Line | Branch | Exec | Source |
---|---|---|---|
1 | /* | ||
2 | * Copyright (c) 2012 Clément Bœsch | ||
3 | * | ||
4 | * This file is part of FFmpeg. | ||
5 | * | ||
6 | * FFmpeg is free software; you can redistribute it and/or | ||
7 | * modify it under the terms of the GNU Lesser General Public | ||
8 | * License as published by the Free Software Foundation; either | ||
9 | * version 2.1 of the License, or (at your option) any later version. | ||
10 | * | ||
11 | * FFmpeg is distributed in the hope that it will be useful, | ||
12 | * but WITHOUT ANY WARRANTY; without even the implied warranty of | ||
13 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU | ||
14 | * Lesser General Public License for more details. | ||
15 | * | ||
16 | * You should have received a copy of the GNU Lesser General Public | ||
17 | * License along with FFmpeg; if not, write to the Free Software | ||
18 | * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA | ||
19 | */ | ||
20 | |||
21 | /** | ||
22 | * @file | ||
23 | * SAMI subtitle decoder | ||
24 | * @see http://msdn.microsoft.com/en-us/library/ms971327.aspx | ||
25 | */ | ||
26 | |||
27 | #include "ass.h" | ||
28 | #include "libavutil/avstring.h" | ||
29 | #include "libavutil/bprint.h" | ||
30 | #include "libavutil/mem.h" | ||
31 | #include "codec_internal.h" | ||
32 | #include "htmlsubtitles.h" | ||
33 | |||
34 | typedef struct { | ||
35 | AVBPrint source; | ||
36 | AVBPrint content; | ||
37 | AVBPrint encoded_source; | ||
38 | AVBPrint encoded_content; | ||
39 | AVBPrint full; | ||
40 | int readorder; | ||
41 | } SAMIContext; | ||
42 | |||
43 | 167 | static int sami_paragraph_to_ass(AVCodecContext *avctx, const char *src) | |
44 | { | ||
45 | 167 | SAMIContext *sami = avctx->priv_data; | |
46 | 167 | int ret = 0; | |
47 | 167 | char *tag = NULL; | |
48 | 167 | char *dupsrc = av_strdup(src); | |
49 | 167 | char *p = dupsrc; | |
50 | 167 | AVBPrint *dst_content = &sami->encoded_content; | |
51 | 167 | AVBPrint *dst_source = &sami->encoded_source; | |
52 | |||
53 |
1/2✗ Branch 0 not taken.
✓ Branch 1 taken 167 times.
|
167 | if (!dupsrc) |
54 | ✗ | return AVERROR(ENOMEM); | |
55 | |||
56 | 167 | av_bprint_clear(&sami->encoded_content); | |
57 | 167 | av_bprint_clear(&sami->content); | |
58 | 167 | av_bprint_clear(&sami->encoded_source); | |
59 | 89 | for (;;) { | |
60 | 256 | char *saveptr = NULL; | |
61 | 256 | int prev_chr_is_space = 0; | |
62 | 256 | AVBPrint *dst = &sami->content; | |
63 | |||
64 | /* parse & extract paragraph tag */ | ||
65 | 256 | p = av_stristr(p, "<P"); | |
66 |
2/2✓ Branch 0 taken 88 times.
✓ Branch 1 taken 168 times.
|
256 | if (!p) |
67 | 88 | break; | |
68 |
2/4✓ Branch 0 taken 168 times.
✗ Branch 1 not taken.
✗ Branch 2 not taken.
✓ Branch 3 taken 168 times.
|
168 | if (p[2] != '>' && !av_isspace(p[2])) { // avoid confusion with tags such as <PRE> |
69 | ✗ | p++; | |
70 | ✗ | continue; | |
71 | } | ||
72 |
1/2✗ Branch 0 not taken.
✓ Branch 1 taken 168 times.
|
168 | if (dst->len) // add a separator with the previous paragraph if there was one |
73 | ✗ | av_bprintf(dst, "\\N"); | |
74 | 168 | tag = av_strtok(p, ">", &saveptr); | |
75 |
2/4✓ Branch 0 taken 168 times.
✗ Branch 1 not taken.
✓ Branch 2 taken 168 times.
