FFmpeg
xwma.c
Go to the documentation of this file.
1 /*
2  * xWMA demuxer
3  * Copyright (c) 2011 Max Horn
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 <inttypes.h>
23 #include <stdint.h>
24 
25 #include "avformat.h"
26 #include "internal.h"
27 #include "riff.h"
28 
29 /*
30  * Demuxer for xWMA, a Microsoft audio container used by XAudio 2.
31  */
32 
33 typedef struct XWMAContext {
34  int64_t data_end;
35 } XWMAContext;
36 
37 static int xwma_probe(const AVProbeData *p)
38 {
39  if (!memcmp(p->buf, "RIFF", 4) && !memcmp(p->buf + 8, "XWMA", 4))
40  return AVPROBE_SCORE_MAX;
41  return 0;
42 }
43 
45 {
46  int64_t size;
47  int ret = 0;
48  uint32_t dpds_table_size = 0;
49  uint32_t *dpds_table = NULL;
50  unsigned int tag;
51  AVIOContext *pb = s->pb;
52  AVStream *st;
53  XWMAContext *xwma = s->priv_data;
54  int i;
55 
56  /* The following code is mostly copied from wav.c, with some
57  * minor alterations.
58  */
59 
60  /* check RIFF header */
61  tag = avio_rl32(pb);
62  if (tag != MKTAG('R', 'I', 'F', 'F'))
63  return AVERROR_INVALIDDATA;
64  avio_rl32(pb); /* file size */
65  tag = avio_rl32(pb);
66  if (tag != MKTAG('X', 'W', 'M', 'A'))
67  return AVERROR_INVALIDDATA;
68 
69  /* parse fmt header */
70  tag = avio_rl32(pb);
71  if (tag != MKTAG('f', 'm', 't', ' '))
72  return AVERROR_INVALIDDATA;
73  size = avio_rl32(pb);
74  st = avformat_new_stream(s, NULL);
75  if (!st)
76  return AVERROR(ENOMEM);
77 
78  ret = ff_get_wav_header(s, pb, st->codecpar, size, 0);
79  if (ret < 0)
80  return ret;
82 
83  /* XWMA encoder only allows a few channel/sample rate/bitrate combinations,
84  * but some create identical files with fake bitrate (1ch 22050hz at
85  * 20/48/192kbps are all 20kbps, with the exact same codec data).
86  * Decoder needs correct bitrate to work, so it's normalized here. */
87  if (st->codecpar->codec_id == AV_CODEC_ID_WMAV2) {
88  int ch = st->codecpar->channels;
89  int sr = st->codecpar->sample_rate;
90  int br = st->codecpar->bit_rate;
91 
92  if (ch == 1) {
93  if (sr == 22050 && (br==48000 || br==192000))
94  br = 20000;
95  else if (sr == 32000 && (br==48000 || br==192000))
96  br = 20000;
97  else if (sr == 44100 && (br==96000 || br==192000))
98  br = 48000;
99  }
100  else if (ch == 2) {
101  if (sr == 22050 && (br==48000 || br==192000))
102  br = 32000;
103  else if (sr == 32000 && (br==192000))
104  br = 48000;
105  }
106 
107  st->codecpar->bit_rate = br;
108  }
109 
110  /* Normally xWMA can only contain WMAv2 with 1/2 channels,
111  * and WMAPRO with 6 channels. */
112  if (st->codecpar->codec_id != AV_CODEC_ID_WMAV2 &&
114  avpriv_request_sample(s, "Unexpected codec (tag %s; id %d)",
116  st->codecpar->codec_id);
117  } else {
118  /* xWMA shouldn't have extradata. But the WMA codecs require it,
119  * so we provide our own fake extradata.
120  *
121  * First, check that there really was no extradata in the header. If
122  * there was, then try to use it, after asking the user to provide a
123  * sample of this unusual file.
124  */
125  if (st->codecpar->extradata_size != 0) {
126  /* Surprise, surprise: We *did* get some extradata. No idea
127  * if it will work, but just go on and try it, after asking
128  * the user for a sample.
