))) + return 0; + +#if defined(_MSC_VER) && _MSC_VER >= 1400 + if (0 != _wfopen_s(&f, wFilename, wMode)) + f = 0; +#else + f = _wfopen(wFilename, wMode); +#endif + +#elif defined(_MSC_VER) && _MSC_VER >= 1400 + if (0 != fopen_s(&f, filename, mode)) + f=0; +#else + f = fopen(filename, mode); +#endif + return f; +} + +static int stbi__start_write_file(stbi__write_context *s, const char *filename) +{ + FILE *f = stbiw__fopen(filename, "wb"); + stbi__start_write_callbacks(s, stbi__stdio_write, (void *) f); + return f != NULL; +} + +static void stbi__end_write_file(stbi__write_context *s) +{ + fclose((FILE *)s->context); +} + +#endif // !STBI_WRITE_NO_STDIO + +typedef unsigned int stbiw_uint32; +typedef int stb_image_write_test[sizeof(stbiw_uint32)==4 ? 1 : -1]; + +static void stbiw__writefv(stbi__write_context *s, const char *fmt, va_list v) +{ + while (*fmt) { + switch (*fmt++) { + case ' ': break; + case '1': { unsigned char x = STBIW_UCHAR(va_arg(v, int)); + s->func(s->context,&x,1); + break; } + case '2': { int x = va_arg(v,int); + unsigned char b[2]; + b[0] = STBIW_UCHAR(x); + b[1] = STBIW_UCHAR(x>>8); + s->func(s->context,b,2); + break; } + case '4': { stbiw_uint32 x = va_arg(v,int); + unsigned char b[4]; + b[0]=STBIW_UCHAR(x); + b[1]=STBIW_UCHAR(x>>8); + b[2]=STBIW_UCHAR(x>>16); + b[3]=STBIW_UCHAR(x>>24); + s->func(s->context,b,4); + break; } + default: + STBIW_ASSERT(0); + return; + } + } +} + +static void stbiw__writef(stbi__write_context *s, const char *fmt, ...) +{ + va_list v; + va_start(v, fmt); + stbiw__writefv(s, fmt, v); + va_end(v); +} + +static void stbiw__write_flush(stbi__write_context *s) +{ + if (s->buf_used) { + s->func(s->context, &s->buffer, s->buf_used); + s->buf_used = 0; + } +} + +static void stbiw__putc(stbi__write_context *s, unsigned char c) +{ + s->func(s->context, &c, 1); +} + +static void stbiw__write1(stbi__write_context *s, unsigned char a) +{ + if ((size_t)s->buf_used + 1 > sizeof(s->buffer)) + stbiw__write_flush(s); + s->buffer[s->buf_used++] = a; +} + +static void stbiw__write3(stbi__write_context *s, unsigned char a, unsigned char b, unsigned char c) +{ + int n; + if ((size_t)s->buf_used + 3 > sizeof(s->buffer)) + stbiw__write_flush(s); + n = s->buf_used; + s->buf_used = n+3; + s->buffer[n+0] = a; + s->buffer[n+1] = b; + s->buffer[n+2] = c; +} + +static void stbiw__write_pixel(stbi__write_context *s, int rgb_dir, int comp, int write_alpha, int expand_mono, unsigned char *d) +{ + unsigned char bg[3] = { 255, 0, 255}, px[3]; + int k; + + if (write_alpha < 0) + stbiw__write1(s, d[comp - 1]); + + switch (comp) { + case 2: // 2 pixels = mono + alpha, alpha is written separately, so same as 1-channel case + case 1: + if (expand_mono) + stbiw__write3(s, d[0], d[0], d[0]); // monochrome bmp + else + stbiw__write1(s, d[0]); // monochrome TGA + break; + case 4: + if (!write_alpha) { + // composite against pink background + for (k = 0; k < 3; ++k) + px[k] = bg[k] + ((d[k] - bg[k]) * d[3]) / 255; + stbiw__write3(s, px[1 - rgb_dir], px[1], px[1 + rgb_dir]); + break; + } + /* FALLTHROUGH */ + case 3: + stbiw__write3(s, d[1 - rgb_dir], d[1], d[1 + rgb_dir]); + break; + } + if (write_alpha > 0) + stbiw__write1(s, d[comp - 1]); +} + +static void stbiw__write_pixels(stbi__write_context *s, int rgb_dir, int vdir, int x, int y, int comp, void *data, int write_alpha, int scanline_pad, int expand_mono) +{ + stbiw_uint32 zero = 0; + int i,j, j_end; + + if (y <= 0) + return; + + if (stbi__flip_vertically_on_write) + vdir *= -1; + + if (vdir < 0) { + j_end = -1; j = y-1; + } else { + j_end = y; j = 0; + } + + for (; j != j_end; j += vdir) { + for (i=0; i < x; ++i) { + unsigned char *d = (unsigned char *) data + (j*x+i)*comp; + stbiw__write_pixel(s, rgb_dir, comp, write_alpha, expand_mono, d); + } + stbiw__write_flush(s); + s->func(s->context, &zero, scanline_pad); + } +} + +static int stbiw__outfile(stbi__write_context *s, int rgb_dir, int vdir, int x, int y, int comp, int expand_mono, void *data, int alpha, int pad, const char *fmt, ...) +{ + if (y < 0 || x < 0) { + return 0; + } else { + va_list v; + va_start(v, fmt); + stbiw__writefv(s, fmt, v); + va_end(v); + stbiw__write_pixels(s,rgb_dir,vdir,x,y,comp,data,alpha,pad, expand_mono); + return 1; + } +} + +static int stbi_write_bmp_core(stbi__write_context *s, int x, int y, int comp, const void *data) +{ + if (comp != 4) { + // write RGB bitmap + int pad = (-x*3) & 3; + return stbiw__outfile(s,-1,-1,x,y,comp,1,(void *) data,0,pad, + "11 4 22 4" "4 44 22 444444", + 'B', 'M', 14+40+(x*3+pad)*y, 0,0, 14+40, // file header + 40, x,y, 1,24, 0,0,0,0,0,0); // bitmap header + } else { + // RGBA bitmaps need a v4 header + // use BI_BITFIELDS mode with 32bpp and alpha mask + // (straight BI_RGB with alpha mask doesn't work in most readers) + return stbiw__outfile(s,-1,-1,x,y,comp,1,(void *)data,1,0, + "11 4 22 4" "4 44 22 444444 4444 4 444 444 444 444", + 'B', 'M', 14+108+x*y*4, 0, 0, 14+108, // file header + 108, x,y, 1,32, 3,0,0,0,0,0, 0xff0000,0xff00,0xff,0xff000000u, 0, 0,0,0, 0,0,0, 0,0,0, 0,0,0); // bitmap V4 header + } +} + +STBIWDEF int stbi_write_bmp_to_func(stbi_write_func *func, void *context, int x, int y, int comp, const void *data) +{ + stbi__write_context s = { 0 }; + stbi__start_write_callbacks(&s, func, context); + return stbi_write_bmp_core(&s, x, y, comp, data); +} + +#ifndef STBI_WRITE_NO_STDIO +STBIWDEF int stbi_write_bmp(char const *filename, int x, int y, int comp, const void *data) +{ + stbi__write_context s = { 0 }; + if (stbi__start_write_file(&s,filename)) { + int r = stbi_write_bmp_core(&s, x, y, comp, data); + stbi__end_write_file(&s); + return r; + } else + return 0; +} +#endif //!STBI_WRITE_NO_STDIO + +static int stbi_write_tga_core(stbi__write_context *s, int x, int y, int comp, void *data) +{ + int has_alpha = (comp == 2 || comp == 4); + int colorbytes = has_alpha ? comp-1 : comp; + int format = colorbytes < 2 ? 3 : 2; // 3 color channels (RGB/RGBA) = 2, 1 color channel (Y/YA) = 3 + + if (y < 0 || x < 0) + return 0; + + if (!stbi_write_tga_with_rle) { + return stbiw__outfile(s, -1, -1, x, y, comp, 0, (void *) data, has_alpha, 0, + "111 221 2222 11", 0, 0, format, 0, 0, 0, 0, 0, x, y, (colorbytes + has_alpha) * 8, has_alpha * 8); + } else { + int i,j,k; + int jend, jdir; + + stbiw__writef(s, "111 221 2222 11", 0,0,format+8, 0,0,0, 0,0,x,y, (colorbytes + has_alpha) * 8, has_alpha * 8); + + if (stbi__flip_vertically_on_write) { + j = 0; + jend = y; + jdir = 1; + } else { + j = y-1; + jend = -1; + jdir = -1; + } + for (; j != jend; j += jdir) { + unsigned char *row = (unsigned char *) data + j * x * comp; + int len; + + for (i = 0; i < x; i += len) { + unsigned char *begin = row + i * comp; + int diff = 1; + len = 1; + + if (i < x - 1) { + ++len; + diff = memcmp(begin, row + (i + 1) * comp, comp); + if (diff) { + const unsigned char *prev = begin; + for (k = i + 2; k < x && len < 128; ++k) { + if (memcmp(prev, row + k * comp, comp)) { + prev += comp; + ++len; + } else { + --len; + break; + } + } + } else { + for (k = i + 2; k < x && len < 128; ++k) { + if (!memcmp(begin, row + k * comp, comp)) { + ++len; + } else { + break; + } + } + } + } + + if (diff) { + unsigned char header = STBIW_UCHAR(len - 1); + stbiw__write1(s, header); + for (k = 0; k < len; ++k) { + stbiw__write_pixel(s, -1, comp, has_alpha, 0, begin + k * comp); + } + } else { + unsigned char header = STBIW_UCHAR(len - 129); + stbiw__write1(s, header); + stbiw__write_pixel(s, -1, comp, has_alpha, 0, begin); + } + } + } + stbiw__write_flush(s); + } + return 1; +} + +STBIWDEF int stbi_write_tga_to_func(stbi_write_func *func, void *context, int x, int y, int comp, const void *data) +{ + stbi__write_context s = { 0 }; + stbi__start_write_callbacks(&s, func, context); + return stbi_write_tga_core(&s, x, y, comp, (void *) data); +} + +#ifndef STBI_WRITE_NO_STDIO +STBIWDEF int stbi_write_tga(char const *filename, int x, int y, int comp, const void *data) +{ + stbi__write_context s = { 0 }; + if (stbi__start_write_file(&s,filename)) { + int r = stbi_write_tga_core(&s, x, y, comp, (void *) data); + stbi__end_write_file(&s); + return r; + } else + return 0; +} +#endif + +// ************************************************************************************************* +// Radiance RGBE HDR writer +// by Baldur Karlsson + +#define stbiw__max(a, b) ((a) > (b) ? (a) : (b)) + +#ifndef STBI_WRITE_NO_STDIO + +static void stbiw__linear_to_rgbe(unsigned char *rgbe, float *linear) +{ + int exponent; + float maxcomp = stbiw__max(linear[0], stbiw__max(linear[1], linear[2])); + + if (maxcomp < 1e-32f) { + rgbe[0] = rgbe[1] = rgbe[2] = rgbe[3] = 0; + } else { + float normalize = (float) frexp(maxcomp, &exponent) * 256.0f/maxcomp; + + rgbe[0] = (unsigned char)(linear[0] * normalize); + rgbe[1] = (unsigned char)(linear[1] * normalize); + rgbe[2] = (unsigned char)(linear[2] * normalize); + rgbe[3] = (unsigned char)(exponent + 128); + } +} + +static void stbiw__write_run_data(stbi__write_context *s, int length, unsigned char databyte) +{ + unsigned char lengthbyte = STBIW_UCHAR(length+128); + STBIW_ASSERT(length+128 <= 255); + s->func(s->context, &lengthbyte, 1); + s->func(s->context, &databyte, 1); +} + +static void stbiw__write_dump_data(stbi__write_context *s, int length, unsigned char *data) +{ + unsigned char lengthbyte = STBIW_UCHAR(length); + STBIW_ASSERT(length <= 128); // inconsistent with spec but consistent with official code + s->func(s->context, &lengthbyte, 1); + s->func(s->context, data, length); +} + +static void stbiw__write_hdr_scanline(stbi__write_context *s, int width, int ncomp, unsigned char *scratch, float *scanline) +{ + unsigned char scanlineheader[4] = { 2, 2, 0, 0 }; + unsigned char rgbe[4]; + float linear[3]; + int x; + + scanlineheader[2] = (width&0xff00)>>8; + scanlineheader[3] = (width&0x00ff); + + /* skip RLE for images too small or large */ + if (width < 8 || width >= 32768) { + for (x=0; x < width; x++) { + switch (ncomp) { + case 4: /* fallthrough */ + case 3: linear[2] = scanline[x*ncomp + 2]; + linear[1] = scanline[x*ncomp + 1]; + linear[0] = scanline[x*ncomp + 0]; + break; + default: + linear[0] = linear[1] = linear[2] = scanline[x*ncomp + 0]; + break; + } + stbiw__linear_to_rgbe(rgbe, linear); + s->func(s->context, rgbe, 4); + } + } else { + int c,r; + /* encode into scratch buffer */ + for (x=0; x < width; x++) { + switch(ncomp) { + case 4: /* fallthrough */ + case 3: linear[2] = scanline[x*ncomp + 2]; + linear[1] = scanline[x*ncomp + 1]; + linear[0] = scanline[x*ncomp + 0]; + break; + default: + linear[0] = linear[1] = linear[2] = scanline[x*ncomp + 0]; + break; + } + stbiw__linear_to_rgbe(rgbe, linear); + scratch[x + width*0] = rgbe[0]; + scratch[x + width*1] = rgbe[1]; + scratch[x + width*2] = rgbe[2]; + scratch[x + width*3] = rgbe[3]; + } + + s->func(s->context, scanlineheader, 4); + + /* RLE each component separately */ + for (c=0; c < 4; c++) { + unsigned char *comp = &scratch[width*c]; + + x = 0; + while (x < width) { + // find first run + r = x; + while (r+2 < width) { + if (comp[r] == comp[r+1] && comp[r] == comp[r+2]) + break; + ++r; + } + if (r+2 >= width) + r = width; + // dump up to first run + while (x < r) { + int len = r-x; + if (len > 128) len = 128; + stbiw__write_dump_data(s, len, &comp[x]); + x += len; + } + // if there's a run, output it + if (r+2 < width) { // same test as what we break out of in search loop, so only true if we break'd + // find next byte after run + while (r < width && comp[r] == comp[x]) + ++r; + // output run up to r + while (x < r) { + int len = r-x; + if (len > 127) len = 127; + stbiw__write_run_data(s, len, comp[x]); + x += len; + } + } + } + } + } +} + +static int stbi_write_hdr_core(stbi__write_context *s, int x, int y, int comp, float *data) +{ + if (y <= 0 || x <= 0 || data == NULL) + return 0; + else { + // Each component is stored separately. Allocate scratch space for full output scanline. + unsigned char *scratch = (unsigned char *) STBIW_MALLOC(x*4); + int i, len; + char buffer[128]; + char header[] = "#?RADIANCE\n# Written by stb_image_write.h\nFORMAT=32-bit_rle_rgbe\n"; + s->func(s->context, header, sizeof(header)-1); + +#ifdef __STDC_LIB_EXT1__ + len = sprintf_s(buffer, sizeof(buffer), "EXPOSURE= 1.0000000000000\n\n-Y %d +X %d\n", y, x); +#else + len = sprintf(buffer, "EXPOSURE= 1.0000000000000\n\n-Y %d +X %d\n", y, x); +#endif + s->func(s->context, buffer, len); + + for(i=0; i < y; i++) + stbiw__write_hdr_scanline(s, x, comp, scratch, data + comp*x*(stbi__flip_vertically_on_write ? y-1-i : i)); + STBIW_FREE(scratch); + return 1; + } +} + +STBIWDEF int stbi_write_hdr_to_func(stbi_write_func *func, void *context, int x, int y, int comp, const float *data) +{ + stbi__write_context s = { 0 }; + stbi__start_write_callbacks(&s, func, context); + return stbi_write_hdr_core(&s, x, y, comp, (float *) data); +} + +STBIWDEF int stbi_write_hdr(char const *filename, int x, int y, int comp, const float *data) +{ + stbi__write_context s = { 0 }; + if (stbi__start_write_file(&s,filename)) { + int r = stbi_write_hdr_core(&s, x, y, comp, (float *) data); + stbi__end_write_file(&s); + return r; + } else + return 0; +} +#endif // STBI_WRITE_NO_STDIO + + +////////////////////////////////////////////////////////////////////////////// +// +// PNG writer +// + +#ifndef STBIW_ZLIB_COMPRESS +// stretchy buffer; stbiw__sbpush() == vector<>::push_back() -- stbiw__sbcount() == vector<>::size() +#define stbiw__sbraw(a) ((int *) (void *) (a) - 2) +#define stbiw__sbm(a) stbiw__sbraw(a)[0] +#define stbiw__sbn(a) stbiw__sbraw(a)[1] + +#define stbiw__sbneedgrow(a,n) ((a)==0 || stbiw__sbn(a)+n >= stbiw__sbm(a)) +#define stbiw__sbmaybegrow(a,n) (stbiw__sbneedgrow(a,(n)) ? stbiw__sbgrow(a,n) : 0) +#define stbiw__sbgrow(a,n) stbiw__sbgrowf((void **) &(a), (n), sizeof(*(a))) + +#define stbiw__sbpush(a, v) (stbiw__sbmaybegrow(a,1), (a)[stbiw__sbn(a)++] = (v)) +#define stbiw__sbcount(a) ((a) ? stbiw__sbn(a) : 0) +#define stbiw__sbfree(a) ((a) ? STBIW_FREE(stbiw__sbraw(a)),0 : 0) + +static void *stbiw__sbgrowf(void **arr, int increment, int itemsize) +{ + int m = *arr ? 