Mercurial > libavformat.hg
view avienc.c @ 4723:a2390c6a35e6 libavformat
Fix index generation in the way that it was supposed to be used. See the
discussion in the ML thread "[PATCH] rmdec.c: merge old/new packet reading
code".
Over time, this code broke somewhat, e.g. seq was never actually written
into (and was thus always 1, therefore the seq condition was always true),
whereas it was supposed to be set to the sequence number of the video slice
in case the video frame is divided over multiple RM packets (slices). The
problem of this is that packets other than those containing the beginning
of a video frame would be indexed as well.
Secondly, flags&2 is supposed to be true for video keyframes and for these
audio packets containing the start of a block. For some codecs (e.g. AAC),
that is every single packet, whereas for others (e.g. cook), that is the
packet containing the first of a series of scrambled packets that are to be
descrambled together. Indexing any of the following would lead to incomplete
and thus useless frames. Problem here is that flags would be reset to 2 to
indicate that the first packet is ready to be returned, and in addition if
no data was left to be returned (which is always true for the first packet),
then we wouldn't actually write the index entry anyway.
All in all, the idea was good and it probably worked at some point, but that
is long ago. This patch should at the very least make it likely for this code
to be executed again at the right times, i.e. the way it was originally
intended to be used.
author | rbultje |
---|---|
date | Sun, 15 Mar 2009 20:14:25 +0000 |
parents | 963e3b76c7a6 |
children | 5748ec2d6705 |
line wrap: on
line source
/* * AVI muxer * Copyright (c) 2000 Fabrice Bellard * * This file is part of FFmpeg. * * FFmpeg is free software; you can redistribute it and/or * modify it under the terms of the GNU Lesser General Public * License as published by the Free Software Foundation; either * version 2.1 of the License, or (at your option) any later version. * * FFmpeg is distributed in the hope that it will be useful, * but WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU * Lesser General Public License for more details. * * You should have received a copy of the GNU Lesser General Public * License along with FFmpeg; if not, write to the Free Software * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA */ #include "avformat.h" #include "avi.h" #include "riff.h" /* * TODO: * - fill all fields if non streamed (nb_frames for example) */ typedef struct AVIIentry { unsigned int flags, pos, len; } AVIIentry; #define AVI_INDEX_CLUSTER_SIZE 16384 typedef struct AVIIndex { int64_t indx_start; int entry; int ents_allocated; AVIIentry** cluster; } AVIIndex; typedef struct { int64_t riff_start, movi_list, odml_list; int64_t frames_hdr_all, frames_hdr_strm[MAX_STREAMS]; int audio_strm_length[MAX_STREAMS]; int riff_id; int packet_count[MAX_STREAMS]; AVIIndex indexes[MAX_STREAMS]; } AVIContext; static inline AVIIentry* avi_get_ientry(AVIIndex* idx, int ent_id) { int cl = ent_id / AVI_INDEX_CLUSTER_SIZE; int id = ent_id % AVI_INDEX_CLUSTER_SIZE; return &idx->cluster[cl][id]; } static int64_t avi_start_new_riff(AVIContext *avi, ByteIOContext *pb, const char* riff_tag, const char* list_tag) { int64_t loff; int i; avi->riff_id++; for (i=0; i<MAX_STREAMS; i++) avi->indexes[i].entry = 0; avi->riff_start = start_tag(pb, "RIFF"); put_tag(pb, riff_tag); loff = start_tag(pb, "LIST"); put_tag(pb, list_tag); return loff; } static char* avi_stream2fourcc(char* tag, int index, enum CodecType type) { tag[0] = '0'; tag[1] = '0' + index; if (type == CODEC_TYPE_VIDEO) { tag[2] = 'd'; tag[3] = 'c'; } else { tag[2] = 'w'; tag[3] = 'b'; } tag[4] = '\0'; return tag; } static void avi_write_info_tag(ByteIOContext *pb, const char *tag, const char *str) { int len = strlen(str); if (len > 0) { len++; put_tag(pb, tag); put_le32(pb, len); put_strz(pb, str); if (len & 1) put_byte(pb, 0); } } static void avi_write_info_tag2(AVFormatContext *s, const char *fourcc, const char *key1, const char *key2) { AVMetadataTag *tag= av_metadata_get(s->metadata, key1, NULL, 0); if(!tag && key2) tag= av_metadata_get(s->metadata, key2, NULL, 0); if(tag) avi_write_info_tag(s->pb, fourcc, tag->value); } static int avi_write_counters(AVFormatContext* s, int riff_id) { ByteIOContext *pb = s->pb; AVIContext *avi = s->priv_data; int n, au_byterate, au_ssize, au_scale, nb_frames = 0; int64_t file_size; AVCodecContext* stream; file_size = url_ftell(pb); for(n = 0; n < s->nb_streams; n++) { assert(avi->frames_hdr_strm[n]); stream = s->streams[n]->codec; url_fseek(pb, avi->frames_hdr_strm[n], SEEK_SET); ff_parse_specific_params(stream, &au_byterate, &au_ssize, &au_scale); if(au_ssize == 0) { put_le32(pb, avi->packet_count[n]); } else { put_le32(pb, avi->audio_strm_length[n] / au_ssize); } if(stream->codec_type == CODEC_TYPE_VIDEO) nb_frames = FFMAX(nb_frames, avi->packet_count[n]); } if(riff_id == 1) { assert(avi->frames_hdr_all); url_fseek(pb, avi->frames_hdr_all, SEEK_SET); put_le32(pb, nb_frames); } url_fseek(pb, file_size, SEEK_SET); return 0; } static int avi_write_header(AVFormatContext *s) { AVIContext *avi = s->priv_data; ByteIOContext *pb = s->pb; int bitrate, n, i, nb_frames, au_byterate, au_ssize, au_scale; AVCodecContext *stream, *video_enc; int64_t list1, list2, strh, strf; /* header list */ avi->riff_id = 0; list1 = avi_start_new_riff(avi, pb, "AVI ", "hdrl"); /* avi header */ put_tag(pb, "avih"); put_le32(pb, 14 * 4); bitrate = 0; video_enc = NULL; for(n=0;n<s->nb_streams;n++) { stream = s->streams[n]->codec; bitrate += stream->bit_rate; if (stream->codec_type == CODEC_TYPE_VIDEO) video_enc = stream; } nb_frames = 0; if(video_enc){ put_le32(pb, (uint32_t)(INT64_C(1000000) * video_enc->time_base.num / video_enc->time_base.den)); } else { put_le32(pb, 0); } put_le32(pb, bitrate / 8); /* XXX: not quite exact */ put_le32(pb, 0); /* padding */ if (url_is_streamed(pb)) put_le32(pb, AVIF_TRUSTCKTYPE | AVIF_ISINTERLEAVED); /* flags */ else put_le32(pb, AVIF_TRUSTCKTYPE | AVIF_HASINDEX | AVIF_ISINTERLEAVED); /* flags */ avi->frames_hdr_all = url_ftell(pb); /* remember this offset to fill later */ put_le32(pb, nb_frames); /* nb frames, filled later */ put_le32(pb, 0); /* initial frame */ put_le32(pb, s->nb_streams); /* nb streams */ put_le32(pb, 1024 * 1024); /* suggested buffer size */ if(video_enc){ put_le32(pb, video_enc->width); put_le32(pb, video_enc->height); } else { put_le32(pb, 0); put_le32(pb, 0); } put_le32(pb, 0); /* reserved */ put_le32(pb, 0); /* reserved */ put_le32(pb, 0); /* reserved */ put_le32(pb, 0); /* reserved */ /* stream list */ for(i=0;i<n;i++) { list2 = start_tag(pb, "LIST"); put_tag(pb, "strl"); stream = s->streams[i]->codec; /* stream generic header */ strh = start_tag(pb, "strh"); switch(stream->codec_type) { case CODEC_TYPE_VIDEO: put_tag(pb, "vids"); break; case CODEC_TYPE_AUDIO: put_tag(pb, "auds"); break; // case CODEC_TYPE_TEXT : put_tag(pb, "txts"); break; case CODEC_TYPE_DATA : put_tag(pb, "dats"); break; } if(stream->codec_type == CODEC_TYPE_VIDEO) put_le32(pb, stream->codec_tag); else put_le32(pb, 1); put_le32(pb, 0); /* flags */ put_le16(pb, 0); /* priority */ put_le16(pb, 0); /* language */ put_le32(pb, 0); /* initial frame */ ff_parse_specific_params(stream, &au_byterate, &au_ssize, &au_scale); put_le32(pb, au_scale); /* scale */ put_le32(pb, au_byterate); /* rate */ av_set_pts_info(s->streams[i], 64, au_scale, au_byterate); put_le32(pb, 0); /* start */ avi->frames_hdr_strm[i] = url_ftell(pb); /* remember this offset to fill later */ if (url_is_streamed(pb)) put_le32(pb, AVI_MAX_RIFF_SIZE); /* FIXME: this may be broken, but who cares */ else put_le32(pb, 0); /* length, XXX: filled later */ /* suggested buffer size */ //FIXME set at the end to largest chunk if(stream->codec_type == CODEC_TYPE_VIDEO) put_le32(pb, 1024 * 1024); else if(stream->codec_type == CODEC_TYPE_AUDIO) put_le32(pb, 12 * 1024); else put_le32(pb, 0); put_le32(pb, -1); /* quality */ put_le32(pb, au_ssize); /* sample size */ put_le32(pb, 0); put_le16(pb, stream->width); put_le16(pb, stream->height); end_tag(pb, strh); if(stream->codec_type != CODEC_TYPE_DATA){ strf = start_tag(pb, "strf"); switch(stream->codec_type) { case CODEC_TYPE_VIDEO: put_bmp_header(pb, stream, codec_bmp_tags, 0); break; case CODEC_TYPE_AUDIO: if (put_wav_header(pb, stream) < 0) { return -1; } break; default: return -1; } end_tag(pb, strf); } if (!url_is_streamed(pb)) { unsigned char tag[5]; int j; /* Starting to lay out AVI OpenDML master index. * We want to make it JUNK entry for now, since we'd * like to get away without making AVI an OpenDML one * for compatibility reasons. */ avi->indexes[i].entry = avi->indexes[i].ents_allocated = 0; avi->indexes[i].indx_start = start_tag(pb, "JUNK"); put_le16(pb, 4); /* wLongsPerEntry */ put_byte(pb, 0); /* bIndexSubType (0 == frame index) */ put_byte(pb, 0); /* bIndexType (0 == AVI_INDEX_OF_INDEXES) */ put_le32(pb, 0); /* nEntriesInUse (will fill out later on) */ put_tag(pb, avi_stream2fourcc(&tag[0], i, stream->codec_type)); /* dwChunkId */ put_le64(pb, 0); /* dwReserved[3] put_le32(pb, 0); Must be 0. */ for (j=0; j < AVI_MASTER_INDEX_SIZE * 2; j++) put_le64(pb, 0); end_tag(pb, avi->indexes[i].indx_start); } if( stream->codec_type == CODEC_TYPE_VIDEO && s->streams[i]->sample_aspect_ratio.num>0 && s->streams[i]->sample_aspect_ratio.den>0){ int vprp= start_tag(pb, "vprp"); AVRational dar = av_mul_q(s->streams[i]->sample_aspect_ratio, (AVRational){stream->width, stream->height}); int num, den; av_reduce(&num, &den, dar.num, dar.den, 0xFFFF); put_le32(pb, 0); //video format = unknown put_le32(pb, 0); //video standard= unknown put_le32(pb, lrintf(1.0/av_q2d(stream->time_base))); put_le32(pb, stream->width ); put_le32(pb, stream->height); put_le16(pb, den); put_le16(pb, num); put_le32(pb, stream->width ); put_le32(pb, stream->height); put_le32(pb, 1); //progressive FIXME put_le32(pb, stream->height); put_le32(pb, stream->width ); put_le32(pb, stream->height); put_le32(pb, stream->width ); put_le32(pb, 0); put_le32(pb, 0); put_le32(pb, 0); put_le32(pb, 0); end_tag(pb, vprp); } end_tag(pb, list2); } if (!url_is_streamed(pb)) { /* AVI could become an OpenDML one, if it grows beyond 2Gb range */ avi->odml_list = start_tag(pb, "JUNK"); put_tag(pb, "odml"); put_tag(pb, "dmlh"); put_le32(pb, 248); for (i = 0; i < 248; i+= 4) put_le32(pb, 0); end_tag(pb, avi->odml_list); } end_tag(pb, list1); list2 = start_tag(pb, "LIST"); put_tag(pb, "INFO"); avi_write_info_tag2(s, "INAM", "Title", NULL); avi_write_info_tag2(s, "IART", "Artist", "Author"); avi_write_info_tag2(s, "ICOP", "Copyright", NULL); avi_write_info_tag2(s, "ICMT", "Comment", NULL); avi_write_info_tag2(s, "IPRD", "Album", NULL); avi_write_info_tag2(s, "IGNR", "Genre", NULL); avi_write_info_tag2(s, "IPRT", "Track", NULL); if(!