X-Git-Url: http://git.osdn.jp/view?a=blobdiff_plain;f=libhb%2Fdecsrtsub.c;h=daf210d6f1358712872877daf376dcb12cd5867e;hb=8d3d6aaf3b18b8694ae556bf8c2d0d1c03a1a210;hp=05fdb7646769baeed028ffb3b4f5189aeff88bd0;hpb=3548036ec5601f2a5033ca7e256ec4fe088482b8;p=handbrake-jp%2Fhandbrake-jp-git.git diff --git a/libhb/decsrtsub.c b/libhb/decsrtsub.c index 05fdb764..daf210d6 100644 --- a/libhb/decsrtsub.c +++ b/libhb/decsrtsub.c @@ -17,6 +17,7 @@ struct start_and_end { enum { k_state_inEntry, + k_state_inEntry_or_new, k_state_potential_new_entry, k_state_timecode, }; @@ -25,6 +26,7 @@ typedef struct srt_entry_s { long offset, duration; long start, stop; char text[1024]; + int pos; } srt_entry_t; /* @@ -32,10 +34,18 @@ typedef struct srt_entry_s { */ struct hb_work_private_s { - hb_job_t *job; - FILE *file; + hb_job_t * job; + FILE * file; + char buf[1024]; + int pos; + int end; + char utf8_buf[2048]; + int utf8_pos; + int utf8_end; + int utf8_bom_skipped; unsigned long current_time; unsigned long number_of_entries; + unsigned long last_entry_number; unsigned long current_state; srt_entry_t current_entry; iconv_t *iconv_context; @@ -44,20 +54,138 @@ struct hb_work_private_s uint64_t stop_time; // In HB time }; -static struct start_and_end read_time_from_string( const char* timeString ) +static int +read_time_from_string( const char* timeString, struct start_and_end *result ) { // for ex. 00:00:15,248 --> 00:00:16,545 long houres1, minutes1, seconds1, milliseconds1, - houres2, minutes2, seconds2, milliseconds2; + houres2, minutes2, seconds2, milliseconds2; + int scanned; - sscanf(timeString, "%ld:%ld:%ld,%ld --> %ld:%ld:%ld,%ld\n", &houres1, &minutes1, &seconds1, &milliseconds1, - &houres2, &minutes2, &seconds2, &milliseconds2); - - struct start_and_end result = { - milliseconds1 + seconds1*1000 + minutes1*60*1000 + houres1*60*60*1000, - milliseconds2 + seconds2*1000 + minutes2*60*1000 + houres2*60*60*1000}; - return result; + scanned = sscanf(timeString, "%ld:%ld:%ld,%ld --> %ld:%ld:%ld,%ld\n", + &houres1, &minutes1, &seconds1, &milliseconds1, + &houres2, &minutes2, &seconds2, &milliseconds2); + if (scanned != 8) + { + return 0; + } + result->start = + milliseconds1 + seconds1*1000 + minutes1*60*1000 + houres1*60*60*1000; + result->end = + milliseconds2 + seconds2*1000 + minutes2*60*1000 + houres2*60*60*1000; + return 1; +} + +static int utf8_fill( hb_work_private_t * pv ) +{ + int bytes, conversion = 0; + size_t out_size; + + /* Align utf8 data to beginning of the buffer so that we can + * fill the buffer to its maximum */ + memmove( pv->utf8_buf, pv->utf8_buf + pv->utf8_pos, pv->utf8_end - pv->utf8_pos ); + pv->utf8_end -= pv->utf8_pos; + pv->utf8_pos = 0; + out_size = 2048 - pv->utf8_end; + while( out_size ) + { + char *p, *q; + size_t in_size, retval; + + if( pv->end == pv->pos ) + { + bytes = fread( pv->buf, 1, 1024, pv->file ); + pv->pos = 0; + pv->end = bytes; + if( bytes == 0 ) + { + if( conversion ) + return 1; + else + return 0; + } + } + + p = pv->buf + pv->pos; + q = pv->utf8_buf + pv->utf8_end; + in_size = pv->end - pv->pos; + + retval = iconv( pv->iconv_context, &p, &in_size, &q, &out_size); + if( q != pv->utf8_buf + pv->utf8_pos ) + conversion = 1; + + pv->utf8_end = q - pv->utf8_buf; + pv->pos = p - pv->buf; + + if ( !pv->utf8_bom_skipped ) + { + uint8_t *buf = (uint8_t*)pv->utf8_buf; + if (buf[0] == 0xef && buf[1] == 