| 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/attributes.h" | ||
| 29 | #include "libavutil/avstring.h" | ||
| 30 | #include "libavutil/bprint.h" | ||
| 31 | #include "libavutil/mem.h" | ||
| 32 | #include "codec_internal.h" | ||
| 33 | #include "htmlsubtitles.h" | ||
| 34 | |||
| 35 | typedef struct { | ||
| 36 | AVBPrint source; | ||
| 37 | AVBPrint content; | ||
| 38 | AVBPrint encoded_source; | ||
| 39 | AVBPrint encoded_content; | ||
| 40 | AVBPrint full; | ||
| 41 | int readorder; | ||
| 42 | } SAMIContext; | ||
| 43 | |||
| 44 | 167 | static int sami_paragraph_to_ass(AVCodecContext *avctx, const char *src) | |
| 45 | { | ||
| 46 | 167 | SAMIContext *sami = avctx->priv_data; | |
| 47 | 167 | int ret = 0; | |
| 48 | 167 | char *tag = NULL; | |
| 49 | 167 | char *dupsrc = av_strdup(src); | |
| 50 | 167 | char *p = dupsrc; | |
| 51 | 167 | AVBPrint *dst_content = &sami->encoded_content; | |
| 52 | 167 | AVBPrint *dst_source = &sami->encoded_source; | |
| 53 | |||
| 54 |
1/2✗ Branch 0 not taken.
✓ Branch 1 taken 167 times.
|
167 | if (!dupsrc) |
| 55 | ✗ | return AVERROR(ENOMEM); | |
| 56 | |||
| 57 | 167 | av_bprint_clear(&sami->encoded_content); | |
| 58 | 167 | av_bprint_clear(&sami->content); | |
| 59 | 167 | av_bprint_clear(&sami->encoded_source); | |
| 60 | 89 | for (;;) { | |
| 61 | 256 | char *saveptr = NULL; | |
| 62 | 256 | int prev_chr_is_space = 0; | |
| 63 | 256 | AVBPrint *dst = &sami->content; | |
| 64 | |||
| 65 | /* parse & extract paragraph tag */ | ||
| 66 | 256 | p = av_stristr(p, "<P"); | |
| 67 |
2/2✓ Branch 0 taken 88 times.
✓ Branch 1 taken 168 times.
|
256 | if (!p) |
| 68 | 88 | break; | |
| 69 |
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> |
| 70 | ✗ | p++; | |
| 71 | ✗ | continue; | |
| 72 | } | ||
| 73 |
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 |
| 74 | ✗ | av_bprintf(dst, "\\N"); | |
| 75 | 168 | tag = av_strtok(p, ">", &saveptr); | |
| 76 |
2/4✓ Branch 0 taken 168 times.
✗ Branch 1 not taken.
✓ Branch 2 taken 168 times.
✗ Branch 3 not taken.
|
168 | if (!tag || !saveptr) |
| 77 | break; | ||
| 78 | 168 | p = saveptr; | |
| 79 | |||
| 80 | /* check if the current paragraph is the "source" (speaker name) */ | ||
| 81 |
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\"")) { |
| 82 | 2 | dst = &sami->source; | |
| 83 | 2 | av_bprint_clear(dst); | |
| 84 | } | ||
| 85 | |||
| 86 | /* if empty event -> skip subtitle */ | ||
| 87 |
2/2✓ Branch 0 taken 77 times.
✓ Branch 1 taken 168 times.
|
245 | while (av_isspace(*p)) |
| 88 | 77 | p++; | |
| 89 |
2/2✓ Branch 0 taken 79 times.
✓ Branch 1 taken 89 times.
|
168 | if (!strncmp(p, " ", 6)) { |
| 90 | 79 | ret = -1; | |
| 91 | 79 | goto end; | |
| 92 | } | ||
| 93 | |||
| 94 | /* extract the text, stripping most of the tags */ | ||
| 95 |
2/2✓ Branch 0 taken 6893 times.
✓ Branch 1 taken 88 times.
|
6981 | while (*p) { |
| 96 |
2/2✓ Branch 0 taken 316 times.
✓ Branch 1 taken 6577 times.
|
6893 | if (*p == '<') { |
| 97 |
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]))) |
| 98 | break; | ||
| 99 | } | ||
| 100 |
2/2✓ Branch 1 taken 137 times.
✓ Branch 2 taken 6755 times.
|
6892 | if (!av_strncasecmp(p, "<BR", 3)) { |
| 101 | 137 | av_bprintf(dst, "\\N"); | |
| 102 | 137 | p++; | |
| 103 |
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 != '>') |
| 104 | 274 | p++; | |
| 105 |
1/2✗ Branch 0 not taken.
✓ Branch 1 taken 137 times.
|
137 | if (!*p) |
| 106 | ✗ | break; | |
| 107 |
1/2✓ Branch 0 taken 137 times.
✗ Branch 1 not taken.
|
137 | if (*p == '>') |
| 108 | 137 | p++; | |
| 109 | 137 | continue; | |
| 110 | } | ||
| 111 |
2/2✓ Branch 0 taken 5951 times.
✓ Branch 1 taken 804 times.
|
6755 | if (!av_isspace(*p)) |
| 112 | 5951 | av_bprint_chars(dst, *p, 1); | |
| 113 |
2/2✓ Branch 0 taken 683 times.
