/*****************************************************************************
* Preamble
*****************************************************************************/
+#ifdef HAVE_CONFIG_H
+# include "config.h"
+#endif
#include "subsdec.h"
+#include <vlc_plugin.h>
/*****************************************************************************
* Local prototypes
static subpicture_t *DecodeBlock ( decoder_t *, block_t ** );
static subpicture_t *ParseText ( decoder_t *, block_t * );
static char *StripTags ( char * );
-static char *CreateHtmlSubtitle ( char * );
+static char *CreateHtmlSubtitle( int *pi_align, char * );
/*****************************************************************************
* Module descriptor.
*****************************************************************************/
-static const char *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" };
+static const char *const ppsz_encodings[] = {
+ "",
+ "system",
+ "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_("System codeset"),
+ 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:
254 = PC 437
*/
-static int pi_justification[] = { 0, 1, 2 };
-static const char *ppsz_justification_text[] = {N_("Center"),N_("Left"),N_("Right")};
+static const int pi_justification[] = { 0, 1, 2 };
+static const char *const ppsz_justification_text[] = {
+ N_("Center"),N_("Left"),N_("Right")};
#define ENCODING_TEXT N_("Subtitles text encoding")
#define ENCODING_LONGTEXT N_("Set the encoding used in text subtitles")
"VLC partly implements this, but you can choose to disable all formatting.")
-vlc_module_begin();
- set_shortname( _("Subtitles"));
- set_description( _("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,
- VLC_FALSE );
- change_integer_list( pi_justification, ppsz_justification_text, 0 );
- add_string( "subsdec-encoding", DEFAULT_NAME, NULL,
- ENCODING_TEXT, ENCODING_LONGTEXT, VLC_FALSE );
- change_string_list( ppsz_encodings, 0, 0 );
- add_bool( "subsdec-autodetect-utf8", VLC_TRUE, NULL,
- AUTODETECT_UTF8_TEXT, AUTODETECT_UTF8_LONGTEXT, VLC_FALSE );
- add_bool( "subsdec-formatted", VLC_TRUE, NULL, FORMAT_TEXT, FORMAT_LONGTEXT,
- VLC_FALSE );
-vlc_module_end();
+ 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 )
+ add_bool( "subsdec-formatted", true, NULL, FORMAT_TEXT, FORMAT_LONGTEXT,
+ false )
+vlc_module_end ()
/*****************************************************************************
* OpenDecoder: probe the decoder and return score
{
decoder_t *p_dec = (decoder_t*)p_this;
decoder_sys_t *p_sys;
- vlc_value_t val;
- if( p_dec->fmt_in.i_codec != VLC_FOURCC('s','u','b','t') &&
- p_dec->fmt_in.i_codec != VLC_FOURCC('s','s','a',' ') )
+ switch( p_dec->fmt_in.i_codec )
{
- return VLC_EGENERIC;
+ 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 */
- if( ( p_dec->p_sys = p_sys =
- (decoder_sys_t *)calloc(1, sizeof(decoder_sys_t)) ) == NULL )
- {
- msg_Err( p_dec, "out of memory" );
+ p_dec->p_sys = p_sys = calloc( 1, sizeof( *p_sys ) );
+ if( p_sys == NULL )
return VLC_ENOMEM;
- }
/* init of p_sys */
p_sys->i_align = 0;
p_sys->iconv_handle = (vlc_iconv_t)-1;
- p_sys->b_autodetect_utf8 = VLC_FALSE;
- p_sys->b_ass = VLC_FALSE;
+ p_sys->b_autodetect_utf8 = false;
+ p_sys->b_ass = false;
p_sys->i_original_height = -1;
p_sys->i_original_width = -1;
TAB_INIT( p_sys->i_ssa_styles, p_sys->pp_ssa_styles );
TAB_INIT( p_sys->i_images, p_sys->pp_images );
char *psz_charset = NULL;
+
/* First try demux-specified encoding */