✗ Branch 3 not taken.
|
168 | if (!tag || !saveptr) |
76 | break; | ||
77 | 168 | p = saveptr; | |
78 | |||
79 | /* check if the current paragraph is the "source" (speaker name) */ | ||
80 |
3/4✓ Branch 1 taken 166 times.
✓ Branch 2 taken 2 times.
✗ Branch 4 not taken.
✓ Branch 5 taken 166 times.
|
168 | if (av_stristr(tag, "ID=Source") || av_stristr(tag, "ID=\"Source\"")) { |
81 | 2 | dst = &sami->source; | |
82 | 2 | av_bprint_clear(dst); | |
83 | } | ||
84 | |||
85 | /* if empty event -> skip subtitle */ | ||
86 |
2/2✓ Branch 0 taken 77 times.
✓ Branch 1 taken 168 times.
|
245 | while (av_isspace(*p)) |
87 | 77 | p++; | |
88 |
2/2✓ Branch 0 taken 79 times.
✓ Branch 1 taken 89 times.
|
168 | if (!strncmp(p, " ", 6)) { |
89 | 79 | ret = -1; | |
90 | 79 | goto end; | |
91 | } | ||
92 | |||
93 | /* extract the text, stripping most of the tags */ | ||
94 |
2/2✓ Branch 0 taken 6893 times.
✓ Branch 1 taken 88 times.
|
6981 | while (*p) { |
95 |
2/2✓ Branch 0 taken 316 times.
✓ Branch 1 taken 6577 times.
|
6893 | if (*p == '<') { |
96 |
5/6✓ Branch 1 taken 2 times.
✓ Branch 2 taken 314 times.
✓ Branch 3 taken 2 times.
✗ Branch 4 not taken.
✓ Branch 5 taken 1 times.
✓ Branch 6 taken 1 times.
|
316 | if (!av_strncasecmp(p, "<P", 2) && (p[2] == '>' || av_isspace(p[2]))) |
97 | break; | ||
98 | } | ||
99 |
2/2✓ Branch 1 taken 137 times.
✓ Branch 2 taken 6755 times.
|
6892 | if (!av_strncasecmp(p, "<BR", 3)) { |
100 | 137 | av_bprintf(dst, "\\N"); | |
101 | 137 | p++; | |
102 |
3/4✓ Branch 0 taken 411 times.
✗ Branch 1 not taken.
✓ Branch 2 taken 274 times.
✓ Branch 3 taken 137 times.
|
411 | while (*p && *p != '>') |
103 | 274 | p++; | |
104 |
1/2✗ Branch 0 not taken.
✓ Branch 1 taken 137 times.
|
137 | if (!*p) |
105 | ✗ | break; | |
106 |
1/2✓ Branch 0 taken 137 times.
✗ Branch 1 not taken.
|
137 | if (*p == '>') |
107 | 137 | p++; | |
108 | 137 | continue; | |
109 | } | ||
110 |
2/2✓ Branch 0 taken 5951 times.
✓ Branch 1 taken 804 times.
|
6755 | if (!av_isspace(*p)) |
111 | 5951 | av_bprint_chars(dst, *p, 1); | |
112 |
2/2✓ Branch 0 taken 683 times.
✓ Branch 1 taken 121 times.
|
804 | else if (!prev_chr_is_space) |
113 | 683 | av_bprint_chars(dst, ' ', 1); | |
114 | 6755 | prev_chr_is_space = av_isspace(*p); | |
115 | 6755 | p++; | |
116 | } | ||
117 | } | ||
118 | |||
119 | 88 | av_bprint_clear(&sami->full); | |
120 |
2/2✓ Branch 0 taken 9 times.
✓ Branch 1 taken 79 times.
|
88 | if (sami->source.len) { |
121 | 9 | ret = ff_htmlmarkup_to_ass(avctx, dst_source, sami->source.str); | |
122 |
1/2✗ Branch 0 not taken.
✓ Branch 1 taken 9 times.
|
9 | if (ret < 0) |
123 | ✗ | goto end; | |
124 | 9 | av_bprintf(&sami->full, "{\\i1}%s{\\i0}\\N", sami->encoded_source.str); | |
125 | } | ||
126 | 88 | ret = ff_htmlmarkup_to_ass(avctx, dst_content, sami->content.str); | |
127 |
1/2✗ Branch 0 not taken.
