FFmpeg
h264_mp4toannexb_bsf.c
Go to the documentation of this file.
1 /*
2  * H.264 MP4 to Annex B byte stream format filter
3  * Copyright (c) 2007 Benoit Fouet <benoit.fouet@free.fr>
4  *
5  * This file is part of FFmpeg.
6  *
7  * FFmpeg is free software; you can redistribute it and/or
8  * modify it under the terms of the GNU Lesser General Public
9  * License as published by the Free Software Foundation; either
10  * version 2.1 of the License, or (at your option) any later version.
11  *
12  * FFmpeg is distributed in the hope that it will be useful,
13  * but WITHOUT ANY WARRANTY; without even the implied warranty of
14  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
15  * Lesser General Public License for more details.
16  *
17  * You should have received a copy of the GNU Lesser General Public
18  * License along with FFmpeg; if not, write to the Free Software
19  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
20  */
21 
22 #include <string.h>
23 
24 #include "libavutil/avassert.h"
25 #include "libavutil/intreadwrite.h"
26 #include "libavutil/mem.h"
27 
28 #include "avcodec.h"
29 #include "bsf.h"
30 #include "bsf_internal.h"
31 #include "bytestream.h"
32 #include "h264.h"
33 
34 typedef struct H264BSFContext {
37  int sps_size;
38  int pps_size;
45 
46 static void count_or_copy(uint8_t **out, uint64_t *out_size,
47  const uint8_t *in, int in_size, int ps, int copy)
48 {
49  uint8_t start_code_size = ps < 0 ? 0 : *out_size == 0 || ps ? 4 : 3;
50 
51  if (copy) {
52  memcpy(*out + start_code_size, in, in_size);
53  if (start_code_size == 4) {
54  AV_WB32(*out, 1);
55  } else if (start_code_size) {
56  (*out)[0] =
57  (*out)[1] = 0;
58  (*out)[2] = 1;
59  }
60  *out += start_code_size + in_size;
61  }
62  *out_size += start_code_size + in_size;
63 }
64 
65 static int h264_extradata_to_annexb(AVBSFContext *ctx, const int padding)
66 {
67  H264BSFContext *s = ctx->priv_data;
68  GetByteContext ogb, *gb = &ogb;
69  uint16_t unit_size;
70  uint32_t total_size = 0;
71  uint8_t *out = NULL, unit_nb, sps_done = 0;
72  static const uint8_t nalu_header[4] = { 0, 0, 0, 1 };
73  int length_size, pps_offset = 0;
74 
76 
77  bytestream2_skipu(gb, 4);
78 
79  /* retrieve length coded size */
80  length_size = (bytestream2_get_byteu(gb) & 0x3) + 1;
81 
82  /* retrieve sps and pps unit(s) */
83  unit_nb = bytestream2_get_byteu(gb) & 0x1f; /* number of sps unit(s) */
84  if (!unit_nb) {
85  goto pps;
86  }
87 
88  while (unit_nb--) {
89  int err;
90 
91  /* possible overread ok due to padding */
92  unit_size = bytestream2_get_be16u(gb);
93  total_size += unit_size + 4;
94  av_assert1(total_size <= INT_MAX - padding);
95  if (bytestream2_get_bytes_left(gb) < unit_size + !sps_done) {
96  av_log(ctx, AV_LOG_ERROR, "Global extradata truncated, "
97  "corrupted stream or invalid MP4/AVCC bitstream\n");
98  av_free(out);
99  return AVERROR_INVALIDDATA;
100  }
101  if ((err = av_reallocp(&out, total_size + padding)) < 0)
102  return err;
103  memcpy(out + total_size - unit_size - 4, nalu_header, 4);
104  bytestream2_get_bufferu(gb, out + total_size - unit_size, unit_size);
105 pps:
106  if (!unit_nb && !sps_done++) {
107  unit_nb = bytestream2_get_byteu(gb); /* number of pps unit(s) */
108  pps_offset = total_size;
109  }
110  }
111 
112  if (out)