+ 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))
+ psz_charset = var_InheritString (p_dec, "subsdec-encoding");
+ msg_Dbg (p_dec, "trying configured character encoding: %s",
+ psz_charset ? psz_charset : "not specified");
+ if (!strcmp (psz_charset, "system"))
{
free (psz_charset);
- psz_charset = NULL;
+ psz_charset = strdup ("");
+ /* ^ iconv() treats "" as nl_langinfo(CODESET) */
}
-
- msg_Dbg (p_dec, "trying configured character encoding: %s",
- 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"))
+ if (var_InheritBool (p_dec, "subsdec-autodetect-utf8"))
{
msg_Dbg (p_dec, "using automatic UTF-8 detection");
- p_sys->b_autodetect_utf8 = VLC_TRUE;
+ p_sys->b_autodetect_utf8 = true;
}
}
+ /* Forth, don't do character decoding, i.e. assume UTF-8 */
if (psz_charset == NULL)
{
psz_charset = strdup ("UTF-8");
- msg_Dbg (p_dec, "trying hard-coded character encoding: %s",
- psz_charset ?: "error");
+ msg_Dbg (p_dec, "using UTF-8 character encoding" );
}
- if (psz_charset == NULL)
- {
- free (p_sys);
- return VLC_ENOMEM;
- }
-
- if (strcasecmp (psz_charset, "UTF-8") && strcasecmp (psz_charset, "utf8"))
+ if ((psz_charset != NULL)
+ && strcasecmp (psz_charset, "UTF-8")
+ && strcasecmp (psz_charset, "utf8"))
{
p_sys->iconv_handle = vlc_iconv_open ("UTF-8", psz_charset);
if (p_sys->iconv_handle == (vlc_iconv_t)(-1))
- msg_Err (p_dec, "cannot convert from %s: %s", psz_charset,
- strerror (errno));
+ msg_Err (p_dec, "cannot convert from %s: %m", psz_charset);
}
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_InheritInteger( p_dec, "subsdec-align" );
- if( p_dec->fmt_in.i_codec == VLC_FOURCC('s','s','a',' ') && var_CreateGetBool( p_dec, "subsdec-formatted" ) )
+ if( p_dec->fmt_in.i_codec == VLC_CODEC_SSA
+ && var_InheritBool( p_dec, "subsdec-formatted" ) )
{
if( p_dec->fmt_in.i_extra > 0 )
ParseSSAHeader( p_dec );
****************************************************************************/
static subpicture_t *DecodeBlock( decoder_t *p_dec, block_t **pp_block )
{
- subpicture_t *p_spu = NULL;
+ subpicture_t *p_spu;
+ block_t *p_block;
- if( !pp_block || *pp_block == NULL ) return NULL;
+ if( !pp_block || *pp_block == NULL )
+ return NULL;
+
+ p_block = *pp_block;
+ if( p_block->i_flags & (BLOCK_FLAG_DISCONTINUITY|BLOCK_FLAG_CORRUPTED) )
+ {
+ block_Release( p_block );
+ return NULL;
+ }
- p_spu = ParseText( p_dec, *pp_block );
+ p_spu = ParseText( p_dec, p_block );
- block_Release( *pp_block );
+ block_Release( p_block );
*pp_block = NULL;
return p_spu;
if( !p_sys->pp_ssa_styles[i] )
continue;
- if( p_sys->pp_ssa_styles[i]->psz_stylename )
- free( p_sys->pp_ssa_styles[i]->psz_stylename );
- if( p_sys->pp_ssa_styles[i]->font_style.psz_fontname )
- free( p_sys->pp_ssa_styles[i]->font_style.psz_fontname );
- if( p_sys->pp_ssa_styles[i] )
- free( p_sys->pp_ssa_styles[i] );
+ free( p_sys->pp_ssa_styles[i]->psz_stylename );
+ free( p_sys->pp_ssa_styles[i]->font_style.psz_fontname );
+ free( p_sys->pp_ssa_styles[i] );
}
TAB_CLEAN( p_sys->i_ssa_styles, p_sys->pp_ssa_styles );
}
continue;
if( p_sys->pp_images[i]->p_pic )
- p_sys->pp_images[i]->p_pic->pf_release( p_sys->pp_images[i]->p_pic );
- if( p_sys->pp_images[i]->psz_filename )
- free( p_sys->pp_images[i]->psz_filename );
+ picture_Release( p_sys->pp_images[i]->p_pic );
+ free( p_sys->pp_images[i]->psz_filename );
free( p_sys->pp_images[i] );
}
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;
}
/* 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 )
{
