X-Git-Url: https://git.sesse.net/?a=blobdiff_plain;f=modules%2Fcodec%2Fsubtitles%2Fsubsdec.c;h=f32849f4ab113b3adfd64334e1cbeb2392a66762;hb=78d1d1105b1b309406eca03d2275daf7e78cdd62;hp=24b9021dfb5b61b4e0574520680c74ddb4bcc83e;hpb=0357e2cf82ad87897a43ea3b5a2540aacff504a7;p=vlc diff --git a/modules/codec/subtitles/subsdec.c b/modules/codec/subtitles/subsdec.c index 24b9021dfb..f32849f4ab 100644 --- a/modules/codec/subtitles/subsdec.c +++ b/modules/codec/subtitles/subsdec.c @@ -50,27 +50,116 @@ static char *CreateHtmlSubtitle( int *pi_align, char * ); * Module descriptor. *****************************************************************************/ static const char *const ppsz_encodings[] = { - DEFAULT_NAME, "ASCII", "UTF-8", "", - "ISO-8859-1", "CP1252", "MacRoman", "MacIceland","ISO-8859-15", "", - "ISO-8859-2", "CP1250", "MacCentralEurope", "MacCroatian", "MacRomania", "", - "ISO-8859-5", "CP1251", "MacCyrillic", "MacUkraine", "KOI8-R", "KOI8-U", "KOI8-RU", "", - "ISO-8859-6", "CP1256", "MacArabic", "", - "ISO-8859-7", "CP1253", "MacGreek", "", - "ISO-8859-8", "CP1255", "MacHebrew", "", - "ISO-8859-9", "CP1254", "MacTurkish", "", - "ISO-8859-13", "CP1257", "", - "ISO-2022-JP", "ISO-2022-JP-1", "ISO-2022-JP-2", "EUC-JP", "SHIFT_JIS", "", - "ISO-2022-CN", "ISO-2022-CN-EXT", "EUC-CN", "EUC-TW", "BIG5", "BIG5-HKSCS", "", - "ISO-2022-KR", "EUC-KR", "", - "MacThai", "KOI8-T", "", - "ISO-8859-3", "ISO-8859-4", "ISO-8859-10", "ISO-8859-14", "ISO-8859-16", "", - "CP850", "CP862", "CP866", "CP874", "CP932", "CP949", "CP950", "CP1133", "CP1258", "", - "Macintosh", "", - "UTF-7", "UTF-16", "UTF-16BE", "UTF-16LE", "UTF-32", "UTF-32BE", "UTF-32LE", - "C99", "JAVA", "UCS-2", "UCS-2BE", "UCS-2LE", "UCS-4", "UCS-4BE", "UCS-4LE", "", - "HZ", "GBK", "GB18030", "JOHAB", "ARMSCII-8", - "Georgian-Academy", "Georgian-PS", "TIS-620", "MuleLao-1", "VISCII", "TCVN", - "HPROMAN8", "NEXTSTEP" }; + "", + "UTF-8", + "UTF-16", + "UTF-16BE", + "UTF-16LE", + "GB18030", + "ISO-8859-15", + "Windows-1252", + "ISO-8859-2", + "Windows-1250", + "ISO-8859-3", + "ISO-8859-10", + "Windows-1251", + "KOI8-R", + "KOI8-U", + "ISO-8859-6", + "Windows-1256", + "ISO-8859-7", + "Windows-1253", + "ISO-8859-8", + "Windows-1255", + "ISO-8859-9", + "Windows-1254", + "ISO-8859-11", + "Windows-874", + "ISO-8859-13", + "Windows-1257", + "ISO-8859-14", + "ISO-8859-16", + "ISO-2022-CN-EXT", + "EUC-CN", + "ISO-2022-JP-2", + "EUC-JP", + "Shift_JIS", + "CP949", + "ISO-2022-KR", + "Big5", + "ISO-2022-TW", + "Big5-HKSCS", + "VISCII", + "Windows-1258", +}; + +static const char *const ppsz_encoding_names[] = { + /* xgettext: + The character encoding name in parenthesis corresponds to that used for + the GetACP translation. "Windows-1252" applies to Western European + languages using the Latin alphabet. */ + N_("Default (Windows-1252)"), + N_("Universal (UTF-8)"), + N_("Universal (UTF-16)"), + N_("Universal (big endian UTF-16)"), + N_("Universal (little endian UTF-16)"), + N_("Universal, Chinese (GB18030)"), + + /* ISO 8859 and the likes */ + /* 1 */ + N_("Western European (Latin-9)"), /* mostly superset of Latin-1 */ + N_("Western European (Windows-1252)"), + /* 2 */ + N_("Eastern European (Latin-2)"), + N_("Eastern European (Windows-1250)"), + /* 3 */ + N_("Esperanto (Latin-3)"), + /* 4 */ + N_("Nordic (Latin-6)"), /* Latin 