113  memset(out + total_size, 0, padding);
114 
115  if (pps_offset) {
116  s->sps = out;
117  s->sps_size = pps_offset;
118  } else {
119  av_log(ctx, AV_LOG_WARNING,
120  "Warning: SPS NALU missing or invalid. "
121  "The resulting stream may not play.\n");
122  }
123  if (pps_offset < total_size) {
124  s->pps = out + pps_offset;
125  s->pps_size = total_size - pps_offset;
126  } else {
127  av_log(ctx, AV_LOG_WARNING,
128  "Warning: PPS NALU missing or invalid. "
129  "The resulting stream may not play.\n");
130  }
131 
132  av_freep(&ctx->par_out->extradata);
133  ctx->par_out->extradata = out;
134  ctx->par_out->extradata_size = total_size;
135 
136  return length_size;
137 }
138 
140 {
141  H264BSFContext *s = ctx->priv_data;
142  int extra_size = ctx->par_in->extradata_size;
143  int ret;
144 
145  /* retrieve sps and pps NAL units from extradata */
146  if (!extra_size ||
147  (extra_size >= 3 && AV_RB24(ctx->par_in->extradata) == 1) ||
148  (extra_size >= 4 && AV_RB32(ctx->par_in->extradata) == 1)) {
149  av_log(ctx, AV_LOG_VERBOSE,
150  "The input looks like it is Annex B already\n");
151  } else if (extra_size >= 7) {
153  if (ret < 0)
154  return ret;
155 
156  s->length_size = ret;
157  s->new_idr = 1;
158  s->idr_sps_seen = 0;
159  s->idr_pps_seen = 0;
160  s->extradata_parsed = 1;
161  } else {
162  av_log(ctx, AV_LOG_ERROR, "Invalid extradata size: %d\n", extra_size);
163  return AVERROR_INVALIDDATA;
164  }
165 
166  return 0;
167 }
168 
170 {
171  H264BSFContext *s = ctx->priv_data;
172  AVPacket *in;
173  uint8_t unit_type, new_idr, sps_seen, pps_seen;
174  const uint8_t *buf;
175  const uint8_t *buf_end;
176  uint8_t *out;
177  uint64_t out_size;
178  int ret;
179 
180  ret = ff_bsf_get_packet(ctx, &in);
181  if (ret < 0)
182  return ret;
183 
184  /* nothing to filter */
185  if (!s->extradata_parsed) {
186  av_packet_move_ref(opkt, in);
187  av_packet_free(&in);
188  return 0;
189  }
190 
191  buf_end = in->data + in->size;
192 
193 #define LOG_ONCE(...) \
194  if (j) \
195  av_log(__VA_ARGS__)
196  for (int j = 0; j < 2; j++) {
197  buf = in->data;
198  new_idr = s->new_idr;
199  sps_seen = s->idr_sps_seen;
200  pps_seen = s->idr_pps_seen;
201  out_size = 0;
202 
203  do {
204  uint32_t nal_size = 0;
205 
206  /* possible overread ok due to padding */
207  for (int i = 0; i < s->length_size; i++)
208  nal_size = (nal_size << 8) | buf[i];
209 
210  buf += s->length_size;
211 
212  /* This check requires the cast as the right side might
213  * otherwise be promoted to an unsigned value. */
214  if ((int64_t)nal_size > buf_end - buf) {
215  ret = AVERROR_INVALIDDATA;
216  goto fail;
217  }
218 
219  if (!nal_size)
220  continue;
221 
222  unit_type = *buf & 0x1f;
223 
224  if (unit_type == H264_NAL_SPS) {
225  sps_seen = new_idr = 1;
226  } else if (unit_type == H264_NAL_PPS) {
227  pps_seen = new_idr = 1;
228  /* if SPS has not been seen yet, prepend the AVCC one to PPS */
229  if (!sps_seen) {
230  if (!s->sps_size) {
231  LOG_ONCE(ctx, AV_LOG_WARNING, "SPS not present in the stream, nor in AVCC, stream may be unreadable\n");
232  } else {
233  count_or_copy(&out, &out_size, s->sps, s->sps_size, -1, j);
234  sps_seen = 1;
235  }
236  }
237  }
238 
239  /* If this is a new IDR picture following an IDR picture, reset the idr flag.
