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
30
31 /*
32 * Demuxer for xWMA, a Microsoft audio container used by XAudio 2.
33 */
34
38
40 {
41 if (!memcmp(
p->buf,
"RIFF", 4) && !memcmp(
p->buf + 8,
"XWMA", 4))
43 return 0;
44 }
45
47 {
50 uint32_t dpds_table_size = 0;
51 uint32_t *dpds_table =
NULL;
57
58 /* The following code is mostly copied from wav.c, with some
59 * minor alterations.
60 */
61
62 /* check RIFF header */
64 if (
tag !=
MKTAG(
'R',
'I',
'F',
'F'))
68 if (
tag !=
MKTAG(
'X',
'W',
'M',
'A'))
70
71 /* parse fmt header */
73 if (
tag !=
MKTAG(
'f',
'm',
't',
' '))
77 if (!st)
79
84
85 /* XWMA encoder only allows a few channel/sample rate/bitrate combinations,
86 * but some create identical files with fake bitrate (1ch 22050hz at
87 * 20/48/192kbps are all 20kbps, with the exact same codec data).
88 * Decoder needs correct bitrate to work, so it's normalized here. */
93
94 if (ch == 1) {
95 if (sr == 22050 && (br==48000 || br==192000))
96 br = 20000;
97 else if (sr == 32000 && (br==48000 || br==192000))
98 br = 20000;
99 else if (sr == 44100 && (br==96000 || br==192000))
100 br = 48000;
101 }
102 else if (ch == 2) {
103 if (sr == 22050 && (br==48000 || br==192000))
104 br = 32000;
105 else if (sr == 32000 && (br==192000))
106 br = 48000;
107 }
108
110 }
111
112 /* Normally xWMA can only contain WMAv2 with 1/2 channels,
113 * and WMAPRO with 6 channels. */
119 } else {
120 /* xWMA shouldn't have extradata. But the WMA codecs require it,
121 * so we provide our own fake extradata.
122 *
123 * First, check that there really was no extradata in the header. If
124 * there was, then try to use it, after asking the user to provide a
125 * sample of this unusual file.
126 */
128 /* Surprise, surprise: We *did* get some extradata. No idea
129 * if it will work, but just go on and try it, after asking
130 * the user for a sample.
131 */
137
141 } else {
144
146 /* setup extradata with our experimentally obtained value */
148 }
149 }
150
155 }
160 }
161
162 /* set the sample rate */
164
165 /* parse the remaining RIFF chunks */
166 for (;;) {
170 }
171 /* read next chunk tag */
174 if (
tag ==
MKTAG(
'd',
'a',
't',
'a')) {
175 /* We assume that the data chunk comes last. */
176 break;
177 }
else if (
tag ==
MKTAG(
'd',
'p',
'd',
's')) {
178 /* Quoting the MSDN xWMA docs on the dpds chunk: "Contains the
179 * decoded packet cumulative data size array, each element is the
180 * number of bytes accumulated after the corresponding xWMA packet
181 * is decoded in order."
182 *
183 * Each packet has size equal to st->codecpar->block_align, which in
184 * all cases I saw so far was always 2230. Thus, we can use the
185 * dpds data to compute a seeking index.
186 */
187
188 /* Error out if there is more than one dpds chunk. */
189 if (dpds_table) {
193 }
194
195 /* Compute the number of entries in the dpds chunk. */
196 if (
size & 3) {
/* Size should be divisible by four */
198 "dpds chunk size %"PRId64
" not divisible by 4\n",
size);
199 }
200 dpds_table_size =
size / 4;
201 if (dpds_table_size == 0 || dpds_table_size >= INT_MAX / 4) {
203 "dpds chunk size %"PRId64
" invalid\n",
size);
205 }
206
207 /* Allocate some temporary storage to keep the dpds data around.
208 * for processing later on.
209 */
211 if (!dpds_table) {
213 }
214
215 for (
i = 0;
i < dpds_table_size; ++
i) {
219 }
222 }
223 }
225 }
226
227 /* Determine overall data length */
231 }
234 } else
236
237
238 if (dpds_table && dpds_table_size) {
240 const uint32_t bytes_per_sample
242
243 /* Estimate the duration from the total number of output bytes. */
244 const uint64_t total_decoded_bytes = dpds_table[dpds_table_size - 1];
245
246 if (!bytes_per_sample) {
248 "Invalid bits_per_coded_sample %d for %d channels\n",
252 }
253
254 st->
duration = total_decoded_bytes / bytes_per_sample;
255
256 /* Use the dpds data to build a seek table. We can only do this after
257 * we know the offset to the data chunk, as we need that to determine
258 * the actual offset to each input block.
259 * Note: If we allowed ourselves to assume that the data chunk always
260 * follows immediately after the dpds block, we could of course guess
261 * the data block's start offset already while reading the dpds chunk.
262 * I decided against that, just in case other chunks ever are
263 * discovered.
264 */
266 for (
i = 0;
i < dpds_table_size; ++
i) {
267 /* From the number of output bytes that would accumulate in the
268 * output buffer after decoding the first (i+1) packets, we compute
269 * an offset / timestamp pair.
270 */
273 dpds_table[
i] / bytes_per_sample,
/* timestamp */
275 0, /* duration */
277 }
279 /* No dpds chunk was present (or only an empty one), so estimate
280 * the total duration using the average bits per sample and the
281 * total data length.
282 */
284 }
285
288
290 }
291
293 {
298
300
304 }
305
306 /* read a single block; the default block size is 2230. */
309
313
316 }
317
325 };
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