6 supersedes Latin 4 */ + /* 5 */ + N_("Cyrillic (Windows-1251)"), /* ISO 8859-5 is not practically used */ + N_("Russian (KOI8-R)"), + N_("Ukrainian (KOI8-U)"), + /* 6 */ + N_("Arabic (ISO 8859-6)"), + N_("Arabic (Windows-1256)"), + /* 7 */ + N_("Greek (ISO 8859-7)"), + N_("Greek (Windows-1253)"), + /* 8 */ + N_("Hebrew (ISO 8859-8)"), + N_("Hebrew (Windows-1255)"), + /* 9 */ + N_("Turkish (ISO 8859-9)"), + N_("Turkish (Windows-1254)"), + /* 10 -> 4 */ + /* 11 */ + N_("Thai (TIS 620-2533/ISO 8859-11)"), + N_("Thai (Windows-874)"), + /* 13 */ + N_("Baltic (Latin-7)"), + N_("Baltic (Windows-1257)"), + /* 12 -> /dev/null */ + /* 14 */ + N_("Celtic (Latin-8)"), + /* 15 -> 1 */ + /* 16 */ + N_("South-Eastern European (Latin-10)"), + /* CJK families */ + N_("Simplified Chinese (ISO-2022-CN-EXT)"), + N_("Simplified Chinese Unix (EUC-CN)"), + N_("Japanese (7-bits JIS/ISO-2022-JP-2)"), + N_("Japanese Unix (EUC-JP)"), + N_("Japanese (Shift JIS)"), + N_("Korean (EUC-KR/CP949)"), + N_("Korean (ISO-2022-KR)"), + N_("Traditional Chinese (Big5)"), + N_("Traditional Chinese Unix (EUC-TW)"), + N_("Hong-Kong Supplementary (HKSCS)"), + /* Other */ + N_("Vietnamese (VISCII)"), + N_("Vietnamese (Windows-1258)"), +}; /* SSA supports charset selection. The following known charsets are used: @@ -113,25 +202,25 @@ static const char *const ppsz_justification_text[] = { "VLC partly implements this, but you can choose to disable all formatting.") -vlc_module_begin(); - set_shortname( N_("Subtitles")); - set_description( N_("Text subtitles decoder") ); - set_capability( "decoder", 50 ); - set_callbacks( OpenDecoder, CloseDecoder ); - set_category( CAT_INPUT ); - set_subcategory( SUBCAT_INPUT_SCODEC ); +vlc_module_begin () + set_shortname( N_("Subtitles")) + set_description( N_("Text subtitles decoder") ) + set_capability( "decoder", 50 ) + set_callbacks( OpenDecoder, CloseDecoder ) + set_category( CAT_INPUT ) + set_subcategory( SUBCAT_INPUT_SCODEC ) add_integer( "subsdec-align", 0, NULL, ALIGN_TEXT, ALIGN_LONGTEXT, - false ); - change_integer_list( pi_justification, ppsz_justification_text, NULL ); - add_string( "subsdec-encoding", DEFAULT_NAME, NULL, - ENCODING_TEXT, ENCODING_LONGTEXT, false ); - change_string_list( ppsz_encodings, 0, 0 ); + false ) + change_integer_list( pi_justification, ppsz_justification_text, NULL ) + add_string( "subsdec-encoding", "", NULL, + ENCODING_TEXT, ENCODING_LONGTEXT, false ) + change_string_list( ppsz_encodings, ppsz_encoding_names, 0 ) add_bool( "subsdec-autodetect-utf8", true, NULL, - AUTODETECT_UTF8_TEXT, AUTODETECT_UTF8_LONGTEXT, false ); + AUTODETECT_UTF8_TEXT, AUTODETECT_UTF8_LONGTEXT, false ) add_bool( "subsdec-formatted", true, NULL, FORMAT_TEXT, FORMAT_LONGTEXT, - false ); -vlc_module_end(); + false ) +vlc_module_end () /***************************************************************************** * OpenDecoder: probe the decoder and return score @@ -143,27 +232,27 @@ static int OpenDecoder( vlc_object_t *p_this ) { decoder_t *p_dec = (decoder_t*)p_this; decoder_sys_t *p_sys; - vlc_value_t val; switch( p_dec->fmt_in.i_codec ) { - case VLC_FOURCC('s','u','b','t'): - case VLC_FOURCC('s','s','a',' '): - case VLC_FOURCC('t','1','4','0'): + case VLC_CODEC_SUBT: + case VLC_CODEC_SSA: + case VLC_CODEC_ITU_T140: break; default: return