X-Git-Url: https://git.sesse.net/?a=blobdiff_plain;f=src%2Ftext%2Fstrings.c;h=a37fe165ead0f408ac1f694240d81a0f4fdc7c45;hb=8524d44bbbfa4a3aabf1176b1d41a295b603180a;hp=532e54c949cfe010539e3ccdbf82beb583bc0633;hpb=e148e58dc2e9db6a2269bdaba653d14d26c0d721;p=vlc diff --git a/src/text/strings.c b/src/text/strings.c index 532e54c949..a37fe165ea 100644 --- a/src/text/strings.c +++ b/src/text/strings.c @@ -2,6 +2,7 @@ * strings.c: String related functions ***************************************************************************** * Copyright (C) 2006 the VideoLAN team + * Copyright (C) 2008-2009 Rémi Denis-Courmont * $Id$ * * Authors: Antoine Cellerier @@ -26,14 +27,16 @@ /***************************************************************************** * Preamble *****************************************************************************/ -#include -#include -#include -#include +#ifdef HAVE_CONFIG_H +# include "config.h" +#endif + +#include #include /* Needed by str_format_time */ #include +#include /* Needed by str_format_meta */ #include @@ -46,88 +49,7 @@ #include /** - * Unescape URI encoded string - * \return decoded duplicated string - */ -char *unescape_URI_duplicate( const char *psz ) -{ - char *psz_dup = strdup( psz ); - unescape_URI( psz_dup ); - return psz_dup; -} - -/** - * Unescape URI encoded string in place - * \return nothing - */ -void unescape_URI( char *psz ) -{ - unsigned char *in = (unsigned char *)psz, *out = in, c; - - while( ( c = *in++ ) != '\0' ) - { - switch( c ) - { - case '%': - { - char val[5], *pval = val; - unsigned long cp; - - switch( c = *in++ ) - { - case '\0': - return; - - case 'u': - case 'U': - if( ( *pval++ = *in++ ) == '\0' ) - return; - if( ( *pval++ = *in++ ) == '\0' ) - return; - c = *in++; - - default: - *pval++ = c; - if( ( *pval++ = *in++ ) == '\0' ) - return; - *pval = '\0'; - } - - cp = strtoul( val, NULL, 0x10 ); - if( cp < 0x80 ) - *out++ = cp; - else - if( cp < 0x800 ) - { - *out++ = (( cp >> 6) | 0xc0); - *out++ = (( cp & 0x3f) | 0x80); - } - else - { - assert( cp < 0x10000 ); - *out++ = (( cp >> 12) | 0xe0); - *out++ = (((cp >> 6) & 0x3f) | 0x80); - *out++ = (( cp & 0x3f) | 0x80); - } - break; - } - - /* + is not a special case - it means plus, not space. */ - - default: - /* Inserting non-ASCII or non-printable characters is unsafe, - * and no sane browser will send these unencoded */ - if( ( c < 32 ) || ( c > 127 ) ) - *out++ = '?'; - else - *out++ = c; - } - } - *out = '\0'; -} - -/** - * Decode encoded URI string + * Decode encoded URI component. See also decode_URI(). * \return decoded duplicated string */ char *decode_URI_duplicate( const char *psz ) @@ -138,13 +60,24 @@ char *decode_URI_duplicate( const char *psz ) } /** - * Decode encoded URI string in place - * \return nothing + * Decode an encoded URI component in place. + * This function does NOT decode entire URIs. + * It decodes components (e.g. host name, directory, file name). + * Decoded URIs do not exist in the real world (see RFC3986 §2.4). + * Complete URIs are always "encoded" (or they are syntaxically invalid). + * + * Note that URI encoding is different from Javascript