tools_common.c 13.3 KB
Newer Older
1 2 3 4 5 6 7 8 9
/*
 *  Copyright (c) 2010 The WebM project authors. All Rights Reserved.
 *
 *  Use of this source code is governed by a BSD-style license
 *  that can be found in the LICENSE file in the root of the source
 *  tree. An additional intellectual property rights grant can be found
 *  in the file PATENTS.  All contributing project authors may
 *  be found in the AUTHORS file in the root of the source tree.
 */
10

11
#include <math.h>
12
#include <stdarg.h>
13
#include <stdio.h>
14
#include <stdlib.h>
15
#include <string.h>
16

17 18
#include "./tools_common.h"

19
#if CONFIG_VP8_ENCODER || CONFIG_VP9_ENCODER || CONFIG_VP10_ENCODER
20 21 22
#include "vpx/vp8cx.h"
#endif

23
#if CONFIG_VP8_DECODER || CONFIG_VP9_DECODER || CONFIG_VP10_DECODER
24 25 26
#include "vpx/vp8dx.h"
#endif

27
#if defined(_WIN32) || defined(__OS2__)
28 29
#include <io.h>
#include <fcntl.h>
30 31 32 33 34 35

#ifdef __OS2__
#define _setmode    setmode
#define _fileno     fileno
#define _O_BINARY   O_BINARY
#endif
36 37
#endif

38 39 40 41 42 43 44 45 46 47 48 49
#define LOG_ERROR(label) do {\
  const char *l = label;\
  va_list ap;\
  va_start(ap, fmt);\
  if (l)\
    fprintf(stderr, "%s: ", l);\
  vfprintf(stderr, fmt, ap);\
  fprintf(stderr, "\n");\
  va_end(ap);\
} while (0)


John Koleszar's avatar
John Koleszar committed
50 51
FILE *set_binary_mode(FILE *stream) {
  (void)stream;
52
#if defined(_WIN32) || defined(__OS2__)
John Koleszar's avatar
John Koleszar committed
53
  _setmode(_fileno(stream), _O_BINARY);
54
#endif
John Koleszar's avatar
John Koleszar committed
55
  return stream;
56
}
57 58 59 60 61 62 63 64 65 66 67 68 69 70

void die(const char *fmt, ...) {
  LOG_ERROR(NULL);
  usage_exit();
}

void fatal(const char *fmt, ...) {
  LOG_ERROR("Fatal");
  exit(EXIT_FAILURE);
}

void warn(const char *fmt, ...) {
  LOG_ERROR("Warning");
}
71

72 73 74 75 76 77 78 79 80
void die_codec(vpx_codec_ctx_t *ctx, const char *s) {
  const char *detail = vpx_codec_error_detail(ctx);

  printf("%s: %s\n", s, vpx_codec_error(ctx));
  if (detail)
    printf("    %s\n", detail);
  exit(EXIT_FAILURE);
}

81 82 83 84 85
int read_yuv_frame(struct VpxInputContext *input_ctx, vpx_image_t *yuv_frame) {
  FILE *f = input_ctx->file;
  struct FileTypeDetectionBuffer *detect = &input_ctx->detect;
  int plane = 0;
  int shortread = 0;
86
  const int bytespp = (yuv_frame->fmt & VPX_IMG_FMT_HIGHBITDEPTH) ? 2 : 1;
87 88 89

  for (plane = 0; plane < 3; ++plane) {
    uint8_t *ptr;
90 91
    const int w = vpx_img_plane_width(yuv_frame, plane);
    const int h = vpx_img_plane_height(yuv_frame, plane);
92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111
    int r;

    /* Determine the correct plane based on the image format. The for-loop
     * always counts in Y,U,V order, but this may not match the order of
     * the data on disk.
     */
    switch (plane) {
      case 1:
        ptr = yuv_frame->planes[
            yuv_frame->fmt == VPX_IMG_FMT_YV12 ? VPX_PLANE_V : VPX_PLANE_U];
        break;
      case 2:
        ptr = yuv_frame->planes[
            yuv_frame->fmt == VPX_IMG_FMT_YV12 ? VPX_PLANE_U : VPX_PLANE_V];
        break;
      default:
        ptr = yuv_frame->planes[plane];
    }

    for (r = 0; r < h; ++r) {
112
      size_t needed = w * bytespp;
113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131
      size_t buf_position = 0;
      const size_t left = detect->buf_read - detect->position;
      if (left > 0) {
        const size_t more = (left < needed) ? left : needed;
        memcpy(ptr, detect->buf + detect->position, more);
        buf_position = more;
        needed -= more;
        detect->position += more;
      }
      if (needed > 0) {
        shortread |= (fread(ptr + buf_position, 1, needed, f) < needed);
      }

      ptr += yuv_frame->stride[plane];
    }
  }

  return shortread;
}
132

133 134
#if CONFIG_ENCODERS

135
static const VpxInterface vpx_encoders[] = {
136 137 138 139
#if CONFIG_VP10_ENCODER
  {"vp10", VP10_FOURCC, &vpx_codec_vp10_cx},
#endif