240  * Just check first_mb_in_slice to be 0 as this is the simplest solution.
241  * This could be checking idr_pic_id instead, but would complexify the parsing. */
242  if (!new_idr && unit_type == H264_NAL_IDR_SLICE && (buf[1] & 0x80))
243  new_idr = 1;
244 
245  /* prepend only to the first type 5 NAL unit of an IDR picture, if no sps/pps are already present */
246  if (new_idr && unit_type == H264_NAL_IDR_SLICE && !sps_seen && !pps_seen) {
247  if (ctx->par_out->extradata)
248  count_or_copy(&out, &out_size, ctx->par_out->extradata,
249  ctx->par_out->extradata_size, -1, j);
250  new_idr = 0;
251  /* if only SPS has been seen, also insert PPS */
252  } else if (new_idr && unit_type == H264_NAL_IDR_SLICE && sps_seen && !pps_seen) {
253  if (!s->pps_size) {
254  LOG_ONCE(ctx, AV_LOG_WARNING, "PPS not present in the stream, nor in AVCC, stream may be unreadable\n");
255  } else {
256  count_or_copy(&out, &out_size, s->pps, s->pps_size, -1, j);
257  }
258  }
259 
260  count_or_copy(&out, &out_size, buf, nal_size,
261  unit_type == H264_NAL_SPS || unit_type == H264_NAL_PPS, j);
262  if (!new_idr && unit_type == H264_NAL_SLICE) {
263  new_idr = 1;
264  sps_seen = 0;
265  pps_seen = 0;
266  }
267 
268  buf += nal_size;
269  } while (buf < buf_end);
270 
271  if (!j) {
272  if (out_size > INT_MAX - AV_INPUT_BUFFER_PADDING_SIZE) {
273  ret = AVERROR_INVALIDDATA;
274  goto fail;
275  }
276  ret = av_new_packet(opkt, out_size);
277  if (ret < 0)
278  goto fail;
279  out = opkt->data;
280  }
281  }
282 #undef LOG_ONCE
283 
284  av_assert1(out_size == opkt->size);
285 
286  s->new_idr = new_idr;
287  s->idr_sps_seen = sps_seen;
288  s->idr_pps_seen = pps_seen;
289 
290  ret = av_packet_copy_props(opkt, in);
291  if (ret < 0)
292  goto fail;
293 
294 fail:
295  if (ret < 0)
296  av_packet_unref(opkt);
297  av_packet_free(&in);
298 
299  return ret;
300 }
301 
303 {
304  H264BSFContext *s = ctx->priv_data;
305 
306  s->idr_sps_seen = 0;
307  s->idr_pps_seen = 0;
308  s->new_idr = s->extradata_parsed;
309 }
310 
311 static const enum AVCodecID codec_ids[] = {
313 };
314 
316  .name = "h264_mp4toannexb",
317  .priv_data_size = sizeof(H264BSFContext),
321  .codec_ids = codec_ids,
322 };
#define NULL
Definition: coverity.c:32
#define AVERROR_INVALIDDATA
Invalid data found when processing input.
Definition: error.h:59
AVCodecParameters * par_out
Parameters of the output stream.
Definition: bsf.h:83
static void copy(const float *p1, float *p2, const int length)
static void flush(AVCodecContext *avctx)
#define AV_LOG_WARNING
Something somehow does not look correct.
Definition: log.h:200
Memory handling functions.
static av_cold int init(AVCodecContext *avctx)
Definition: avrndec.c:35
The bitstream filter state.
Definition: bsf.h:49
int size
Definition: packet.h:364
static av_always_inline void bytestream2_init(GetByteContext *g, const uint8_t *buf, int buf_size)
Definition: bytestream.h:133
int out_size
Definition: movenc.c:55
static int h264_mp4toannexb_init(AVBSFContext *ctx)
void * priv_data
Opaque filter-specific private data.
Definition: bsf.h:70
static av_always_inline unsigned int bytestream2_get_bufferu(GetByteContext *g, uint8_t *dst, unsigned int size)
Definition: bytestream.h:273
void av_packet_free(AVPacket **pkt)
Free the packet, if the packet is reference counted, it will be unreferenced first.
