FFmpeg: libavcodec/libtheoraenc.c Source File

FFmpeg

[フレーム]

libavcodec

libtheoraenc.c

Go to the documentation of this file.

1 /*

3 *

4 * This file is part of FFmpeg.

5 *

6 * FFmpeg is free software; you can redistribute it and/or

7 * modify it under the terms of the GNU Lesser General Public

8 * License as published by the Free Software Foundation; either

9 * version 2.1 of the License, or (at your option) any later version.

10 *

11 * FFmpeg is distributed in the hope that it will be useful,

12 * but WITHOUT ANY WARRANTY; without even the implied warranty of

13 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU

14 * Lesser General Public License for more details.

15 *

16 * You should have received a copy of the GNU Lesser General Public

17 * License along with FFmpeg; if not, write to the Free Software

18 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA

19 */

21 /**

22 * @file

23 * @brief Theora encoder using libtheora.

24 * @author Paul Richards <paul.richards@gmail.com>

25 *

26 * A lot of this is copy / paste from other output codecs in

27 * libavcodec or pure guesswork (or both).

28 *

29 * I have used t_ prefixes on variables which are libtheora types

30 * and o_ prefixes on variables which are libogg types.

31 */

33 /* FFmpeg includes */

34 #include "libavutil/common.h"

35 #include "libavutil/intreadwrite.h"

36 #include "libavutil/pixdesc.h"

37 #include "libavutil/log.h"

38 #include "libavutil/base64.h"

39 #include "avcodec.h"

40 #include "internal.h"

42 /* libtheora includes */

43 #include <theora/theoraenc.h>

45 typedef struct TheoraContext {

46 th_enc_ctx *t_state;

47 uint8_t *stats;

48 int stats_size;

49 int stats_offset;

50 int uv_hshift;

51 int uv_vshift;

52 int keyframe_mask;

53 } TheoraContext;

55 /** Concatenate an ogg_packet into the extradata. */

56 static int concatenate_packet(unsigned int* offset,

57 AVCodecContext* avc_context,

58 const ogg_packet* packet)

59 {

60 const char* message = NULL;

61 uint8_t* newdata = NULL;

62 int newsize = avc_context->extradata_size + 2 + packet->bytes;

63 int ret;

65 if (packet->bytes < 0) {

66 message = "ogg_packet has negative size";

67 ret = AVERROR_INVALIDDATA;

68 } else if (packet->bytes > 0xffff) {

69 message = "ogg_packet is larger than 65535 bytes";

70 ret = AVERROR_INVALIDDATA;

71 } else if (newsize < avc_context->extradata_size) {

72 message = "extradata_size would overflow";

73 ret = AVERROR_INVALIDDATA;

74 } else {

75 newdata = av_realloc(avc_context->extradata, newsize);

76 if (!newdata)

77 message = "av_realloc failed";

78 ret = AVERROR(ENOMEM);

79 }

80 if (message) {

81 av_log(avc_context, AV_LOG_ERROR, "concatenate_packet failed: %s\n", message);

82 return ret;

83 }

85 avc_context->extradata = newdata;

86 avc_context->extradata_size = newsize;

87 AV_WB16(avc_context->extradata + (*offset), packet->bytes);

88 *offset += 2;

89 memcpy(avc_context->extradata + (*offset), packet->packet, packet->bytes);

90 (*offset) += packet->bytes;

91 return 0;

92 }

94 static int get_stats(AVCodecContext *avctx, int eos)

95 {

96 #ifdef TH_ENCCTL_2PASS_OUT

97 TheoraContext *h = avctx->priv_data;

98 uint8_t *buf;

99 int bytes;

100

101 bytes = th_encode_ctl(h->t_state, TH_ENCCTL_2PASS_OUT, &buf, sizeof(buf));

102 if (bytes < 0) {

103 av_log(avctx, AV_LOG_ERROR, "Error getting first pass stats\n");

104 return AVERROR_EXTERNAL;