140 141 142 143 144 145 146 147 148
#if CONFIG_VP8_ENCODER
  {"vp8", VP8_FOURCC, &vpx_codec_vp8_cx},
#endif

#if CONFIG_VP9_ENCODER
  {"vp9", VP9_FOURCC, &vpx_codec_vp9_cx},
#endif
};

149
int get_vpx_encoder_count(void) {
150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168
  return sizeof(vpx_encoders) / sizeof(vpx_encoders[0]);
}

const VpxInterface *get_vpx_encoder_by_index(int i) {
  return &vpx_encoders[i];
}

const VpxInterface *get_vpx_encoder_by_name(const char *name) {
  int i;

  for (i = 0; i < get_vpx_encoder_count(); ++i) {
    const VpxInterface *encoder = get_vpx_encoder_by_index(i);
    if (strcmp(encoder->name, name) == 0)
      return encoder;
  }

  return NULL;
}

169 170 171 172
#endif  // CONFIG_ENCODERS

#if CONFIG_DECODERS

173
static const VpxInterface vpx_decoders[] = {
174
#if CONFIG_VP8_DECODER
175
  {"vp8", VP8_FOURCC, &vpx_codec_vp8_dx},
176
#endif
177

178
#if CONFIG_VP9_DECODER
179
  {"vp9", VP9_FOURCC, &vpx_codec_vp9_dx},
180
#endif
181 182 183 184

#if CONFIG_VP10_DECODER
  {"vp10", VP10_FOURCC, &vpx_codec_vp10_dx},
#endif
185 186
};

187
int get_vpx_decoder_count(void) {
188 189 190 191 192 193 194 195 196 197 198 199 200 201
  return sizeof(vpx_decoders) / sizeof(vpx_decoders[0]);
}

const VpxInterface *get_vpx_decoder_by_index(int i) {
  return &vpx_decoders[i];
}

const VpxInterface *get_vpx_decoder_by_name(const char *name) {
  int i;

  for (i = 0; i < get_vpx_decoder_count(); ++i) {
     const VpxInterface *const decoder = get_vpx_decoder_by_index(i);
     if (strcmp(decoder->name, name) == 0)
       return decoder;
202
  }
203 204 205 206 207 208 209 210 211 212 213 214 215

  return NULL;
}

const VpxInterface *get_vpx_decoder_by_fourcc(uint32_t fourcc) {
  int i;

  for (i = 0; i < get_vpx_decoder_count(); ++i) {
    const VpxInterface *const decoder = get_vpx_decoder_by_index(i);
    if (decoder->fourcc == fourcc)
      return decoder;
  }

216 217 218
  return NULL;
}

219 220
#endif  // CONFIG_DECODERS

221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236
// TODO(dkovalev): move this function to vpx_image.{c, h}, so it will be part
// of vpx_image_t support
int vpx_img_plane_width(const vpx_image_t *img, int plane) {
  if (plane > 0 && img->x_chroma_shift > 0)
    return (img->d_w + 1) >> img->x_chroma_shift;
  else
    return img->d_w;
}

int vpx_img_plane_height(const vpx_image_t *img, int plane) {
  if (plane > 0 &&  img->y_chroma_shift > 0)
    return (img->d_h + 1) >> img->y_chroma_shift;
  else
    return img->d_h;
}

237
void vpx_img_write(const vpx_image_t *img, FILE *file) {
238
  int plane;
239 240 241 242

  for (plane = 0; plane < 3; ++plane) {
    const unsigned char *buf = img->planes[plane];
    const int stride = img->stride[plane];
243 244
    const int w = vpx_img_plane_width(img, plane) *
        ((img->fmt & VPX_IMG_FMT_HIGHBITDEPTH) ? 2 : 1);
245 246
    const int h = vpx_img_plane_height(img, plane);
    int y;
247