129  */
130  avpriv_request_sample(s, "Unexpected extradata (%d bytes)",
131  st->codecpar->extradata_size);
132  } else if (st->codecpar->codec_id == AV_CODEC_ID_WMAPRO) {
133  if ((ret = ff_alloc_extradata(st->codecpar, 18)) < 0)
134  return ret;
135 
136  memset(st->codecpar->extradata, 0, st->codecpar->extradata_size);
138  st->codecpar->extradata[14] = 224;
139  } else {
140  if ((ret = ff_alloc_extradata(st->codecpar, 6)) < 0)
141  return ret;
142 
143  memset(st->codecpar->extradata, 0, st->codecpar->extradata_size);
144  /* setup extradata with our experimentally obtained value */
145  st->codecpar->extradata[4] = 31;
146  }
147  }
148 
149  if (!st->codecpar->channels) {
150  av_log(s, AV_LOG_WARNING, "Invalid channel count: %d\n",
151  st->codecpar->channels);
152  return AVERROR_INVALIDDATA;
153  }
154  if (!st->codecpar->bits_per_coded_sample) {
155  av_log(s, AV_LOG_WARNING, "Invalid bits_per_coded_sample: %d\n",
157  return AVERROR_INVALIDDATA;
158  }
159 
160  /* set the sample rate */
161  avpriv_set_pts_info(st, 64, 1, st->codecpar->sample_rate);
162 
163  /* parse the remaining RIFF chunks */
164  for (;;) {
165  if (pb->eof_reached) {
166  ret = AVERROR_EOF;
167  goto fail;
168  }
169  /* read next chunk tag */
170  tag = avio_rl32(pb);
171  size = avio_rl32(pb);
172  if (tag == MKTAG('d', 'a', 't', 'a')) {
173  /* We assume that the data chunk comes last. */
174  break;
175  } else if (tag == MKTAG('d','p','d','s')) {
176  /* Quoting the MSDN xWMA docs on the dpds chunk: "Contains the
177  * decoded packet cumulative data size array, each element is the
178  * number of bytes accumulated after the corresponding xWMA packet
179  * is decoded in order."
180  *
181  * Each packet has size equal to st->codecpar->block_align, which in
182  * all cases I saw so far was always 2230. Thus, we can use the
183  * dpds data to compute a seeking index.
184  */
185 
186  /* Error out if there is more than one dpds chunk. */
187  if (dpds_table) {
188  av_log(s, AV_LOG_ERROR, "two dpds chunks present\n");
189  ret = AVERROR_INVALIDDATA;
190  goto fail;
191  }
192 
193  /* Compute the number of entries in the dpds chunk. */
194  if (size & 3) { /* Size should be divisible by four */
196  "dpds chunk size %"PRId64" not divisible by 4\n", size);
197  }
198  dpds_table_size = size / 4;
199  if (dpds_table_size == 0 || dpds_table_size >= INT_MAX / 4) {
200  av_log(s, AV_LOG_ERROR,
201  "dpds chunk size %"PRId64" invalid\n", size);
202  return AVERROR_INVALIDDATA;
203  }
204 
205  /* Allocate some temporary storage to keep the dpds data around.
206  * for processing later on.
207  */
208  dpds_table = av_malloc_array(dpds_table_size, sizeof(uint32_t));
209  if (!dpds_table) {
210  return AVERROR(ENOMEM);
211  }
212 
213  for (i = 0; i < dpds_table_size; ++i) {
214  if (avio_feof(pb)) {
215  ret = AVERROR_INVALIDDATA;
216  goto fail;
217  }
218  dpds_table[i] = avio_rl32(pb);
219  size -= 4;
220  }
221  }
222  avio_skip(pb, size);
223  }
224 
225  /* Determine overall data length */
226  if (size < 0) {
227  ret = AVERROR_INVALIDDATA;
228  goto fail;
229  }
230  if (!size) {
231  xwma->data_end = INT64_MAX;
232  } else
233  xwma->data_end = avio_tell(pb) + size;
234 
235 
236  if (dpds_table && dpds_table_size) {
237  int64_t cur_pos;
238  const uint32_t bytes_per_sample
239  = (st->codecpar->channels * st->codecpar->bits_per_coded_sample) >> 3;
240 
241  /* Estimate the duration from the total number of output bytes. */
242  const uint64_t total_decoded_bytes = dpds_table[dpds_table_size - 1];
243 
244  if (!bytes_per_sample) {
245  av_log(s, AV_LOG_ERROR,
246  "Invalid bits_per_coded_sample %d for %d channels\n",
248  ret = AVERROR_INVALIDDATA;
249  goto fail;
250  }
251 
252  st->duration = total_decoded_bytes / bytes_per_sample;
253 
254  /* Use the dpds data to build a seek table. We can only do this after
255  * we know the offset to the data chunk, as we need that to determine
256  * the actual offset to each input block.