✓ Branch 1 taken 88 times.
|
88 | if (ret < 0) |
128 | ✗ | goto end; | |
129 | 88 | av_bprintf(&sami->full, "%s", sami->encoded_content.str); | |
130 | |||
131 | 167 | end: | |
132 | 167 | av_free(dupsrc); | |
133 | 167 | return ret; | |
134 | } | ||
135 | |||
136 | 167 | static int sami_decode_frame(AVCodecContext *avctx, AVSubtitle *sub, | |
137 | int *got_sub_ptr, const AVPacket *avpkt) | ||
138 | { | ||
139 | 167 | const char *ptr = avpkt->data; | |
140 | 167 | SAMIContext *sami = avctx->priv_data; | |
141 | |||
142 |
2/4✓ Branch 0 taken 167 times.
✗ Branch 1 not taken.
✓ Branch 2 taken 167 times.
✗ Branch 3 not taken.
|
167 | if (ptr && avpkt->size > 0) { |
143 | 167 | int ret = sami_paragraph_to_ass(avctx, ptr); | |
144 |
2/2✓ Branch 0 taken 79 times.
✓ Branch 1 taken 88 times.
|
167 | if (ret < 0) |
145 | 79 | return ret; | |
146 | // TODO: pass escaped sami->encoded_source.str as source | ||
147 | 88 | ret = ff_ass_add_rect(sub, sami->full.str, sami->readorder++, 0, NULL, NULL); | |
148 |
1/2✗ Branch 0 not taken.
✓ Branch 1 taken 88 times.
|
88 | if (ret < 0) |
149 | ✗ | return ret; | |
150 | } | ||
151 | 88 | *got_sub_ptr = sub->num_rects > 0; | |
152 | 88 | return avpkt->size; | |
153 | } | ||
154 | |||
155 | 4 | static av_cold int sami_init(AVCodecContext *avctx) | |
156 | { | ||
157 | 4 | SAMIContext *sami = avctx->priv_data; | |
158 | 4 | av_bprint_init(&sami->source, 0, 2048); | |
159 | 4 | av_bprint_init(&sami->content, 0, 2048); | |
160 | 4 | av_bprint_init(&sami->encoded_source, 0, 2048); | |
161 | 4 | av_bprint_init(&sami->encoded_content, 0, 2048); | |
162 | 4 | av_bprint_init(&sami->full, 0, 2048); | |
163 | 4 | return ff_ass_subtitle_header_default(avctx); | |
164 | } | ||
165 | |||
166 | 4 | static av_cold int sami_close(AVCodecContext *avctx) | |
167 | { | ||
168 | 4 | SAMIContext *sami = avctx->priv_data; | |
169 | 4 | av_bprint_finalize(&sami->source, NULL); | |
170 | 4 | av_bprint_finalize(&sami->content, NULL); | |
171 | 4 | av_bprint_finalize(&sami->encoded_source, NULL); | |
172 | 4 | av_bprint_finalize(&sami->encoded_content, NULL); | |
173 | 4 | av_bprint_finalize(&sami->full, NULL); | |
174 | 4 | return 0; | |
175 | } | ||
176 | |||
177 | ✗ | static void sami_flush(AVCodecContext *avctx) | |
178 | { | ||
179 | ✗ | SAMIContext *sami = avctx->priv_data; | |
180 | ✗ | if (!(avctx->flags2 & AV_CODEC_FLAG2_RO_FLUSH_NOOP)) | |
181 | ✗ | sami->readorder = 0; | |
182 | ✗ | } | |
183 | |||
184 | const FFCodec ff_sami_decoder = { | ||
185 | .p.name = "sami", | ||
186 | CODEC_LONG_NAME("SAMI subtitle"), | ||
187 | .p.type = AVMEDIA_TYPE_SUBTITLE, | ||
188 | .p.id = AV_CODEC_ID_SAMI, | ||
189 | .priv_data_size = sizeof(SAMIContext), | ||
190 | .init = sami_init, | ||
191 | .close = sami_close, | ||
192 | FF_CODEC_DECODE_SUB_CB(sami_decode_frame), | ||
193 | .flush = sami_flush, | ||
194 | }; | ||
195 |