FFmpeg
idcinvideo.c
Go to the documentation of this file.
1 /*
2  * id Quake II CIN Video Decoder
3  * Copyright (C) 2003 The FFmpeg project
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 /**
23  * @file
24  * id Quake II Cin Video Decoder by Dr. Tim Ferguson
25  * For more information about the id CIN format, visit:
26  * http://www.csse.monash.edu.au/~timf/
27  *
28  * This video decoder outputs PAL8 colorspace data. Interacting with this
29  * decoder is a little involved. During initialization, the demuxer must
30  * transmit the 65536-byte Huffman table(s) to the decoder via extradata.
31  * Then, whenever a palette change is encountered while demuxing the file,
32  * the demuxer must use the same extradata space to transmit an
33  * AVPaletteControl structure.
34  *
35  * id CIN video is purely Huffman-coded, intraframe-only codec. It achieves
36  * a little more compression by exploiting the fact that adjacent pixels
37  * tend to be similar.
38  *
39  * Note that this decoder could use libavcodec's optimized VLC facilities
40  * rather than naive, tree-based Huffman decoding. However, there are 256
41  * Huffman tables. Plus, the VLC bit coding order is right -> left instead
42  * or left -> right, so all of the bits would have to be reversed. Further,
43  * the original Quake II implementation likely used a similar naive
44  * decoding algorithm and it worked fine on much lower spec machines.
45  */
46 
47 #include <stdio.h>
48 #include <stdlib.h>
49 #include <string.h>
50 
51 #include "avcodec.h"
52 #include "codec_internal.h"
53 #include "decode.h"
54 #include "internal.h"
55 #include "libavutil/internal.h"
56 
57 #define HUFFMAN_TABLE_SIZE 64 * 1024
58 #define HUF_TOKENS 256
59 #define PALETTE_COUNT 256
60 
61 typedef struct hnode {
62  int count;
63  unsigned char used;
64  int children[2];
65 } hnode;
66 
67 typedef struct IdcinContext {
68 
70 
71  const unsigned char *buf;
72  int size;
73 
75  int num_huff_nodes[256];
76 
77  uint32_t pal[256];
78 } IdcinContext;
79 
80 /**
81  * Find the lowest probability node in a Huffman table, and mark it as
82  * being assigned to a higher probability.
83  * @return the node index of the lowest unused node, or -1 if all nodes
84  * are used.
85  */
86 static int huff_smallest_node(hnode *hnodes, int num_hnodes) {
87  int i;
88  int best, best_node;
89 
90  best = 99999999;
91  best_node = -1;
92  for(i = 0; i < num_hnodes; i++) {
93  if(hnodes[i].used)
94  continue;
95  if(!hnodes[i].count)
96  continue;
97  if(hnodes[i].count < best) {
98  best = hnodes[i].count;
99  best_node = i;
100  }
101  }
102 
103  if(best_node == -1)
104  return -1;
105  hnodes[best_node].used = 1;
106  return best_node;
107 }
108 
109 /*
110  * Build the Huffman tree using the generated/loaded probabilities histogram.
111  *
112  * On completion:
113  * huff_nodes[prev][i < HUF_TOKENS] - are the nodes at the base of the tree.
114  * huff_nodes[prev][i >= HUF_TOKENS] - are used to construct the tree.
115  * num_huff_nodes[prev] - contains the index to the root node of the tree.
116  * That is: huff_nodes[prev][num_huff_nodes[prev]] is the root node.