257  * Note: If we allowed ourselves to assume that the data chunk always
258  * follows immediately after the dpds block, we could of course guess
259  * the data block's start offset already while reading the dpds chunk.
260  * I decided against that, just in case other chunks ever are
261  * discovered.
262  */
263  cur_pos = avio_tell(pb);
264  for (i = 0; i < dpds_table_size; ++i) {
265  /* From the number of output bytes that would accumulate in the
266  * output buffer after decoding the first (i+1) packets, we compute
267  * an offset / timestamp pair.
268  */
270  cur_pos + (i+1) * st->codecpar->block_align, /* pos */
271  dpds_table[i] / bytes_per_sample, /* timestamp */
272  st->codecpar->block_align, /* size */
273  0, /* duration */
275  }
276  } else if (st->codecpar->bit_rate) {
277  /* No dpds chunk was present (or only an empty one), so estimate
278  * the total duration using the average bits per sample and the
279  * total data length.
280  */
281  st->duration = (size<<3) * st->codecpar->sample_rate / st->codecpar->bit_rate;
282  }
283 
284 fail:
285  av_free(dpds_table);
286 
287  return ret;
288 }
289 
291 {
292  int ret, size;
293  int64_t left;
294  AVStream *st;
295  XWMAContext *xwma = s->priv_data;
296 
297  st = s->streams[0];
298 
299  left = xwma->data_end - avio_tell(s->pb);
300  if (left <= 0) {
301  return AVERROR_EOF;
302  }
303 
304  /* read a single block; the default block size is 2230. */
305  size = (st->codecpar->block_align > 1) ? st->codecpar->block_align : 2230;
306  size = FFMIN(size, left);
307 
308  ret = av_get_packet(s->pb, pkt, size);
309  if (ret < 0)
310  return ret;
311 
312  pkt->stream_index = 0;
313  return ret;
314 }
315 
317  .name = "xwma",
318  .long_name = NULL_IF_CONFIG_SMALL("Microsoft xWMA"),
319  .priv_data_size = sizeof(XWMAContext),
323 };
#define NULL
Definition: coverity.c:32
Bytestream IO Context.
Definition: avio.h:161
#define AVERROR_INVALIDDATA
Invalid data found when processing input.
Definition: error.h:59
int av_add_index_entry(AVStream *st, int64_t pos, int64_t timestamp, int size, int distance, int flags)
Add an index entry into a sorted list.
Definition: utils.c:1989
#define AV_LOG_WARNING
Something somehow does not look correct.
Definition: log.h:200
void avpriv_set_pts_info(AVStream *s, int pts_wrap_bits, unsigned int pts_num, unsigned int pts_den)
Set the time base and wrapping info for a given stream.
Definition: utils.c:4899
#define avpriv_request_sample(...)
enum AVCodecID codec_id
Specific type of the encoded data (the codec used).
Definition: codec_par.h:60
static int xwma_read_packet(AVFormatContext *s, AVPacket *pkt)
Definition: xwma.c:290
int64_t avio_skip(AVIOContext *s, int64_t offset)
Skip given number of bytes forward.
Definition: aviobuf.c:341
static AVPacket pkt
Format I/O context.
Definition: avformat.h:1239
int64_t data_end
Definition: xwma.c:34
enum AVStreamParseType need_parsing
Definition: avformat.h:1088
AVStream * avformat_new_stream(AVFormatContext *s, const AVCodec *c)
Add a new stream to a media file.
Definition: utils.c:4480
AVStream ** streams
A list of all streams in the file.
Definition: avformat.h:1307
uint32_t tag
Definition: movenc.c:1597
#define AVERROR_EOF
End of file.
Definition: error.h:55
int av_get_packet(AVIOContext *s, AVPacket *pkt, int size)
Allocate and read the payload of a packet and initialize its fields with default values.
Definition: utils.c:310
ptrdiff_t size
Definition: opengl_enc.c:100
static av_always_inline int64_t avio_tell(AVIOContext *s)
ftell() equivalent for AVIOContext.
Definition: avio.h:557
#define av_log(a,...)
int ff_get_wav_header(AVFormatContext *s, AVIOContext *pb, AVCodecParameters *par, int size, int big_endian)
Definition: riffdec.c:91
int64_t bit_rate
The average bitrate of the encoded data (in bits per second).