0xbb && buf[2] == 0xbf) + { + pv->utf8_pos = 3; + } + pv->utf8_bom_skipped = 1; + } + + if( ( retval == -1 ) && ( errno == EINVAL ) ) + { + /* Incomplete multibyte sequence, read more data */ + memmove( pv->buf, p, pv->end - pv->pos ); + pv->end -= pv->pos; + pv->pos = 0; + bytes = fread( pv->buf + pv->end, 1, 1024 - pv->end, pv->file ); + if( bytes == 0 ) + { + if( !conversion ) + return 0; + else + return 1; + } + pv->end += bytes; + } else if ( ( retval == -1 ) && ( errno == EILSEQ ) ) + { + hb_error( "Invalid byte for codeset in input, discard byte" ); + /* Try the next byte of the input */ + pv->pos++; + } else if ( ( retval == -1 ) && ( errno == E2BIG ) ) + { + /* buffer full */ + return conversion; + } + } + return 1; +} + +static int get_line( hb_work_private_t * pv, char *buf, int size ) +{ + int i; + char c; + + /* Find newline in converted UTF-8 buffer */ + for( i = 0; i < size - 1; i++ ) + { + if( pv->utf8_pos >= pv->utf8_end ) + { + if( !utf8_fill( pv ) ) + { + if( i ) + return 1; + else + return 0; + } + } + c = pv->utf8_buf[pv->utf8_pos++]; + if( c == '\n' ) + { + buf[i] = '\n'; + buf[i+1] = '\0'; + return 1; + } + buf[i] = c; + } + buf[0] = '\0'; + return 1; } /* @@ -65,21 +193,31 @@ static struct start_and_end read_time_from_string( const char* timeString ) */ static hb_buffer_t *srt_read( hb_work_private_t *pv ) { - char line_buffer[1024]; + int reprocess = 0, resync = 0; if( !pv->file ) { return NULL; } - while( fgets( line_buffer, sizeof( line_buffer ), pv->file ) ) + while( reprocess || get_line( pv, line_buffer, sizeof( line_buffer ) ) ) { + reprocess = 0; switch (pv->current_state) { case k_state_timecode: { - struct start_and_end timing = read_time_from_string( line_buffer ); + struct start_and_end timing; + int result; + + result = read_time_from_string( line_buffer, &timing ); + if (!result) + { + resync = 1; + pv->current_state = k_state_potential_new_entry; + continue; + } pv->current_entry.duration = timing.end - timing.start; pv->current_entry.offset = timing.start - pv->current_time; @@ -89,15 +227,40 @@ static hb_buffer_t *srt_read( hb_work_private_t *pv ) pv->current_entry.stop = timing.end; pv->current_state = k_state_inEntry; - continue; + continue; + } + + case k_state_inEntry_or_new: + { + char *endpoint; + long entry_number; + /* + * Is this really new next entry begin? + */ + entry_number = strtol(line_buffer, &endpoint, 10); + if (endpoint == line_buffer || + (endpoint && *endpoint != '\n' && *endpoint != '\r')) + { + /* + * Doesn't resemble an entry number + * must still be in an entry + */ + if (!resync) + { + reprocess = 1; + pv->current_state = k_state_inEntry; + } + continue; + } + reprocess = 1; + pv->current_state = k_state_potential_new_entry; + break; } - + case k_state_inEntry: { - char *p, *q; - size_t in_size; - size_t out_size; - size_t retval; + char *q; + int size, len; // If the current line is empty, we assume this is the // seperation betwene two entries. In case we are wrong, @@ -107,102 +270,185 @@ static hb_buffer_t *srt_read( hb_work_private_t *pv ) continue; } - - for( q = pv->current_entry.text; (q < pv->current_entry.text+1024) && *q; q++); - - p = line_buffer; - - in_size = strlen(line_buffer); - out_size = (pv->current_entry.text+1024) - q; - - retval = iconv( pv->iconv_context, &p, &in_size, &q, &out_size); - *q = '\0'; - - if( ( retval == -1 ) && ( errno == EINVAL ) ) - { - hb_error( "Invalid shift sequence" ); - } else if ( ( retval == -1 ) && ( errno == EILSEQ ) ) - { - hb_error( "Invalid byte for codeset in input, %"PRId64" bytes discarded", (int64_t)in_size); - } else if ( ( retval == -1 ) && ( errno == E2BIG ) ) - { - hb_error( "Not enough space in output buffer"); - } - - break; + q = pv->current_entry.text + pv->current_entry.pos; + len = strlen( line_buffer ); + size = MIN(1024 - pv->current_entry.pos - 1, len ); + memcpy(q, line_buffer, size); + pv->current_entry.pos += size; + pv->current_entry.text[pv->current_entry.pos] = '\0'; + break; } - + case k_state_potential_new_entry: { - const char endpoint[] = "\0"; - const unsigned long potential_entry_number = strtol(line_buffer, (char**)&endpoint, 10); + char *endpoint; + long entry_number; hb_buffer_t *buffer = NULL; /* * Is this really new next entry begin? */ - if (potential_entry_number == pv->number_of_entries + 1) { + entry_number = strtol(line_buffer, &endpoint, 10); + if (!resync && (*line_buffer == '\n' || *line_buffer == '\r')) + { /* - * We found the next entry - or a really rare error condition + * Well.. looks like we are in the wrong mode.. lets add the + * newline we misinterpreted... */ - if( *pv->current_entry.text ) + strncat(pv->current_entry.text, " ", 1024); + pv->current_state = k_state_inEntry_or_new; + continue; + } + if (endpoint == line_buffer || + (endpoint && *endpoint != '\n' && *endpoint != '\r')) + { + /* + * Well.. looks like we are in the wrong mode.. lets add the + * line we misinterpreted... + */ + if (!resync) { - long length; - char *p; - uint64_t start_time = ( pv->current_entry.start + - pv->subtitle->config.offset ) * 90; - uint64_t stop_time = ( pv->current_entry.stop + - pv->subtitle->config.offset ) * 90; - - if( !( start_time > pv->start_time && stop_time < pv->stop_time ) ) - { - hb_deep_log( 3, "Discarding SRT at time start %"PRId64", stop %"PRId64, start_time, stop_time); - memset( &pv->current_entry, 0, sizeof( srt_entry_t ) ); - ++(pv->number_of_entries); - pv->current_state = k_state_timecode; - continue; - } + reprocess = 1; + pv->current_state = k_state_inEntry; + } + continue; + } + /* + * We found the next entry - or a really rare error condition + */ + pv->last_entry_number = entry_number; + resync = 0; + if( *pv->current_entry.text ) + { + long length; + char *p, *q; + int line = 1; + uint64_t start_time = ( pv->current_entry.start + + pv->subtitle->config.offset ) * 90; + uint64_t stop_time = ( pv->current_entry.stop + + pv->subtitle->config.offset ) * 90; + + if( !( start_time > pv->start_time && stop_time < pv->stop_time ) ) + { + hb_deep_log( 3, "Discarding SRT at time start %"PRId64", stop %"PRId64, start_time, stop_time); + memset( &pv->current_entry, 0, sizeof( srt_entry_t ) ); + ++(pv->number_of_entries); + pv->current_state = k_state_timecode; + continue; + } - length = strlen( pv->current_entry.text ); + length = strlen( pv->current_entry.text ); - for( p = pv->current_entry.text; *p; p++) + for( q = p = pv->current_entry.text; *p; p++) + { + if( *p == '\n' ) { - if( *p == '\n' || *p == '\r' ) + if ( line == 1 ) + { + *q = *p; + line = 2; + } + else { - *p = ' '; + *q = ' '; } + q++; } - - buffer = hb_buffer_init( length + 1 ); - - if( buffer ) + else if( *p != '\r' ) { - buffer->start = start_time - pv->start_time; - buffer->stop = stop_time - pv->start_time; - - memcpy( buffer->data, pv->current_entry.text, length + 1 ); + *q = *p; + q++; + } + else + { + length--; } } - memset( &pv->current_entry, 0, sizeof( srt_entry_t ) ); - ++(pv->number_of_entries); - pv->current_state = k_state_timecode; + *q = '\0'; + + buffer = hb_buffer_init( length + 1 ); + if( buffer ) { - return buffer; + buffer->start = start_time - pv->start_time; + buffer->stop = stop_time - pv->start_time; + + memcpy( buffer->data, pv->current_entry.text, length + 1 ); } - continue; - } else { - /* - * Well.. looks like we are in the wrong mode.. lets add the - * newline we misinterpreted... - */ - strncat(pv->current_entry.text, " ", 1024); - pv->current_state = k_state_inEntry; } - - break; + memset( &pv->current_entry, 0, sizeof( srt_entry_t ) ); + ++(pv->number_of_entries); + pv->current_state = k_state_timecode; + if( buffer ) + { + return buffer; + } + continue; + } + } + } + + hb_buffer_t *buffer = NULL; + if( *pv->current_entry.text ) + { + long length; + char *p, *q; + int line = 1; + uint64_t start_time = ( pv->current_entry.start + + pv->subtitle->config.offset ) * 90; + uint64_t stop_time = ( pv->current_entry.stop + + pv->subtitle->config.offset ) * 90; + + if( !( start_time > pv->start_time && stop_time < pv->stop_time ) ) + { + hb_deep_log( 3, "Discarding SRT at time start %"PRId64", stop %"PRId64, start_time, stop_time); + memset( &pv->current_entry, 0, sizeof( srt_entry_t ) ); + return NULL; + } + + length = strlen( pv->current_entry.text ); + + for( q = p = pv->current_entry.text; *p; p++) + { + if( *p == '\n' ) + { + if ( line == 1 ) + { + *q = *p; + line = 2; + } + else + { + *q = ' '; + } + q++; + } + else if( *p != '\r' ) + { + *q = *p; + q++; + } + else + { + length--; + } } + *q = '\0'; + + buffer = hb_buffer_init( length + 1 ); + + if( buffer ) + { + buffer->start = start_time - pv->start_time; + buffer->stop = stop_time - pv->start_time; + + memcpy( buffer->data, pv->current_entry.text, length + 1 ); } } + memset( &pv->current_entry, 0, sizeof( srt_entry_t ) ); + if( buffer ) + { + return buffer; + } return NULL; } @@ -226,10 +472,9 @@ static int decsrtInit( hb_work_object_t * w, hb_job_t * job ) buffer = hb_buffer_init( 0 ); hb_fifo_push( w->fifo_in, buffer); - pv->file = fopen( w->subtitle->config.src_filename, "r" ); - pv->current_state = k_state_potential_new_entry; pv->number_of_entries = 0; + pv->last_entry_number = 0; pv->current_time = 0; pv->subtitle = w->subtitle; @@ -249,19 +494,22 @@ static int decsrtInit( hb_work_object_t * w, hb_job_t * job ) retval = 0; } } - chapter = hb_list_item( title->list_chapter, i - 1 ); - - if( chapter ) + pv->stop_time = pv->start_time; + for( i = job->chapter_start; i <= job->chapter_end; ++i ) { - pv->stop_time = pv->start_time + chapter->duration; - } else { - hb_error( "Could not locate chapter %d for SRT stop time", i ); - retval = 0; + chapter = hb_list_item( title->list_chapter, i - 1 ); + if( chapter ) + { + pv->stop_time += chapter->duration; + } else { + hb_error( "Could not locate chapter %d for SRT start time", i ); + retval = 0; + } } hb_deep_log( 3, "SRT Start time %"PRId64", stop time %"PRId64, pv->start_time, pv->stop_time); - pv->iconv_context = iconv_open( "utf8", pv->subtitle->config.src_codeset ); + pv->iconv_context = iconv_open( "utf-8", pv->subtitle->config.src_codeset ); if( pv->iconv_context == (iconv_t) -1 )