Definition: avpacket.c:64
uint8_t
uint64_t_TMPL AV_WL64 unsigned int_TMPL AV_WL32 unsigned int_TMPL AV_WL24 unsigned int_TMPL AV_WL16 uint64_t_TMPL AV_WB64 unsigned int_TMPL AV_RB32
Definition: bytestream.h:87
const char * name
Definition: bsf.h:99
uint8_t * data
Definition: packet.h:363
void av_packet_move_ref(AVPacket *dst, AVPacket *src)
Move every field in src to dst and reset src.
Definition: avpacket.c:664
static int h264_extradata_to_annexb(AVBSFContext *ctx, const int padding)
static av_always_inline void bytestream2_skipu(GetByteContext *g, unsigned int size)
Definition: bytestream.h:170
#define AV_LOG_VERBOSE
Detailed information.
Definition: log.h:210
#define av_log(a,...)
static av_always_inline int bytestream2_get_bytes_left(GetByteContext *g)
Definition: bytestream.h:154
H.264 common definitions.
int av_new_packet(AVPacket *pkt, int size)
Allocate the payload of a packet and initialize its fields with default values.
Definition: avpacket.c:88
AVCodecID
Identify the syntax and semantics of the bitstream.
Definition: codec_id.h:46
#define AV_LOG_ERROR
Something went wrong and cannot losslessly be recovered.
Definition: log.h:194
filter_frame For filters that do not use the this method is called when a frame is pushed to the filter s input It can be called at any time except in a reentrant way If the input frame is enough to produce then the filter should push the output frames on the output link immediately As an exception to the previous rule if the input frame is enough to produce several output frames then the filter needs output only at least one per link The additional frames can be left buffered in the filter
static void h264_mp4toannexb_flush(AVBSFContext *ctx)
simple assert() macros that are a bit more flexible than ISO C assert().
#define fail()
Definition: checkasm.h:123
int extradata_size
Size of the extradata content in bytes.
Definition: codec_par.h:78
int av_packet_copy_props(AVPacket *dst, const AVPacket *src)
Copy only "properties" fields from src to dst.
Definition: avpacket.c:572
static void count_or_copy(uint8_t **out, uint64_t *out_size, const uint8_t *in, int in_size, int ps, int copy)
#define av_assert1(cond)
assert() equivalent, that does not lie in speed critical code.
Definition: avassert.h:53
AVFormatContext * ctx
Definition: movenc.c:48
#define s(width, name)
Definition: cbs_vp9.c:257
int av_reallocp(void *ptr, size_t size)
Allocate, reallocate, or free a block of memory through a pointer to a pointer.
Definition: mem.c:161
Libavcodec external API header.
uint64_t_TMPL AV_WL64 unsigned int_TMPL AV_WL32 unsigned int_TMPL AV_WL24 unsigned int_TMPL AV_WL16 uint64_t_TMPL AV_WB64 unsigned int_TMPL AV_WB32 unsigned int_TMPL AV_RB24
Definition: bytestream.h:87
void av_packet_unref(AVPacket *pkt)
Wipe the packet.
Definition: avpacket.c:606
#define LOG_ONCE(...)