105 }

106 if (!eos) {

107 h->stats = av_fast_realloc(h->stats, &h->stats_size,

108 h->stats_offset + bytes);

109 memcpy(h->stats + h->stats_offset, buf, bytes);

110 h->stats_offset += bytes;

111 } else {

112 int b64_size = AV_BASE64_SIZE(h->stats_offset);

113 // libtheora generates a summary header at the end

114 memcpy(h->stats, buf, bytes);

115 avctx->stats_out = av_malloc(b64_size);

116 av_base64_encode(avctx->stats_out, b64_size, h->stats, h->stats_offset);

117 }

118 return 0;

119 #else

120 av_log(avctx, AV_LOG_ERROR, "libtheora too old to support 2pass\n");

121 return AVERROR(ENOSUP);

122 #endif

123 }

124

125 // libtheora won't read the entire buffer we give it at once, so we have to

126 // repeatedly submit it...

127 static int submit_stats(AVCodecContext *avctx)

128 {

129 #ifdef TH_ENCCTL_2PASS_IN

130 TheoraContext *h = avctx->priv_data;

131 int bytes;

132 if (!h->stats) {

133 if (!avctx->stats_in) {

134 av_log(avctx, AV_LOG_ERROR, "No statsfile for second pass\n");

135 return AVERROR(EINVAL);

136 }

137 h->stats_size = strlen(avctx->stats_in) * 3/4;

138 h->stats = av_malloc(h->stats_size);

139 h->stats_size = av_base64_decode(h->stats, avctx->stats_in, h->stats_size);

140 }

141 while (h->stats_size - h->stats_offset > 0) {

142 bytes = th_encode_ctl(h->t_state, TH_ENCCTL_2PASS_IN,

143 h->stats + h->stats_offset,

144 h->stats_size - h->stats_offset);

145 if (bytes < 0) {

146 av_log(avctx, AV_LOG_ERROR, "Error submitting stats\n");

147 return AVERROR_EXTERNAL;

148 }

149 if (!bytes)

150 return 0;

151 h->stats_offset += bytes;

152 }

153 return 0;

154 #else

155 av_log(avctx, AV_LOG_ERROR, "libtheora too old to support 2pass\n");

156 return AVERROR(ENOSUP);

157 #endif

158 }

159

160 static av_cold int encode_init(AVCodecContext* avc_context)

161 {

162 th_info t_info;

163 th_comment t_comment;

164 ogg_packet o_packet;

165 unsigned int offset;

166 TheoraContext *h = avc_context->priv_data;

167 uint32_t gop_size = avc_context->gop_size;

168 int ret;

169

170 /* Set up the theora_info struct */

171 th_info_init(&t_info);

172 t_info.frame_width = FFALIGN(avc_context->width, 16);

173 t_info.frame_height = FFALIGN(avc_context->height, 16);

174 t_info.pic_width = avc_context->width;

175 t_info.pic_height = avc_context->height;

176 t_info.pic_x = 0;

177 t_info.pic_y = 0;

178 /* Swap numerator and denominator as time_base in AVCodecContext gives the

179 * time period between frames, but theora_info needs the framerate. */

180 t_info.fps_numerator = avc_context->time_base.den;

181 t_info.fps_denominator = avc_context->time_base.num;

182 if (avc_context->sample_aspect_ratio.num) {

183 t_info.aspect_numerator = avc_context->sample_aspect_ratio.num;

184 t_info.aspect_denominator = avc_context->sample_aspect_ratio.den;

185 } else {

186 t_info.aspect_numerator = 1;

187 t_info.aspect_denominator = 1;

188 }

189

190 if (avc_context->color_primaries == AVCOL_PRI_BT470M)

191 t_info.colorspace = TH_CS_ITU_REC_470M;

192 else if (avc_context->color_primaries == AVCOL_PRI_BT470BG)

193 t_info.colorspace = TH_CS_ITU_REC_470BG;

194 else

195 t_info.colorspace = TH_CS_UNSPECIFIED;