Definition: codec_par.h:89
static int xwma_read_header(AVFormatContext *s)
Definition: xwma.c:44
#define AVINDEX_KEYFRAME
Definition: avformat.h:816
#define AV_LOG_ERROR
Something went wrong and cannot losslessly be recovered.
Definition: log.h:194
AVInputFormat ff_xwma_demuxer
Definition: xwma.c:316
unsigned int avio_rl32(AVIOContext *s)
Definition: aviobuf.c:754
#define NULL_IF_CONFIG_SMALL(x)
Return NULL if CONFIG_SMALL is true, otherwise the argument without modification. ...
Definition: internal.h:115
#define av_fourcc2str(fourcc)
Definition: avutil.h:348
#define fail()
Definition: checkasm.h:123
int extradata_size
Size of the extradata content in bytes.
Definition: codec_par.h:78
unsigned char * buf
Buffer must have AVPROBE_PADDING_SIZE of extra allocated bytes filled with zero.
Definition: avformat.h:443
int block_align
Audio only.
Definition: codec_par.h:177
int ff_alloc_extradata(AVCodecParameters *par, int size)
Allocate extradata with additional AV_INPUT_BUFFER_PADDING_SIZE at end which is always set to 0...
Definition: utils.c:3285
#define FFMIN(a, b)
Definition: common.h:96
internal header for RIFF based (de)muxers do NOT include this in end user applications ...
#define s(width, name)
Definition: cbs_vp9.c:257
static int read_header(FFV1Context *f)
Definition: ffv1dec.c:527
Stream structure.
Definition: avformat.h:880
static int read_packet(void *opaque, uint8_t *buf, int buf_size)
Definition: avio_reading.c:42
AVIOContext * pb
I/O context.
Definition: avformat.h:1281
Tag MUST be and< 10hcoeff half pel interpolation filter coefficients, hcoeff[0] are the 2 middle coefficients[1] are the next outer ones and so on, resulting in a filter like:...eff[2], hcoeff[1], hcoeff[0], hcoeff[0], hcoeff[1], hcoeff[2]...the sign of the coefficients is not explicitly stored but alternates after each coeff and coeff[0] is positive, so...,+,-,+,-,+,+,-,+,-,+,...hcoeff[0] is not explicitly stored but found by subtracting the sum of all stored coefficients with signs from 32 hcoeff[0]=32-hcoeff[1]-hcoeff[2]-...a good choice for hcoeff and htaps is htaps=6 hcoeff={40,-10, 2}an alternative which requires more computations at both encoder and decoder side and may or may not be better is htaps=8 hcoeff={42,-14, 6,-2}ref_frames minimum of the number of available reference frames and max_ref_frames for example the first frame after a key frame always has ref_frames=1spatial_decomposition_type wavelet type 0 is a 9/7 symmetric compact integer wavelet 1 is a 5/3 symmetric compact integer wavelet others are reserved stored as delta from last, last is reset to 0 if always_reset||keyframeqlog quality(logarithmic quantizer scale) stored as delta from last, last is reset to 0 if always_reset||keyframemv_scale stored as delta from last, last is reset to 0 if always_reset||keyframe FIXME check that everything works fine if this changes between framesqbias dequantization bias stored as delta from last, last is reset to 0 if always_reset||keyframeblock_max_depth maximum depth of the block tree stored as delta from last, last is reset to 0 if always_reset||keyframequant_table quantization tableHighlevel bitstream structure:==============================--------------------------------------------|Header|--------------------------------------------|------------------------------------|||Block0||||split?||||yes no||||.........intra?||||:Block01:yes no||||:Block02:.................||||:Block03::y DC::ref index:||||:Block04::cb DC::motion x:||||.........:cr DC::motion y:||||.................|||------------------------------------||------------------------------------|||Block1|||...|--------------------------------------------|------------------------------------|||Y subbands||Cb subbands||Cr subbands||||------||------||------|||||LL0||HL0||||LL0||HL0||||LL0||HL0|||||------||------||------||||------||------||------|||||LH0||HH0||||LH0||HH0||||LH0||HH0|||||------||------||------||||------||------||------|||||HL1||LH1||||HL1||LH1||||HL1||LH1|||||------||------||------||||------||------||------|||||HH1||HL2||||HH1||HL2||||HH1||HL2|||||...||...||...|||------------------------------------|--------------------------------------------Decoding process:=================------------|||Subbands|------------||||------------|Intra DC||||LL0 subband prediction------------|\Dequantization-------------------\||Reference frames|\IDWT|--------------|Motion\|||Frame 0||Frame 1||Compensation.OBMC v-------|--------------|--------------.\------> Frame n output Frame Frame<----------------------------------/|...