117  */
118 static av_cold void huff_build_tree(IdcinContext *s, int prev) {
119  hnode *node, *hnodes;
120  int num_hnodes, i;
121 
122  num_hnodes = HUF_TOKENS;
123  hnodes = s->huff_nodes[prev];
124  for(i = 0; i < HUF_TOKENS * 2; i++)
125  hnodes[i].used = 0;
126 
127  while (1) {
128  node = &hnodes[num_hnodes]; /* next free node */
129 
130  /* pick two lowest counts */
131  node->children[0] = huff_smallest_node(hnodes, num_hnodes);
132  if(node->children[0] == -1)
133  break; /* reached the root node */
134 
135  node->children[1] = huff_smallest_node(hnodes, num_hnodes);
136  if(node->children[1] == -1)
137  break; /* reached the root node */
138 
139  /* combine nodes probability for new node */
140  node->count = hnodes[node->children[0]].count +
141  hnodes[node->children[1]].count;
142  num_hnodes++;
143  }
144 
145  s->num_huff_nodes[prev] = num_hnodes - 1;
146 }
147 
149 {
150  IdcinContext *s = avctx->priv_data;
151  int i, j, histogram_index = 0;
152  unsigned char *histograms;
153 
154  s->avctx = avctx;
155  avctx->pix_fmt = AV_PIX_FMT_PAL8;
156 
157  /* make sure the Huffman tables make it */
158  if (s->avctx->extradata_size != HUFFMAN_TABLE_SIZE) {
159  av_log(s->avctx, AV_LOG_ERROR, " id CIN video: expected extradata size of %d\n", HUFFMAN_TABLE_SIZE);
160  return -1;
161  }
162 
163  /* build the 256 Huffman decode trees */
164  histograms = (unsigned char *)s->avctx->extradata;
165  for (i = 0; i < 256; i++) {
166  for(j = 0; j < HUF_TOKENS; j++)
167  s->huff_nodes[i][j].count = histograms[histogram_index++];
168  huff_build_tree(s, i);
169  }
170 
171  return 0;
172 }
173 
175 {
176  hnode *hnodes;
177  long x, y;
178  int prev;
179  unsigned char v = 0;
180  int bit_pos, node_num, dat_pos;
181 
182  prev = bit_pos = dat_pos = 0;
183  for (y = 0; y < (frame->linesize[0] * s->avctx->height);
184  y += frame->linesize[0]) {
185  for (x = y; x < y + s->avctx->width; x++) {
186  node_num = s->num_huff_nodes[prev];
187  hnodes = s->huff_nodes[prev];
188 
189  while(node_num >= HUF_TOKENS) {
190  if(!bit_pos) {
191  if(dat_pos >= s->size) {
192  av_log(s->avctx, AV_LOG_ERROR, "Huffman decode error.\n");
193  return -1;
194  }
195  bit_pos = 8;
196  v = s->buf[dat_pos++];
197  }
198 
199  node_num = hnodes[node_num].children[v & 0x01];
200  v = v >> 1;
201  bit_pos--;
202  }
203 
204  frame->data[0][x] = node_num;
205  prev = node_num;
206  }
207  }
208 
209  return 0;
210 }
211 
213  int *got_frame, AVPacket *avpkt)
214 {
215  const uint8_t *buf = avpkt->data;
216  int buf_size = avpkt->size;
217  IdcinContext *s = avctx->priv_data;
218  int ret;
219 
220  s->buf = buf;
221  s->size = buf_size;
222 
223  if ((ret = ff_get_buffer(avctx, frame, 0)) < 0)
224  return ret;
225 
226  if (idcin_decode_vlcs(s, frame))
227  return AVERROR_INVALIDDATA;
228 
229  frame->palette_has_changed = ff_copy_palette(s->pal, avpkt, avctx);
230  /* make the palette available on the way out */
231  memcpy(frame->data[1], s->pal, AVPALETTE_SIZE);
232 
233  *got_frame = 1;
234 
235  /* report that the buffer was completely consumed */
236  return buf_size;
237 }
238 
239 static const FFCodecDefault idcin_defaults[] = {
240  { "max_pixels", "320*240" },
241  { NULL },
242 };
243 
245  .p.name = "idcinvideo",
246  .p.long_name = NULL_IF_CONFIG_SMALL("id Quake II CIN video"),
247  .p.type = AVMEDIA_TYPE_VIDEO,
248  .p.id = AV_CODEC_ID_IDCIN,
249  .priv_data_size = sizeof(IdcinContext),
252  .p.capabilities = AV_CODEC_CAP_DR1,
253  .defaults = idcin_defaults,
254  .caps_internal = FF_CODEC_CAP_INIT_THREADSAFE,
255 };
HUF_TOKENS
#define HUF_TOKENS
Definition: idcinvideo.c:58
AV_CODEC_ID_IDCIN
@ AV_CODEC_ID_IDCIN
Definition: codec_id.h:97
hnode::used
unsigned char used
Definition: idcinvideo.c:63
IdcinContext::pal
uint32_t pal[256]
Definition: idcinvideo.c:77
AVFrame
This structure describes decoded (raw) audio or video data.
Definition: frame.h:325
internal.h
AVPacket::data
uint8_t * data
Definition: packet.h:374
FFCodec
Definition: codec_internal.h:112
IdcinContext::num_huff_nodes
int num_huff_nodes[256]
Definition: idcinvideo.c:75
init
static int init
Definition: av_tx.c:47
FFCodecDefault
Definition: codec_internal.h:82
FFCodec::p
AVCodec p
The public AVCodec.
Definition: codec_internal.h:116
IdcinContext::buf
const unsigned char * buf
Definition: idcinvideo.c:71
IdcinContext::huff_nodes
hnode huff_nodes[256][HUF_TOKENS *2]
Definition: idcinvideo.c:74
AV_LOG_ERROR
#define AV_LOG_ERROR
Something went wrong and cannot losslessly be recovered.