196

197 if (avc_context->pix_fmt == AV_PIX_FMT_YUV420P)

198 t_info.pixel_fmt = TH_PF_420;

199 else if (avc_context->pix_fmt == AV_PIX_FMT_YUV422P)

200 t_info.pixel_fmt = TH_PF_422;

201 else if (avc_context->pix_fmt == AV_PIX_FMT_YUV444P)

202 t_info.pixel_fmt = TH_PF_444;

203 else {

204 av_log(avc_context, AV_LOG_ERROR, "Unsupported pix_fmt\n");

205 return AVERROR(EINVAL);

206 }

207 avcodec_get_chroma_sub_sample(avc_context->pix_fmt, &h->uv_hshift, &h->uv_vshift);

208

209 if (avc_context->flags & CODEC_FLAG_QSCALE) {

210 /* to be constant with the libvorbis implementation, clip global_quality to 0 - 10

211 Theora accepts a quality parameter p, which is:

212 * 0 <= p <=63

213 * an int value

214 */

215 t_info.quality = av_clipf(avc_context->global_quality / (float)FF_QP2LAMBDA, 0, 10) * 6.3;

216 t_info.target_bitrate = 0;

217 } else {

218 t_info.target_bitrate = avc_context->bit_rate;

219 t_info.quality = 0;

220 }

221

222 /* Now initialise libtheora */

223 h->t_state = th_encode_alloc(&t_info);

224 if (!h->t_state) {

225 av_log(avc_context, AV_LOG_ERROR, "theora_encode_init failed\n");

226 return AVERROR_EXTERNAL;

227 }

228

229 h->keyframe_mask = (1 << t_info.keyframe_granule_shift) - 1;

230 /* Clear up theora_info struct */

231 th_info_clear(&t_info);

232

233 if (th_encode_ctl(h->t_state, TH_ENCCTL_SET_KEYFRAME_FREQUENCY_FORCE,

234 &gop_size, sizeof(gop_size))) {

235 av_log(avc_context, AV_LOG_ERROR, "Error setting GOP size\n");

236 return AVERROR_EXTERNAL;

237 }

238

239 // need to enable 2 pass (via TH_ENCCTL_2PASS_) before encoding headers

240 if (avc_context->flags & CODEC_FLAG_PASS1) {

241 if ((ret = get_stats(avc_context, 0)) < 0)

242 return ret;

243 } else if (avc_context->flags & CODEC_FLAG_PASS2) {

244 if ((ret = submit_stats(avc_context)) < 0)

245 return ret;

246 }

247

248 /*

249 Output first header packet consisting of theora

250 header, comment, and tables.

251

252 Each one is prefixed with a 16bit size, then they

253 are concatenated together into libavcodec's extradata.

254 */

255 offset = 0;

256

257 /* Headers */

258 th_comment_init(&t_comment);

259

260 while (th_encode_flushheader(h->t_state, &t_comment, &o_packet))

261 if ((ret = concatenate_packet(&offset, avc_context, &o_packet)) < 0)

262 return ret;

263

264 th_comment_clear(&t_comment);

265

266 /* Set up the output AVFrame */

267 avc_context->coded_frame= avcodec_alloc_frame();

268

269 return 0;

270 }

271

272 static int encode_frame(AVCodecContext* avc_context, AVPacket *pkt,

273 const AVFrame *frame, int *got_packet)

274 {

275 th_ycbcr_buffer t_yuv_buffer;

276 TheoraContext *h = avc_context->priv_data;

277 ogg_packet o_packet;

278 int result, i, ret;

279

280 // EOS, finish and get 1st pass stats if applicable

281 if (!frame) {

282 th_encode_packetout(h->t_state, 1, &o_packet);

283 if (avc_context->flags & CODEC_FLAG_PASS1)

284 if ((ret = get_stats(avc_context, 1)) < 0)

285 return ret;

286 return 0;