escaping. Especially, + * white spaces and Unicode non-ASCII code points are encoded differently. + * + * \return psz on success, NULL if it was not properly encoded */ -void decode_URI( char *psz ) +char *decode_URI( char *psz ) { unsigned char *in = (unsigned char *)psz, *out = in, c; + if( psz == NULL ) + return NULL; + while( ( c = *in++ ) != '\0' ) { switch( c ) @@ -155,14 +88,14 @@ void decode_URI( char *psz ) if( ( ( hex[0] = *in++ ) == 0 ) || ( ( hex[1] = *in++ ) == 0 ) ) - return; + return NULL; hex[2] = '\0'; *out++ = (unsigned char)strtoul( hex, NULL, 0x10 ); break; } - case '+': + case '+': /* This is HTTP forms, not URI decoding... */ *out++ = ' '; break; @@ -177,59 +110,200 @@ void decode_URI( char *psz ) } *out = '\0'; EnsureUTF8( psz ); + return psz; } -static inline int isurlsafe( int c ) +static inline bool isurisafe( int c ) { + /* These are the _unreserved_ URI characters (RFC3986 §2.3) */ return ( (unsigned char)( c - 'a' ) < 26 ) || ( (unsigned char)( c - 'A' ) < 26 ) || ( (unsigned char)( c - '0' ) < 10 ) - /* Hmm, we should not encode character that are allowed in URLs - * (even if they are not URL-safe), nor URL-safe characters. - * We still encode some of them because of Microsoft's crap browser. - */ - || ( strchr( "-_.", c ) != NULL ); + || ( strchr( "-._~", c ) != NULL ); } -static inline char url_hexchar( int c ) +static char *encode_URI_bytes (const char *psz_uri, size_t len) { - return ( c < 10 ) ? c + '0' : c + 'A' - 10; + char *psz_enc = malloc (3 * len + 1), *out = psz_enc; + if (psz_enc == NULL) + return NULL; + + for (size_t i = 0; i < len; i++) + { + static const char hex[16] = "0123456789ABCDEF"; + uint8_t c = *psz_uri; + + if( isurisafe( c ) ) + *out++ = c; + /* This is URI encoding, not HTTP forms: + * Space is encoded as '%20', not '+'. */ + else + { + *out++ = '%'; + *out++ = hex[c >> 4]; + *out++ = hex[c & 0xf]; + } + psz_uri++; + } + *out++ = '\0'; + + out = realloc (psz_enc, out - psz_enc); + return out ? out : psz_enc; /* realloc() can fail (safe) */ } /** - * encode_URI_component - * Encodes an URI component. + * Encodes an URI component (RFC3986 §2). * - * @param psz_url nul-terminated UTF-8 representation of the component. + * @param psz_uri nul-terminated UTF-8 representation of the component. * Obviously, you can't pass an URI containing a nul character, but you don't * want to do that, do you? * - * @return encoded string (must be free()'d) + * @return encoded string (must be free()'d), or NULL for ENOMEM. */ -char *encode_URI_component( const char *psz_url ) +char *encode_URI_component( const char *psz_uri ) { - char psz_enc[3 * strlen( psz_url ) + 1], *out = psz_enc; - const uint8_t *in; + return encode_URI_bytes (psz_uri, strlen (psz_uri)); +} - for( in = (const uint8_t *)psz_url; *in; in++ ) - { - uint8_t c = *in; - if( isurlsafe( c ) ) - *out++ = (char)c; - else - if ( c == ' ') - *out++ = '+'; - else - { - *out++ = '%'; - *out++ = url_hexchar( c >> 4 ); - *out++ = url_hexchar( c & 0xf ); - } - } - *out++ = '\0'; +static const struct xml_entity_s +{ + char psz_entity[8]; + char psz_char[4]; +} xml_entities[] = { + /* Important: this list has to be in alphabetical