Definition: log.h:180
av_cold
#define av_cold
Definition: attributes.h:90
huff_build_tree
static av_cold void huff_build_tree(IdcinContext *s, int prev)
Definition: idcinvideo.c:118
FF_CODEC_DECODE_CB
#define FF_CODEC_DECODE_CB(func)
Definition: codec_internal.h:254
s
#define s(width, name)
Definition: cbs_vp9.c:256
ff_idcin_decoder
const FFCodec ff_idcin_decoder
Definition: idcinvideo.c:244
decode.h
IdcinContext
Definition: idcinvideo.c:67
NULL
#define NULL
Definition: coverity.c:32
huff_smallest_node
static int huff_smallest_node(hnode *hnodes, int num_hnodes)
Find the lowest probability node in a Huffman table, and mark it as being assigned to a higher probab...
Definition: idcinvideo.c:86
AVPALETTE_SIZE
#define AVPALETTE_SIZE
Definition: pixfmt.h:32
IdcinContext::size
int size
Definition: idcinvideo.c:72
HUFFMAN_TABLE_SIZE
#define HUFFMAN_TABLE_SIZE
Definition: idcinvideo.c:57
ff_get_buffer
int ff_get_buffer(AVCodecContext *avctx, AVFrame *frame, int flags)
Get a buffer for a frame.
Definition: decode.c:1403
AV_CODEC_CAP_DR1
#define AV_CODEC_CAP_DR1
Codec uses get_buffer() or get_encode_buffer() for allocating buffers and supports custom allocators.
Definition: codec.h:52
AVPacket::size
int size
Definition: packet.h:375
NULL_IF_CONFIG_SMALL
#define NULL_IF_CONFIG_SMALL(x)
Return NULL if CONFIG_SMALL is true, otherwise the argument without modification.
Definition: internal.h:116
codec_internal.h
for
for(k=2;k<=8;++k)
Definition: h264pred_template.c:425
hnode::children
int children[2]
Definition: idcinvideo.c:64
idcin_decode_frame
static int idcin_decode_frame(AVCodecContext *avctx, AVFrame *frame, int *got_frame, AVPacket *avpkt)
Definition: idcinvideo.c:212
i
#define i(width, name, range_min, range_max)
Definition: cbs_h2645.c:271
hnode::count
int count
Definition: idcinvideo.c:62
internal.h
idcin_defaults
static const FFCodecDefault idcin_defaults[]
Definition: idcinvideo.c:239
FF_CODEC_CAP_INIT_THREADSAFE
#define FF_CODEC_CAP_INIT_THREADSAFE
The codec does not modify any global variables in the init function, allowing to call the init functi...
Definition: codec_internal.h:31
AVCodec::name
const char * name
Name of the codec implementation.
Definition: codec.h:203
AVCodecContext::pix_fmt
enum AVPixelFormat pix_fmt
Pixel format, see AV_PIX_FMT_xxx.
Definition: avcodec.h:599
avcodec.h
AV_PIX_FMT_PAL8
@ AV_PIX_FMT_PAL8
8 bits with AV_PIX_FMT_RGB32 palette
Definition: pixfmt.h:77
ret
ret
Definition: filter_design.txt:187
frame
these buffered frames must be flushed immediately if a new input produces new the filter must not call request_frame to get more It must just process the frame or queue it The task of requesting more frames is left to the filter s request_frame method or the application If a filter has several the filter must be ready for frames arriving randomly on any input any filter with several inputs will most likely require some kind of queuing mechanism It is perfectly acceptable to have a limited queue and to drop frames when the inputs are too unbalanced request_frame For filters that do not use the this method is called when a frame is wanted on an output For a it should directly call filter_frame on the corresponding output For a if there are queued frames already one of these frames should be pushed If the filter should request a frame on one of its repeatedly until at least one frame has been pushed Return or at least make progress towards producing a frame
Definition: filter_design.txt:264
AVCodecContext
main external API structure.
Definition: avcodec.h:389
idcin_decode_init
static av_cold int idcin_decode_init(AVCodecContext *avctx)
Definition: idcinvideo.c:148
IdcinContext::avctx
AVCodecContext * avctx
Definition: idcinvideo.c:69
AVMEDIA_TYPE_VIDEO
@ AVMEDIA_TYPE_VIDEO
Definition: avutil.h:201
hnode
Definition: idcinvideo.c:61
AVPacket
This structure stores compressed data.
Definition: packet.h:351
AVCodecContext::priv_data
void * priv_data
Definition: avcodec.h:416
idcin_decode_vlcs
static int idcin_decode_vlcs(IdcinContext *s, AVFrame *frame)
Definition: idcinvideo.c:174
av_log
#define av_log(a,...)
Definition: tableprint_vlc.h:27
AVERROR_INVALIDDATA
#define AVERROR_INVALIDDATA
Invalid data found when processing input.
Definition: error.h:61
ff_copy_palette
int ff_copy_palette(void *dst, const AVPacket *src, void *logctx)
Check whether the side-data of src contains a palette of size AVPALETTE_SIZE; if so,...
Definition: decode.c:1624