287 }

288

289 /* Copy planes to the theora yuv_buffer */

290 for (i = 0; i < 3; i++) {

291 t_yuv_buffer[i].width = FFALIGN(avc_context->width, 16) >> (i && h->uv_hshift);

292 t_yuv_buffer[i].height = FFALIGN(avc_context->height, 16) >> (i && h->uv_vshift);

293 t_yuv_buffer[i].stride = frame->linesize[i];

294 t_yuv_buffer[i].data = frame->data[i];

295 }

296

297 if (avc_context->flags & CODEC_FLAG_PASS2)

298 if ((ret = submit_stats(avc_context)) < 0)

299 return ret;

300

301 /* Now call into theora_encode_YUVin */

302 result = th_encode_ycbcr_in(h->t_state, t_yuv_buffer);

303 if (result) {

304 const char* message;

305 switch (result) {

306 case -1:

307 message = "differing frame sizes";

308 break;

309 case TH_EINVAL:

310 message = "encoder is not ready or is finished";

311 break;

312 default:

313 message = "unknown reason";

314 break;

315 }

316 av_log(avc_context, AV_LOG_ERROR, "theora_encode_YUVin failed (%s) [%d]\n", message, result);

317 return AVERROR_EXTERNAL;

318 }

319

320 if (avc_context->flags & CODEC_FLAG_PASS1)

321 if ((ret = get_stats(avc_context, 0)) < 0)

322 return ret;

323

324 /* Pick up returned ogg_packet */

325 result = th_encode_packetout(h->t_state, 0, &o_packet);

326 switch (result) {

327 case 0:

328 /* No packet is ready */

329 return 0;

330 case 1:

331 /* Success, we have a packet */

332 break;

333 default:

334 av_log(avc_context, AV_LOG_ERROR, "theora_encode_packetout failed [%d]\n", result);

335 return AVERROR_EXTERNAL;

336 }

337

338 /* Copy ogg_packet content out to buffer */

339 if ((ret = ff_alloc_packet2(avc_context, pkt, o_packet.bytes)) < 0)

340 return ret;

341 memcpy(pkt->data, o_packet.packet, o_packet.bytes);

342

343 // HACK: assumes no encoder delay, this is true until libtheora becomes

344 // multithreaded (which will be disabled unless explicitly requested)

345 pkt->pts = pkt->dts = frame->pts;

346 avc_context->coded_frame->key_frame = !(o_packet.granulepos & h->keyframe_mask);

347 if (avc_context->coded_frame->key_frame)

348 pkt->flags |= AV_PKT_FLAG_KEY;

349 *got_packet = 1;

350

351 return 0;

352 }

353

354 static av_cold int encode_close(AVCodecContext* avc_context)

355 {

356 TheoraContext *h = avc_context->priv_data;

357

358 th_encode_free(h->t_state);

359 av_freep(&h->stats);

360 av_freep(&avc_context->coded_frame);

361 av_freep(&avc_context->stats_out);

362 av_freep(&avc_context->extradata);

363 avc_context->extradata_size = 0;

364

365 return 0;

366 }

367

368 /** AVCodec struct exposed to libavcodec */

369 AVCodec ff_libtheora_encoder = {

370 .name = "libtheora",

371 .type = AVMEDIA_TYPE_VIDEO,

372 .id = AV_CODEC_ID_THEORA,

373 .priv_data_size = sizeof(TheoraContext),

374 .init = encode_init,

375 .close = encode_close,

376 .encode2 = encode_frame,

377 .capabilities = CODEC_CAP_DELAY, // needed to get the statsfile summary

378 .pix_fmts = (const enum AVPixelFormat[]){

379 AV_PIX_FMT_YUV420P, AV_PIX_FMT_YUV422P, AV_PIX_FMT_YUV444P, AV_PIX_FMT_NONE

380 },

381 .long_name = NULL_IF_CONFIG_SMALL("libtheora Theora"),

382 };

Generated on Sat May 25 2013 03:58:36 for FFmpeg by doxygen 1.8.2