if (EnsureUTF8( psz_subtitle ) == NULL)
{
- msg_Err( p_dec, _("failed to convert subtitle encoding.\n"
+ msg_Err( p_dec, "failed to convert subtitle encoding.\n"
"Try manually setting a character-encoding "
- "before you open the file.") );
+ "before you open the file." );
}
}
else
{
msg_Dbg( p_dec, "invalid UTF-8 sequence: "
"disabling UTF-8 subtitles autodetection" );
- p_sys->b_autodetect_utf8 = VLC_FALSE;
+ p_sys->b_autodetect_utf8 = false;
}
}
{
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;
if( ( ret == (size_t)(-1) ) || inbytes_left )
{
free( psz_new_subtitle );
- msg_Err( p_dec, _("failed to convert subtitle encoding.\n"
+ msg_Err( p_dec, "failed to convert subtitle encoding.\n"
"Try manually setting a character-encoding "
- "before you open the file.") );
+ "before you open the file." );
return NULL;
}
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 = p_dec->pf_spu_buffer_new( p_dec );
+ p_spu = decoder_NewSubpicture( p_dec, NULL );
if( !p_spu )
{
msg_Warn( p_dec, "can't get spu buffer" );
- if( psz_subtitle ) free( psz_subtitle );
+ free( psz_subtitle );
return NULL;
}
- p_spu->b_pausable = VLC_TRUE;
-
/* 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 = p_spu->pf_create_region( VLC_OBJECT(p_dec), &fmt );
+ p_spu->p_region = subpicture_region_New( &fmt );
if( !p_spu->p_region )
{
msg_Err( p_dec, "cannot allocate SPU region" );
- if( psz_subtitle ) free( psz_subtitle );
- p_dec->pf_spu_buffer_del( p_dec, p_spu );
+ free( psz_subtitle );
+ decoder_DeleteSubpicture( p_dec, p_spu );
return NULL;
}
/* 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;
- p_spu->i_x = p_sys->i_align ? 20 : 0;
- p_spu->i_y = 10;
+ p_spu->p_region->i_x = p_sys->i_align ? 20 : 0;
+ p_spu->p_region->i_y = 10;
/* Remove formatting from string */
p_spu->p_region->psz_text = StripTags( psz_subtitle );
- if( var_CreateGetBool( p_dec, "subsdec-formatted" ) )
+ if( var_InheritBool( p_dec, "subsdec-formatted" ) )
{
- p_spu->p_region->psz_html = CreateHtmlSubtitle( psz_subtitle );
+ p_spu->p_region->psz_html = CreateHtmlSubtitle( &p_spu->p_region->i_align, psz_subtitle );
}
p_spu->i_start = p_block->i_pts;
p_spu->i_stop = p_block->i_pts + p_block->i_length;
p_spu->b_ephemer = (p_block->i_length == 0);
- p_spu->b_absolute = VLC_FALSE;
+ p_spu->b_absolute = false;
}
else
{
/* Decode SSA/USF strings */
- if( p_dec->fmt_in.i_codec == VLC_FOURCC('s','s','a',' ') )
- ParseSSAString( p_dec, psz_subtitle, p_spu );
+ ParseSSAString( p_dec, psz_subtitle, p_spu );
p_spu->i_start = p_block->i_pts;
p_spu->i_stop = p_block->i_pts + p_block->i_length;
p_spu->b_ephemer = (p_block->i_length == 0);
- p_spu->b_absolute = VLC_FALSE;
+ p_spu->b_absolute = false;
p_spu->i_original_picture_width = p_sys->i_original_width;
p_spu->i_original_picture_height = p_sys->i_original_height;
}
- if( psz_subtitle ) free( psz_subtitle );
+ free( psz_subtitle );
return p_spu;
}
return p_newline;
}
-/* Function now handles tags which has attribute values, and tries
+/* Function now handles tags with attribute values, and tries
* to deal with &' commands too. It no longer modifies the string
* in place, so that the original text can be reused
*/
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;
}
* 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: <br>, <i>, <b>, <u>
+ * Basic: <br>, <i>, <b>, <u>, <s>
* Extended: <font>
* Attributes: face
* family
* returned, and the rendering engine will fall back to the
* plain text version of the subtitle.