order (psz_entity-wise) */ + { "AElig;", "Æ" }, + { "Aacute;", "Á" }, + { "Acirc;", "Â" }, + { "Agrave;", "À" }, + { "Aring;", "Å" }, + { "Atilde;", "Ã" }, + { "Auml;", "Ä" }, + { "Ccedil;", "Ç" }, + { "Dagger;", "‡" }, + { "ETH;", "Ð" }, + { "Eacute;", "É" }, + { "Ecirc;", "Ê" }, + { "Egrave;", "È" }, + { "Euml;", "Ë" }, + { "Iacute;", "Í" }, + { "Icirc;", "Î" }, + { "Igrave;", "Ì" }, + { "Iuml;", "Ï" }, + { "Ntilde;", "Ñ" }, + { "OElig;", "Œ" }, + { "Oacute;", "Ó" }, + { "Ocirc;", "Ô" }, + { "Ograve;", "Ò" }, + { "Oslash;", "Ø" }, + { "Otilde;", "Õ" }, + { "Ouml;", "Ö" }, + { "Scaron;", "Š" }, + { "THORN;", "Þ" }, + { "Uacute;", "Ú" }, + { "Ucirc;", "Û" }, + { "Ugrave;", "Ù" }, + { "Uuml;", "Ü" }, + { "Yacute;", "Ý" }, + { "Yuml;", "Ÿ" }, + { "aacute;", "á" }, + { "acirc;", "â" }, + { "acute;", "´" }, + { "aelig;", "æ" }, + { "agrave;", "à" }, + { "amp;", "&" }, + { "apos;", "'" }, + { "aring;", "å" }, + { "atilde;", "ã" }, + { "auml;", "ä" }, + { "bdquo;", "„" }, + { "brvbar;", "¦" }, + { "ccedil;", "ç" }, + { "cedil;", "¸" }, + { "cent;", "¢" }, + { "circ;", "ˆ" }, + { "copy;", "©" }, + { "curren;", "¤" }, + { "dagger;", "†" }, + { "deg;", "°" }, + { "divide;", "÷" }, + { "eacute;", "é" }, + { "ecirc;", "ê" }, + { "egrave;", "è" }, + { "eth;", "ð" }, + { "euml;", "ë" }, + { "euro;", "€" }, + { "frac12;", "½" }, + { "frac14;", "¼" }, + { "frac34;", "¾" }, + { "gt;", ">" }, + { "hellip;", "…" }, + { "iacute;", "í" }, + { "icirc;", "î" }, + { "iexcl;", "¡" }, + { "igrave;", "ì" }, + { "iquest;", "¿" }, + { "iuml;", "ï" }, + { "laquo;", "«" }, + { "ldquo;", "“" }, + { "lsaquo;", "‹" }, + { "lsquo;", "‘" }, + { "lt;", "<" }, + { "macr;", "¯" }, + { "mdash;", "—" }, + { "micro;", "µ" }, + { "middot;", "·" }, + { "nbsp;", "\xc2\xa0" }, + { "ndash;", "–" }, + { "not;", "¬" }, + { "ntilde;", "ñ" }, + { "oacute;", "ó" }, + { "ocirc;", "ô" }, + { "oelig;", "œ" }, + { "ograve;", "ò" }, + { "ordf;", "ª" }, + { "ordm;", "º" }, + { "oslash;", "ø" }, + { "otilde;", "õ" }, + { "ouml;", "ö" }, + { "para;", "¶" }, + { "permil;", "‰" }, + { "plusmn;", "±" }, + { "pound;", "£" }, + { "quot;", "\"" }, + { "raquo;", "»" }, + { "rdquo;", "”" }, + { "reg;", "®" }, + { "rsaquo;", "›" }, + { "rsquo;", "’" }, + { "sbquo;", "‚" }, + { "scaron;", "š" }, + { "sect;", "§" }, + { "shy;", "­" }, + { "sup1;", "¹" }, + { "sup2;", "²" }, + { "sup3;", "³" }, + { "szlig;", "ß" }, + { "thorn;", "þ" }, + { "tilde;", "˜" }, + { "times;", "×" }, + { "trade;", "™" }, + { "uacute;", "ú" }, + { "ucirc;", "û" }, + { "ugrave;", "ù" }, + { "uml;", "¨" }, + { "uuml;", "ü" }, + { "yacute;", "ý" }, + { "yen;", "¥" }, + { "yuml;", "ÿ" }, +}; + +static int cmp_entity (const void *key, const void *elem) +{ + const struct xml_entity_s *ent = elem; + const char *name = key; - return strdup( psz_enc ); + return strncmp (name, ent->psz_entity, strlen (ent->psz_entity)); } /** @@ -242,30 +316,73 @@ void resolve_xml_special_chars( char *psz_value ) while ( *psz_value ) { - if( !strncmp( psz_value, "<", 4 ) ) - { - *p_pos = '<'; - psz_value += 4; - } - else if( !strncmp( psz_value, ">", 4 ) ) - { - *p_pos = '>'; - psz_value += 4; - } - else if( !strncmp( psz_value, "&", 