2*stbiw__sbm(*arr)+increment : increment+1; + void *p = STBIW_REALLOC_SIZED(*arr ? stbiw__sbraw(*arr) : 0, *arr ? (stbiw__sbm(*arr)*itemsize + sizeof(int)*2) : 0, itemsize * m + sizeof(int)*2); + STBIW_ASSERT(p); + if (p) { + if (!*arr) ((int *) p)[1] = 0; + *arr = (void *) ((int *) p + 2); + stbiw__sbm(*arr) = m; + } + return *arr; +} + +static unsigned char *stbiw__zlib_flushf(unsigned char *data, unsigned int *bitbuffer, int *bitcount) +{ + while (*bitcount >= 8) { + stbiw__sbpush(data, STBIW_UCHAR(*bitbuffer)); + *bitbuffer >>= 8; + *bitcount -= 8; + } + return data; +} + +static int stbiw__zlib_bitrev(int code, int codebits) +{ + int res=0; + while (codebits--) { + res = (res << 1) | (code & 1); + code >>= 1; + } + return res; +} + +static unsigned int stbiw__zlib_countm(unsigned char *a, unsigned char *b, int limit) +{ + int i; + for (i=0; i < limit && i < 258; ++i) + if (a[i] != b[i]) break; + return i; +} + +static unsigned int stbiw__zhash(unsigned char *data) +{ + stbiw_uint32 hash = data[0] + (data[1] << 8) + (data[2] << 16); + hash ^= hash << 3; + hash += hash >> 5; + hash ^= hash << 4; + hash += hash >> 17; + hash ^= hash << 25; + hash += hash >> 6; + return hash; +} + +#define stbiw__zlib_flush() (out = stbiw__zlib_flushf(out, &bitbuf, &bitcount)) +#define stbiw__zlib_add(code,codebits) \ + (bitbuf |= (code) << bitcount, bitcount += (codebits), stbiw__zlib_flush()) +#define stbiw__zlib_huffa(b,c) stbiw__zlib_add(stbiw__zlib_bitrev(b,c),c) +// default huffman tables +#define stbiw__zlib_huff1(n) stbiw__zlib_huffa(0x30 + (n), 8) +#define stbiw__zlib_huff2(n) stbiw__zlib_huffa(0x190 + (n)-144, 9) +#define stbiw__zlib_huff3(n) stbiw__zlib_huffa(0 + (n)-256,7) +#define stbiw__zlib_huff4(n) stbiw__zlib_huffa(0xc0 + (n)-280,8) +#define stbiw__zlib_huff(n) ((n) <= 143 ? stbiw__zlib_huff1(n) : (n) <= 255 ? stbiw__zlib_huff2(n) : (n) <= 279 ? stbiw__zlib_huff3(n) : stbiw__zlib_huff4(n)) +#define stbiw__zlib_huffb(n) ((n) <= 143 ? stbiw__zlib_huff1(n) : stbiw__zlib_huff2(n)) + +#define stbiw__ZHASH 16384 + +#endif // STBIW_ZLIB_COMPRESS + +STBIWDEF unsigned char * stbi_zlib_compress(unsigned char *data, int data_len, int *out_len, int quality) +{ +#ifdef STBIW_ZLIB_COMPRESS + // user provided a zlib compress implementation, use that + return STBIW_ZLIB_COMPRESS(data, data_len, out_len, quality); +#else // use builtin + static unsigned short lengthc[] = { 3,4,5,6,7,8,9,10,11,13,15,17,19,23,27,31,35,43,51,59,67,83,99,115,131,163,195,227,258, 259 }; + static unsigned char lengtheb[]= { 0,0,0,0,0,0,0, 0, 1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3, 4, 4, 4, 4, 5, 5, 5, 5, 0 }; + static unsigned short distc[] = { 1,2,3,4,5,7,9,13,17,25,33,49,65,97,129,193,257,385,513,769,1025,1537,2049,3073,4097,6145,8193,12289,16385,24577, 32768 }; + static unsigned char disteb[] = { 0,0,0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,12,12,13,13 }; + unsigned int bitbuf=0; + int i,j, bitcount=0; + unsigned char *out = NULL; + unsigned char ***hash_table = (unsigned char***) STBIW_MALLOC(stbiw__ZHASH * sizeof(unsigned char**)); + if (hash_table == NULL) + return NULL; + if (quality < 5) quality = 5; + + stbiw__sbpush(out, 0x78); // DEFLATE 32K window + stbiw__sbpush(out, 0x5e); // FLEVEL = 1 + stbiw__zlib_add(1,1); // BFINAL = 1 + stbiw__zlib_add(1,2); // BTYPE = 1 -- fixed huffman + + for (i=0; i < stbiw__ZHASH; ++i) + hash_table[i] = NULL; + + i=0; + while (i < data_len-3) { + // hash next 3 bytes of data to be compressed + int h = stbiw__zhash(data+i)&(stbiw__ZHASH-1), best=3; + unsigned char *bestloc = 0; + unsigned char **hlist = hash_table[h]; + int n = stbiw__sbcount(hlist); + for (j=0; j < n; ++j) { + if (hlist[j]-data > i-32768) { // if entry lies within window + int d = stbiw__zlib_countm(hlist[j], data+i, data_len-i); + if (d >= best) { best=d; bestloc=hlist[j]; } + } + } + // when hash table entry is too long, delete half the entries + if (hash_table[h] && stbiw__sbn(hash_table[h]) == 2*quality) { + STBIW_MEMMOVE(hash_table[h], hash_table[h]+quality, sizeof(hash_table[h][0])*quality); + stbiw__sbn(hash_table[h]) = quality; + } + stbiw__sbpush(hash_table[h],data+i); + + if (bestloc) { + // "lazy matching" - check match at *next* byte, and if it's better, do cur byte as literal + h = stbiw__zhash(data+i+1)&(stbiw__ZHASH-1); + hlist = hash_table[h]; + n = stbiw__sbcount(hlist); + for (j=0; j < n; ++j) { + if (hlist[j]-data > i-32767) { + int e = stbiw__zlib_countm(hlist[j], data+i+1, data_len-i-1); + if (e > best) { // if next match is better, bail on current match + bestloc = NULL; + break; + } + } + } + } + + if (bestloc) { + int d = (int) (data+i - bestloc); // distance back + STBIW_ASSERT(d <= 32767 && best <= 258); + for (j=0; best > lengthc[j+1]-1; ++j); + stbiw__zlib_huff(j+257); + if (lengtheb[j]) stbiw__zlib_add(best - lengthc[j], lengtheb[j]); + for (j=0; d > distc[j+1]-1; ++j); + stbiw__zlib_add(stbiw__zlib_bitrev(j,5),5); + if (disteb[j]) stbiw__zlib_add(d - distc[j], disteb[j]); + i += best; + } else { + stbiw__zlib_huffb(data[i]); + ++i; + } + } + // write out final bytes + for (;i < data_len; ++i) + stbiw__zlib_huffb(data[i]); + stbiw__zlib_huff(256); // end of block + // pad with 0 bits to byte boundary + while (bitcount) + stbiw__zlib_add(0,1); + + for (i=0; i < stbiw__ZHASH; ++i) + (void) stbiw__sbfree(hash_table[i]); + STBIW_FREE(hash_table); + + // store uncompressed instead if compression was worse + if (stbiw__sbn(out) > data_len + 2 + ((data_len+32766)/32767)*5) { + stbiw__sbn(out) = 2; // truncate to DEFLATE 32K window and FLEVEL = 1 + for (j = 0; j < data_len;) { + int blocklen = data_len - j; + if (blocklen > 32767) blocklen = 32767; + stbiw__sbpush(out, data_len - j == blocklen); // BFINAL = ?, BTYPE = 0 -- no compression + stbiw__sbpush(out, STBIW_UCHAR(blocklen)); // LEN + stbiw__sbpush(out, STBIW_UCHAR(blocklen >> 8)); + stbiw__sbpush(out, STBIW_UCHAR(~blocklen)); // NLEN + stbiw__sbpush(out, STBIW_UCHAR(~blocklen >> 8)); + memcpy(out+stbiw__sbn(out), data+j, blocklen); + stbiw__sbn(out) += blocklen; + j += blocklen; + } + } + + { + // compute adler32 on input + unsigned int s1=1, s2=0; + int blocklen = (int) (data_len % 5552); + j=0; + while (j < data_len) { + for (i=0; i < blocklen; ++i) { s1 += data[j+i]; s2 += s1; } + s1 %= 65521; s2 %= 65521; + j += blocklen; + blocklen = 5552; + } + stbiw__sbpush(out, STBIW_UCHAR(s2 >> 8)); + stbiw__sbpush(out, STBIW_UCHAR(s2)); + stbiw__sbpush(out, STBIW_UCHAR(s1 >> 8)); + stbiw__sbpush(out, STBIW_UCHAR(s1)); + } + *out_len = stbiw__sbn(out); + // make returned pointer freeable + STBIW_MEMMOVE(stbiw__sbraw(out), out, *out_len); + return (unsigned char *) stbiw__sbraw(out); +#endif // STBIW_ZLIB_COMPRESS +} + +static unsigned int stbiw__crc32(unsigned char *buffer, int len) +{ +#ifdef STBIW_CRC32 + return STBIW_CRC32(buffer, len); +#else + static unsigned int crc_table[256] = + { + 0x00000000, 0x77073096, 0xEE0E612C, 0x990951BA, 0x076DC419, 0x706AF48F, 0xE963A535, 0x9E6495A3, + 0x0eDB8832, 0x79DCB8A4, 0xE0D5E91E, 0x97D2D988, 0x09B64C2B, 0x7EB17CBD, 0xE7B82D07, 0x90BF1D91, + 0x1DB71064, 0x6AB020F2, 0xF3B97148, 0x84BE41DE, 0x1ADAD47D, 0x6DDDE4EB, 0xF4D4B551, 0x83D385C7, + 0x136C9856, 0x646BA8C0, 0xFD62F97A, 0x8A65C9EC, 0x14015C4F, 0x63066CD9, 0xFA0F3D63, 0x8D080DF5, + 0x3B6E20C8, 0x4C69105E, 0xD56041E4, 0xA2677172, 0x3C03E4D1, 0x4B04D447, 0xD20D85FD, 0xA50AB56B, + 0x35B5A8FA, 0x42B2986C, 0xDBBBC9D6, 0xACBCF940, 0x32D86CE3, 0x45DF5C75, 0xDCD60DCF, 0xABD13D59, + 0x26D930AC, 0x51DE003A, 0xC8D75180, 0xBFD06116, 0x21B4F4B5, 0x56B3C423, 0xCFBA9599, 0xB8BDA50F, + 0x2802B89E, 0x5F058808, 0xC60CD9B2, 0xB10BE924, 0x2F6F7C87, 0x58684C11, 0xC1611DAB, 0xB6662D3D, + 0x76DC4190, 0x01DB7106, 0x98D220BC, 0xEFD5102A, 0x71B18589, 0x06B6B51F, 0x9FBFE4A5, 0xE8B8D433, + 0x7807C9A2, 0x0F00F934, 0x9609A88E, 0xE10E9818, 0x7F6A0DBB, 0x086D3D2D, 0x91646C97, 0xE6635C01, + 0x6B6B51F4, 0x1C6C6162, 0x856530D8, 0xF262004E, 0x6C0695ED, 0x1B01A57B, 0x8208F4C1, 0xF50FC457, + 0x65B0D9C6, 0x12B7E950, 0x8BBEB8EA, 0xFCB9887C, 0x62DD1DDF, 0x15DA2D49, 0x8CD37CF3, 0xFBD44C65, + 0x4DB26158, 0x3AB551CE, 0xA3BC0074, 0xD4BB30E2, 0x4ADFA541, 0x3DD895D7, 0xA4D1C46D, 0xD3D6F4FB, + 0x4369E96A, 0x346ED9FC, 0xAD678846, 0xDA60B8D0, 0x44042D73, 0x33031DE5, 0xAA0A4C5F, 0xDD0D7CC9, + 0x5005713C, 0x270241AA, 0xBE0B1010, 0xC90C2086, 0x5768B525, 0x206F85B3, 0xB966D409, 0xCE61E49F, + 0x5EDEF90E, 0x29D9C998, 0xB0D09822, 0xC7D7A8B4, 0x59B33D17, 0x2EB40D81, 0xB7BD5C3B, 0xC0BA6CAD, + 0xEDB88320, 0x9ABFB3B6, 0x03B6E20C, 0x74B1D29A, 0xEAD54739, 0x9DD277AF, 0x04DB2615, 0x73DC1683, + 0xE3630B12, 0x94643B84, 0x0D6D6A3E, 0x7A6A5AA8, 0xE40ECF0B, 0x9309FF9D, 0x0A00AE27, 0x7D079EB1, + 0xF00F9344, 0x8708A3D2, 0x1E01F268, 0x6906C2FE, 0xF762575D, 0x806567CB, 0x196C3671, 0x6E6B06E7, + 0xFED41B76, 0x89D32BE0, 0x10DA7A5A, 0x67DD4ACC, 0xF9B9DF6F, 0x8EBEEFF9, 0x17B7BE43, 0x60B08ED5, + 0xD6D6A3E8, 0xA1D1937E, 0x38D8C2C4, 0x4FDFF252, 0xD1BB67F1, 0xA6BC5767, 0x3FB506DD, 0x48B2364B, + 0xD80D2BDA, 0xAF0A1B4C, 0x36034AF6, 0x41047A60, 0xDF60EFC3, 0xA867DF55, 0x316E8EEF, 0x4669BE79, + 0xCB61B38C, 0xBC66831A, 0x256FD2A0, 0x5268E236, 0xCC0C7795, 0xBB0B4703, 0x220216B9, 0x5505262F, + 0xC5BA3BBE, 0xB2BD0B28, 0x2BB45A92, 0x5CB36A04, 0xC2D7FFA7, 0xB5D0CF31, 0x2CD99E8B, 0x5BDEAE1D, + 0x9B64C2B0, 0xEC63F226, 0x756AA39C, 0x026D930A, 0x9C0906A9, 0xEB0E363F, 0x72076785, 0x05005713, + 0x95BF4A82, 0xE2B87A14, 0x7BB12BAE, 0x0CB61B38, 0x92D28E9B, 0xE5D5BE0D, 0x7CDCEFB7, 0x0BDBDF21, + 0x86D3D2D4, 0xF1D4E242, 0x68DDB3F8, 0x1FDA836E, 0x81BE16CD, 0xF6B9265B, 0x6FB077E1, 0x18B74777, + 0x88085AE6, 0xFF0F6A70, 0x66063BCA, 0x11010B5C, 0x8F659EFF, 0xF862AE69, 0x616BFFD3, 0x166CCF45, + 0xA00AE278, 0xD70DD2EE, 0x4E048354, 0x3903B3C2, 0xA7672661, 0xD06016F7, 0x4969474D, 0x3E6E77DB, + 0xAED16A4A, 0xD9D65ADC, 0x40DF0B66, 0x37D83BF0, 0xA9BCAE53, 0xDEBB9EC5, 0x47B2CF7F, 0x30B5FFE9, + 0xBDBDF21C, 0xCABAC28A, 0x53B39330, 0x24B4A3A6, 0xBAD03605, 0xCDD70693, 0x54DE5729, 0x23D967BF, + 0xB3667A2E, 0xC4614AB8, 0x5D681B02, 0x2A6F2B94, 0xB40BBE37, 0xC30C8EA1, 0x5A05DF1B, 0x2D02EF8D + }; + + unsigned int crc = ~0u; + int i; + for (i=0; i < len; ++i) + crc = (crc >> 8) ^ crc_table[buffer[i] ^ (crc & 0xff)]; + return ~crc; +#endif +} + +#define stbiw__wpng4(o,a,b,c,d) ((o)[0]=STBIW_UCHAR(a),(o)[1]=STBIW_UCHAR(b),(o)[2]=STBIW_UCHAR(c),(o)[3]=STBIW_UCHAR(d),(o)+=4) +#define stbiw__wp32(data,v) stbiw__wpng4(data, (v)>>24,(v)>>16,(v)>>8,(v)); +#define stbiw__wptag(data,s) stbiw__wpng4(data, s[0],s[1],s[2],s[3]) + +static void stbiw__wpcrc(unsigned char **data, int len) +{ + unsigned int crc = stbiw__crc32(*data - len - 4, len+4); + stbiw__wp32(*data, crc); +} + +static unsigned char stbiw__paeth(int a, int b, int c) +{ + int p = a + b - c, pa = abs(p-a), pb = abs(p-b), pc = abs(p-c); + if (pa <= pb && pa <= pc) return STBIW_UCHAR(a); + if (pb <= pc) return STBIW_UCHAR(b); + return STBIW_UCHAR(c); +} + +// @OPTIMIZE: provide an option that always forces left-predict or paeth predict +static void stbiw__encode_png_line(unsigned char *pixels, int stride_bytes, int width, int height, int y, int n, int filter_type, signed char *line_buffer) +{ + static int mapping[] = { 0,1,2,3,4 }; + static int firstmap[] = { 0,1,0,5,6 }; + int *mymap = (y != 0) ? mapping : firstmap; + int i; + int type = mymap[filter_type]; + unsigned char *z = pixels + stride_bytes * (stbi__flip_vertically_on_write ? height-1-y : y); + int signed_stride = stbi__flip_vertically_on_write ? -stride_bytes : stride_bytes; + + if (type==0) { + memcpy(line_buffer, z, width*n); + return; + } + + // first loop isn't optimized since it's just one pixel + for (i = 0; i < n; ++i) { + switch (type) { + case 1: line_buffer[i] = z[i]; break; + case 2: line_buffer[i] = z[i] - z[i-signed_stride]; break; + case 3: line_buffer[i] = z[i] - (z[i-signed_stride]>>1); break; + case 4: line_buffer[i] = (signed char) (z[i] - stbiw__paeth(0,z[i-signed_stride],0)); break; + case 5: line_buffer[i] = z[i]; break; + case 6: line_buffer[i] = z[i]; break; + } + } + switch (type) { + case 1: for (i=n; i < width*n; ++i) line_buffer[i] = z[i] - z[i-n]; break; + case 2: for (i=n; i < width*n; ++i) line_buffer[i] = z[i] - z[i-signed_stride]; break; + case 3: for (i=n; i < width*n; ++i) line_buffer[i] = z[i] - ((z[i-n] + z[i-signed_stride])>>1); break; + case 4: for (i=n; i < width*n; ++i) line_buffer[i] = z[i] - stbiw__paeth(z[i-n], z[i-signed_stride], z[i-signed_stride-n]); break; + case 5: for (i=n; i < width*n; ++i) line_buffer[i] = z[i] - (z[i-n]>>1); break; + case 6: for (i=n; i < width*n; ++i) line_buffer[i] = z[i] - stbiw__paeth(z[i-n], 0,0); break; + } +} + +STBIWDEF unsigned char *stbi_write_png_to_mem(const unsigned char *pixels, int stride_bytes, int x, int y, int n, int *out_len) +{ + int force_filter = stbi_write_force_png_filter; + int ctype[5] = { -1, 0, 4, 2, 6 }; + unsigned char sig[8] = { 137,80,78,71,13,10,26,10 }; + unsigned char *out,*o, *filt, *zlib; + signed char *line_buffer; + int j,zlen; + + if (stride_bytes == 0) + stride_bytes = x * n; + + if (force_filter >= 5) { + force_filter = -1; + } + + filt = (unsigned char *) STBIW_MALLOC((x*n+1) * y); if (!filt) return 0; + line_buffer = (signed char *) STBIW_MALLOC(x * n); if (!line_buffer) { STBIW_FREE(filt); return 0; } + for (j=0; j < y; ++j) { + int filter_type; + if (force_filter > -1) { + filter_type = force_filter; + stbiw__encode_png_line((unsigned char*)(pixels), stride_bytes, x, y, j, n, force_filter, line_buffer); + } else { // Estimate the best filter by running through all of them: + int best_filter = 0, best_filter_val = 0x7fffffff, est, i; + for (filter_type = 0; filter_type < 5; filter_type++) { + stbiw__encode_png_line((unsigned char*)(pixels), stride_bytes, x, y, j, n, filter_type, line_buffer); + + // Estimate the entropy of the line using this filter; the less, the better. + est = 0; + for (i = 0; i < x*n; ++i) { + est += abs((signed char) line_buffer[i]); + } + if (est < best_filter_val) { + best_filter_val = est; + best_filter = filter_type; + } + } + if (filter_type != best_filter) { // If the last iteration already got us the best filter, don't redo it + stbiw__encode_png_line((unsigned char*)(pixels), stride_bytes, x, y, j, n, best_filter, line_buffer); + filter_type = best_filter; + } + } + // when we get here, filter_type contains the filter type, and line_buffer contains the data + filt[j*(x*n+1)] = (unsigned char) filter_type; + STBIW_MEMMOVE(filt+j*(x*n+1)+1, line_buffer, x*n); + } + STBIW_FREE(line_buffer); + zlib = stbi_zlib_compress(filt, y*( x*n+1), &zlen, stbi_write_png_compression_level); + STBIW_FREE(filt); + if (!zlib) return 0; + + // each tag requires 12 bytes of overhead + out = (unsigned char *) STBIW_MALLOC(8 + 12+13 + 12+zlen + 12); + if (!out) return 0; + *out_len = 8 + 12+13 + 12+zlen + 12; + + o=out; + STBIW_MEMMOVE(o,sig,8); o+= 8; + stbiw__wp32(o, 13); // header length + stbiw__wptag(o, "IHDR"); + stbiw__wp32(o, x); + stbiw__wp32(o, y); + *o++ = 8; + *o++ = STBIW_UCHAR(ctype[n]); + *o++ = 0; + *o++ = 0; + *o++ = 0; + stbiw__wpcrc(&o,13); + + stbiw__wp32(o, zlen); + stbiw__wptag(o, "IDAT"); + STBIW_MEMMOVE(o, zlib, zlen); + o += zlen; + STBIW_FREE(zlib); + stbiw__wpcrc(&o, zlen); + + stbiw__wp32(o,0); + stbiw__wptag(o, "IEND"); + stbiw__wpcrc(&o,0); + + STBIW_ASSERT(o == out + *out_len); + + return out; +} + +#ifndef STBI_WRITE_NO_STDIO +STBIWDEF int stbi_write_png(char const *filename, int x, int y, int comp, const void *data, int stride_bytes) +{ + FILE *f; + int len; + unsigned char *png = stbi_write_png_to_mem((const unsigned char *) data, stride_bytes, x, y, comp, &len); + if (png == NULL) return 0; + + f = stbiw__fopen(filename, "wb"); + if (!f) { STBIW_FREE(png); return 0; } + fwrite(png, 1, len, f); + fclose(f); + STBIW_FREE(png); + return 1; +} +#endif + +STBIWDEF int stbi_write_png_to_func(stbi_write_func *func, void *context, int x, int y, int comp, const void *data, int stride_bytes) +{ + int len; + unsigned char *png = stbi_write_png_to_mem((const unsigned char *) data, stride_bytes, x, y, comp, &len); + if (png == NULL) return 0; + func(context, png, len); + STBIW_FREE(png); + return 1; +} + + +/* *************************************************************************** + * + * JPEG writer + * + * This is based on Jon Olick's jo_jpeg.cpp: + * public domain Simple, Minimalistic JPEG writer - http://www.jonolick.com/code.html + */ + +static const unsigned char stbiw__jpg_ZigZag[] = { 0,1,5,6,14,15,27,28,2,4,7,13,16,26,29,42,3,8,12,17,25,30,41,43,9,11,18, + 24,31,40,44,53,10,19,23,32,39,45,52,54,20,22,33,38,46,51,55,60,21,34,37,47,50,56,59,61,35,36,48,49,57,58,62,63 }; + +static void stbiw__jpg_writeBits(stbi__write_context *s, int *bitBufP, int *bitCntP, const unsigned short *bs) { + int bitBuf = *bitBufP, bitCnt = *bitCntP; + bitCnt += bs[1]; + bitBuf |= bs[0] << (24 - bitCnt); + while(bitCnt >= 8) { + unsigned char c = (bitBuf >> 16) & 255; + stbiw__putc(s, c); + if(c == 255) { + stbiw__putc(s, 0); + } + bitBuf <<= 8; + bitCnt -= 8; + } + *bitBufP = bitBuf; + *bitCntP = bitCnt; +} + +static void stbiw__jpg_DCT(float *d0p, float *d1p, float *d2p, float *d3p, float *d4p, float *d5p, float *d6p, float *d7p) { + float d0 = *d0p, d1 = *d1p, d2 = *d2p, d3 = *d3p, d4 = *d4p, d5 = *d5p, d6 = *d6p, d7 = *d7p; + float z1, z2, z3, z4, z5, z11, z13; + + float tmp0 = d0 + d7; + float tmp7 = d0 - d7; + float tmp1 = d1 + d6; + float tmp6 = d1 - d6; + float tmp2 = d2 + d5; + float tmp5 = d2 - d5; + float tmp3 = d3 + d4; + float tmp4 = d3 - d4; + + // Even part + float tmp10 = tmp0 + tmp3; // phase 2 + float tmp13 = tmp0 - tmp3; + float tmp11 = tmp1 + tmp2; + float tmp12 = tmp1 - tmp2; + + d0 = tmp10 + tmp11; // phase 3 + d4 = tmp10 - tmp11; + + z1 = (tmp12 + tmp13) * 0.707106781f; // c4 + d2 = tmp13 + z1; // phase 5 + d6 = tmp13 - z1; + + // Odd part + tmp10 = tmp4 + tmp5; // phase 2 + tmp11 = tmp5 + tmp6; + tmp12 = tmp6 + tmp7; + + // The rotator is modified from fig 4-8 to avoid extra negations. + z5 = (tmp10 - tmp12) * 0.382683433f; // c6 + z2 = tmp10 * 0.541196100f + z5; // c2-c6 + z4 = tmp12 * 1.306562965f + z5; // c2+c6 + z3 = tmp11 * 0.707106781f; // c4 + + z11 = tmp7 + z3; // phase 5 + z13 = tmp7 - z3; + + *d5p = z13 + z2; // phase 6 + *d3p = z13 - z2; + *d1p = z11 + z4; + *d7p = z11 - z4; + + *d0p = d0; *d2p = d2; *d4p = d4; *d6p = d6; +} + +static void stbiw__jpg_calcBits(int val, unsigned short bits[2]) { + int tmp1 = val < 0 ? -val : val; + val = val < 0 ? val-1 : val; + bits[1] = 1; + while(tmp1 >>= 1) { + ++bits[1]; + } + bits[0] = val & ((1<<bits[1])-1); +} + +static int stbiw__jpg_processDU(stbi__write_context *s, int *bitBuf, int *bitCnt, float *CDU, int du_stride, float *fdtbl, int DC, const unsigned short HTDC[256][2], const unsigned short HTAC[256][2]) { + const unsigned short EOB[2] = { HTAC[0x00][0], HTAC[0x00][1] }; + const unsigned short M16zeroes[2] = { HTAC[0xF0][0], HTAC[0xF0][1] }; + int dataOff, i, j, n, diff, end0pos, x, y; + int DU[64]; + + // DCT rows + for(dataOff=0, n=du_stride*8; dataOff<n; dataOff+=du_stride) { + stbiw__jpg_DCT(&CDU[dataOff], &CDU[dataOff+1], &CDU[dataOff+2], &CDU[dataOff+3], &CDU[dataOff+4], &CDU[dataOff+5], &CDU[dataOff+6], &CDU[dataOff+7]); + } + // DCT columns + for(dataOff=0; dataOff<8; ++dataOff) { + stbiw__jpg_DCT(&CDU[dataOff], &CDU[dataOff+du_stride], &CDU[dataOff+du_stride*2], &CDU[dataOff+du_stride*3], &CDU[dataOff+du_stride*4], + &CDU[dataOff+du_stride*5], &CDU[dataOff+du_stride*6], &CDU[dataOff+du_stride*7]); + } + // Quantize/descale/zigzag the coefficients + for(y = 0, j=0; y < 8; ++y) { + for(x = 0; x < 8; ++x,++j) { + float v; + i = y*du_stride+x; + v = CDU[i]*fdtbl[j]; + // DU[stbiw__jpg_ZigZag[j]] = (int)(v < 0 ? ceilf(v - 0.5f) : floorf(v + 0.5f)); + // ceilf() and floorf() are C99, not C89, but I /think/ they're not needed here anyway? + DU[stbiw__jpg_ZigZag[j]] = (int)(v < 0 ? v - 0.5f : v + 0.5f); + } + } + + // Encode DC + diff = DU[0] - DC; + if (diff == 0) { + stbiw__jpg_writeBits(s, bitBuf, bitCnt, HTDC[0]); + } else { + unsigned short bits[2]; + stbiw__jpg_calcBits(diff, bits); + stbiw__jpg_writeBits(s, bitBuf, bitCnt, HTDC[bits[1]]); + stbiw__jpg_writeBits(s, bitBuf, bitCnt, bits); + } + // Encode ACs + end0pos = 63; + for(; (end0pos>0)&&(DU[end0pos]==0); --end0pos) { + } + // end0pos = first element in reverse order !=0 + if(end0pos == 0) { + stbiw__jpg_writeBits(s, bitBuf, bitCnt, EOB); + return DU[0]; + } + for(i = 1; i <= end0pos; ++i) { + int startpos = i; + int nrzeroes; + unsigned short bits[2]; + for (; DU[i]==0 && i<=end0pos; ++i) { + } + nrzeroes = i-startpos; + if ( nrzeroes >= 16 ) { + int lng = nrzeroes>>4; + int nrmarker; + for (nrmarker=1; nrmarker <= lng; ++nrmarker) + stbiw__jpg_writeBits(s, bitBuf, bitCnt, M16zeroes); + nrzeroes &= 15; + } + stbiw__jpg_calcBits(DU[i], bits); + stbiw__jpg_writeBits(s, bitBuf, bitCnt, HTAC[(nrzeroes<<4)+bits[1]]); + stbiw__jpg_writeBits(s, bitBuf, bitCnt, bits); + } + if(end0pos != 63) { + stbiw__jpg_writeBits(s, bitBuf, bitCnt, EOB); + } + return DU[0]; +} + +static int stbi_write_jpg_core(stbi__write_context *s, int width, int height, int comp, const void* data, int quality) { + // Constants that don't pollute global namespace + static const unsigned char std_dc_luminance_nrcodes[] = {0,0,1,5,1,1,1,1,1,1,0,0,0,0,0,0,0}; + static const unsigned char std_dc_luminance_values[] = {0,1,2,3,4,5,6,7,8,9,10,11}; + static const unsigned char std_ac_luminance_nrcodes[] = {0,0,2,1,3,3,2,4,3,5,5,4,4,0,0,1,0x7d}; + static const unsigned char std_ac_luminance_values[] = { + 0x01,0x02,0x03,0x00,0x04,0x11,0x05,0x12,0x21,0x31,0x41,0x06,0x13,0x51,0x61,0x07,0x22,0x71,0x14,0x32,0x81,0x91,0xa1,0x08, + 0x23,0x42,0xb1,0xc1,0x15,0x52,0xd1,0xf0,0x24,0x33,0x62,0x72,0x82,0x09,0x0a,0x16,0x17,0x18,0x19,0x1a,0x25,0x26,0x27,0x28, + 0x29,0x2a,0x34,0x35,0x36,0x37,0x38,0x39,0x3a,0x43,0x44,0x45,0x46,0x47,0x48,0x49,0x4a,0x53,0x54,0x55,0x56,0x57,0x58,0x59, + 0x5a,0x63,0x64,0x65,0x66,0x67,0x68,0x69,0x6a,0x73,0x74,0x75,0x76,0x77,0x78,0x79,0x7a,0x83,0x84,0x85,0x86,0x87,0x88,0x89, + 0x8a,0x92,0x93,0x94,0x95,0x96,0x97,0x98,0x99,0x9a,0xa2,0xa3,0xa4,0xa5,0xa6,0xa7,0xa8,0xa9,0xaa,0xb2,0xb3,0xb4,0xb5,0xb6, + 0xb7,0xb8,0xb9,0xba,0xc2,0xc3,0xc4,0xc5,0xc6,0xc7,0xc8,0xc9,0xca,0xd2,0xd3,0xd4,0xd5,0xd6,0xd7,0xd8,0xd9,0xda,0xe1,0xe2, + 0xe3,0xe4,0xe5,0xe6,0xe7,0xe8,0xe9,0xea,0xf1,0xf2,0xf3,0xf4,0xf5,0xf6,0xf7,0xf8,0xf9,0xfa + }; + static const unsigned char std_dc_chrominance_nrcodes[] = {0,0,3,1,1,1,1,1,1,1,1,1,0,0,0,0,0}; + static const unsigned char std_dc_chrominance_values[] = {0,1,2,3,4,5,6,7,8,9,10,11}; + static const unsigned char std_ac_chrominance_nrcodes[] = {0,0,2,1,2,4,4,3,4,7,5,4,4,0,1,2,0x77}; + static const unsigned char std_ac_chrominance_values[] = { + 0x00,0x01,0x02,0x03,0x11,0x04,0x05,0x21,0x31,0x06,0x12,0x41,0x51,0x07,0x61,0x71,0x13,0x22,0x32,0x81,0x08,0x14,0x42,0x91, + 0xa1,0xb1,0xc1,0x09,0x23,0x33,0x52,0xf0,0x15,0x62,0x72,0xd1,0x0a,0x16,0x24,0x34,0xe1,0x25,0xf1,0x17,0x18,0x19,0x1a,0x26, + 0x27,0x28,0x29,0x2a,0x35,0x36,0x37,0x38,0x39,0x3a,0x43,0x44,0x45,0x46,0x47,0x48,0x49,0x4a,0x53,0x54,0x55,0x56,0x57,0x58, + 0x59,0x5a,0x63,0x64,0x65,0x66,0x67,0x68,0x69,0x6a,0x73,0x74,0x75,0x76,0x77,0x78,0x79,0x7a,0x82,0x83,0x84,0x85,0x86,0x87, + 0x88,0x89,0x8a,0x92,0x93,0x94,0x95,0x96,0x97,0x98,0x99,0x9a,0xa2,0xa3,0xa4,0xa5,0xa6,0xa7,0xa8,0xa9,0xaa,0xb2,0xb3,0xb4, + 0xb5,0xb6,0xb7,0xb8,0xb9,0xba,0xc2,0xc3,0xc4,0xc5,0xc6,0xc7,0xc8,0xc9,0xca,0xd2,0xd3,0xd4,0xd5,0xd6,0xd7,0xd8,0xd9,0xda, + 0xe2,0xe3,0xe4,0xe5,0xe6,0xe7,0xe8,0xe9,0xea,0xf2,0xf3,0xf4,0xf5,0xf6,0xf7,0xf8,0xf9,0xfa + }; + // Huffman tables + static const unsigned short YDC_HT[256][2] = { {0,2},{2,3},{3,3},{4,3},{5,3},{6,3},{14,4},{30,5},{62,6},{126,7},{254,8},{510,9}}; + static const unsigned short UVDC_HT[256][2] = { {0,2},{1,2},{2,2},{6,3},{14,4},{30,5},{62,6},{126,7},{254,8},{510,9},{1022,10},{2046,11}}; + static const unsigned short YAC_HT[256][2] = { + {10,4},{0,2},{1,2},{4,3},{11,4},{26,5},{120,7},{248,8},{1014,10},{65410,16},{65411,16},{0,0},{0,0},{0,0},{0,0},{0,0},{0,0}, + {12,4},{27,5},{121,7},{502,9},{2038,11},{65412,16},{65413,16},{65414,16},{65415,16},{65416,16},{0,0},{0,0},{0,0},{0,0},{0,0},{0,0}, + {28,5},{249,8},{1015,10},{4084,12},{65417,16},{65418,16},{65419,16},{65420,16},{65421,16},{65422,16},{0,0},{0,0},{0,0},{0,0},{0,0},{0,0}, + {58,6},{503,9},{4085,12},{65423,16},{65424,16},{65425,16},{65426,16},{65427,16},{65428,16},{65429,16},{0,0},{0,0},{0,0},{0,0},{0,0},{0,0}, + {59,6},{1016,10},{65430,16},{65431,16},{65432,16},{65433,16},{65434,16},{65435,16},{65436,16},{65437,16},{0,0},{0,0},{0,0},{0,0},{0,0},{0,0}, + {122,7},{2039,11},{65438,16},{65439,16},{65440,16},{65441,16},{65442,16},{65443,16},{65444,16},{65445,16},{0,0},{0,0},{0,0},{0,0},{0,0},{0,0}, + {123,7},{4086,12},{65446,16},{65447,16},{65448,16},{65449,16},{65450,16},{65451,16},{65452,16},{65453,16},{0,0},{0,0},{0,0},{0,0},{0,0},{0,0}, + {250,8},{4087,12},{65454,16},{65455,16},{65456,16},{65457,16},{65458,16},{65459,16},{65460,16},{65461,16},{0,0},{0,0},{0,0},{0,0},{0,0},{0,0}, + {504,9},{32704,15},{65462,16},{65463,16},{65464,16},{65465,16},{65466,16},{65467,16},{65468,16},{65469,16},{0,0},{0,0},{0,0},{0,0},{0,0},{0,0}, + {505,9},{65470,16},{65471,16},{65472,16},{65473,16},{65474,16},{65475,16},{65476,16},{65477,16},{65478,16},{0,0},{0,0},{0,0},{0,0},{0,0},{0,0}, + {506,9},{65479,16},{65480,16},{65481,16},{65482,16},{65483,16},{65484,16},{65485,16},{65486,16},{65487,16},{0,0},{0,0},{0,0},{0,0},{0,0},{0,0}, + {1017,10},{65488,16},{65489,16},{65490,16},{65491,16},{65492,16},{65493,16},{65494,16},{65495,16},{65496,16},{0,0},{0,0},{0,0},{0,0},{0,0},{0,0}, + {1018,10},{65497,16},{65498,16},{65499,16},{65500,16},{65501,16},{65502,16},{65503,16},{65504,16},{65505,16},{0,0},{0,0},{0,0},{0,0},{0,0},{0,0}, + {2040,11},{65506,16},{65507,16},{65508,16},{65509,16},{65510,16},{65511,16},{65512,16},{65513,16},{65514,16},{0,0},{0,0},{0,0},{0,0},{0,0},{0,0}, + {65515,16},{65516,16},{65517,16},{65518,16},{65519,16},{65520,16},{65521,16},{65522,16},{65523,16},{65524,16},{0,0},{0,0},{0,0},{0,0},{0,0}, + {2041,11},{65525,16},{65526,16},{65527,16},{65528,16},{65529,16},{65530,16},{65531,16},{65532,16},{65533,16},{65534,16},{0,0},{0,0},{0,0},{0,0},{0,0} + }; + static const unsigned short UVAC_HT[256][2] = { + {0,2},{1,2},{4,3},{10,4},{24,5},{25,5},{56,6},{120,7},{500,9},{1014,10},{4084,12},{0,0},{0,0},{0,0},{0,0},{0,0},{0,0}, + {11,4},{57,6},{246,8},{501,9},{2038,11},{4085,12},{65416,16},{65417,16},{65418,16},{65419,16},{0,0},{0,0},{0,0},{0,0},{0,0},{0,0}, + {26,5},{247,8},{1015,10},{4086,12},{32706,15},{65420,16},{65421,16},{65422,16},{65423,16},{65424,16},{0,0},{0,0},{0,0},{0,0},{0,0},{0,0}, + {27,5},{248,8},{1016,10},{4087,12},{65425,16},{65426,16},{65427,16},{65428,16},{65429,16},{65430,16},{0,0},{0,0},{0,0},{0,0},{0,0},{0,0}, + {58,6},{502,9},{65431,16},{65432,16},{65433,16},{65434,16},{65435,16},{65436,16},{65437,16},{65438,16},{0,0},{0,0},{0,0},{0,0},{0,0},{0,0}, + {59,6},{1017,10},{65439,16},{65440,16},{65441,16},{65442,16},{65443,16},{65444,16},{65445,16},{65446,16},{0,0},{0,0},{0,0},{0,0},{0,0},{0,0}, + {121,7},{2039,11},{65447,16},{65448,16},{65449,16},{65450,16},{65451,16},{65452,16},{65453,16},{65454,16},{0,0},{0,0},{0,0},{0,0},{0,0},{0,0}, + {122,7},{2040,11},{65455,16},{65456,16},{65457,16},{65458,16},{65459,16},{65460,16},{65461,16},{65462,16},{0,0},{0,0},{0,0},{0,0},{0,0},{0,0}, + {249,8},{65463,16},{65464,16},{65465,16},{65466,16},{65467,16},{65468,16},{65469,16},{65470,16},{65471,16},{0,0},{0,0},{0,0},{0,0},{0,0},{0,0}, + {503,9},{65472,16},{65473,16},{65474,16},{65475,16},{65476,16},{65477,16},{65478,16},{65479,16},{65480,16},{0,0},{0,0},{0,0},{0,0},{0,0},{0,0}, + {504,9},{65481,16},{65482,16},{65483,16},{65484,16},{65485,16},{65486,16},{65487,16},{65488,16},{65489,16},{0,0},{0,0},{0,0},{0,0},{0,0},{0,0}, + {505,9},{65490,16},{65491,16},{65492,16},{65493,16},{65494,16},{65495,16},{65496,16},{65497,16},{65498,16},{0,0},{0,0},{0,0},{0,0},{0,0},{0,0}, + {506,9},{65499,16},{65500,16},{65501,16},{65502,16},{65503,16},{65504,16},{65505,16},{65506,16},{65507,16},{0,0},{0,0},{0,0},{0,0},{0,0},{0,0}, + {2041,11},{65508,16},{65509,16},{65510,16},{65511,16},{65512,16},{65513,16},{65514,16},{65515,16},{65516,16},{0,0},{0,0},{0,0},{0,0},{0,0},{0,0}, + {16352,14},{65517,16},{65518,16},{65519,16},{65520,16},{65521,16},{65522,16},{65523,16},{65524,16},{65525,16},{0,0},{0,0},{0,0},{0,0},{0,0}, + {1018,10},{32707,15},{65526,16},{65527,16},{65528,16},{65529,16},{65530,16},{65531,16},{65532,16},{65533,16},{65534,16},{0,0},{0,0},{0,0},{0,0},{0,0} + }; + static const int YQT[] = {16,11,10,16,24,40,51,61,12,12,14,19,26,58,60,55,14,13,16,24,40,57,69,56,14,17,22,29,51,87,80,62,18,22, + 37,56,68,109,103,77,24,35,55,64,81,104,113,92,49,64,78,87,103,121,120,101,72,92,95,98,112,100,103,99}; + static const int UVQT[] = {17,18,24,47,99,99,99,99,18,21,26,66,99,99,99,99,24,26,56,99,99,99,99,99,47,66,99,99,99,99,99,99, + 99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99,99}; + static const float aasf[] = { 1.0f * 2.828427125f, 1.387039845f * 2.828427125f, 1.306562965f * 2.828427125f, 1.175875602f * 2.828427125f, + 1.0f * 2.828427125f, 0.785694958f * 2.828427125f, 0.541196100f * 2.828427125f, 0.275899379f * 2.828427125f }; + + int row, col, i, k, subsample; + float fdtbl_Y[64], fdtbl_UV[64]; + unsigned char YTable[64], UVTable[64]; + + if(!data || !width || !height || comp > 4 || comp < 1) { + return 0; + } + + quality = quality ? quality : 90; + subsample = quality <= 90 ? 