(s->streams[0]->codec->flags & CODEC_FLAG_BITEXACT)) avi_write_info_tag(pb, "ISFT", LIBAVFORMAT_IDENT); end_tag(pb, list2); /* some padding for easier tag editing */ list2 = start_tag(pb, "JUNK"); for (i = 0; i < 1016; i += 4) put_le32(pb, 0); end_tag(pb, list2); avi->movi_list = start_tag(pb, "LIST"); put_tag(pb, "movi"); put_flush_packet(pb); return 0; } static int avi_write_ix(AVFormatContext *s) { ByteIOContext *pb = s->pb; AVIContext *avi = s->priv_data; char tag[5]; char ix_tag[] = "ix00"; int i, j; assert(!url_is_streamed(pb)); if (avi->riff_id > AVI_MASTER_INDEX_SIZE) return -1; for (i=0;i<s->nb_streams;i++) { int64_t ix, pos; avi_stream2fourcc(&tag[0], i, s->streams[i]->codec->codec_type); ix_tag[3] = '0' + i; /* Writing AVI OpenDML leaf index chunk */ ix = url_ftell(pb); put_tag(pb, &ix_tag[0]); /* ix?? */ put_le32(pb, avi->indexes[i].entry * 8 + 24); /* chunk size */ put_le16(pb, 2); /* wLongsPerEntry */ put_byte(pb, 0); /* bIndexSubType (0 == frame index) */ put_byte(pb, 1); /* bIndexType (1 == AVI_INDEX_OF_CHUNKS) */ put_le32(pb, avi->indexes[i].entry); /* nEntriesInUse */ put_tag(pb, &tag[0]); /* dwChunkId */ put_le64(pb, avi->movi_list);/* qwBaseOffset */ put_le32(pb, 0); /* dwReserved_3 (must be 0) */ for (j=0; j<avi->indexes[i].entry; j++) { AVIIentry* ie = avi_get_ientry(&avi->indexes[i], j); put_le32(pb, ie->pos + 8); put_le32(pb, ((uint32_t)ie->len & ~0x80000000) | (ie->flags & 0x10 ? 0 : 0x80000000)); } put_flush_packet(pb); pos = url_ftell(pb); /* Updating one entry in the AVI OpenDML master index */ url_fseek(pb, avi->indexes[i].indx_start - 8, SEEK_SET); put_tag(pb, "indx"); /* enabling this entry */ url_fskip(pb, 8); put_le32(pb, avi->riff_id); /* nEntriesInUse */ url_fskip(pb, 16*avi->riff_id); put_le64(pb, ix); /* qwOffset */ put_le32(pb, pos - ix); /* dwSize */ put_le32(pb, avi->indexes[i].entry); /* dwDuration */ url_fseek(pb, pos, SEEK_SET); } return 0; } static int avi_write_idx1(AVFormatContext *s) { ByteIOContext *pb = s->pb; AVIContext *avi = s->priv_data; int64_t idx_chunk; int i; char tag[5]; if (!url_is_streamed(pb)) { AVIIentry* ie = 0, *tie; int entry[MAX_STREAMS]; int empty, stream_id = -1; idx_chunk = start_tag(pb, "idx1"); memset(&entry[0], 0, sizeof(entry)); do { empty = 1; for (i=0; i<s->nb_streams; i++) { if (avi->indexes[i].entry <= entry[i]) continue; tie = avi_get_ientry(&avi->indexes[i], entry[i]); if (empty || tie->pos < ie->pos) { ie = tie; stream_id = i; } empty = 0; } if (!empty) { avi_stream2fourcc(&tag[0], stream_id, s->streams[stream_id]->codec->codec_type); put_tag(pb, &tag[0]); put_le32(pb, ie->flags); put_le32(pb, ie->pos); put_le32(pb, ie->len); entry[stream_id]++; } } while (!empty); end_tag(pb, idx_chunk); avi_write_counters(s, avi->riff_id); } return 0; } static int avi_write_packet(AVFormatContext *s, AVPacket *pkt) { AVIContext *avi = s->priv_data; ByteIOContext *pb = s->pb; unsigned char tag[5]; unsigned int flags=0; const int