*/
-static char *CreateHtmlSubtitle( char *psz_subtitle )
+static void HtmlNPut( char **ppsz_html, const char *psz_text, int i_max )
{
- char psz_tagStack[ 100 ];
- size_t i_buf_size = strlen( psz_subtitle ) + 100;
- char *psz_html_start = malloc( i_buf_size );
+ const int i_len = strlen(psz_text);
- psz_tagStack[ 0 ] = '\0';
+ strncpy( *ppsz_html, psz_text, i_max );
+ *ppsz_html += __MIN(i_max,i_len);
+}
+
+static void HtmlPut( char **ppsz_html, const char *psz_text )
+{
+ strcpy( *ppsz_html, psz_text );
+ *ppsz_html += strlen(psz_text);
+}
+static void HtmlCopy( char **ppsz_html, char **ppsz_subtitle, const char *psz_text )
+{
+ HtmlPut( ppsz_html, psz_text );
+ *ppsz_subtitle += strlen(psz_text);
+}
- if( psz_html_start != NULL )
+static char *CreateHtmlSubtitle( int *pi_align, char *psz_subtitle )
+{
+ /* */
+ 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 )
{
- char *psz_html = psz_html_start;
+ free( psz_tag );
+ return NULL;
+ }
+ psz_html[0] = '\0';
+
+ bool b_has_align = false;
- strcpy( psz_html, "<text>" );
- psz_html += 6;
+ HtmlPut( &psz_html, "<text>" );
- while( *psz_subtitle )
+ /* */
+ while( *psz_subtitle )
+ {
+ if( *psz_subtitle == '\n' )
+ {
+ HtmlPut( &psz_html, "<br/>" );
+ psz_subtitle++;
+ }
+ else if( *psz_subtitle == '<' )
{
- if( *psz_subtitle == '\n' )
+ if( !strncasecmp( psz_subtitle, "<br/>", 5 ))
{
- strcpy( psz_html, "<br/>" );
- psz_html += 5;
- psz_subtitle++;
+ HtmlCopy( &psz_html, &psz_subtitle, "<br/>" );
}
- else if( *psz_subtitle == '<' )
+ else if( !strncasecmp( psz_subtitle, "<b>", 3 ) )
{
- if( !strncasecmp( psz_subtitle, "<br/>", 5 ))
- {
- strcpy( psz_html, "<br/>" );
- psz_html += 5;
- psz_subtitle += 5;
- }
- else if( !strncasecmp( psz_subtitle, "<b>", 3 ) )
- {
- strcpy( psz_html, "<b>" );
- strcat( psz_tagStack, "b" );
- psz_html += 3;
- psz_subtitle += 3;
- }
- else if( !strncasecmp( psz_subtitle, "<i>", 3 ) )
- {
- strcpy( psz_html, "<i>" );
- strcat( psz_tagStack, "i" );
- psz_html += 3;
- psz_subtitle += 3;
- }
- else if( !strncasecmp( psz_subtitle, "<u>", 3 ) )
- {
- strcpy( psz_html, "<u>" );
- strcat( psz_tagStack, "u" );
- psz_html += 3;
- psz_subtitle += 3;
- }
- else if( !strncasecmp( psz_subtitle, "<font ", 6 ))
- {
- const char *psz_attribs[] = { "face=\"", "family=\"", "size=\"",
- "color=\"", "outline-color=\"", "shadow-color=\"",
- "outline-level=\"", "shadow-level=\"", "back-color=\"",
- "alpha=\"", NULL };
+ HtmlCopy( &psz_html, &psz_subtitle, "<b>" );
+ strcat( psz_tag, "b" );
+ }
+ else if( !strncasecmp( psz_subtitle, "<i>", 3 ) )
+ {
+ HtmlCopy( &psz_html, &psz_subtitle, "<i>" );
+ strcat( psz_tag, "i" );
+ }
+ else if( !strncasecmp( psz_subtitle, "<u>", 3 ) )
+ {
+ HtmlCopy( &psz_html, &psz_subtitle, "<u>" );
+ strcat( psz_tag, "u" );
+ }
+ else if( !strncasecmp( psz_subtitle, "<s>", 3 ) )
+ {
+ HtmlCopy( &psz_html, &psz_subtitle, "<s>" );
+ strcat( psz_tag, "s" );
+ }
+ else if( !strncasecmp( psz_subtitle, "<font ", 6 ))
+ {