5 ) ) - { - *p_pos = '&'; - psz_value += 5; - } - else if( !strncmp( psz_value, """, 6 ) ) + if( *psz_value == '&' ) { - *p_pos = '\"'; - psz_value += 6; - } - else if( !strncmp( psz_value, "'", 6 ) ) - { - *p_pos = '\''; - psz_value += 6; + if( psz_value[1] == '#' ) + { /* &#xxx; Unicode code point */ + char *psz_end; + unsigned long cp = strtoul( psz_value+2, &psz_end, 10 ); + if( *psz_end == ';' ) + { + psz_value = psz_end + 1; + if( cp == 0 ) + (void)0; /* skip nuls */ + else + if( cp <= 0x7F ) + { + *p_pos = cp; + } + else + /* Unicode code point outside ASCII. + * &#xxx; representation is longer than UTF-8 :) */ + if( cp <= 0x7FF ) + { + *p_pos++ = 0xC0 | (cp >> 6); + *p_pos = 0x80 | (cp & 0x3F); + } + else + if( cp <= 0xFFFF ) + { + *p_pos++ = 0xE0 | (cp >> 12); + *p_pos++ = 0x80 | ((cp >> 6) & 0x3F); + *p_pos = 0x80 | (cp & 0x3F); + } + else + if( cp <= 0x1FFFFF ) /* Outside the BMP */ + { /* Unicode stops at 10FFFF, but who cares? */ + *p_pos++ = 0xF0 | (cp >> 18); + *p_pos++ = 0x80 | ((cp >> 12) & 0x3F); + *p_pos++ = 0x80 | ((cp >> 6) & 0x3F); + *p_pos = 0x80 | (cp & 0x3F); + } + } + else + { + /* Invalid entity number */ + *p_pos = *psz_value; + psz_value++; + } + } + else + { /* Well-known XML entity */ + const struct xml_entity_s *ent; + + ent = bsearch (psz_value + 1, xml_entities, + sizeof (xml_entities) / sizeof (*ent), + sizeof (*ent), cmp_entity); + if (ent != NULL) + { + size_t olen = strlen (ent->psz_char); + memcpy (p_pos, ent->psz_char, olen); + p_pos += olen - 1; + psz_value += strlen (ent->psz_entity) + 1; + } + else + { /* No match */ + *p_pos = *psz_value; + psz_value++; + } + } } else { @@ -285,47 +402,36 @@ void resolve_xml_special_chars( char *psz_value ) */ char *convert_xml_special_chars( const char *psz_content ) { - char *psz_temp = malloc( 6 * strlen( psz_content ) + 1 ); - const char *p_from = psz_content; + assert( psz_content ); + + const size_t len = strlen( psz_content ); + char *const psz_temp = malloc( 6 * len + 1 ); char *p_to = psz_temp; - while ( *p_from ) + if( psz_temp == NULL ) + return NULL; + for( size_t i = 0; i < len; i++ ) { - if ( *p_from == '<' ) - { - strcpy( p_to, "<" ); - p_to += 4; - } - else if ( *p_from == '>' ) - { - strcpy( p_to, ">" ); - p_to += 4; - } - else if ( *p_from == '&' ) - { - strcpy( p_to, "&" ); - p_to += 5; - } - else if( *p_from == '\"' ) - { - strcpy( p_to, """ ); - p_to += 6; - } - else if( *p_from == '\'' ) - { - strcpy( p_to, "'" ); - p_to += 6; - } - else + const char *str; + char c = psz_content[i]; + + switch ( c ) { - *p_to = *p_from; - p_to++; + case '\"': str = "quot"; break; + case '&': str = "amp"; break; + case '\'': str = "#39"; break; + case '<': str = "lt"; break; + case '>': str = "gt"; break; + default: + *(p_to++) = c; + continue; } - p_from++; + p_to += sprintf( p_to, "&%s;", str ); } - *p_to = '\0'; + *(p_to++) = '\0'; - return psz_temp; + p_to = realloc( psz_temp, p_to - psz_temp ); + return p_to ? p_to : psz_temp; /* cannot fail */ } /* Base64 encoding */ @@ -410,7 +516,7 @@ size_t vlc_b64_decode_binary_to_buffer( uint8_t *p_dst, size_t i_dst, const char int i_level; int i_last; - for( i_level = 0, i_last = 0; i_dst > 0 && *p != '\0'; i_dst--, p++ ) + for( i_level = 0, i_last = 0; (size_t)( p_dst - p_start ) < i_dst && *p != '\0'; p++ ) { const int c = b64[(unsigned int)*p]; if( c == -1 ) @@ -462,70 +568,83 @@ char *vlc_b64_decode( const char *psz_src ) return p_dst; } -/**************************************************************************** - * String formating functions - ****************************************************************************/ +/** + * Formats current time into a heap-allocated string. + * @param tformat time format (as with C strftime()) + * @return an allocated string (must be free()'d), or NULL on memory error. + */ char *str_format_time( const char *tformat ) { - char buffer[255]; time_t curtime; -#if defined(HAVE_LOCALTIME_R) struct tm loctime; -#else - struct tm *loctime; -#endif + + if (strcmp (tformat, "") == 0) + return strdup (""); /* corner case w.r.t. strftime() return value */ /* Get the current time. */ - curtime = time( NULL ); + time( &curtime ); /* Convert it to local time representation. */ -#if defined(HAVE_LOCALTIME_R) localtime_r( &curtime, &loctime ); - strftime( buffer, 255, tformat, &loctime ); -#else - loctime = localtime( &curtime ); - strftime( buffer, 255, tformat, loctime ); -#endif - return strdup( buffer ); + for (size_t buflen = strlen (tformat) + 32;; buflen += 32) + { + char *str = malloc (buflen); + if (str == NULL) + return NULL; + + size_t len = strftime (str, buflen, tformat, &loctime); + if (len > 0) + { + char *ret = realloc (str, len + 1); + return ret ? ret : str; /* <- this cannot fail */ + } + } + assert (0); } -#define INSERT_STRING( check, string ) \ - if( check && string ) \ +#define INSERT_STRING( string ) \ + if( string != NULL ) \ { \ int len = strlen( string ); \ - dst = realloc( dst, \ - i_size = i_size + len + 1 ); \ - strncpy( d, string, len+1 ); \ + dst = xrealloc( dst, i_size = i_size + len );\ + memcpy( (dst+d), string, len ); \ d += len; \ + free( string ); \ } \ - else \ + else if( !b_empty_if_na ) \ { \ - *d = '-'; \ + *(dst+d) = '-'; \ d++; \ + } \ + +/* same than INSERT_STRING, except that string won't be freed */ +#define INSERT_STRING_NO_FREE( string ) \ + { \ + int len = strlen( string ); \ + dst = xrealloc( dst, i_size = i_size + len );\ + memcpy( dst+d, string, len ); \ + d += len; \ } char *__str_format_meta( vlc_object_t *p_object, const char *string ) { const char *s = string; - char *dst = malloc( 1000 ); - char *d = dst; - int b_is_format = 0; + bool b_is_format = false; + bool b_empty_if_na = false; char buf[10]; - int i_size = strlen( string ); + int i_size = strlen( string ) + 1; /* +1 to store '\0' */ + char *dst = strdup( string ); + if( !dst ) return NULL; + int d = 0; - playlist_t *p_playlist = pl_Yield( p_object ); - input_thread_t *p_input = p_playlist->p_input; + playlist_t *p_playlist = pl_Hold( p_object ); + input_thread_t *p_input = playlist_CurrentInput( p_playlist ); input_item_t *p_item = NULL; pl_Release( p_object ); if( p_input ) { - vlc_object_yield( p_input ); p_item = input_GetItem(p_input); - if( p_item ) - vlc_mutex_lock( &p_item->lock ); } - sprintf( dst, string ); - while( *s ) { if( b_is_format ) @@ -533,71 +652,106 @@ char *__str_format_meta( vlc_object_t *p_object, const char *string ) switch( *s ) { case 'a': - INSERT_STRING( p_item && p_item->p_meta, - p_item->p_meta->psz_artist ); + if( p_item ) + { + INSERT_STRING( input_item_GetArtist( p_item ) ); + } break; case 'b': - INSERT_STRING( p_item && p_item->p_meta, - p_item->p_meta->psz_album ); + if( p_item ) + { + INSERT_STRING( input_item_GetAlbum( p_item ) ); + } break; case 'c': - INSERT_STRING( p_item && p_item->p_meta, - p_item->p_meta->psz_copyright ); + if( p_item ) + { + INSERT_STRING( input_item_GetCopyright( p_item ) ); + } break; case 'd': - INSERT_STRING( p_item && p_item->p_meta, - p_item->p_meta->psz_description ); + if( p_item ) + { + INSERT_STRING( input_item_GetDescription( p_item ) ); + } break; case 'e': - INSERT_STRING( p_item && p_item->p_meta, - p_item->p_meta->psz_encodedby ); + if( p_item ) + { + INSERT_STRING( input_item_GetEncodedBy( p_item ) ); + } + break; + case 'f': + if( p_item && p_item->p_stats ) + { + vlc_mutex_lock( &p_item->p_stats->lock ); + snprintf( buf, 10, "%d", + p_item->p_stats->i_displayed_pictures ); + vlc_mutex_unlock( &p_item->p_stats->lock ); + } + else + { + sprintf( buf, b_empty_if_na ? "" : "-" ); + } + INSERT_STRING_NO_FREE( buf ); break; case 'g': - INSERT_STRING( p_item && p_item->p_meta, - p_item->p_meta->psz_genre ); + if( p_item ) + { + INSERT_STRING( input_item_GetGenre( p_item ) ); + } break; case 'l': - INSERT_STRING( p_item && p_item->p_meta, - p_item->p_meta->psz_language ); + if( p_item ) + { + INSERT_STRING( input_item_GetLanguage( p_item ) ); + } break; case 'n': - INSERT_STRING( p_item && p_item->p_meta, - p_item->p_meta->psz_tracknum ); + if( p_item ) + { + INSERT_STRING( input_item_GetTrackNum( p_item ) ); + } break; case 'p': - INSERT_STRING( p_item && p_item->p_meta, - p_item->p_meta->psz_nowplaying ); + if( p_item ) + { + INSERT_STRING( input_item_GetNowPlaying( p_item ) ); + } break; case 'r': - INSERT_STRING( p_item && p_item->p_meta, - p_item->p_meta->psz_rating ); + if( p_item ) + { + INSERT_STRING( input_item_GetRating( p_item ) ); + } break; case 's': { - char *lang; + char *lang = NULL; if( p_input ) - { - lang = var_GetString( p_input, "sub-language" ); - } - else - { - lang = strdup( "-" ); - } - INSERT_STRING( 1, lang ); - free( lang ); + lang = var_GetNonEmptyString( p_input, "sub-language" ); + if( lang == NULL ) + lang = strdup( b_empty_if_na ? "" : "-" ); + INSERT_STRING( lang ); break; } case 't': - INSERT_STRING( p_item && p_item->p_meta, - p_item->p_meta->psz_title ); + if( p_item ) + { + INSERT_STRING( input_item_GetTitle( p_item ) ); + } break; case 'u': - INSERT_STRING( p_item && p_item->p_meta, - p_item->p_meta->psz_url ); + if( p_item ) + { + INSERT_STRING( input_item_GetURL( p_item ) ); + } break; case 'A': - INSERT_STRING( p_item && p_item->p_meta, - p_item->p_meta->psz_date ); + if( p_item ) + { + INSERT_STRING( input_item_GetDate( p_item ) ); + } break; case 'B': if( p_input ) @@ -607,9 +761,9 @@ char *__str_format_meta( vlc_object_t *p_object, const char *string ) } else { - sprintf( buf, "-" ); + sprintf( buf, b_empty_if_na ? "" : "-" ); } - INSERT_STRING( 1, buf ); + INSERT_STRING_NO_FREE( buf ); break; case 'C': if( p_input ) @@ -619,26 +773,30 @@ char *__str_format_meta( vlc_object_t *p_object, const char *string ) } else { - sprintf( buf, "-" ); + sprintf( buf, b_empty_if_na ? "" : "-" ); } - INSERT_STRING( 1, buf ); + INSERT_STRING_NO_FREE( buf ); break; case 'D': if( p_item ) { - sprintf( buf, "%02d:%02d:%02d", - (int)(p_item->i_duration/(3600000000)), - (int)((p_item->i_duration/(60000000))%60), - (int)((p_item->i_duration/1000000)%60) ); + mtime_t i_duration = input_item_GetDuration( p_item ); + snprintf( buf, 10, "%02d:%02d:%02d", + (int)(i_duration/(3600000000)), + (int)((i_duration/(60000000))%60), + (int)((i_duration/1000000)%60) ); } else { - sprintf( buf, "--:--:--" ); + snprintf( buf, 10, b_empty_if_na ? "" : "--:--:--" ); } - INSERT_STRING( 1, buf ); + INSERT_STRING_NO_FREE( buf ); break; case 'F': - INSERT_STRING( p_item, p_item->psz_uri ); + if( p_item ) + { + INSERT_STRING( input_item_GetURI( p_item ) ); + } break; case 'I': if( p_input ) @@ -648,40 +806,41 @@ char *__str_format_meta( vlc_object_t *p_object, const char *string ) } else { - sprintf( buf, "-" ); + sprintf( buf, b_empty_if_na ? "" : "-" ); } - INSERT_STRING( 1, buf ); + INSERT_STRING_NO_FREE( buf ); break; case 'L': if( p_item && p_input ) { - sprintf( buf, "%02d:%02d:%02d", - (int)((p_item->i_duration-p_input->i_time)/(3600000000)), - (int)(((p_item->i_duration-p_input->i_time)/(60000000))%60), - (int)(((p_item->i_duration-p_input->i_time)/1000000)%60) ); + mtime_t i_duration = input_item_GetDuration( p_item ); + int64_t i_time = var_GetTime( p_input, "time" ); + snprintf( buf, 10, "%02d:%02d:%02d", + (int)( ( i_duration - i_time ) / 3600000000 ), + (int)( ( ( i_duration - i_time ) / 60000000 ) % 60 ), + (int)( ( ( i_duration - i_time ) / 1000000 ) % 60 ) ); } else { - sprintf( buf, "--:--:--" ); + snprintf( buf, 10, b_empty_if_na ? "" : "--:--:--" ); } - INSERT_STRING( 1, buf ); + INSERT_STRING_NO_FREE( buf ); break; case 'N': - INSERT_STRING( p_item, p_item->psz_name ); + if( p_item ) + { + INSERT_STRING( input_item_GetName( p_item ) ); + } break; case 'O': { - char *lang; + char *lang = NULL; if( p_input ) - { - lang = var_GetString( p_input, "audio-language" ); - } - else - { - lang = strdup( "-" ); - } - INSERT_STRING( 1, lang ); - free( lang ); + lang = var_GetNonEmptyString( p_input, + "audio-language" ); + if( lang == NULL ) + lang = strdup( b_empty_if_na ? "" : "-" ); + INSERT_STRING( lang ); break; } case 'P': @@ -692,21 +851,21 @@ char *__str_format_meta( vlc_object_t *p_object, const char *string ) } else { - sprintf( buf, "--.-%%" ); + snprintf( buf, 10, b_empty_if_na ? "" : "--.-%%" ); } - INSERT_STRING( 1, buf ); + INSERT_STRING_NO_FREE( buf ); break; case 'R': if( p_input ) { - int r = var_GetInteger( p_input, "rate" ); - snprintf( buf, 10, "%d.