1 : 0; + quality = quality < 1 ? 1 : quality > 100 ? 100 : quality; + quality = quality < 50 ? 5000 / quality : 200 - quality * 2; + + for(i = 0; i < 64; ++i) { + int uvti, yti = (YQT[i]*quality+50)/100; + YTable[stbiw__jpg_ZigZag[i]] = (unsigned char) (yti < 1 ? 1 : yti > 255 ? 255 : yti); + uvti = (UVQT[i]*quality+50)/100; + UVTable[stbiw__jpg_ZigZag[i]] = (unsigned char) (uvti < 1 ? 1 : uvti > 255 ? 255 : uvti); + } + + for(row = 0, k = 0; row < 8; ++row) { + for(col = 0; col < 8; ++col, ++k) { + fdtbl_Y[k] = 1 / (YTable [stbiw__jpg_ZigZag[k]] * aasf[row] * aasf[col]); + fdtbl_UV[k] = 1 / (UVTable[stbiw__jpg_ZigZag[k]] * aasf[row] * aasf[col]); + } + } + + // Write Headers + { + static const unsigned char head0[] = { 0xFF,0xD8,0xFF,0xE0,0,0x10,'J','F','I','F',0,1,1,0,0,1,0,1,0,0,0xFF,0xDB,0,0x84,0 }; + static const unsigned char head2[] = { 0xFF,0xDA,0,0xC,3,1,0,2,0x11,3,0x11,0,0x3F,0 }; + const unsigned char head1[] = { 0xFF,0xC0,0,0x11,8,(unsigned char)(height>>8),STBIW_UCHAR(height),(unsigned char)(width>>8),STBIW_UCHAR(width), + 3,1,(unsigned char)(subsample?0x22:0x11),0,2,0x11,1,3,0x11,1,0xFF,0xC4,0x01,0xA2,0 }; + s->func(s->context, (void*)head0, sizeof(head0)); + s->func(s->context, (void*)YTable, sizeof(YTable)); + stbiw__putc(s, 1); + s->func(s->context, UVTable, sizeof(UVTable)); + s->func(s->context, (void*)head1, sizeof(head1)); + s->func(s->context, (void*)(std_dc_luminance_nrcodes+1), sizeof(std_dc_luminance_nrcodes)-1); + s->func(s->context, (void*)std_dc_luminance_values, sizeof(std_dc_luminance_values)); + stbiw__putc(s, 0x10); // HTYACinfo + s->func(s->context, (void*)(std_ac_luminance_nrcodes+1), sizeof(std_ac_luminance_nrcodes)-1); + s->func(s->context, (void*)std_ac_luminance_values, sizeof(std_ac_luminance_values)); + stbiw__putc(s, 1); // HTUDCinfo + s->func(s->context, (void*)(std_dc_chrominance_nrcodes+1), sizeof(std_dc_chrominance_nrcodes)-1); + s->func(s->context, (void*)std_dc_chrominance_values, sizeof(std_dc_chrominance_values)); + stbiw__putc(s, 0x11); // HTUACinfo + s->func(s->context, (void*)(std_ac_chrominance_nrcodes+1), sizeof(std_ac_chrominance_nrcodes)-1); + s->func(s->context, (void*)std_ac_chrominance_values, sizeof(std_ac_chrominance_values)); + s->func(s->context, (void*)head2, sizeof(head2)); + } + + // Encode 8x8 macroblocks + { + static const unsigned short fillBits[] = {0x7F, 7}; + int DCY=0, DCU=0, DCV=0; + int bitBuf=0, bitCnt=0; + // comp == 2 is grey+alpha (alpha is ignored) + int ofsG = comp > 2 ? 1 : 0, ofsB = comp > 2 ? 2 : 0; + const unsigned char *dataR = (const unsigned char *)data; + const unsigned char *dataG = dataR + ofsG; + const unsigned char *dataB = dataR + ofsB; + int x, y, pos; + if(subsample) { + for(y = 0; y < height; y += 16) { + for(x = 0; x < width; x += 16) { + float Y[256], U[256], V[256]; + for(row = y, pos = 0; row < y+16; ++row) { + // row >= height => use last input row + int clamped_row = (row < height) ? row : height - 1; + int base_p = (stbi__flip_vertically_on_write ? (height-1-clamped_row) : clamped_row)*width*comp; + for(col = x; col < x+16; ++col, ++pos) { + // if col >= width => use pixel from last input column + int p = base_p + ((col < width) ? col : (width-1))*comp; + float r = dataR[p], g = dataG[p], b = dataB[p]; + Y[pos]= +0.29900f*r + 0.58700f*g + 0.11400f*b - 128; + U[pos]= -0.16874f*r - 0.33126f*g + 0.50000f*b; + V[pos]= +0.50000f*r - 0.41869f*g - 0.08131f*b; + } + } + DCY = stbiw__jpg_processDU(s, &bitBuf, &bitCnt, Y+0, 16, fdtbl_Y, DCY, YDC_HT, YAC_HT); + DCY = stbiw__jpg_processDU(s, &bitBuf, &bitCnt, Y+8, 16, fdtbl_Y, DCY, YDC_HT, YAC_HT); + DCY = stbiw__jpg_processDU(s, &bitBuf, &bitCnt, Y+128, 16, fdtbl_Y, DCY, YDC_HT, YAC_HT); + DCY = stbiw__jpg_processDU(s, &bitBuf, &bitCnt, Y+136, 16, fdtbl_Y, DCY, YDC_HT, YAC_HT); + + // subsample U,V + { + float subU[64], subV[64]; + int yy, xx; + for(yy = 0, pos = 0; yy < 8; ++yy) { + for(xx = 0; xx < 8; ++xx, ++pos) { + int j = yy*32+xx*2; + subU[pos] = (U[j+0] + U[j+1] + U[j+16] + U[j+17]) * 0.25f; + subV[pos] = (V[j+0] + V[j+1] + V[j+16] + V[j+17]) * 0.25f; + } + } + DCU = stbiw__jpg_processDU(s, &bitBuf, &bitCnt, subU, 8, fdtbl_UV, DCU, UVDC_HT, UVAC_HT); + DCV = stbiw__jpg_processDU(s, &bitBuf, &bitCnt, subV, 8, fdtbl_UV, DCV, UVDC_HT, UVAC_HT); + } + } + } + } else { + for(y = 0; y < height; y += 8) { + for(x = 0; x < width; x += 8) { + float Y[64], U[64], V[64]; + for(row = y, pos = 0; row < y+8; ++row) { + // row >= height => use last input row + int clamped_row = (row < height) ? row : height - 1; + int base_p = (stbi__flip_vertically_on_write ? (height-1-clamped_row) : clamped_row)*width*comp; + for(col = x; col < x+8; ++col, ++pos) { + // if col >= width => use pixel from last input column + int p = base_p + ((col < width) ? col : (width-1))*comp; + float r = dataR[p], g = dataG[p], b = dataB[p]; + Y[pos]= +0.29900f*r + 0.58700f*g + 0.11400f*b - 128; + U[pos]= -0.16874f*r - 0.33126f*g + 0.50000f*b; + V[pos]= +0.50000f*r - 0.41869f*g - 0.08131f*b; + } + } + + DCY = stbiw__jpg_processDU(s, &bitBuf, &bitCnt, Y, 8, fdtbl_Y, DCY, YDC_HT, YAC_HT); + DCU = stbiw__jpg_processDU(s, &bitBuf, &bitCnt, U, 8, fdtbl_UV, DCU, UVDC_HT, UVAC_HT); + DCV = stbiw__jpg_processDU(s, &bitBuf, &bitCnt, V, 8, fdtbl_UV, DCV, UVDC_HT, UVAC_HT); + } + } + } + + // Do the bit alignment of the EOI marker + stbiw__jpg_writeBits(s, &bitBuf, &bitCnt, fillBits); + } + + // EOI + stbiw__putc(s, 0xFF); + stbiw__putc(s, 0xD9); + + return 1; +} + +STBIWDEF int stbi_write_jpg_to_func(stbi_write_func *func, void *context, int x, int y, int comp, const void *data, int quality) +{ + stbi__write_context s = { 0 }; + stbi__start_write_callbacks(&s, func, context); + return stbi_write_jpg_core(&s, x, y, comp, (void *) data, quality); +} + + +#ifndef STBI_WRITE_NO_STDIO +STBIWDEF int stbi_write_jpg(char const *filename, int x, int y, int comp, const void *data, int quality) +{ + stbi__write_context s = { 0 }; + if (stbi__start_write_file(&s,filename)) { + int r = stbi_write_jpg_core(&s, x, y, comp, data, quality); + stbi__end_write_file(&s); + return r; + } else + return 0; +} +#endif + +#endif // STB_IMAGE_WRITE_IMPLEMENTATION + +/* Revision history + 1.16 (2021-07-11) + make Deflate code emit uncompressed blocks when it would otherwise expand + support writing BMPs with alpha channel + 1.15 (2020-07-13) unknown + 1.14 (2020-02-02) updated JPEG writer to downsample chroma channels + 1.13 + 1.12 + 1.11 (2019-08-11) + + 1.10 (2019-02-07) + support utf8 filenames in Windows; fix warnings and platform ifdefs + 1.09 (2018-02-11) + fix typo in zlib quality API, improve STB_I_W_STATIC in C++ + 1.08 (2018-01-29) + add stbi__flip_vertically_on_write, external zlib, zlib quality, choose PNG filter + 1.07 (2017-07-24) + doc fix + 1.06 (2017-07-23) + writing JPEG (using Jon Olick's code) + 1.05 ??? + 1.04 (2017-03-03) + monochrome BMP expansion + 1.03 ??? + 1.02 (2016-04-02) + avoid allocating large structures on the stack + 1.01 (2016-01-16) + STBIW_REALLOC_SIZED: support allocators with no realloc support + avoid race-condition in crc initialization + minor compile issues + 1.00 (2015-09-14) + installable file IO function + 0.99 (2015-09-13) + warning fixes; TGA rle support + 0.98 (2015-04-08) + added STBIW_MALLOC, STBIW_ASSERT etc + 0.97 (2015-01-18) + fixed HDR asserts, rewrote HDR rle logic + 0.96 (2015-01-17) + add HDR output + fix monochrome BMP + 0.95 (2014-08-17) + add monochrome TGA output + 0.94 (2014-05-31) + rename private functions to avoid conflicts with stb_image.h + 0.93 (2014-05-27) + warning fixes + 0.92 (2010-08-01) + casts to unsigned char to fix warnings + 0.91 (2010-07-17) + first public release + 0.90 first internal release +*/ + +/* +------------------------------------------------------------------------------ +This software is available under 2 licenses -- choose whichever you prefer. +------------------------------------------------------------------------------ +ALTERNATIVE A - MIT License +Copyright (c) 2017 Sean Barrett +Permission is hereby granted, free of charge, to any person obtaining a copy of +this software and associated documentation files (the "Software"), to deal in +the Software without restriction, including without limitation the rights to +use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies +of the Software, and to permit persons to whom the Software is furnished to do +so, subject to the following conditions: +The above copyright notice and this permission notice shall be included in all +copies or substantial portions of the Software. +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +SOFTWARE. +------------------------------------------------------------------------------ +ALTERNATIVE B - Public Domain (www.unlicense.org) +This is free and unencumbered software released into the public domain. +Anyone is free to copy, modify, publish, use, compile, sell, or distribute this +software, either in source code form or as a compiled binary, for any purpose, +commercial or non-commercial, and by any means. +In jurisdictions that recognize copyright laws, the author or authors of this +software dedicate any and all copyright interest in the software to the public +domain. We make this dedication for the benefit of the public at large and to +the detriment of our heirs and successors. We intend this dedication to be an +overt act of relinquishment in perpetuity of all present and future rights to +this software under copyright law. +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN +ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION +WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. +------------------------------------------------------------------------------ +*/ diff --git a/include/capture/capture.h b/include/capture/capture.h index dc5b7ac..634eee0 100644 --- a/include/capture/capture.h +++ b/include/capture/capture.h @@ -13,31 +13,39 @@ typedef struct AVMasteringDisplayMetadata AVMasteringDisplayMetadata; typedef struct AVContentLightMetadata AVContentLightMetadata; typedef struct gsr_capture gsr_capture; +typedef struct { + int width; + int height; + int fps; + AVCodecContext *video_codec_context; /* can be NULL */ + AVFrame *frame; /* can be NULL, but will never be NULL if |video_codec_context| is set */ +} gsr_capture_metadata; + struct gsr_capture { /* These methods should not be called manually. Call gsr_capture_* instead */ - int (*start)(gsr_capture *cap, AVCodecContext *video_codec_context, AVFrame *frame); + int (*start)(gsr_capture *cap, gsr_capture_metadata *capture_metadata); void (*on_event)(gsr_capture *cap, gsr_egl *egl); /* can be NULL */ void (*tick)(gsr_capture *cap); /* can be NULL. If there is an event then |on_event| is called before this */ bool (*should_stop)(gsr_capture *cap, bool *err); /* can be NULL. If NULL, return false */ - int (*capture)(gsr_capture *cap, AVFrame *frame, gsr_color_conversion *color_conversion); + int (*capture)(gsr_capture *cap, gsr_capture_metadata *capture_metadata, gsr_color_conversion *color_conversion); bool (*uses_external_image)(gsr_capture *cap); /* can be NULL. If NULL, return false */ bool (*set_hdr_metadata)(gsr_capture *cap, AVMasteringDisplayMetadata *mastering_display_metadata, AVContentLightMetadata *light_metadata); /* can be NULL. If NULL, return false */ uint64_t (*get_window_id)(gsr_capture *cap); /* can be NULL. Returns 0 if unknown */ bool (*is_damaged)(gsr_capture *cap); /* can be NULL */ void (*clear_damage)(gsr_capture *cap); /* can be NULL */ - void (*destroy)(gsr_capture *cap, AVCodecContext *video_codec_context); + void (*destroy)(gsr_capture *cap); void *priv; /* can be NULL */ bool started; }; -int gsr_capture_start(gsr_capture *cap, AVCodecContext *video_codec_context, AVFrame *frame); +int gsr_capture_start(gsr_capture *cap, gsr_capture_metadata *capture_metadata); void gsr_capture_on_event(gsr_capture *cap, gsr_egl *egl); void gsr_capture_tick(gsr_capture *cap); bool gsr_capture_should_stop(gsr_capture *cap, bool *err); -int gsr_capture_capture(gsr_capture *cap, AVFrame *frame, gsr_color_conversion *color_conversion); +int gsr_capture_capture(gsr_capture *cap, gsr_capture_metadata *capture_metadata, gsr_color_conversion *color_conversion); bool gsr_capture_uses_external_image(gsr_capture *cap); bool gsr_capture_set_hdr_metadata(gsr_capture *cap, AVMasteringDisplayMetadata *mastering_display_metadata, AVContentLightMetadata *light_metadata); -void gsr_capture_destroy(gsr_capture *cap, AVCodecContext *video_codec_context); +void gsr_capture_destroy(gsr_capture *cap); #endif /* GSR_CAPTURE_CAPTURE_H */ diff --git a/include/encoder/video/image.h b/include/encoder/video/image.h