+ const char *psz_attribs[] = { "face=", "family=", "size=",
+ "color=", "outline-color=", "shadow-color=",
+ "outline-level=", "shadow-level=", "back-color=",
+ "alpha=", NULL };
+
+ HtmlCopy( &psz_html, &psz_subtitle, "<font " );
+ strcat( psz_tag, "f" );
- strcpy( psz_html, "<font " );
- strcat( psz_tagStack, "f" );
- psz_html += 6;
- psz_subtitle += 6;
+ while( *psz_subtitle != '>' )
+ {
+ int k;
- while( *psz_subtitle != '>' )
+ for( k=0; psz_attribs[ k ]; k++ )
{
- int k;
+ int i_len = strlen( psz_attribs[ k ] );
- for( k=0; psz_attribs[ k ]; k++ )
+ if( !strncasecmp( psz_subtitle, psz_attribs[k], i_len ) )
{
- int i_len = strlen( psz_attribs[ k ] );
+ /* */
+ HtmlPut( &psz_html, psz_attribs[k] );
+ psz_subtitle += i_len;
- if( !strncasecmp( psz_subtitle, psz_attribs[ k ], i_len ))
+ /* */
+ if( *psz_subtitle == '"' )
{
- i_len += strcspn( psz_subtitle + i_len, "\"" ) + 1;
-
- strncpy( psz_html, psz_subtitle, i_len );
- psz_html += i_len;
- psz_subtitle += i_len;
- break;
+ psz_subtitle++;
+ i_len = strcspn( psz_subtitle, "\"" );
}
- }
- if( psz_attribs[ k ] == NULL )
- {
- /* Jump over unrecognised tag */
- int i_len = strcspn( psz_subtitle, "\"" ) + 1;
+ else
+ {
+ i_len = strcspn( psz_subtitle, " \t>" );
+ }
+ HtmlPut( &psz_html, "\"" );
+ if( !strcmp( psz_attribs[ k ], "color=" ) && *psz_subtitle >= '0' && *psz_subtitle <= '9' )
+ HtmlPut( &psz_html, "#" );
+ HtmlNPut( &psz_html, psz_subtitle, i_len );
+ HtmlPut( &psz_html, "\"" );
- i_len += strcspn( psz_subtitle + i_len, "\"" ) + 1;
psz_subtitle += i_len;
+ if( *psz_subtitle == '\"' )
+ psz_subtitle++;
+ break;
}
- while (*psz_subtitle == ' ')
- *psz_html++ = *psz_subtitle++;
}
- *psz_html++ = *psz_subtitle++;
- }
- else if( !strncmp( psz_subtitle, "</", 2 ))
- {
- vlc_bool_t b_match = VLC_FALSE;
- int i_len = strlen( psz_tagStack ) - 1;
- char *psz_lastTag = NULL;
-
- if( i_len >= 0 )
+ if( psz_attribs[ k ] == NULL )
{
- psz_lastTag = psz_tagStack + i_len;
- i_len = 0;
-
- switch( *psz_lastTag )
+ /* Jump over unrecognised tag */
+ int i_len = strcspn( psz_subtitle, "\"" );
+ if( psz_subtitle[i_len] == '\"' )
{
- case 'b':
- b_match = !strncasecmp( psz_subtitle, "</b>", 4 );
- i_len = 4;
- break;
- case 'i':
- b_match = !strncasecmp( psz_subtitle, "</i>", 4 );
- i_len = 4;
- break;
- case 'u':
- b_match = !strncasecmp( psz_subtitle, "</u>", 4 );
- i_len = 4;
- break;
- case 'f':
- b_match = !strncasecmp( psz_subtitle, "</font>", 7 );
- i_len = 7;
- break;
+ i_len += 1 + strcspn( &psz_subtitle[i_len + 1], "\"" );
+ if( psz_subtitle[i_len] == '\"' )
+ i_len++;
}
+ psz_subtitle += i_len;
}
- if( ! b_match )
+ while (*psz_subtitle == ' ')
+ *psz_html++ = *psz_subtitle++;
+ }
+ *psz_html++ = *psz_subtitle++;
+ }
+ else if( !strncmp( psz_subtitle, "</", 2 ))
+ {
+ bool b_match = false;
+ bool b_ignore = false;
+ int i_len = strlen( psz_tag ) - 1;
+ char *psz_lastTag = NULL;
+
+ if( i_len >= 0 )
+ {
+ psz_lastTag = psz_tag + i_len;
+ i_len = 0;
+
+ switch( *psz_lastTag )
{
- /* Not well formed -- kill everything */