stream_index= pkt->stream_index; AVCodecContext *enc= s->streams[stream_index]->codec; int size= pkt->size; // av_log(s, AV_LOG_DEBUG, "%"PRId64" %d %d\n", pkt->dts, avi->packet_count[stream_index], stream_index); while(enc->block_align==0 && pkt->dts != AV_NOPTS_VALUE && pkt->dts > avi->packet_count[stream_index]){ AVPacket empty_packet; av_init_packet(&empty_packet); empty_packet.size= 0; empty_packet.data= NULL; empty_packet.stream_index= stream_index; avi_write_packet(s, &empty_packet); // av_log(s, AV_LOG_DEBUG, "dup %"PRId64" %d\n", pkt->dts, avi->packet_count[stream_index]); } avi->packet_count[stream_index]++; // Make sure to put an OpenDML chunk when the file size exceeds the limits if (!url_is_streamed(pb) && (url_ftell(pb) - avi->riff_start > AVI_MAX_RIFF_SIZE)) { avi_write_ix(s); end_tag(pb, avi->movi_list); if (avi->riff_id == 1) avi_write_idx1(s); end_tag(pb, avi->riff_start); avi->movi_list = avi_start_new_riff(avi, pb, "AVIX", "movi"); } avi_stream2fourcc(&tag[0], stream_index, enc->codec_type); if(pkt->flags&PKT_FLAG_KEY) flags = 0x10; if (enc->codec_type == CODEC_TYPE_AUDIO) { avi->audio_strm_length[stream_index] += size; } if (!url_is_streamed(s->pb)) { AVIIndex* idx = &avi->indexes[stream_index]; int cl = idx->entry / AVI_INDEX_CLUSTER_SIZE; int id = idx->entry % AVI_INDEX_CLUSTER_SIZE; if (idx->ents_allocated <= idx->entry) { idx->cluster = av_realloc(idx->cluster, (cl+1)*sizeof(void*)); if (!idx->cluster) return -1; idx->cluster[cl] = av_malloc(AVI_INDEX_CLUSTER_SIZE*sizeof(AVIIentry)); if (!idx->cluster[cl]) return -1; idx->ents_allocated += AVI_INDEX_CLUSTER_SIZE; } idx->cluster[cl][id].flags = flags; idx->cluster[cl][id].pos = url_ftell(pb) - avi->movi_list; idx->cluster[cl][id].len = size; idx->entry++; } put_buffer(pb, tag, 4); put_le32(pb, size); put_buffer(pb, pkt->data, size); if (size & 1) put_byte(pb, 0); put_flush_packet(pb); return 0; } static int avi_write_trailer(AVFormatContext *s) { AVIContext *avi = s->priv_data; ByteIOContext *pb = s->pb; int res = 0; int i, j, n, nb_frames; int64_t file_size; if (!url_is_streamed(pb)){ if (avi->riff_id == 1) { end_tag(pb, avi->movi_list); res = avi_write_idx1(s); end_tag(pb, avi->riff_start); } else { avi_write_ix(s); end_tag(pb, avi->movi_list); end_tag(pb, avi->riff_start); file_size = url_ftell(pb); url_fseek(pb, avi->odml_list - 8, SEEK_SET); put_tag(pb, "LIST"); /* Making this AVI OpenDML one */ url_fskip(pb, 16); for (n=nb_frames=0;n<s->nb_streams;n++) { AVCodecContext *stream = s->streams[n]->codec; if (stream->codec_type == CODEC_TYPE_VIDEO) { if (nb_frames < avi->packet_count[n]) nb_frames = avi->packet_count[n]; } else { if (stream->codec_id == CODEC_ID_MP2 || stream->codec_id == CODEC_ID_MP3) { nb_frames += avi->packet_count[n]; } } } put_le32(pb, nb_frames); url_fseek(pb, file_size, SEEK_SET); avi_write_counters(s, avi->riff_id); } } put_flush_packet(pb); for (i=0; i<MAX_STREAMS; i++) { for (j=0; j<avi->indexes[i].ents_allocated/AVI_INDEX_CLUSTER_SIZE; j++) av_free(avi->indexes[i].cluster[j]); av_free(avi->indexes[i].cluster); avi->indexes[i].cluster = NULL; avi->indexes[i].ents_allocated = avi->indexes[i].entry = 0; } return res; } AVOutputFormat avi_muxer = { "avi", NULL_IF_CONFIG_SMALL("AVI format"), "video/x-msvideo", "avi", sizeof(AVIContext), CODEC_ID_MP2, CODEC_ID_MPEG4, avi_write_header, avi_write_packet, avi_write_trailer, .codec_tag= (const AVCodecTag* const []){codec_bmp_tags, codec_wav_tags, 0}, .flags= AVFMT_VARIABLE_FPS, };