248 249 250 251 252 253
    for (y = 0; y < h; ++y) {
      fwrite(buf, 1, w, file);
      buf += stride;
    }
  }
}
254 255 256 257 258 259 260

int vpx_img_read(vpx_image_t *img, FILE *file) {
  int plane;

  for (plane = 0; plane < 3; ++plane) {
    unsigned char *buf = img->planes[plane];
    const int stride = img->stride[plane];
261 262
    const int w = vpx_img_plane_width(img, plane) *
        ((img->fmt & VPX_IMG_FMT_HIGHBITDEPTH) ? 2 : 1);
263
    const int h = vpx_img_plane_height(img, plane);
264 265 266
    int y;

    for (y = 0; y < h; ++y) {
267
      if (fread(buf, 1, w, file) != (size_t)w)
268 269 270 271 272 273 274 275
        return 0;
      buf += stride;
    }
  }

  return 1;
}

276 277 278 279 280 281 282 283 284 285 286
// TODO(dkovalev) change sse_to_psnr signature: double -> int64_t
double sse_to_psnr(double samples, double peak, double sse) {
  static const double kMaxPSNR = 100.0;

  if (sse > 0.0) {
    const double psnr = 10.0 * log10(samples * peak * peak / sse);
    return psnr > kMaxPSNR ? kMaxPSNR : psnr;
  } else {
    return kMaxPSNR;
  }
}
287 288

// TODO(debargha): Consolidate the functions below into a separate file.
289
#if CONFIG_VP9_HIGHBITDEPTH
290 291 292 293 294
static void highbd_img_upshift(vpx_image_t *dst, vpx_image_t *src,
                               int input_shift) {
  // Note the offset is 1 less than half.
  const int offset = input_shift > 0 ? (1 << (input_shift - 1)) - 1 : 0;
  int plane;
295
  if (dst->d_w != src->d_w || dst->d_h != src->d_h ||
296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311
      dst->x_chroma_shift != src->x_chroma_shift ||
      dst->y_chroma_shift != src->y_chroma_shift ||
      dst->fmt != src->fmt || input_shift < 0) {
    fatal("Unsupported image conversion");
  }
  switch (src->fmt) {
    case VPX_IMG_FMT_I42016:
    case VPX_IMG_FMT_I42216:
    case VPX_IMG_FMT_I44416:
    case VPX_IMG_FMT_I44016:
      break;
    default:
      fatal("Unsupported image conversion");
      break;
  }
  for (plane = 0; plane < 3; plane++) {
312 313
    int w = src->d_w;
    int h = src->d_h;
314 315
    int x, y;
    if (plane) {
316 317
      w = (w + src->x_chroma_shift) >> src->x_chroma_shift;
      h = (h + src->y_chroma_shift) >> src->y_chroma_shift;
318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334
    }
    for (y = 0; y < h; y++) {
      uint16_t *p_src =
          (uint16_t *)(src->planes[plane] + y * src->stride[plane]);
      uint16_t *p_dst =
          (uint16_t *)(dst->planes[plane] + y * dst->stride[plane]);
      for (x = 0; x < w; x++)
        *p_dst++ = (*p_src++ << input_shift) + offset;
    }
  }
}

static void lowbd_img_upshift(vpx_image_t *dst, vpx_image_t *src,
                              int input_shift) {
  // Note the offset is 1 less than half.
  const int offset = input_shift > 0 ? (1 << (input_shift - 1)) - 1 : 0;
  int plane;
335
  if (dst->d_w != src->d_w || dst->d_h != src->d_h ||
336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352
      dst->x_chroma_shift != src->x_chroma_shift ||
      dst->y_chroma_shift != src->y_chroma_shift ||
      dst->fmt != src->fmt + VPX_IMG_FMT_HIGHBITDEPTH ||
      input_shift < 0) {
    fatal("Unsupported image conversion");
  }
  switch (src->fmt) {
    case VPX_IMG_FMT_I420:
    case VPX_IMG_FMT_I422:
    case VPX_IMG_FMT_I444:
    case VPX_IMG_FMT_I440:
      break;
    default:
      fatal("Unsupported image conversion");
      break;
  }
  for (plane = 0; plane < 3; plane++) {
353 354
    int w = src->d_w;
    int h = src->d_h;
355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402
    int x, y;
    if (plane) {
      w = (w + src->x_chroma_shift) >> src->x_chroma_shift;
      h = (h + src->y_chroma_shift) >> src->y_chroma_shift;
    }
    for (y = 0; y < h; y++) {
      uint8_t *p_src = src->planes[plane] + y * src->stride[plane];
      uint16_t *p_dst =
          (uint16_t *)(dst->planes[plane] + y * dst->stride[plane]);
      for (x = 0; x < w; x++) {
        *p_dst++ = (*p_src++ << input_shift) + offset;
      }
    }
  }
}

void vpx_img_upshift(vpx_image_t *dst, vpx_image_t *src,
                     int input_shift) {
  if (src->fmt & VPX_IMG_FMT_HIGHBITDEPTH) {
    highbd_img_upshift(dst, src, input_shift);
  } else {
    lowbd_img_upshift(dst, src, input_shift);
  }
}