- free( psz_html_start );
- psz_html_start = NULL;
+ case 'b':
+ b_match = !strncasecmp( psz_subtitle, "</b>", 4 );
+ i_len = 4;
+ break;
+ case 'i':
+ b_match = !strncasecmp( psz_subtitle, "</i>", 4 );
+ i_len = 4;
+ break;
+ case 'u':
+ b_match = !strncasecmp( psz_subtitle, "</u>", 4 );
+ i_len = 4;
+ break;
+ case 's':
+ b_match = !strncasecmp( psz_subtitle, "</s>", 4 );
+ i_len = 4;
+ break;
+ case 'f':
+ b_match = !strncasecmp( psz_subtitle, "</font>", 7 );
+ i_len = 7;
+ break;
+ case 'I':
+ i_len = strcspn( psz_subtitle, ">" );
+ b_match = psz_subtitle[i_len] == '>';
+ b_ignore = true;
+ if( b_match )
+ i_len++;
break;
}
- *psz_lastTag = '\0';
- strncpy( psz_html, psz_subtitle, i_len );
- psz_html += i_len;
- psz_subtitle += i_len;
}
- else
+ if( !b_match )
{
- psz_subtitle += strcspn( psz_subtitle, ">" );
+ /* Not well formed -- kill everything */
+ free( psz_html_start );
+ psz_html_start = NULL;
+ break;
}
+ *psz_lastTag = '\0';
+ if( !b_ignore )
+ HtmlNPut( &psz_html, psz_subtitle, i_len );
+
+ psz_subtitle += i_len;
}
- else if( *psz_subtitle == '&' )
+ else if( ( psz_subtitle[1] < 'a' || psz_subtitle[1] > 'z' ) &&
+ ( psz_subtitle[1] < 'A' || psz_subtitle[1] > 'Z' ) )
{
- if( !strncasecmp( psz_subtitle, "<", 4 ))
- {
- strcpy( psz_html, "<" );
- psz_html += 4;
- psz_subtitle += 4;
- }
- else if( !strncasecmp( psz_subtitle, ">", 4 ))
+ /* We have a single < */
+ HtmlPut( &psz_html, "<" );
+ psz_subtitle++;
+ }
+ else
+ {
+ /* We have an unknown tag or a single < */
+
+ /* Search for the next tag or end of tag or end of string */
+ char *psz_stop = psz_subtitle + 1 + strcspn( &psz_subtitle[1], "<>" );
+ char *psz_closing = strstr( psz_subtitle, "/>" );
+
+ if( psz_closing && psz_closing < psz_stop )
{
- strcpy( psz_html, ">" );
- psz_html += 4;
- psz_subtitle += 4;
+ /* We have a self closed tag, remove it */
+ psz_subtitle = &psz_closing[2];
}
- else if( !strncasecmp( psz_subtitle, "&", 5 ))
+ else if( *psz_stop == '>' )
{
- strcpy( psz_html, "&" );
- psz_html += 5;
- psz_subtitle += 5;
+ char psz_match[256];
+
+ snprintf( psz_match, sizeof(psz_match), "</%s", &psz_subtitle[1] );
+ psz_match[strcspn( psz_match, " \t>" )] = '\0';
+
+ if( strstr( psz_subtitle, psz_match ) )
+ {
+ /* We have the closing tag, ignore it TODO */
+ psz_subtitle = &psz_stop[1];
+ strcat( psz_tag, "I" );
+ }
+ else
+ {
+ int i_len = psz_stop + 1 - psz_subtitle;
+
+ /* Copy the whole data */
+ for( ; i_len > 0; i_len--, psz_subtitle++ )
+ {
+ if( *psz_subtitle == '<' )
+ HtmlPut( &psz_html, "<" );
+ else if( *psz_subtitle == '>' )
+ HtmlPut( &psz_html, ">" );
+ else
+ *psz_html++ = *psz_subtitle;
+ }
+ }
}
else
{
- strcpy( psz_html, "&" );
- psz_html += 5;
+ /* We have a single < */
+ HtmlPut( &psz_html, "<" );
psz_subtitle++;
}
}
+ }
+ else if( *psz_subtitle == '&' )
+ {
+ if( !strncasecmp( psz_subtitle, "<", 4 ))
+ {
+ HtmlCopy( &psz_html, &psz_subtitle, "<" );
+ }
+ else if( !strncasecmp( psz_subtitle, ">", 4 ))
+ {