deleted file mode 100644 index 76c7bd4..0000000 --- a/include/encoder/video/image.h +++ /dev/null @@ -1,15 +0,0 @@ -#ifndef GSR_ENCODER_VIDEO_IMAGE_H -#define GSR_ENCODER_VIDEO_IMAGE_H - -#include "video.h" - -typedef struct gsr_egl gsr_egl; - -typedef struct { - gsr_egl *egl; - gsr_color_depth color_depth; -} gsr_video_encoder_image_params; - -gsr_video_encoder* gsr_video_encoder_image_create(const gsr_video_encoder_image_params *params); - -#endif /* GSR_ENCODER_VIDEO_IMAGE_H */ diff --git a/include/image_writer.h b/include/image_writer.h new file mode 100644 index 0000000..79f549e --- /dev/null +++ b/include/image_writer.h @@ -0,0 +1,31 @@ +#ifndef GSR_IMAGE_WRITER_H +#define GSR_IMAGE_WRITER_H + +#include <stdbool.h> + +typedef struct gsr_egl gsr_egl; + +typedef enum { + GSR_IMAGE_FORMAT_JPEG, + GSR_IMAGE_FORMAT_PNG +} gsr_image_format; + +typedef enum { + GSR_IMAGE_WRITER_SOURCE_OPENGL +} gsr_image_writer_source; + +typedef struct { + gsr_image_writer_source source; + gsr_egl *egl; + int width; + int height; + unsigned int texture; +} gsr_image_writer; + +bool gsr_image_writer_init(gsr_image_writer *self, gsr_image_writer_source source, gsr_egl *egl, int width, int height); +void gsr_image_writer_deinit(gsr_image_writer *self); + +/* Quality is between 1 and 100 where 100 is the max quality. Quality doesn't apply to lossless formats */ +bool gsr_image_writer_write_to_file(gsr_image_writer *self, const char *filepath, gsr_image_format image_format, int quality); + +#endif /* GSR_IMAGE_WRITER_H */ diff --git a/meson.build b/meson.build index a88e13d..7b1ea04 100644 --- a/meson.build +++ b/meson.build @@ -18,7 +18,6 @@ src = [ 'src/encoder/video/vaapi.c', 'src/encoder/video/vulkan.c', 'src/encoder/video/software.c', - 'src/encoder/video/image.c', 'src/codec_query/nvenc.c', 'src/codec_query/vaapi.c', 'src/codec_query/vulkan.c', @@ -36,6 +35,7 @@ src = [ 'src/library_loader.c', 'src/cursor.c', 'src/damage.c', + 'src/image_writer.c', 'src/sound.cpp', 'src/main.cpp', ] diff --git a/src/capture/capture.c b/src/capture/capture.c index 2a4a689..bc95300 100644 --- a/src/capture/capture.c +++ b/src/capture/capture.c @@ -1,9 +1,9 @@ #include "../../include/capture/capture.h" #include <assert.h> -int gsr_capture_start(gsr_capture *cap, AVCodecContext *video_codec_context, AVFrame *frame) { +int gsr_capture_start(gsr_capture *cap, gsr_capture_metadata *capture_metadata) { assert(!cap->started); - int res = cap->start(cap, video_codec_context, frame); + int res = cap->start(cap, capture_metadata); if(res == 0) cap->started = true; @@ -29,9 +29,9 @@ bool gsr_capture_should_stop(gsr_capture *cap, bool *err) { return false; } -int gsr_capture_capture(gsr_capture *cap, AVFrame *frame, gsr_color_conversion *color_conversion) { +int gsr_capture_capture(gsr_capture *cap, gsr_capture_metadata *capture_metadata, gsr_color_conversion *color_conversion) { assert(cap->started); - return cap->capture(cap, frame, color_conversion); + return cap->capture(cap, capture_metadata, color_conversion); } bool gsr_capture_uses_external_image(gsr_capture *cap) { @@ -48,6 +48,6 @@ bool gsr_capture_set_hdr_metadata(gsr_capture *cap, AVMasteringDisplayMetadata * return false; } -void gsr_capture_destroy(gsr_capture *cap, AVCodecContext *video_codec_context) { - cap->destroy(cap, video_codec_context); +void gsr_capture_destroy(gsr_capture *cap) { + cap->destroy(cap); } diff --git a/src/capture/kms.c b/src/capture/kms.c index ae0c36f..266d4e6 100644 --- a/src/capture/kms.c +++ b/src/capture/kms.c @@ -55,7 +55,6 @@ typedef struct { bool is_x11; gsr_cursor x11_cursor; - AVCodecContext *video_codec_context; bool performance_error_shown; bool fast_path_failed; bool mesa_supports_compute_only_vaapi_copy; @@ -177,7 +176,7 @@ static vec2i rotate_capture_size_if_rotated(gsr_capture_kms *self, vec2i capture return capture_size; } -static int gsr_capture_kms_start(gsr_capture *cap, AVCodecContext *video_codec_context, AVFrame *frame) { +static int gsr_capture_kms_start(gsr_capture *cap, gsr_capture_metadata *capture_metadata) { gsr_capture_kms *self = cap->priv; gsr_capture_kms_create_input_texture_ids(self); @@ -219,17 +218,14 @@ static int gsr_capture_kms_start(gsr_capture *cap, AVCodecContext *video_codec_c else self->capture_size = rotate_capture_size_if_rotated(self, monitor.size); - /* Disable vsync */ - self->params.egl->eglSwapInterval(self->params.egl->egl_display, 0); - if(self->params.output_resolution.x == 0 && self->params.output_resolution.y == 0) { self->params.output_resolution = self->capture_size; - video_codec_context->width = FFALIGN(self->capture_size.x, 2); - video_codec_context->height = FFALIGN(self->capture_size.y, 2); + capture_metadata->width = FFALIGN(self->capture_size.x, 2); + capture_metadata->height = FFALIGN(self->capture_size.y, 2); } else { self->params.output_resolution = scale_keep_aspect_ratio(self->capture_size, self->params.output_resolution); - video_codec_context->width = FFALIGN(self->params.output_resolution.x, 2); - video_codec_context->height = FFALIGN(self->params.output_resolution.y, 2); + capture_metadata->width = FFALIGN(self->params.output_resolution.x, 2); + capture_metadata->height = FFALIGN(self->params.output_resolution.y, 2); } self->fast_path_failed = self->params.egl->gpu_info.vendor == GSR_GPU_VENDOR_AMD && !gl_driver_version_greater_than(&self->params.egl->gpu_info, 24, 0, 9); @@ -243,10 +239,6 @@ static int gsr_capture_kms_start(gsr_capture *cap, AVCodecContext *video_codec_c self->mesa_supports_compute_only_vaapi_copy = self->params.egl->gpu_info.vendor == GSR_GPU_VENDOR_AMD && gl_driver_version_greater_than(&self->params.egl->gpu_info, 24, 3, 6); - frame->width = video_codec_context->width; - frame->height = video_codec_context->height; - - self->video_codec_context = video_codec_context; self->last_time_monitor_check = clock_get_monotonic_seconds(); return 0; } @@ -617,7 +609,7 @@ static void gsr_capture_kms_fail_fast_path_if_not_fast(gsr_capture_kms *self, ui } } -static int gsr_capture_kms_capture(gsr_capture *cap, AVFrame *frame, gsr_color_conversion *color_conversion) { +static int gsr_capture_kms_capture(gsr_capture *cap, gsr_capture_metadata *capture_metadata, gsr_color_conversion *color_conversion) { gsr_capture_kms *self = cap->priv; gsr_capture_kms_cleanup_kms_fds(self); @@ -648,7 +640,7 @@ static int gsr_capture_kms_capture(gsr_capture *cap, AVFrame *frame, gsr_color_c if(drm_fd->has_hdr_metadata && self->params.hdr && hdr_metadata_is_supported_format(&drm_fd->hdr_metadata)) gsr_kms_set_hdr_metadata(self, drm_fd); - if(!self->performance_error_shown && self->monitor_rotation != GSR_MONITOR_ROT_0 && video_codec_context_is_vaapi(self->video_codec_context) && self->params.egl->gpu_info.vendor == GSR_GPU_VENDOR_AMD) { + if(!self->performance_error_shown && self->monitor_rotation != GSR_MONITOR_ROT_0 && video_codec_context_is_vaapi(capture_metadata->video_codec_context) && self->params.egl->gpu_info.vendor == GSR_GPU_VENDOR_AMD) { self->performance_error_shown = true; self->fast_path_failed = true; fprintf(stderr, "gsr warning: gsr_capture_kms_capture: the monitor you are recording is rotated, composition will have to be used." @@ -664,7 +656,7 @@ static int gsr_capture_kms_capture(gsr_capture *cap, AVFrame *frame, gsr_color_c output_size = scale_keep_aspect_ratio(self->capture_size, output_size); const float texture_rotation = monitor_rotation_to_radians(self->monitor_rotation); - const vec2i target_pos = { max_int(0, frame->width / 2 - output_size.x / 2), max_int(0, frame->height / 2 - output_size.y / 2) }; + const vec2i target_pos = { max_int(0, capture_metadata->width / 2 - output_size.x / 2), max_int(0, capture_metadata->height / 2 - output_size.y / 2) }; gsr_capture_kms_update_capture_size_change(self, color_conversion, target_pos, drm_fd); vec2i capture_pos = self->capture_pos; @@ -675,7 +667,7 @@ static int gsr_capture_kms_capture(gsr_capture *cap, AVFrame *frame, gsr_color_c self->params.egl->glFinish(); /* Fast opengl free path */ - if(!self->fast_path_failed && self->monitor_rotation == GSR_MONITOR_ROT_0 && video_codec_context_is_vaapi(self->video_codec_context) && self->params.egl->gpu_info.vendor == GSR_GPU_VENDOR_AMD) { + if(!self->fast_path_failed && self->monitor_rotation == GSR_MONITOR_ROT_0 && video_codec_context_is_vaapi(capture_metadata->video_codec_context) && self->params.egl->gpu_info.vendor == GSR_GPU_VENDOR_AMD) { int fds[4]; uint32_t offsets[4]; uint32_t pitches[4]; @@ -686,7 +678,7 @@ static int gsr_capture_kms_capture(gsr_capture *cap, AVFrame *frame, gsr_color_c pitches[i] = drm_fd->dma_buf[i].pitch; modifiers[i] = drm_fd->modifier; } - if(!vaapi_copy_drm_planes_to_video_surface(self->video_codec_context, frame, (vec2i){capture_pos.x, capture_pos.y}, self->capture_size, target_pos, output_size, drm_fd->pixel_format, (vec2i){drm_fd->width, drm_fd->height}, fds, offsets, pitches, modifiers, drm_fd->num_dma_bufs)) { + if(!vaapi_copy_drm_planes_to_video_surface(capture_metadata->video_codec_context, capture_metadata->frame, (vec2i){capture_pos.x, capture_pos.y}, self->capture_size, target_pos, output_size, drm_fd->pixel_format, (vec2i){drm_fd->width, drm_fd->height}, fds, offsets, pitches, modifiers, drm_fd->num_dma_bufs)) { fprintf(stderr, "gsr error: gsr_capture_kms_capture: vaapi_copy_drm_planes_to_video_surface failed, falling back to opengl copy. Please report this as an issue at https://github.com/dec05eba/gpu-screen-recorder-issues\n"); self->fast_path_failed = true; } @@ -777,8 +769,7 @@ static bool gsr_capture_kms_set_hdr_metadata(gsr_capture *cap, AVMasteringDispla // self->damaged = false; // } -static void gsr_capture_kms_destroy(gsr_capture *cap, AVCodecContext *video_codec_context) { - (void)video_codec_context; +static void gsr_capture_kms_destroy(gsr_capture *cap) { gsr_capture_kms *self = cap->priv; if(cap->priv) { gsr_capture_kms_stop(self); diff --git a/src/capture/nvfbc.c b/src/capture/nvfbc.c index 676d269..af79e0d 100644 --- a/src/capture/nvfbc.c +++ b/src/capture/nvfbc.c @@ -133,31 +133,6 @@ static bool gsr_capture_nvfbc_load_library(gsr_capture *cap) { return true; } -/* TODO: check for glx swap control extension string (GLX_EXT_swap_control, etc) */ -static void set_vertical_sync_enabled(gsr_egl *egl, int enabled) { - int result = 0; - - if(egl->glXSwapIntervalEXT) { - assert(gsr_window_get_display_server(egl->window) == GSR_DISPLAY_SERVER_X11); - Display *display = gsr_window_get_display(egl->window); - const Window window = (Window)gsr_window_get_window(egl->window); - egl->glXSwapIntervalEXT(display, window, enabled ? 1 : 0); - } else if(egl->glXSwapIntervalMESA) { - result = egl->glXSwapIntervalMESA(enabled ? 1 : 0); - } else if(egl->glXSwapIntervalSGI) { - result = egl->glXSwapIntervalSGI(enabled ? 1 : 0); - } else { - static int warned = 0; - if (!warned) { - warned = 1; - fprintf(stderr, "gsr warning: setting vertical sync not supported\n"); - } - } - - if(result != 0) - fprintf(stderr, "gsr warning: setting vertical sync failed\n"); -} - static void gsr_capture_nvfbc_destroy_session(gsr_capture_nvfbc *self) { if(self->fbc_handle_created && self->capture_session_created) { NVFBC_DESTROY_CAPTURE_SESSION_PARAMS destroy_capture_params; @@ -311,7 +286,7 @@ static void gsr_capture_nvfbc_stop(gsr_capture_nvfbc *self) { } } -static int gsr_capture_nvfbc_start(gsr_capture *cap, AVCodecContext *video_codec_context, AVFrame *frame) { +static int gsr_capture_nvfbc_start(gsr_capture *cap, gsr_capture_metadata *capture_metadata) { gsr_capture_nvfbc *self = cap->priv; if(!gsr_capture_nvfbc_load_library(cap)) @@ -357,27 +332,21 @@ static int gsr_capture_nvfbc_start(gsr_capture *cap, AVCodecContext *video_codec } if(self->capture_region) { - video_codec_context->width = FFALIGN(self->width, 2); - video_codec_context->height = FFALIGN(self->height, 2); + capture_metadata->width = FFALIGN(self->width, 2); + capture_metadata->height = FFALIGN(self->height, 2); } else { - video_codec_context->width = FFALIGN(self->tracking_width, 2); - video_codec_context->height = FFALIGN(self->tracking_height, 2); + capture_metadata->width = FFALIGN(self->tracking_width, 2); + capture_metadata->height = FFALIGN(self->tracking_height, 2); } if(self->params.output_resolution.x == 0 && self->params.output_resolution.y == 0) { - self->params.output_resolution = (vec2i){video_codec_context->width, video_codec_context->height}; + self->params.output_resolution = (vec2i){capture_metadata->width, capture_metadata->height}; } else { - self->params.output_resolution = scale_keep_aspect_ratio((vec2i){video_codec_context->width, video_codec_context->height}, self->params.output_resolution); - video_codec_context->width = FFALIGN(self->params.output_resolution.x, 2); - video_codec_context->height = FFALIGN(self->params.output_resolution.y, 2); + self->params.output_resolution = scale_keep_aspect_ratio((vec2i){capture_metadata->width, capture_metadata->height}, self->params.output_resolution); + capture_metadata->width = FFALIGN(self->params.output_resolution.x, 2); + capture_metadata->height = FFALIGN(self->params.output_resolution.y, 2); } - frame->width = video_codec_context->width; - frame->height = video_codec_context->height; - - /* Disable vsync */ - set_vertical_sync_enabled(self->params.egl, 0); - return 0; error_cleanup: @@ -385,7 +354,7 @@ static int gsr_capture_nvfbc_start(gsr_capture *cap, AVCodecContext *video_codec return -1; } -static int gsr_capture_nvfbc_capture(gsr_capture *cap, AVFrame *frame, gsr_color_conversion *color_conversion) { +static int gsr_capture_nvfbc_capture(gsr_capture *cap, gsr_capture_metadata *capture_metadata, gsr_color_conversion *color_conversion) { gsr_capture_nvfbc *self = cap->priv; const double nvfbc_recreate_retry_time_seconds = 1.0; @@ -416,7 +385,7 @@ static int gsr_capture_nvfbc_capture(gsr_capture *cap, AVFrame *frame, gsr_color vec2i output_size = is_scaled ? self->params.output_resolution : frame_size; output_size = scale_keep_aspect_ratio(frame_size, output_size); - const vec2i target_pos = { max_int(0, frame->width / 2 - output_size.x / 2), max_int(0, frame->height / 2 - output_size.y / 2) }; + const vec2i target_pos = { max_int(0, capture_metadata->width / 2 - output_size.x / 2), max_int(0, capture_metadata->height / 2 - output_size.y / 2) }; NVFBC_FRAME_GRAB_INFO frame_info; memset(&frame_info, 0, sizeof(frame_info)); @@ -450,8 +419,7 @@ static int gsr_capture_nvfbc_capture(gsr_capture *cap, AVFrame *frame, gsr_color return 0; } -static void gsr_capture_nvfbc_destroy(gsr_capture *cap, AVCodecContext *video_codec_context) { - (void)video_codec_context; +static void gsr_capture_nvfbc_destroy(gsr_capture *cap) { gsr_capture_nvfbc *self = cap->priv; gsr_capture_nvfbc_stop(self); free(cap->priv); diff --git a/src/capture/portal.c b/src/capture/portal.c index 27486fd..cfbfbcd 100644 --- a/src/capture/portal.c +++ b/src/capture/portal.c @@ -25,7 +25,6 @@ typedef struct { gsr_pipewire_video_dmabuf_data dmabuf_data[GSR_PIPEWIRE_VIDEO_DMABUF_MAX_PLANES]; int num_dmabuf_data; - AVCodecContext *video_codec_context; bool fast_path_failed; bool mesa_supports_compute_only_vaapi_copy; } gsr_capture_portal; @@ -257,7 +256,7 @@ static bool gsr_capture_portal_get_frame_dimensions(gsr_capture_portal *self) { return false; } -static int gsr_capture_portal_start(gsr_capture *cap, AVCodecContext *video_codec_context, AVFrame *frame) { +static int gsr_capture_portal_start(gsr_capture *cap, gsr_capture_metadata *capture_metadata) { gsr_capture_portal *self = cap->priv; gsr_capture_portal_create_input_textures(self); @@ -286,7 +285,7 @@ static int gsr_capture_portal_start(gsr_capture *cap, AVCodecContext *video_code fprintf(stderr, "gsr info: gsr_capture_portal_start: setting up pipewire\n"); /* TODO: support hdr when pipewire supports it */ /* gsr_pipewire closes the pipewire fd, even on failure */ - if(!gsr_pipewire_video_init(&self->pipewire, pipewire_fd, pipewire_node, video_codec_context->framerate.num, self->params.record_cursor, self->params.egl)) { + if(!gsr_pipewire_video_init(&self->pipewire, pipewire_fd, pipewire_node, capture_metadata->fps, self->params.record_cursor, self->params.egl)) { fprintf(stderr, "gsr error: gsr_capture_portal_start: failed to setup pipewire with fd: %d, node: %" PRIu32 "\n", pipewire_fd, pipewire_node); gsr_capture_portal_stop(self); return -1; @@ -298,17 +297,14 @@ static int gsr_capture_portal_start(gsr_capture *cap, AVCodecContext *video_code return -1; } - /* Disable vsync */ - self->params.egl->eglSwapInterval(self->params.egl->egl_display, 0); - if(self->params.output_resolution.x == 0 && self->params.output_resolution.y == 0) { self->params.output_resolution = self->capture_size; - video_codec_context->width = FFALIGN(self->capture_size.x, 2); - video_codec_context->height = FFALIGN(self->capture_size.y, 2); + capture_metadata->width = FFALIGN(self->capture_size.x, 2); + capture_metadata->height = FFALIGN(self->capture_size.y, 2); } else { self->params.output_resolution = scale_keep_aspect_ratio(self->capture_size, self->params.output_resolution); - video_codec_context->width = FFALIGN(self->params.output_resolution.x, 2); - video_codec_context->height = FFALIGN(self->params.output_resolution.y, 2); + capture_metadata->width = FFALIGN(self->params.output_resolution.x, 2); + capture_metadata->height = FFALIGN(self->params.output_resolution.y, 2); } self->fast_path_failed = self->params.egl->gpu_info.vendor == GSR_GPU_VENDOR_AMD && !gl_driver_version_greater_than(&self->params.egl->gpu_info, 24, 0, 9); @@ -317,10 +313,6 @@ static int gsr_capture_portal_start(gsr_capture *cap, AVCodecContext *video_code self->mesa_supports_compute_only_vaapi_copy = self->params.egl->gpu_info.vendor == GSR_GPU_VENDOR_AMD && gl_driver_version_greater_than(&self->params.egl->gpu_info, 24, 3, 6); - frame->width = video_codec_context->width; - frame->height = video_codec_context->height; - - self->video_codec_context = video_codec_context; return 0; } @@ -338,8 +330,7 @@ static void gsr_capture_portal_fail_fast_path_if_not_fast(gsr_capture_portal *se } } -static int gsr_capture_portal_capture(gsr_capture *cap, AVFrame *frame, gsr_color_conversion *color_conversion) { - (void)frame; +static int gsr_capture_portal_capture(gsr_capture *cap, gsr_capture_metadata *capture_metadata, gsr_color_conversion *color_conversion) { (void)color_conversion; gsr_capture_portal *self = cap->priv; @@ -365,7 +356,7 @@ static int gsr_capture_portal_capture(gsr_capture *cap, AVFrame *frame, gsr_colo vec2i output_size = is_scaled ? self->params.output_resolution : self->capture_size; output_size = scale_keep_aspect_ratio(self->capture_size, output_size); - const vec2i target_pos = { max_int(0, frame->width / 2 - output_size.x / 2), max_int(0, frame->height / 2 - output_size.y / 2) }; + const vec2i target_pos = { max_int(0, capture_metadata->width / 2 - output_size.x / 2), max_int(0, capture_metadata->height / 2 - output_size.y / 2) }; self->params.egl->glFlush(); self->params.egl->glFinish(); @@ -373,7 +364,7 @@ static int gsr_capture_portal_capture(gsr_capture *cap, AVFrame *frame, gsr_colo // TODO: Handle region crop /* Fast opengl free path */ - if(!self->fast_path_failed && video_codec_context_is_vaapi(self->video_codec_context) && self->params.egl->gpu_info.vendor == GSR_GPU_VENDOR_AMD) { + if(!self->fast_path_failed && video_codec_context_is_vaapi(capture_metadata->video_codec_context) && self->params.egl->gpu_info.vendor == GSR_GPU_VENDOR_AMD) { int fds[4]; uint32_t offsets[4]; uint32_t pitches[4]; @@ -384,7 +375,7 @@ static int gsr_capture_portal_capture(gsr_capture *cap, AVFrame *frame, gsr_colo pitches[i] = self->dmabuf_data[i].stride; modifiers[i] = pipewire_modifiers; } - if(!vaapi_copy_drm_planes_to_video_surface(self->video_codec_context, frame, (vec2i){region.x, region.y}, self->capture_size, target_pos, output_size, pipewire_fourcc, self->capture_size, fds, offsets, pitches, modifiers, self->num_dmabuf_data)) { + if(!vaapi_copy_drm_planes_to_video_surface(capture_metadata->video_codec_context, capture_metadata->frame, (vec2i){region.x, region.y}, self->capture_size, target_pos, output_size, pipewire_fourcc, self->capture_size, fds, offsets, pitches, modifiers, self->num_dmabuf_data)) { fprintf(stderr, "gsr error: gsr_capture_portal_capture: vaapi_copy_drm_planes_to_video_surface failed, falling back to opengl copy. Please report this as an issue at https://github.com/dec05eba/gpu-screen-recorder-issues\n"); self->fast_path_failed = true; } @@ -442,8 +433,7 @@ static void gsr_capture_portal_clear_damage(gsr_capture *cap) { gsr_pipewire_video_clear_damage(&self->pipewire); } -static void gsr_capture_portal_destroy(gsr_capture *cap, AVCodecContext *video_codec_context) { - (void)video_codec_context; +static void gsr_capture_portal_destroy(gsr_capture *cap) { gsr_capture_portal *self = cap->priv; if(cap->priv) { gsr_capture_portal_stop(self); diff --git a/src/capture/xcomposite.c b/src/capture/xcomposite.c index 5cef71d..94e691b 100644 --- a/src/capture/xcomposite.c +++ b/src/capture/xcomposite.c @@ -31,7 +31,6 @@ typedef struct { double window_resize_timer; WindowTexture window_texture; - AVCodecContext *video_codec_context; Atom net_active_window_atom; @@ -64,7 +63,7 @@ static Window get_focused_window(Display *display, Atom net_active_window_atom) return None; } -static int gsr_capture_xcomposite_start(gsr_capture *cap, AVCodecContext *video_codec_context, AVFrame *frame) { +static int gsr_capture_xcomposite_start(gsr_capture *cap, gsr_capture_metadata *capture_metadata) { gsr_capture_xcomposite *self = cap->priv; if(self->params.follow_focused) { @@ -95,8 +94,6 @@ static int gsr_capture_xcomposite_start(gsr_capture *cap, AVCodecContext *video_ // TODO: Get select and add these on top of it and then restore at the end. Also do the same in other xcomposite XSelectInput(self->display, self->window, StructureNotifyMask | ExposureMask); - /* Disable vsync */ - self->params.egl->eglSwapInterval(self->params.egl->egl_display, 0); if(window_texture_init(&self->window_texture, self->display, self->window, self->params.egl) != 0 && !self->params.follow_focused) { fprintf(stderr, "gsr error: gsr_capture_xcomposite_start: failed to get window texture for window %ld\n", (long)self->window); return -1; @@ -117,21 +114,17 @@ static int gsr_capture_xcomposite_start(gsr_capture *cap, AVCodecContext *video_ if(self->params.output_resolution.x == 0 && self->params.output_resolution.y == 0) { self->params.output_resolution = self->texture_size; - video_codec_context->width = FFALIGN(self->texture_size.x, 2); - video_codec_context->height = FFALIGN(self->texture_size.y, 2); + capture_metadata->width = FFALIGN(self->texture_size.x, 2); + capture_metadata->height = FFALIGN(self->texture_size.y, 2); } else { - video_codec_context->width = FFALIGN(self->params.output_resolution.x, 2); - video_codec_context->height = FFALIGN(self->params.output_resolution.y, 2); + capture_metadata->width = FFALIGN(self->params.output_resolution.x, 2); + capture_metadata->height = FFALIGN(self->params.output_resolution.y, 2); } self->fast_path_failed = self->params.egl->gpu_info.vendor == GSR_GPU_VENDOR_AMD && !gl_driver_version_greater_than(&self->params.egl->gpu_info, 24, 0, 9); if(self->fast_path_failed) fprintf(stderr, "gsr warning: gsr_capture_kms_start: your amd driver (mesa) version is known to be buggy (<= version 24.0.9), falling back to opengl copy\n"); - frame->width = video_codec_context->width; - frame->height = video_codec_context->height; - - self->video_codec_context = video_codec_context; self->window_resize_timer = clock_get_monotonic_seconds(); return 0; } @@ -255,9 +248,8 @@ static bool gsr_capture_xcomposite_should_stop(gsr_capture *cap, bool *err) { return false; } -static int gsr_capture_xcomposite_capture(gsr_capture *cap, AVFrame *frame, gsr_color_conversion *color_conversion) { +static int gsr_capture_xcomposite_capture(gsr_capture *cap, gsr_capture_metadata *capture_metdata, gsr_color_conversion *color_conversion) { gsr_capture_xcomposite *self = cap->priv; - (void)frame; if(self->clear_background) { self->clear_background = false; @@ -268,14 +260,14 @@ static int gsr_capture_xcomposite_capture(gsr_capture *cap, AVFrame *frame, gsr_ vec2i output_size = is_scaled ? self->params.output_resolution : self->texture_size; output_size = scale_keep_aspect_ratio(self->texture_size, output_size); - const vec2i target_pos = { max_int(0, frame->width / 2 - output_size.x / 2), max_int(0, frame->height / 2 - output_size.y / 2) }; + const vec2i target_pos = { max_int(0, capture_metdata->width / 2 - output_size.x / 2), max_int(0, capture_metdata->height / 2 - output_size.y / 2) }; self->params.egl->glFlush(); self->params.egl->glFinish(); /* Fast opengl free path */ - if(!self->fast_path_failed && video_codec_context_is_vaapi(self->video_codec_context) && self->params.egl->gpu_info.vendor == GSR_GPU_VENDOR_AMD) { - if(!vaapi_copy_egl_image_to_video_surface(self->params.egl, self->window_texture.image, (vec2i){0, 0}, self->texture_size, target_pos, output_size, self->video_codec_context, frame)) { + if(!self->fast_path_failed && video_codec_context_is_vaapi(capture_metdata->video_codec_context) && self->params.egl->gpu_info.vendor == GSR_GPU_VENDOR_AMD) { + if(!vaapi_copy_egl_image_to_video_surface(self->params.egl, self->window_texture.image, (vec2i){0, 0}, self->texture_size, target_pos, output_size, capture_metdata->video_codec_context, capture_metdata->frame)) { fprintf(stderr, "gsr error: gsr_capture_xcomposite_capture: vaapi_copy_egl_image_to_video_surface failed, falling back to opengl copy. Please report this as an issue at https://github.com/dec05eba/gpu-screen-recorder-issues\n"); self->fast_path_failed = true; } @@ -325,8 +317,7 @@ static uint64_t gsr_capture_xcomposite_get_window_id(gsr_capture *cap) { return self->window; } -static void gsr_capture_xcomposite_destroy(gsr_capture *cap, AVCodecContext *video_codec_context) { - (void)video_codec_context; +static void gsr_capture_xcomposite_destroy(gsr_capture *cap) { if(cap->priv) { gsr_capture_xcomposite_stop(cap->priv); free(cap->priv); @@ -355,6 +355,44 @@ static void debug_callback(unsigned int source, unsigned int type, unsigned int fprintf(stderr, "gsr info: gl callback: %s type = 0x%x, severity = 0x%x, message = %s\n", type == GL_DEBUG_TYPE_ERROR ? "** GL ERROR **" : "", type, severity, message); } +/* TODO: check for glx swap control extension string (GLX_EXT_swap_control, etc) */ +static void set_vertical_sync_enabled(gsr_egl *egl, int enabled) { + int result = 0; + + if(egl->glXSwapIntervalEXT) { + assert(gsr_window_get_display_server(egl->window) == GSR_DISPLAY_SERVER_X11); + Display *display = gsr_window_get_display(egl->window); + const Window window = (Window)gsr_window_get_window(egl->window); + egl->glXSwapIntervalEXT(display, window, enabled ? 1 : 0); + } else if(egl->glXSwapIntervalMESA) { + result = egl->glXSwapIntervalMESA(enabled ? 1 : 0); + } else if(egl->glXSwapIntervalSGI) { + result = egl->glXSwapIntervalSGI(enabled ? 