uint8_t pi<< 24) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_U8, uint8_t,(*(const uint8_t *) pi-0x80)*(1.0f/(1<< 7))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_U8, uint8_t,(*(const uint8_t *) pi-0x80)*(1.0/(1<< 7))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_S16, int16_t,(*(const int16_t *) pi >> 8)+0x80) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_S16, int16_t,*(const int16_t *) pi *(1.0f/(1<< 15))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_S16, int16_t,*(const int16_t *) pi *(1.0/(1<< 15))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_S32, int32_t,(*(const int32_t *) pi >> 24)+0x80) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_S32, int32_t,*(const int32_t *) pi *(1.0f/(1U<< 31))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_S32, int32_t,*(const int32_t *) pi *(1.0/(1U<< 31))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_FLT, float, av_clip_uint8(lrintf(*(const float *) pi *(1<< 7))+0x80)) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S16, int16_t, AV_SAMPLE_FMT_FLT, float, av_clip_int16(lrintf(*(const float *) pi *(1<< 15)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S32, int32_t, AV_SAMPLE_FMT_FLT, float, av_clipl_int32(llrintf(*(const float *) pi *(1U<< 31)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_DBL, double, av_clip_uint8(lrint(*(const double *) pi *(1<< 7))+0x80)) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S16, int16_t, AV_SAMPLE_FMT_DBL, double, av_clip_int16(lrint(*(const double *) pi *(1<< 15)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S32, int32_t, AV_SAMPLE_FMT_DBL, double, av_clipl_int32(llrint(*(const double *) pi *(1U<< 31))))#define SET_CONV_FUNC_GROUP(ofmt, ifmt) static void set_generic_function(AudioConvert *ac){}void ff_audio_convert_free(AudioConvert **ac){if(!*ac) return;ff_dither_free(&(*ac) ->dc);av_freep(ac);}AudioConvert *ff_audio_convert_alloc(AVAudioResampleContext *avr, enum AVSampleFormat out_fmt, enum AVSampleFormat in_fmt, int channels, int sample_rate, int apply_map){AudioConvert *ac;int in_planar, out_planar;ac=av_mallocz(sizeof(*ac));if(!ac) return NULL;ac->avr=avr;ac->out_fmt=out_fmt;ac->in_fmt=in_fmt;ac->channels=channels;ac->apply_map=apply_map;if(avr->dither_method!=AV_RESAMPLE_DITHER_NONE &&av_get_packed_sample_fmt(out_fmt)==AV_SAMPLE_FMT_S16 &&av_get_bytes_per_sample(in_fmt) > 2){ac->dc=ff_dither_alloc(avr, out_fmt, in_fmt, channels, sample_rate, apply_map);if(!ac->dc){av_free(ac);return NULL;}return ac;}in_planar=ff_sample_fmt_is_planar(in_fmt, channels);out_planar=ff_sample_fmt_is_planar(out_fmt, channels);if(in_planar==out_planar){ac->func_type=CONV_FUNC_TYPE_FLAT;ac->planes=in_planar?ac->channels:1;}else if(in_planar) ac->func_type=CONV_FUNC_TYPE_INTERLEAVE;else ac->func_type=CONV_FUNC_TYPE_DEINTERLEAVE;set_generic_function(ac);if(ARCH_AARCH64) ff_audio_convert_init_aarch64(ac);if(ARCH_ARM) ff_audio_convert_init_arm(ac);if(ARCH_X86) ff_audio_convert_init_x86(ac);return ac;}int ff_audio_convert(AudioConvert *ac, AudioData *out, AudioData *in){int use_generic=1;int len=in->nb_samples;int p;if(ac->dc){av_log(ac->avr, AV_LOG_TRACE,"%d samples - audio_convert: %s to %s (dithered)\n", len, av_get_sample_fmt_name(ac->in_fmt), av_get_sample_fmt_name(ac->out_fmt));return ff_convert_dither(ac-> in
#define AV_WB32(p, v)
Definition: intreadwrite.h:419
static enum AVCodecID codec_ids[]
int ff_bsf_get_packet(AVBSFContext *ctx, AVPacket **pkt)
Called by the bitstream filters to get the next packet for filtering.
Definition: bsf.c:231
#define AV_INPUT_BUFFER_PADDING_SIZE
Required number of additionally allocated bytes at the end of the input bitstream for decoding...
Definition: avcodec.h:215
const AVBitStreamFilter ff_h264_mp4toannexb_bsf
#define av_free(p)
static int h264_mp4toannexb_filter(AVBSFContext *ctx, AVPacket *opkt)
uint8_t * extradata
Extra binary data needed for initializing the decoder, codec-dependent.
Definition: codec_par.h:74
FILE * out
Definition: movenc.c:54
#define av_freep(p)
This structure stores compressed data.
Definition: packet.h:340
AVCodecParameters * par_in
Parameters of the input stream.
Definition: bsf.h:77
int i
Definition: input.c:407