+ HtmlCopy( &psz_html, &psz_subtitle, ">" );
+ }
+ else if( !strncasecmp( psz_subtitle, "&", 5 ))
+ {
+ HtmlCopy( &psz_html, &psz_subtitle, "&" );
+ }
else
{
- *psz_html = *psz_subtitle;
- if( psz_html > psz_html_start )
- {
- /* Check for double whitespace */
- if((( *psz_html == ' ' ) ||
- ( *psz_html == '\t' )) &&
- (( *(psz_html-1) == ' ' ) ||
- ( *(psz_html-1) == '\t' )))
- {
- strcpy( psz_html, NO_BREAKING_SPACE );
- psz_html += strlen( NO_BREAKING_SPACE ) - 1;
- }
- }
- psz_html++;
+ HtmlPut( &psz_html, "&" );
psz_subtitle++;
}
-
- if( ( size_t )( psz_html - psz_html_start ) > i_buf_size - 10 )
+ }
+ else if( *psz_subtitle == '>' )
+ {
+ HtmlPut( &psz_html, ">" );
+ psz_subtitle++;
+ }
+ else if( psz_subtitle[0] == '{' && psz_subtitle[1] == '\\' &&
+ strchr( psz_subtitle, '}' ) )
+ {
+ /* Check for forced alignment */
+ if( !b_has_align &&
+ !strncmp( psz_subtitle, "{\\an", 4 ) && psz_subtitle[4] >= '1' && psz_subtitle[4] <= '9' && psz_subtitle[5] == '}' )
{
- int i_len = psz_html - psz_html_start;
+ static const int pi_vertical[3] = { SUBPICTURE_ALIGN_BOTTOM, 0, SUBPICTURE_ALIGN_TOP };
+ static const int pi_horizontal[3] = { SUBPICTURE_ALIGN_LEFT, 0, SUBPICTURE_ALIGN_RIGHT };
+ const int i_id = psz_subtitle[4] - '1';
- i_buf_size += 100;
- psz_html_start = realloc( psz_html_start, i_buf_size );
- psz_html = psz_html_start + i_len;
- *psz_html = '\0';
+ b_has_align = true;
+ *pi_align = pi_vertical[i_id/3] | pi_horizontal[i_id%3];
}
+ /* TODO fr -> rotation */
+
+ /* 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, "<br/>" );
+ 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;
+ if( psz_html > psz_html_start )
+ {
+ /* Check for double whitespace */
+ if( ( *psz_html == ' ' || *psz_html == '\t' ) &&
+ ( *(psz_html-1) == ' ' || *(psz_html-1) == '\t' ) )
+ {
+ HtmlPut( &psz_html, NO_BREAKING_SPACE );
+ psz_html--;
+ }
+ }
+ psz_html++;
+ psz_subtitle++;
}
- strcpy( psz_html, "</text>" );
- psz_html += 7;
- if( psz_tagStack[ 0 ] != '\0' )
+ if( ( size_t )( psz_html - psz_html_start ) > i_buf_size - 50 )
{
- /* Not well formed -- kill everything */
- free( psz_html_start );
- psz_html_start = NULL;
+ const int i_len = psz_html - psz_html_start;
+
+ i_buf_size += 200;
+ 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];
}
- else if( psz_html_start )
+ }
+ if( psz_html_start )
+ {
+ static const char *psz_text_close = "</text>";
+ 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 )
{
- /* Shrink the memory requirements */
- psz_html_start = realloc( psz_html_start, psz_html - psz_html_start + 1 );
+ 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, "</b>" );
+ break;
+ case 'i':
+ HtmlPut( &psz_html, "</i>" );
+ break;
+ case 'u':
+ HtmlPut( &psz_html, "</u>" );
+ break;
+ case 's':
+ HtmlPut( &psz_html, "</s>" );
+ break;
+ case 'f':
+ HtmlPut( &psz_html, "/font>" );
+ break;
+ case 'I':
+ break;
+ }
+
+ *psz_last = '\0';
+ }
+ HtmlPut( &psz_html, psz_text_close );
}
}
+ free( psz_tag );
+
return psz_html_start;
}
+