1 : 0); + } else { + static int warned = 0; + if (!warned) { + warned = 1; + fprintf(stderr, "gsr warning: setting vertical sync not supported\n"); + } + } + + if(result != 0) + fprintf(stderr, "gsr warning: setting vertical sync failed\n"); +} + +static void gsr_egl_disable_vsync(gsr_egl *self) { + switch(self->context_type) { + case GSR_GL_CONTEXT_TYPE_EGL: { + self->eglSwapInterval(self->egl_display, 0); + break; + } + case GSR_GL_CONTEXT_TYPE_GLX: { + set_vertical_sync_enabled(self, 0); + break; + } + } +} + bool gsr_egl_load(gsr_egl *self, gsr_window *window, bool is_monitor_capture, bool enable_debug) { memset(self, 0, sizeof(gsr_egl)); self->context_type = GSR_GL_CONTEXT_TYPE_EGL; @@ -416,6 +454,7 @@ bool gsr_egl_load(gsr_egl *self, gsr_window *window, bool is_monitor_capture, bo self->glDebugMessageCallback(debug_callback, NULL); } + gsr_egl_disable_vsync(self); return true; fail: diff --git a/src/encoder/video/image.c b/src/encoder/video/image.c deleted file mode 100644 index 5f48b7b..0000000 --- a/src/encoder/video/image.c +++ /dev/null @@ -1,130 +0,0 @@ -#include "../../../include/encoder/video/image.h" -#include "../../../include/egl.h" - -#include <libavcodec/avcodec.h> -#include <libavutil/frame.h> - -#include <stdlib.h> - -#define LINESIZE_ALIGNMENT 4 - -typedef struct { - gsr_video_encoder_image_params params; - - unsigned int target_texture; -} gsr_video_encoder_image; - -static unsigned int gl_create_texture(gsr_egl *egl, int width, int height, int internal_format, unsigned int format) { - unsigned int texture_id = 0; - egl->glGenTextures(1, &texture_id); - egl->glBindTexture(GL_TEXTURE_2D, texture_id); - egl->glTexImage2D(GL_TEXTURE_2D, 0, internal_format, width, height, 0, format, GL_UNSIGNED_BYTE, NULL); - - egl->glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_S, GL_CLAMP_TO_EDGE); - egl->glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_T, GL_CLAMP_TO_EDGE); - egl->glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_NEAREST); - egl->glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_NEAREST); - - egl->glBindTexture(GL_TEXTURE_2D, 0); - return texture_id; -} - -static bool gsr_video_encoder_image_setup_textures(gsr_video_encoder_image *self, AVCodecContext *video_codec_context, AVFrame *frame) { - int res = av_frame_get_buffer(frame, LINESIZE_ALIGNMENT); - if(res < 0) { - fprintf(stderr, "gsr error: gsr_video_encoder_image_setup_textures: av_frame_get_buffer failed: %d\n", res); - return false; - } - - res = av_frame_make_writable(frame); - if(res < 0) { - fprintf(stderr, "gsr error: gsr_video_encoder_image_setup_textures: av_frame_make_writable failed: %d\n", res); - return false; - } - - self->target_texture = gl_create_texture(self->params.egl, video_codec_context->width, video_codec_context->height, self->params.color_depth == GSR_COLOR_DEPTH_8_BITS ? GL_RGB8 : GL_RGB16, GL_RGB); - if(self->target_texture == 0) { - fprintf(stderr, "gsr error: gsr_capture_kms_setup_cuda_textures: failed to create opengl texture\n"); - return false; - } - - return true; -} - -static void gsr_video_encoder_image_stop(gsr_video_encoder_image *self, AVCodecContext *video_codec_context); - -static bool gsr_video_encoder_image_start(gsr_video_encoder *encoder, AVCodecContext *video_codec_context, AVFrame *frame) { - gsr_video_encoder_image *self = encoder->priv; - - video_codec_context->width = FFALIGN(video_codec_context->width, LINESIZE_ALIGNMENT); - video_codec_context->height = FFALIGN(video_codec_context->height, 2); - - frame->width = video_codec_context->width; - frame->height = video_codec_context->height; - - if(!gsr_video_encoder_image_setup_textures(self, video_codec_context, frame)) { - gsr_video_encoder_image_stop(self, video_codec_context); - return false; - } - - return true; -} - -void gsr_video_encoder_image_stop(gsr_video_encoder_image *self, AVCodecContext *video_codec_context) { - (void)video_codec_context; - self->params.egl->glDeleteTextures(1, &self->target_texture); - self->target_texture = 0; -} - -static void gsr_video_encoder_image_copy_textures_to_frame(gsr_video_encoder *encoder, AVFrame *frame, gsr_color_conversion *color_conversion) { - gsr_video_encoder_image *self = encoder->priv; - // TODO: hdr support - self->params.egl->glBindTexture(GL_TEXTURE_2D, self->target_texture); - // We could use glGetTexSubImage and then we wouldn't have to use a specific linesize (LINESIZE_ALIGNMENT) that adds padding, - // but glGetTexSubImage is only available starting from opengl 4.5. - self->params.egl->glGetTexImage(GL_TEXTURE_2D, 0, GL_RGB, GL_UNSIGNED_BYTE, frame->data[0]); - self->params.egl->glBindTexture(GL_TEXTURE_2D, 0); - // cap_kms->kms.base.egl->eglSwapBuffers(cap_kms->kms.base.egl->egl_display, cap_kms->kms.base.egl->egl_surface); - - self->params.egl->glFlush(); - self->params.egl->glFinish(); -} - -static void gsr_video_encoder_image_get_textures(gsr_video_encoder *encoder, unsigned int *textures, int *num_textures, gsr_destination_color *destination_color) { - gsr_video_encoder_image *self = encoder->priv; - textures[0] = self->target_texture; - *num_textures = 1; - // TODO: 10-bit support - //*destination_color = self->params.color_depth == GSR_COLOR_DEPTH_10_BITS ? GSR_DESTINATION_COLOR_P010 : GSR_DESTINATION_COLOR_NV12; - *destination_color = GSR_DESTINATION_COLOR_RGB8; -} - -static void gsr_video_encoder_image_destroy(gsr_video_encoder *encoder, AVCodecContext *video_codec_context) { - gsr_video_encoder_image_stop(encoder->priv, video_codec_context); - free(encoder->priv); - free(encoder); -} - -gsr_video_encoder* gsr_video_encoder_image_create(const gsr_video_encoder_image_params *params) { - gsr_video_encoder *encoder = calloc(1, sizeof(gsr_video_encoder)); - if(!encoder) - return NULL; - - gsr_video_encoder_image *encoder_image = calloc(1, sizeof(gsr_video_encoder_image)); - if(!encoder_image) { - free(encoder); - return NULL; - } - - encoder_image->params = *params; - - *encoder = (gsr_video_encoder) { - .start = gsr_video_encoder_image_start, - .copy_textures_to_frame = gsr_video_encoder_image_copy_textures_to_frame, - .get_textures = gsr_video_encoder_image_get_textures, - .destroy = gsr_video_encoder_image_destroy, - .priv = encoder_image - }; - - return encoder; -} diff --git a/src/image_writer.c b/src/image_writer.c new file mode 100644 index 0000000..d01a66c --- /dev/null +++ b/src/image_writer.c @@ -0,0 +1,85 @@ +#include "../include/image_writer.h" +#include "../include/egl.h" + +#define STB_IMAGE_WRITE_IMPLEMENTATION +#include "../external/stb_image_write.h" + +#include <stdlib.h> +#include <stdint.h> +#include <stdio.h> +#include <assert.h> + +static unsigned int gl_create_texture(gsr_egl *egl, int width, int height, int internal_format, unsigned int format) { + unsigned int texture_id = 0; + egl->glGenTextures(1, &texture_id); + egl->glBindTexture(GL_TEXTURE_2D, texture_id); + egl->glTexImage2D(GL_TEXTURE_2D, 0, internal_format, width, height, 0, format, GL_UNSIGNED_BYTE, NULL); + + egl->glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_S, GL_CLAMP_TO_EDGE); + egl->glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_T, GL_CLAMP_TO_EDGE); + egl->glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_NEAREST); + egl->glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_NEAREST); + + egl->glBindTexture(GL_TEXTURE_2D, 0); + return texture_id; +} + +/* TODO: Support hdr/10-bit */ +bool gsr_image_writer_init(gsr_image_writer *self, gsr_image_writer_source source, gsr_egl *egl, int width, int height) { + assert(source == GSR_IMAGE_WRITER_SOURCE_OPENGL); + self->source = source; + self->egl = egl; + self->width = width; + self->height = height; + self->texture = gl_create_texture(self->egl, self->width, self->height, GL_RGB8, GL_RGB); /* TODO: use GL_RGB16 instead of GL_RGB8 for hdr/10-bit */ + if(self->texture == 0) { + fprintf(stderr, "gsr error: gsr_image_writer_init: failed to create texture\n"); + return false; + } + return true; +} + +void gsr_image_writer_deinit(gsr_image_writer *self) { + if(self->texture) { + self->egl->glDeleteTextures(1, &self->texture); + self->texture = 0; + } +} + +bool gsr_image_writer_write_to_file(gsr_image_writer *self, const char *filepath, gsr_image_format image_format, int quality) { + if(quality < 1) + quality = 1; + else if(quality > 100) + quality = 100; + + uint8_t *frame_data = malloc(self->width * self->height * 3); + if(!frame_data) { + fprintf(stderr, "gsr error: gsr_image_writer_write_to_file: failed to allocate memory for image frame\n"); + return false; + } + + // TODO: hdr support + self->egl->glBindTexture(GL_TEXTURE_2D, self->texture); + // We could use glGetTexSubImage, but it's only available starting from opengl 4.5 + self->egl->glGetTexImage(GL_TEXTURE_2D, 0, GL_RGB, GL_UNSIGNED_BYTE, frame_data); + self->egl->glBindTexture(GL_TEXTURE_2D, 0); + + self->egl->glFlush(); + self->egl->glFinish(); + + bool success = false; + switch(image_format) { + case GSR_IMAGE_FORMAT_JPEG: + success = stbi_write_jpg(filepath, self->width, self->height, 3, frame_data, quality); + break; + case GSR_IMAGE_FORMAT_PNG: + success = stbi_write_png(filepath, self->width, self->height, 3, frame_data, 0); + break; + } + + if(!success) + fprintf(stderr, "gsr error: gsr_image_writer_write_to_file: failed to write image data to output file %s\n", filepath); + + free(frame_data); + return success; +} diff --git a/src/main.cpp b/src/main.cpp index 2eb348d..10dc5a6 100644 --- a/src/main.cpp +++ b/src/main.cpp @@ -13,7 +13,6 @@ extern "C" { #include "../include/encoder/video/vaapi.h" #include "../include/encoder/video/vulkan.h" #include "../include/encoder/video/software.h" -#include "../include/encoder/video/image.h" #include "../include/codec_query/nvenc.h" #include "../include/codec_query/vaapi.h" #include "../include/codec_query/vulkan.h" @@ -23,6 +22,7 @@ extern "C" { #include "../include/utils.h" #include "../include/damage.h" #include "../include/color_conversion.h" +#include "../include/image_writer.h" } #include <assert.h> @@ -113,9 +113,7 @@ enum class VideoCodec { VP8, VP9, H264_VULKAN, - HEVC_VULKAN, - JPEG, - PNG + HEVC_VULKAN }; enum class AudioCodec { @@ -219,16 +217,6 @@ static bool video_codec_is_vulkan(VideoCodec video_codec) { } } -static bool video_codec_is_image(VideoCodec video_codec) { - switch(video_codec) { - case VideoCodec::JPEG: - case VideoCodec::PNG: - return true; - default: - return false; - } -} - struct PacketData { PacketData() {} PacketData(const PacketData&) = delete; @@ -593,22 +581,7 @@ static AVCodecContext *create_video_codec_context(AVPixelFormat pix_fmt, if (codec_context->codec_id == AV_CODEC_ID_MPEG1VIDEO) codec_context->mb_decision = 2; - if(video_codec_is_image(video_codec)) { - switch(video_quality) { - case VideoQuality::MEDIUM: - codec_context->compression_level = 8; - break; - case VideoQuality::HIGH: - codec_context->compression_level = 6; - break; - case VideoQuality::VERY_HIGH: - codec_context->compression_level = 4; - break; - case VideoQuality::ULTRA: - codec_context->compression_level = 2; - break; - } - } else if(!use_software_video_encoder && vendor != GSR_GPU_VENDOR_NVIDIA && bitrate_mode != BitrateMode::CBR) { + if(!use_software_video_encoder && vendor != GSR_GPU_VENDOR_NVIDIA && bitrate_mode != BitrateMode::CBR) { // 8 bit / 10 bit = 80%, and increase it even more const float quality_multiply = hdr ? (8.0f/10.0f * 0.7f) : 1.0f; if(codec_context->codec_id == AV_CODEC_ID_AV1 || codec_context->codec_id == AV_CODEC_ID_H264 || codec_context->codec_id == AV_CODEC_ID_HEVC) { @@ -743,15 +716,6 @@ static AVFrame* create_audio_frame(AVCodecContext *audio_codec_context) { return frame; } -static void open_video_image(AVCodecContext *codec_context) { - AVDictionary *options = nullptr; - int ret = avcodec_open2(codec_context, codec_context->codec, &options); - if (ret < 0) { - fprintf(stderr, "Error: Could not open video codec: %s\n", av_error_to_string(ret)); - _exit(1); - } -} - static void dict_set_profile(AVCodecContext *codec_context, gsr_gpu_vendor vendor, gsr_color_depth color_depth, AVDictionary **options) { #if LIBAVCODEC_VERSION_INT < AV_VERSION_INT(61, 17, 100) if(codec_context->codec_id == AV_CODEC_ID_H264) { @@ -1295,7 +1259,8 @@ static void usage_full() { printf(" %s -w screen -f 60 -a \"app:firefox|app:csgo\" -o \"$HOME/Videos/video.mp4\"\n", program_name); printf(" %s -w screen -f 60 -a \"app-inverse:firefox|app-inverse:csgo\" -o \"$HOME/Videos/video.mp4\"\n", program_name); printf(" %s -w screen -f 60 -a \"default-input|app-inverse:Brave\" -o \"$HOME/Videos/video.mp4\"\n", program_name); - printf(" %s -w screen -f 60 -o \"$HOME/Pictures/image.jpg\"\n", program_name); + printf(" %s -w screen -o \"$HOME/Pictures/image.jpg\"\n", program_name); + printf(" %s -w screen -q medium -o \"$HOME/Pictures/image.jpg\"\n", program_name); //fprintf(stderr, " gpu-screen-recorder -w screen -f 60 -q ultra -pixfmt yuv444 -o video.mp4\n"); fflush(stdout); _exit(1); @@ -1895,14 +1860,6 @@ fail: static gsr_video_encoder* create_video_encoder(gsr_egl *egl, bool overclock, gsr_color_depth color_depth, bool use_software_video_encoder, VideoCodec video_codec) { gsr_video_encoder *video_encoder = nullptr; - if(video_codec_is_image(video_codec)) { - gsr_video_encoder_image_params params; - params.egl = egl; - params.color_depth = color_depth; - video_encoder = gsr_video_encoder_image_create(¶ms); - return video_encoder; - } - if(use_software_video_encoder) { gsr_video_encoder_software_params params; params.egl = egl; @@ -2052,10 +2009,6 @@ static const AVCodec* get_ffmpeg_video_codec(VideoCodec video_codec, gsr_gpu_ven return avcodec_find_encoder_by_name("h264_vulkan"); case VideoCodec::HEVC_VULKAN: return avcodec_find_encoder_by_name("hevc_vulkan"); - case VideoCodec::JPEG: - return avcodec_find_encoder_by_name("libopenjpeg"); - case VideoCodec::PNG: - return avcodec_find_encoder_by_name("png"); } return nullptr; } @@ -2126,10 +2079,6 @@ static void list_supported_video_codecs(gsr_egl *egl, bool wayland) { puts("vp8"); if(supported_video_codecs.vp9.supported) puts("vp9"); - if(avcodec_find_encoder_by_name("libopenjpeg")) - puts("jpeg"); - if(avcodec_find_encoder_by_name("png")) - puts("png"); //if(supported_video_codecs_vulkan.h264.supported) // puts("h264_vulkan"); //if(supported_video_codecs_vulkan.hevc.supported) @@ -2247,6 +2196,9 @@ static void info_command() { list_gpu_info(&egl); puts("section=video_codecs"); list_supported_video_codecs(&egl, wayland); + puts("section=image_formats"); + puts("jpeg"); + puts("png"); puts("section=capture_options"); list_supported_capture_options(window, egl.card_path, list_monitors); @@ -2404,7 +2356,7 @@ static void validate_monitor_get_valid(const gsr_egl *egl, std::string &window_s } } -static gsr_capture* create_capture_impl(std::string &window_str, vec2i output_resolution, bool wayland, gsr_egl *egl, int fps, VideoCodec video_codec, gsr_color_range color_range, +static gsr_capture* create_capture_impl(std::string &window_str, vec2i output_resolution, bool wayland, gsr_egl *egl, int fps, bool hdr, gsr_color_range color_range, bool