VLC_EGENERIC; } p_dec->pf_decode_sub = DecodeBlock; + p_dec->fmt_out.i_cat = SPU_ES; + p_dec->fmt_out.i_codec = 0; /* Allocate the memory needed to store the decoder's structure */ - p_dec->p_sys = p_sys = malloc( sizeof( *p_sys ) ); + p_dec->p_sys = p_sys = calloc( 1, sizeof( *p_sys ) ); if( p_sys == NULL ) return VLC_ENOMEM; /* init of p_sys */ - memset( p_sys, 0, sizeof( *p_sys ) ); p_sys->i_align = 0; p_sys->iconv_handle = (vlc_iconv_t)-1; p_sys->b_autodetect_utf8 = false; @@ -176,36 +265,43 @@ static int OpenDecoder( vlc_object_t *p_this ) char *psz_charset = NULL; /* First try demux-specified encoding */ - if( p_dec->fmt_in.i_codec == VLC_FOURCC('t','1','4','0') ) + if( p_dec->fmt_in.i_codec == VLC_CODEC_ITU_T140 ) psz_charset = strdup( "UTF-8" ); /* IUT T.140 is always using UTF-8 */ else if( p_dec->fmt_in.subs.psz_encoding && *p_dec->fmt_in.subs.psz_encoding ) { psz_charset = strdup (p_dec->fmt_in.subs.psz_encoding); msg_Dbg (p_dec, "trying demuxer-specified character encoding: %s", - p_dec->fmt_in.subs.psz_encoding ?: "not specified"); + p_dec->fmt_in.subs.psz_encoding ? + p_dec->fmt_in.subs.psz_encoding : "not specified"); } /* Second, try configured encoding */ if (psz_charset == NULL) { psz_charset = var_CreateGetNonEmptyString (p_dec, "subsdec-encoding"); - if ((psz_charset != NULL) && !strcasecmp (psz_charset, DEFAULT_NAME)) - { - free (psz_charset); - psz_charset = NULL; - } - msg_Dbg (p_dec, "trying configured character encoding: %s", - psz_charset ?: "not specified"); + psz_charset ? psz_charset : "not specified"); } /* Third, try "local" encoding with optional UTF-8 autodetection */ if (psz_charset == NULL) { - psz_charset = strdup (GetFallbackEncoding ()); + /* xgettext: + The Windows ANSI code page most commonly used for this language. + VLC uses this as a guess of the subtitle files character set + (if UTF-8 and UTF-16 autodetection fails). + Western European languages normally use "CP1252", which is a + Microsoft-variant of ISO 8859-1. That suits the Latin alphabet. + Other scripts use other code pages. + + This MUST be a valid iconv character set. If unsure, please refer + the VideoLAN translators mailing list. */ + const char *acp = vlc_pgettext("GetACP", "CP1252"); + + psz_charset = strdup (acp); msg_Dbg (p_dec, "trying default character encoding: %s", - psz_charset ?: "not specified"); + psz_charset ? psz_charset : "not specified"); if (var_CreateGetBool (p_dec, "subsdec-autodetect-utf8")) { @@ -231,11 +327,9 @@ static int OpenDecoder( vlc_object_t *p_this ) } free (psz_charset); - var_Create( p_dec, "subsdec-align", VLC_VAR_INTEGER | VLC_VAR_DOINHERIT ); - var_Get( p_dec, "subsdec-align", &val ); - p_sys->i_align = val.i_int; + p_sys->i_align = var_CreateGetInteger( p_dec, "subsdec-align" ); - if( p_dec->fmt_in.i_codec == VLC_FOURCC('s','s','a',' ') + if( p_dec->fmt_in.i_codec == VLC_CODEC_SSA && var_CreateGetBool( p_dec, "subsdec-formatted" ) ) { if( p_dec->fmt_in.i_extra > 0 ) @@ -329,7 +423,7 @@ static subpicture_t *ParseText( decoder_t *p_dec, block_t *p_block ) video_format_t fmt; /* We cannot display a subpicture with no date */ - if( p_block->i_pts == 0 ) + if( p_block->i_pts <= VLC_TS_INVALID ) { msg_Warn( p_dec, "subtitle without a date" ); return NULL; @@ -345,10 +439,11 @@ static subpicture_t *ParseText( decoder_t *p_dec, block_t *p_block ) } /* Should be resiliant against bad subtitles */ - psz_subtitle = strndup( (const char *)p_block->p_buffer, - p_block->i_buffer ); + psz_subtitle = malloc( p_block->i_buffer + 1 ); if( psz_subtitle == NULL ) return NULL; + memcpy( psz_subtitle, p_block->p_buffer, p_block->i_buffer ); + psz_subtitle[p_block->i_buffer] = '\0'; if( p_sys->iconv_handle == (vlc_iconv_t)-1 ) { @@ -376,7 +471,7 @@ static subpicture_t *ParseText( decoder_t *p_dec, block_t *p_block ) { size_t inbytes_left = strlen( psz_subtitle ); size_t outbytes_left = 6 * inbytes_left; - char *psz_new_subtitle = malloc( outbytes_left + 1 ); + char *psz_new_subtitle = xmalloc( outbytes_left + 1 ); char *psz_convert_buffer_out = psz_new_subtitle; const char *psz_convert_buffer_in = psz_subtitle; @@ -398,11 +493,13 @@ static subpicture_t *ParseText( decoder_t *p_dec, block_t *p_block ) psz_subtitle = realloc( psz_new_subtitle, psz_convert_buffer_out - psz_new_subtitle ); + if( !psz_subtitle ) + psz_subtitle = psz_new_subtitle; } } /* Create the subpicture unit */ - p_spu = decoder_NewSubpicture( p_dec ); + p_spu = decoder_NewSubpicture( p_dec, NULL ); if( !p_spu ) { msg_Warn( p_dec, "can't get spu buffer" ); @@ -412,8 +509,7 @@ static subpicture_t *ParseText( decoder_t *p_dec, block_t *p_block ) /* Create a new subpicture region */ memset( &fmt, 0, sizeof(video_format_t) ); - fmt.i_chroma = VLC_FOURCC('T','E','X','T'); - fmt.i_aspect = 0; + fmt.i_chroma = VLC_CODEC_TEXT; fmt.i_width = fmt.i_height = 0; fmt.i_x_offset = fmt.i_y_offset = 0; p_spu->p_region = subpicture_region_New( &fmt ); @@ -426,7 +522,7 @@ static subpicture_t *ParseText( decoder_t *p_dec, block_t *p_block ) } /* Decode and format the subpicture unit */ - if( p_dec->fmt_in.i_codec != VLC_FOURCC('s','s','a',' ') ) + if( p_dec->fmt_in.i_codec != VLC_CODEC_SSA ) { /* Normal text subs, easy markup */ p_spu->p_region->i_align = SUBPICTURE_ALIGN_BOTTOM | p_sys->i_align; @@ -539,7 +635,8 @@ static char *StripTags( char *psz_subtitle ) psz_subtitle++; } *psz_text = '\0'; - psz_text_start = realloc( psz_text_start, strlen( psz_text_start ) + 1 ); + char *psz = realloc( psz_text_start, strlen( psz_text_start ) + 1 ); + if( psz ) psz_text_start = psz; return psz_text_start; } @@ -554,7 +651,7 @@ static char *StripTags( char *psz_subtitle ) * to be carrying style information. Over time people have used them that way. * In the absence of specifications from which to work, the tags supported * have been restricted to the simple set permitted by the USF DTD, ie. : - * Basic:
, , , + * Basic:
, , , , * Extended: * Attributes: face * family @@ -597,24 +694,27 @@ static void HtmlCopy( char **ppsz_html, char **ppsz_subtitle, const char *psz_te static char *CreateHtmlSubtitle( int *pi_align, char *psz_subtitle ) { - char *psz_tag = malloc( ( strlen( psz_subtitle ) / 3 ) + 1 ); - if( !psz_tag ) return NULL; - size_t i_buf_size = strlen( psz_subtitle ) + 100; - char *psz_html_start = malloc( i_buf_size ); - bool b_has_align = false; - + /* */ + char *psz_tag = malloc( ( strlen( psz_subtitle ) / 3 ) + 1 ); + if( !psz_tag ) + return NULL; psz_tag[ 0 ] = '\0'; + /* */ + //Oo + 100 ??? + size_t i_buf_size = strlen( psz_subtitle ) + 100; + char *psz_html_start = malloc( i_buf_size ); + char *psz_html = psz_html_start; if( psz_html_start == NULL ) { free( psz_tag ); return NULL; } - - char *psz_html = psz_html_start; + psz_html[0] = '\0'; - strcpy( psz_html, "" ); - psz_html += 6; + bool b_has_align = false; + + HtmlPut( &psz_html, "" ); /* */ while( *psz_subtitle ) @@ -645,6 +745,11 @@ static char *CreateHtmlSubtitle( int *pi_align, char *psz_subtitle ) HtmlCopy( &psz_html, &psz_subtitle, "" ); strcat( psz_tag, "u" ); } + else if( !strncasecmp( psz_subtitle, "", 3 ) ) + { + HtmlCopy( &psz_html, &psz_subtitle, "" ); + strcat( psz_tag, "s" ); + } else if( !strncasecmp( psz_subtitle, "", 4 ); i_len = 4; break; + case 's': + b_match = !strncasecmp( psz_subtitle, "", 4 ); + i_len = 4; + break; case 'f': b_match = !strncasecmp( psz_subtitle, "", 7 ); i_len = 7; @@ -857,6 +970,31 @@ static char *CreateHtmlSubtitle( int *pi_align, char *psz_subtitle ) /* Hide {\stupidity} */ psz_subtitle = strchr( psz_subtitle, '}' ) + 1; } + else if( psz_subtitle[0] == '{' && psz_subtitle[1] == 'Y' + && psz_subtitle[2] == ':' && strchr( psz_subtitle, '}' ) ) + { + /* Hide {Y:stupidity} */ + psz_subtitle = strchr( psz_subtitle, '}' ) + 1; + } + else if( psz_subtitle[0] == '\\' && psz_subtitle[1] ) + { + if( psz_subtitle[1] == 'N' || psz_subtitle[1] == 'n' ) + { + HtmlPut( &psz_html, "
" ); + psz_subtitle += 2; + } + else if( psz_subtitle[1] == 'h' ) + { + /* Non breakable space */ + HtmlPut( &psz_html, NO_BREAKING_SPACE ); + psz_subtitle += 2; + } + else + { + HtmlPut( &psz_html, "\\" ); + psz_subtitle++; + } + } else { *psz_html = *psz_subtitle; @@ -876,27 +1014,61 @@ static char *CreateHtmlSubtitle( int *pi_align, char *psz_subtitle ) if( ( size_t )( psz_html - psz_html_start ) > i_buf_size - 50 ) { - int i_len = psz_html - psz_html_start; + const int i_len = psz_html - psz_html_start; i_buf_size += 200; - psz_html_start = realloc( psz_html_start, i_buf_size ); - psz_html = psz_html_start + i_len; - *psz_html = '\0'; + char *psz_new = realloc( psz_html_start, i_buf_size ); + if( !psz_new ) + break; + psz_html_start = psz_new; + psz_html = &psz_new[i_len]; } } - strcpy( psz_html, "" ); - psz_html += 7; - - if( psz_tag[ 0 ] != '\0' ) + if( psz_html_start ) { - /* Not well formed -- kill everything */ - free( psz_html_start ); - psz_html_start = NULL; - } - else if( psz_html_start ) - { - /* Shrink the memory requirements */ - psz_html_start = realloc( psz_html_start, psz_html - psz_html_start + 1 ); + static const char *psz_text_close = ""; + static const char *psz_tag_long = "/font>"; + + /* Realloc for closing tags and shrink memory */ + const size_t i_length = (size_t)( psz_html - psz_html_start ); + + const size_t i_size = i_length + strlen(psz_tag_long) * strlen(psz_tag) + strlen(psz_text_close) + 1; + char *psz_new = realloc( psz_html_start, i_size ); + if( psz_new ) + { + psz_html_start = psz_new; + psz_html = &psz_new[i_length]; + + /* Close not well formed subtitle */ + while( *psz_tag ) + { + /* */ + char *psz_last = &psz_tag[strlen(psz_tag)-1]; + switch( *psz_last ) + { + case 'b': + HtmlPut( &psz_html, "" ); + break; + case 'i': + HtmlPut( &psz_html, "" ); + break; + case 'u': + HtmlPut( &psz_html, "" ); + break; + case 's': + HtmlPut( &psz_html, "" ); + break; + case 'f': + HtmlPut( &psz_html, "/font>" ); + break; + case 'I': + break; + } + + *psz_last = '\0'; + } + HtmlPut( &psz_html, psz_text_close ); + } } free( psz_tag );