✓ Branch 1 taken 121 times.
|
804 | else if (!prev_chr_is_space) |
| 114 | 683 | av_bprint_chars(dst, ' ', 1); | |
| 115 | 6755 | prev_chr_is_space = av_isspace(*p); | |
| 116 | 6755 | p++; | |
| 117 | } | ||
| 118 | } | ||
| 119 | |||
| 120 | 88 | av_bprint_clear(&sami->full); | |
| 121 |
2/2✓ Branch 0 taken 9 times.
✓ Branch 1 taken 79 times.
|
88 | if (sami->source.len) { |
| 122 | 9 | ret = ff_htmlmarkup_to_ass(avctx, dst_source, sami->source.str); | |
| 123 |
1/2✗ Branch 0 not taken.
✓ Branch 1 taken 9 times.
|
9 | if (ret < 0) |
| 124 | ✗ | goto end; | |
| 125 | 9 | av_bprintf(&sami->full, "{\\i1}%s{\\i0}\\N", sami->encoded_source.str); | |
| 126 | } | ||
| 127 | 88 | ret = ff_htmlmarkup_to_ass(avctx, dst_content, sami->content.str); | |
| 128 |
1/2✗ Branch 0 not taken.
✓ Branch 1 taken 88 times.
|
88 | if (ret < 0) |
| 129 | ✗ | goto end; | |
| 130 | 88 | av_bprintf(&sami->full, "%s", sami->encoded_content.str); | |
| 131 | |||
| 132 | 167 | end: | |
| 133 | 167 | av_free(dupsrc); | |
| 134 | 167 | return ret; | |
| 135 | } | ||
| 136 | |||
| 137 | 167 | static int sami_decode_frame(AVCodecContext *avctx, AVSubtitle *sub, | |
| 138 | int *got_sub_ptr, const AVPacket *avpkt) | ||
| 139 | { | ||
| 140 | 167 | const char *ptr = avpkt->data; | |
| 141 | 167 | SAMIContext *sami = avctx->priv_data; | |
| 142 | |||
| 143 |
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) { |
| 144 | 167 | int ret = sami_paragraph_to_ass(avctx, ptr); | |
| 145 |
2/2✓ Branch 0 taken 79 times.
✓ Branch 1 taken 88 times.
|
167 | if (ret < 0) |
| 146 | 79 | return ret; | |
| 147 | // TODO: pass escaped sami->encoded_source.str as source | ||
| 148 | 88 | ret = ff_ass_add_rect(sub, sami->full.str, sami->readorder++, 0, NULL, NULL); | |
| 149 |
1/2✗ Branch 0 not taken.
✓ Branch 1 taken 88 times.
|
88 | if (ret < 0) |
| 150 | ✗ | return ret; | |
| 151 | } | ||
| 152 | 88 | *got_sub_ptr = sub->num_rects > 0; | |
| 153 | 88 | return avpkt->size; | |
| 154 | } | ||
| 155 | |||
| 156 | 4 | static av_cold int sami_init(AVCodecContext *avctx) | |
| 157 | { | ||
| 158 | 4 | SAMIContext *sami = avctx->priv_data; | |
| 159 | 4 | av_bprint_init(&sami->source, 0, 2048); | |
| 160 | 4 | av_bprint_init(&sami->content, 0, 2048); | |
| 161 | 4 | av_bprint_init(&sami->encoded_source, 0, 2048); | |
| 162 | 4 | av_bprint_init(&sami->encoded_content, 0, 2048); | |
| 163 | 4 | av_bprint_init(&sami->full, 0, 2048); | |
| 164 | 4 | return ff_ass_subtitle_header_default(avctx); | |
| 165 | } | ||
| 166 | |||
| 167 | 4 | static av_cold int sami_close(AVCodecContext *avctx) | |
| 168 | { | ||
| 169 | 4 | SAMIContext *sami = avctx->priv_data; | |
| 170 | 4 | av_bprint_finalize(&sami->source, NULL); | |
| 171 | 4 | av_bprint_finalize(&sami->content, NULL); | |
| 172 | 4 | av_bprint_finalize(&sami->encoded_source, NULL); | |
| 173 | 4 | av_bprint_finalize(&sami->encoded_content, NULL); | |
| 174 | 4 | av_bprint_finalize(&sami->full, NULL); | |
| 175 | 4 | return 0; | |
| 176 | } | ||
| 177 | |||
| 178 | ✗ | static av_cold void sami_flush(AVCodecContext *avctx) | |
| 179 | { | ||
| 180 | ✗ | SAMIContext *sami = avctx->priv_data; | |
| 181 | ✗ | if (!(avctx->flags2 & AV_CODEC_FLAG2_RO_FLUSH_NOOP)) | |
| 182 | ✗ | sami->readorder = 0; | |
| 183 | ✗ | } | |
| 184 | |||
| 185 | const FFCodec ff_sami_decoder = { | ||
| 186 | .p.name = "sami", | ||
| 187 | CODEC_LONG_NAME("SAMI subtitle"), | ||
| 188 | .p.type = AVMEDIA_TYPE_SUBTITLE, | ||
| 189 | .p.id = AV_CODEC_ID_SAMI, | ||
| 190 | .priv_data_size = sizeof(SAMIContext), | ||
| 191 | .init = sami_init, | ||
| 192 | .close = sami_close, | ||
| 193 | FF_CODEC_DECODE_SUB_CB(sami_decode_frame), | ||
| 194 | .flush = sami_flush, | ||
| 195 | }; | ||
| 196 |