record_cursor, bool restore_portal_session, const char *portal_session_token_filepath, gsr_color_depth color_depth) { @@ -2478,7 +2430,7 @@ static gsr_capture* create_capture_impl(std::string &window_str, vec2i output_re kms_params.color_depth = color_depth; kms_params.color_range = color_range; kms_params.record_cursor = record_cursor; - kms_params.hdr = video_codec_is_hdr(video_codec); + kms_params.hdr = hdr; kms_params.fps = fps; kms_params.output_resolution = output_resolution; capture = gsr_capture_kms_create(&kms_params); @@ -2516,11 +2468,107 @@ static gsr_capture* create_capture_impl(std::string &window_str, vec2i output_re return capture; } +static gsr_color_range image_format_to_color_range(gsr_image_format image_format) { + switch(image_format) { + case GSR_IMAGE_FORMAT_JPEG: return GSR_COLOR_RANGE_LIMITED; + case GSR_IMAGE_FORMAT_PNG: return GSR_COLOR_RANGE_FULL; + } + assert(false); + return GSR_COLOR_RANGE_FULL; +} + +static int video_quality_to_image_quality_value(VideoQuality video_quality) { + switch(video_quality) { + case VideoQuality::MEDIUM: + return 60; + case VideoQuality::HIGH: + return 70; + case VideoQuality::VERY_HIGH: + return 80; + case VideoQuality::ULTRA: + return 95; + } + assert(false); + return 80; +} + +// TODO: 10-bit and hdr. +static void capture_image_to_file(const char *filepath, std::string &window_str, vec2i output_resolution, bool wayland, gsr_egl *egl, gsr_image_format image_format, + bool record_cursor, bool restore_portal_session, const char *portal_session_token_filepath, VideoQuality video_quality) { + const gsr_color_range color_range = image_format_to_color_range(image_format); + const int fps = 60; + gsr_capture *capture = create_capture_impl(window_str, output_resolution, wayland, egl, fps, false, color_range, record_cursor, restore_portal_session, portal_session_token_filepath, GSR_COLOR_DEPTH_8_BITS); + + gsr_capture_metadata capture_metadata; + capture_metadata.width = 0; + capture_metadata.height = 0; + capture_metadata.fps = fps; + capture_metadata.video_codec_context = nullptr; + capture_metadata.frame = nullptr; + + int capture_result = gsr_capture_start(capture, &capture_metadata); + if(capture_result != 0) { + fprintf(stderr, "gsr error: gsr_capture_start failed\n"); + _exit(capture_result); + } + + gsr_image_writer image_writer; + if(!gsr_image_writer_init(&image_writer, GSR_IMAGE_WRITER_SOURCE_OPENGL, egl, capture_metadata.width, capture_metadata.height)) { + fprintf(stderr, "gsr error: gsr_image_write_gl_init failed\n"); + _exit(1); + } + + gsr_color_conversion_params color_conversion_params; + memset(&color_conversion_params, 0, sizeof(color_conversion_params)); + color_conversion_params.color_range = color_range; + color_conversion_params.egl = egl; + color_conversion_params.load_external_image_shader = gsr_capture_uses_external_image(capture); + + color_conversion_params.destination_textures[0] = image_writer.texture; + color_conversion_params.num_destination_textures = 1; + color_conversion_params.destination_color = GSR_DESTINATION_COLOR_RGB8; + + gsr_color_conversion color_conversion; + if(gsr_color_conversion_init(&color_conversion, &color_conversion_params) != 0) { + fprintf(stderr, "gsr error: gsr_capture_kms_setup_vaapi_textures: failed to create color conversion\n"); + _exit(1); + } + + gsr_color_conversion_clear(&color_conversion); + + bool should_stop_error = false; + egl->glClear(0); + + while(true) { + should_stop_error = false; + if(gsr_capture_should_stop(capture, &should_stop_error)) + break; + + // It can fail, for example when capturing portal and the target is a monitor that hasn't been updated. + // Desktop portal wont refresh the image until there is an update. + // TODO: Find out if there is a way to force update desktop portal image. + // This can also happen for example if the system suspends and the monitor to capture's framebuffer is gone, or if the target window disappeared. + if(gsr_capture_capture(capture, &capture_metadata, &color_conversion) == 0) + break; + + usleep(30 * 1000); // 30 ms + } + + gsr_egl_swap_buffers(egl); + + const int image_quality = video_quality_to_image_quality_value(video_quality); + if(!gsr_image_writer_write_to_file(&image_writer, filepath, image_format, image_quality)) { + fprintf(stderr, "gsr error: failed to write opengl texture to image output file %s\n", filepath); + _exit(1); + } + + gsr_image_writer_deinit(&image_writer); + gsr_capture_destroy(capture); + _exit(should_stop_error ? 3 : 0); +} + static AVPixelFormat get_pixel_format(VideoCodec video_codec, gsr_gpu_vendor vendor, bool use_software_video_encoder) { - if(video_codec_is_image(video_codec)) { - // TODO: hdr - return AV_PIX_FMT_RGB24; - } else if(use_software_video_encoder) { + if(use_software_video_encoder) { return AV_PIX_FMT_NV12; } else { if(video_codec_is_vulkan(video_codec)) @@ -2744,8 +2792,6 @@ static const char* video_codec_to_string(VideoCodec video_codec) { case VideoCodec::VP9: return "vp9"; case VideoCodec::H264_VULKAN: return "h264_vulkan"; case VideoCodec::HEVC_VULKAN: return "hevc_vulkan"; - case VideoCodec::JPEG: return "jpeg"; - case VideoCodec::PNG: return "png"; } return ""; } @@ -2763,8 +2809,6 @@ static bool video_codec_only_supports_low_power_mode(const gsr_supported_video_c case VideoCodec::VP9: return supported_video_codecs.vp9.low_power; case VideoCodec::H264_VULKAN: return supported_video_codecs.h264.low_power; case VideoCodec::HEVC_VULKAN: return supported_video_codecs.hevc.low_power; // TODO: hdr, 10 bit - case VideoCodec::JPEG: return false; - case VideoCodec::PNG: return false; } return false; } @@ -2840,11 +2884,6 @@ static const AVCodec* pick_video_codec(VideoCodec *video_codec, gsr_egl *egl, bo video_codec_f = get_ffmpeg_video_codec(*video_codec, egl->gpu_info.vendor); break; } - case VideoCodec::JPEG: - case VideoCodec::PNG: { - video_codec_f = get_ffmpeg_video_codec(*video_codec, egl->gpu_info.vendor); - break; - } } if(!video_codec_auto && !video_codec_f && !is_flv) { @@ -2906,12 +2945,6 @@ static const AVCodec* pick_video_codec(VideoCodec *video_codec, gsr_egl *egl, bo video_codec_f = get_ffmpeg_video_codec(*video_codec, egl->gpu_info.vendor); break; } - case VideoCodec::JPEG: - case VideoCodec::PNG: { - // TODO: - //assert(false); - break; - } } } @@ -3085,18 +3118,15 @@ static AudioDeviceData create_application_audio_audio_input(const MergedAudioInp } #endif -static void set_video_codec_for_image_output(const char *filename, VideoCodec *video_codec, const char **video_codec_to_use) { - const bool video_codec_auto = strcmp(*video_codec_to_use, "auto") == 0; +static bool get_image_format_from_filename(const char *filename, gsr_image_format *image_format) { if(string_ends_with(filename, ".jpg") || string_ends_with(filename, ".jpeg")) { - if(!video_codec_auto) - fprintf(stderr, "Warning: expected -k option to be set to 'auto' (or not specified) for jpeg output\n"); - *video_codec = VideoCodec::JPEG; - *video_codec_to_use = "jpeg"; + *image_format = GSR_IMAGE_FORMAT_JPEG; + return true; } else if(string_ends_with(filename, ".png")) { - if(!video_codec_auto) - fprintf(stderr, "Warning: expected -k option to be set to 'auto' (or not specified) for png output\n"); - *video_codec = VideoCodec::PNG; - *video_codec_to_use = "png"; + *image_format = GSR_IMAGE_FORMAT_PNG; + return true; + } else { + return false; } } @@ -3739,10 +3769,15 @@ int main(int argc, char **argv) { const bool is_output_piped = strcmp(filename, "/dev/stdout") == 0; - set_video_codec_for_image_output(filename, &video_codec, &video_codec_to_use); - if(video_codec_is_image(video_codec) && !audio_input_arg.values.empty()) { - fprintf(stderr, "Error: can't record audio (-a) when taking a screenshot\n"); - _exit(1); + gsr_image_format image_format; + if(get_image_format_from_filename(filename, &image_format)) { + if(!audio_input_arg.values.empty()) { + fprintf(stderr, "Error: can't record audio (-a) when taking a screenshot\n"); + _exit(1); + } + + capture_image_to_file(filename, window_str, output_resolution, wayland, &egl, image_format, record_cursor, restore_portal_session, portal_session_token_filepath, quality); + _exit(0); } AVFormatContext *av_format_context; @@ -3771,13 +3806,12 @@ int main(int argc, char **argv) { const double target_fps = 1.0 / (double)fps; const bool uses_amix = merged_audio_inputs_should_use_amix(requested_audio_inputs); - if(!video_codec_is_image(video_codec)) - audio_codec = select_audio_codec_with_fallback(audio_codec, file_extension, uses_amix); + audio_codec = select_audio_codec_with_fallback(audio_codec, file_extension, uses_amix); bool low_power = false; const AVCodec *video_codec_f = select_video_codec_with_fallback(&video_codec, video_codec_to_use, file_extension.c_str(), use_software_video_encoder, &egl, &low_power); const gsr_color_depth color_depth = video_codec_to_bit_depth(video_codec); - gsr_capture *capture = create_capture_impl(window_str, output_resolution, wayland, &egl, fps, video_codec, color_range, record_cursor, restore_portal_session, portal_session_token_filepath, color_depth); + gsr_capture *capture = create_capture_impl(window_str, output_resolution, wayland, &egl, fps, video_codec_is_hdr(video_codec), color_range, record_cursor, restore_portal_session, portal_session_token_filepath, color_depth); // (Some?) livestreaming services require at least one audio track to work. // If not audio is provided then create one silent audio track. @@ -3809,20 +3843,32 @@ int main(int argc, char **argv) { _exit(1); } video_frame->format = video_codec_context->pix_fmt; - video_frame->width = video_codec_context->width; - video_frame->height = video_codec_context->height; + video_frame->width = 0; + video_frame->height = 0; video_frame->color_range = video_codec_context->color_range; video_frame->color_primaries = video_codec_context->color_primaries; video_frame->color_trc = video_codec_context->color_trc; video_frame->colorspace = video_codec_context->colorspace; video_frame->chroma_location = video_codec_context->chroma_sample_location; - int capture_result = gsr_capture_start(capture, video_codec_context, video_frame); + gsr_capture_metadata capture_metadata; + capture_metadata.width = 0; + capture_metadata.height = 0; + capture_metadata.fps = fps; + capture_metadata.video_codec_context = video_codec_context; + capture_metadata.frame = video_frame; + + int capture_result = gsr_capture_start(capture, &capture_metadata); if(capture_result != 0) { fprintf(stderr, "gsr error: gsr_capture_start failed\n"); _exit(capture_result); } + video_codec_context->width = capture_metadata.width; + video_codec_context->height = capture_metadata.height; + video_frame->width = capture_metadata.width; + video_frame->height = capture_metadata.height; + gsr_video_encoder *video_encoder = create_video_encoder(&egl, overclock, color_depth, use_software_video_encoder, video_codec); if(!video_encoder) { fprintf(stderr, "Error: failed to create video encoder\n"); @@ -3849,9 +3895,7 @@ int main(int argc, char **argv) { gsr_color_conversion_clear(&color_conversion); - if(video_codec_is_image(video_codec)) { - open_video_image(video_codec_context); - } else if(use_software_video_encoder) { + if(use_software_video_encoder) { open_video_software(video_codec_context, quality, pixel_format, hdr, color_depth, bitrate_mode); } else { open_video_hardware(video_codec_context, quality, very_old_gpu, egl.gpu_info.vendor, pixel_format, hdr, color_depth, bitrate_mode, video_codec, low_power); @@ -3941,8 +3985,6 @@ int main(int argc, char **argv) { if(replay_buffer_size_secs == -1) { AVDictionary *options = nullptr; av_dict_set(&options, "strict", "experimental", 0); - if(video_codec_is_image(video_codec)) - av_dict_set(&options, "update", "true", 0); //av_dict_set_int(&av_format_context->metadata, "video_full_range_flag", 1, 0); int ret = avformat_write_header(av_format_context, &options); @@ -4192,7 +4234,6 @@ int main(int argc, char **argv) { double last_capture_seconds = record_start_time; bool wait_until_frame_time_elapsed = false; - const bool is_image_output = video_codec_is_image(video_codec); while(running) { const double frame_start = clock_get_monotonic_seconds(); @@ -4272,7 +4313,7 @@ int main(int argc, char **argv) { // TODO: Dont do this if no damage? egl.glClear(0); - gsr_capture_capture(capture, video_frame, &color_conversion); + gsr_capture_capture(capture, &capture_metadata, &color_conversion); gsr_egl_swap_buffers(&egl); gsr_video_encoder_copy_textures_to_frame(video_encoder, video_frame, &color_conversion); @@ -4300,10 +4341,6 @@ int main(int argc, char **argv) { // TODO: Move to separate thread because this could write to network (for example when livestreaming) receive_frames(video_codec_context, VIDEO_STREAM_INDEX, video_stream, video_frame->pts, av_format_context, replay_start_time, frame_data_queue, replay_buffer_size_secs, frames_erased, write_output_mutex, paused_time_offset); - if(is_image_output) { - running = 0; - break; - } } else { fprintf(stderr, "Error: avcodec_send_frame failed, error: %s\n", av_error_to_string(ret)); } @@ -4409,7 +4446,7 @@ int main(int argc, char **argv) { gsr_damage_deinit(&damage); gsr_color_conversion_deinit(&color_conversion); gsr_video_encoder_destroy(video_encoder, video_codec_context); - gsr_capture_destroy(capture, video_codec_context); + gsr_capture_destroy(capture); #ifdef GSR_APP_AUDIO gsr_pipewire_audio_deinit(&pipewire_audio); #endif diff --git a/src/utils.c b/src/utils.c index 61ca856..82c51b5 100644 --- a/src/utils.c +++ b/src/utils.c @@ -635,9 +635,12 @@ static VADisplay video_codec_context_get_vaapi_display(AVCodecContext *video_cod } bool video_codec_context_is_vaapi(AVCodecContext *video_codec_context) { + if(!video_codec_context) + return false; + AVBufferRef *hw_frames_ctx = video_codec_context->hw_frames_ctx; if(!hw_frames_ctx) - return NULL; + return false; AVHWFramesContext *hw_frame_context = (AVHWFramesContext*)hw_frames_ctx->data; AVHWDeviceContext *device_context = (AVHWDeviceContext*)hw_frame_context->device_ctx; |