libavcodec/samidec.c

   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 static int sami_paragraph_to_ass(AVCodecContext *avctx, const char *src)
  44 {
  45     SAMIContext *sami = avctx->priv_data;
  46     int ret = 0;
  47     char *tag = NULL;
  48     char *dupsrc = av_strdup(src);
  49     char *p = dupsrc;
  50     AVBPrint *dst_content = &sami->encoded_content;
  51     AVBPrint *dst_source = &sami->encoded_source;
  52
  53     if (!dupsrc)
  54         return AVERROR(ENOMEM);
  55
  56     av_bprint_clear(&sami->encoded_content);
  57     av_bprint_clear(&sami->content);
  58     av_bprint_clear(&sami->encoded_source);
  59     for (;;) {
  60         char *saveptr = NULL;
  61         int prev_chr_is_space = 0;
  62         AVBPrint *dst = &sami->content;
  63
  64         /* parse & extract paragraph tag */
  65         p = av_stristr(p, "<P");
  66         if (!p)
  67             break;
  68         if (p[2] != '>' && !av_isspace(p[2])) { // avoid confusion with tags such as <PRE>
  69             p++;
  70             continue;
  71         }
  72         if (dst->len) // add a separator with the previous paragraph if there was one
  73             av_bprintf(dst, "\\N");
  74         tag = av_strtok(p, ">", &saveptr);
  75         if (!tag || !saveptr)
  76             break;
  77         p = saveptr;
  78
  79         /* check if the current paragraph is the "source" (speaker name) */
  80         if (av_stristr(tag, "ID=Source") || av_stristr(tag, "ID=\"Source\"")) {
  81             dst = &sami->source;
  82             av_bprint_clear(dst);
  83         }
  84
  85         /* if empty event -> skip subtitle */
  86         while (av_isspace(*p))
  87             p++;
  88         if (!strncmp(p, "&nbsp;", 6)) {
  89             ret = -1;
  90             goto end;
  91         }
  92
  93         /* extract the text, stripping most of the tags */
  94         while (*p) {
  95             if (*p == '<') {
  96                 if (!av_strncasecmp(p, "<P", 2) && (p[2] == '>' || av_isspace(p[2])))
  97                     break;
  98             }
  99             if (!av_strncasecmp(p, "<BR", 3)) {
 100                 av_bprintf(dst, "\\N");
 101                 p++;
 102                 while (*p && *p != '>')
 103                     p++;
 104                 if (!*p)
 105                     break;
 106                 if (*p == '>')
 107                     p++;
 108                 continue;
 109             }
 110             if (!av_isspace(*p))
 111                 av_bprint_chars(dst, *p, 1);
 112             else if (!prev_chr_is_space)
 113                 av_bprint_chars(dst, ' ', 1);
 114             prev_chr_is_space = av_isspace(*p);
 115             p++;
 116         }
 117     }
 118
 119     av_bprint_clear(&sami->full);
 120     if (sami->source.len) {
 121         ret = ff_htmlmarkup_to_ass(avctx, dst_source, sami->source.str);
 122         if (ret < 0)
 123             goto end;
 124         av_bprintf(&sami->full, "{\\i1}%s{\\i0}\\N", sami->encoded_source.str);
 125     }
 126     ret = ff_htmlmarkup_to_ass(avctx, dst_content, sami->content.str);
 127     if (ret < 0)
 128         goto end;
 129     av_bprintf(&sami->full, "%s", sami->encoded_content.str);
 130
 131 end:
 132     av_free(dupsrc);
 133     return ret;
 134 }
 135
 136 static int sami_decode_frame(AVCodecContext *avctx, AVSubtitle *sub,
 137                              int *got_sub_ptr, const AVPacket *avpkt)
 138 {
 139     const char *ptr = avpkt->data;
 140     SAMIContext *sami = avctx->priv_data;
 141
 142     if (ptr && avpkt->size > 0) {
 143         int ret = sami_paragraph_to_ass(avctx, ptr);
 144         if (ret < 0)
 145             return ret;
 146         // TODO: pass escaped sami->encoded_source.str as source
 147         ret = ff_ass_add_rect(sub, sami->full.str, sami->readorder++, 0, NULL, NULL);
 148         if (ret < 0)
 149             return ret;
 150     }
 151     *got_sub_ptr = sub->num_rects > 0;
 152     return avpkt->size;
 153 }
 154
 155 static av_cold int sami_init(AVCodecContext *avctx)
 156 {
 157     SAMIContext *sami = avctx->priv_data;
 158     av_bprint_init(&sami->source,  0, 2048);
 159     av_bprint_init(&sami->content, 0, 2048);
 160     av_bprint_init(&sami->encoded_source,  0, 2048);
 161     av_bprint_init(&sami->encoded_content, 0, 2048);
 162     av_bprint_init(&sami->full,    0, 2048);
 163     return ff_ass_subtitle_header_default(avctx);
 164 }
 165
 166 static av_cold int sami_close(AVCodecContext *avctx)
 167 {
 168     SAMIContext *sami = avctx->priv_data;
 169     av_bprint_finalize(&sami->source,  NULL);
 170     av_bprint_finalize(&sami->content, NULL);
 171     av_bprint_finalize(&sami->encoded_source,  NULL);
 172     av_bprint_finalize(&sami->encoded_content, NULL);
 173     av_bprint_finalize(&sami->full,    NULL);
 174     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 };