void vpx_img_truncate_16_to_8(vpx_image_t *dst, vpx_image_t *src) {
  int plane;
  if (dst->fmt + VPX_IMG_FMT_HIGHBITDEPTH != src->fmt ||
      dst->d_w != src->d_w || dst->d_h != src->d_h ||
      dst->x_chroma_shift != src->x_chroma_shift ||
      dst->y_chroma_shift != src->y_chroma_shift) {
    fatal("Unsupported image conversion");
  }
  switch (dst->fmt) {
    case VPX_IMG_FMT_I420:
    case VPX_IMG_FMT_I422:
    case VPX_IMG_FMT_I444:
    case VPX_IMG_FMT_I440:
      break;
    default:
      fatal("Unsupported image conversion");
      break;
  }
  for (plane = 0; plane < 3; plane++) {
    int w = src->d_w;
    int h = src->d_h;
    int x, y;
    if (plane) {
403 404
      w = (w + src->x_chroma_shift) >> src->x_chroma_shift;
      h = (h + src->y_chroma_shift) >> src->y_chroma_shift;
405 406 407 408 409 410
    }
    for (y = 0; y < h; y++) {
      uint16_t *p_src =
          (uint16_t *)(src->planes[plane] + y * src->stride[plane]);
      uint8_t *p_dst = dst->planes[plane] + y * dst->stride[plane];
      for (x = 0; x < w; x++) {
Yaowu Xu's avatar
Yaowu Xu committed
411
        *p_dst++ = (uint8_t)(*p_src++);
412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501
      }
    }
  }
}

static void highbd_img_downshift(vpx_image_t *dst, vpx_image_t *src,
                                 int down_shift) {
  int plane;
  if (dst->d_w != src->d_w || dst->d_h != src->d_h ||
      dst->x_chroma_shift != src->x_chroma_shift ||
      dst->y_chroma_shift != src->y_chroma_shift ||
      dst->fmt != src->fmt || down_shift < 0) {
    fatal("Unsupported image conversion");
  }
  switch (src->fmt) {
    case VPX_IMG_FMT_I42016:
    case VPX_IMG_FMT_I42216:
    case VPX_IMG_FMT_I44416:
    case VPX_IMG_FMT_I44016:
      break;
    default:
      fatal("Unsupported image conversion");
      break;
  }
  for (plane = 0; plane < 3; plane++) {
    int w = src->d_w;
    int h = src->d_h;
    int x, y;
    if (plane) {
      w = (w + src->x_chroma_shift) >> src->x_chroma_shift;
      h = (h + src->y_chroma_shift) >> src->y_chroma_shift;
    }
    for (y = 0; y < h; y++) {
      uint16_t *p_src =
          (uint16_t *)(src->planes[plane] + y * src->stride[plane]);
      uint16_t *p_dst =
          (uint16_t *)(dst->planes[plane] + y * dst->stride[plane]);
      for (x = 0; x < w; x++)
        *p_dst++ = *p_src++ >> down_shift;
    }
  }
}

static void lowbd_img_downshift(vpx_image_t *dst, vpx_image_t *src,
                                int down_shift) {
  int plane;
  if (dst->d_w != src->d_w || dst->d_h != src->d_h ||
      dst->x_chroma_shift != src->x_chroma_shift ||
      dst->y_chroma_shift != src->y_chroma_shift ||
      src->fmt != dst->fmt + VPX_IMG_FMT_HIGHBITDEPTH ||
      down_shift < 0) {
    fatal("Unsupported image conversion");
  }
  switch (dst->fmt) {
    case VPX_IMG_FMT_I420:
    case VPX_IMG_FMT_I422:
    case VPX_IMG_FMT_I444:
    case VPX_IMG_FMT_I440:
      break;
    default:
      fatal("Unsupported image conversion");
      break;
  }
  for (plane = 0; plane < 3; plane++) {
    int w = src->d_w;
    int h = src->d_h;
    int x, y;
    if (plane) {
      w = (w + src->x_chroma_shift) >> src->x_chroma_shift;
      h = (h + src->y_chroma_shift) >> src->y_chroma_shift;
    }
    for (y = 0; y < h; y++) {
      uint16_t *p_src =
          (uint16_t *)(src->planes[plane] + y * src->stride[plane]);
      uint8_t *p_dst = dst->planes[plane] + y * dst->stride[plane];
      for (x = 0; x < w; x++) {
        *p_dst++ = *p_src++ >> down_shift;
      }
    }
  }
}

void vpx_img_downshift(vpx_image_t *dst, vpx_image_t *src,
                       int down_shift) {
  if (dst->fmt & VPX_IMG_FMT_HIGHBITDEPTH) {
    highbd_img_downshift(dst, src, down_shift);
  } else {
    lowbd_img_downshift(dst, src, down_shift);
  }
}
502
#endif  // CONFIG_VP9_HIGHBITDEPTH