|-------------------Range Coder:============Binary Range Coder:-------------------The implemented range coder is an adapted version based upon"Range encoding: an algorithm for removing redundancy from a digitised message."by G.N.N.Martin.The symbols encoded by the Snow range coder are bits(0|1).The associated probabilities are not fix but change depending on the symbol mix seen so far.bit seen|new state---------+-----------------------------------------------0|256-state_transition_table[256-old_state];1|state_transition_table[old_state];state_transition_table={0, 0, 0, 0, 0, 0, 0, 0, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 190, 191, 192, 194, 194, 195, 196, 197, 198, 199, 200, 201, 202, 202, 204, 205, 206, 207, 208, 209, 209, 210, 211, 212, 213, 215, 215, 216, 217, 218, 219, 220, 220, 222, 223, 224, 225, 226, 227, 227, 229, 229, 230, 231, 232, 234, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 248, 0, 0, 0, 0, 0, 0, 0};FIXME Range Coding of integers:-------------------------FIXME Neighboring Blocks:===================left and top are set to the respective blocks unless they are outside of the image in which case they are set to the Null block top-left is set to the top left block unless it is outside of the image in which case it is set to the left block if this block has no larger parent block or it is at the left side of its parent block and the top right block is not outside of the image then the top right block is used for top-right else the top-left block is used Null block y, cb, cr are 128 level, ref, mx and my are 0 Motion Vector Prediction:=========================1.the motion vectors of all the neighboring blocks are scaled to compensate for the difference of reference frames scaled_mv=(mv *(256 *(current_reference+1)/(mv.reference+1))+128)> the median of the scaled left
Definition: snow.txt:206
This structure contains the data a format has to probe a file.
Definition: avformat.h:441
static int read_probe(const AVProbeData *pd)
Definition: jvdec.c:55
int64_t duration
Decoding: duration of the stream, in stream time base.
Definition: avformat.h:929
int sample_rate
Audio only.
Definition: codec_par.h:170
#define AVPROBE_SCORE_MAX
maximum score
Definition: avformat.h:453
Main libavformat public API header.
#define av_free(p)
int eof_reached
true if was unable to read due to error or eof
Definition: avio.h:239
void * priv_data
Format private data.
Definition: avformat.h:1267
static int xwma_probe(const AVProbeData *p)
Definition: xwma.c:37
int bits_per_coded_sample
The number of bits per sample in the codedwords.
Definition: codec_par.h:102
uint8_t * extradata
Extra binary data needed for initializing the decoder, codec-dependent.
Definition: codec_par.h:74
int channels
Audio only.
Definition: codec_par.h:166
const char * name
A comma separated list of short names for the format.
Definition: avformat.h:652
AVCodecParameters * codecpar
Codec parameters associated with this stream.
Definition: avformat.h:1045
#define av_malloc_array(a, b)
int avio_feof(AVIOContext *s)
Similar to feof() but also returns nonzero on read errors.
Definition: aviobuf.c:368
uint32_t codec_tag
Additional information about the codec (corresponds to the AVI FOURCC).
Definition: codec_par.h:64
int stream_index
Definition: packet.h:365
Filter the word “frame” indicates either a video frame or a group of audio as stored in an AVFrame structure Format for each input and each output the list of supported formats For video that means pixel format For audio that means channel sample they are references to shared objects When the negotiation mechanism computes the intersection of the formats supported at each end of a all references to both lists are replaced with a reference to the intersection And when a single format is eventually chosen for a link amongst the remaining all references to the list are updated That means that if a filter requires that its input and output have the same format amongst a supported all it has to do is use a reference to the same list of formats query_formats can leave some formats unset and return AVERROR(EAGAIN) to cause the negotiation mechanism toagain later.That can be used by filters with complex requirements to use the format negotiated on one link to set the formats supported on another.Frame references ownership and permissions
#define MKTAG(a, b, c, d)
Definition: common.h:405
This structure stores compressed data.
Definition: packet.h:340
int i
Definition: input.c:407