%d", r/1000, r%1000 ); + float f = var_GetFloat( p_input, "rate" ); + snprintf( buf, 10, "%.3f", f ); } else { - sprintf( buf, "-" ); + sprintf( buf, b_empty_if_na ? "" : "-" ); } - INSERT_STRING( 1, buf ); + INSERT_STRING_NO_FREE( buf ); break; case 'S': if( p_input ) @@ -716,70 +875,77 @@ char *__str_format_meta( vlc_object_t *p_object, const char *string ) } else { - sprintf( buf, "-" ); + sprintf( buf, b_empty_if_na ? "" : "-" ); } - INSERT_STRING( 1, buf ); + INSERT_STRING_NO_FREE( buf ); break; case 'T': if( p_input ) { - sprintf( buf, "%02d:%02d:%02d", - (int)(p_input->i_time/(3600000000)), - (int)((p_input->i_time/(60000000))%60), - (int)((p_input->i_time/1000000)%60) ); + int64_t i_time = var_GetTime( p_input, "time" ); + snprintf( buf, 10, "%02d:%02d:%02d", + (int)( i_time / ( 3600000000 ) ), + (int)( ( i_time / ( 60000000 ) ) % 60 ), + (int)( ( i_time / 1000000 ) % 60 ) ); } else { - sprintf( buf, "--:--:--" ); + snprintf( buf, 10, b_empty_if_na ? "" : "--:--:--" ); } - INSERT_STRING( 1, buf ); + INSERT_STRING_NO_FREE( buf ); break; case 'U': - INSERT_STRING( p_item && p_item->p_meta, - p_item->p_meta->psz_publisher ); + if( p_item ) + { + INSERT_STRING( input_item_GetPublisher( p_item ) ); + } break; case 'V': { audio_volume_t volume; aout_VolumeGet( p_object, &volume ); snprintf( buf, 10, "%d", volume ); - INSERT_STRING( 1, buf ); + INSERT_STRING_NO_FREE( buf ); break; } case '_': - *d = '\n'; + *(dst+d) = '\n'; d++; break; + case ' ': + b_empty_if_na = true; + break; + default: - *d = *s; + *(dst+d) = *s; d++; break; } - b_is_format = 0; + if( *s != ' ' ) + b_is_format = false; } else if( *s == '$' ) { - b_is_format = 1; + b_is_format = true; + b_empty_if_na = false; } else { - *d = *s; + *(dst+d) = *s; d++; } s++; } - *d = '\0'; + *(dst+d) = '\0'; if( p_input ) - { vlc_object_release( p_input ); - if( p_item ) - vlc_mutex_unlock( &p_item->lock ); - } return dst; } +#undef INSERT_STRING +#undef INSERT_STRING_NO_FREE /** * Apply str format time and str format meta @@ -796,14 +962,34 @@ char *__str_format( vlc_object_t *p_this, const char *psz_src ) /** * Remove forbidden characters from filenames (including slashes) */ -void filename_sanitize( char *str ) +char* filename_sanitize( const char *str_origin ) { + char *str = strdup( str_origin ); + char *str_base = str; + if( *str == '.' && (str[1] == '\0' || (str[1] == '.' && str[2] == '\0' ) ) ) + { + while( *str ) + { + *str = '_'; + str++; + } + return str_base; + } + +#if defined( WIN32 ) + // Change leading spaces into underscores + while( *str && *str == ' ' ) + *str++ = '_'; +#endif + while( *str ) { switch( *str ) { case '/': -#ifdef WIN32 +#if defined( __APPLE__ ) + case ':': +#elif defined( WIN32 ) case '\\': case '*': case '"': @@ -817,6 +1003,19 @@ void filename_sanitize( char *str ) } str++; } + +#if defined( WIN32 ) + // Change trailing spaces into underscores + str--; + while( str != str_base ) + { + if( *str != ' ' ) + break; + *str-- = '_'; + } +#endif + + return str_base; } /** @@ -825,20 +1024,215 @@ void filename_sanitize( char *str ) void path_sanitize( char *str ) { #ifdef WIN32 + /* check drive prefix if path is absolute */ + if( (((unsigned char)(str[0] - 'A') < 26) + || ((unsigned char)(str[0] - 'a') < 26)) && (':' == str[1]) ) + str += 2; +#endif while( *str ) { - switch( *str ) +#if defined( __APPLE__ ) + if( *str == ':' ) + *str = '_'; +#elif defined( WIN32 ) + if( strchr( "*\"?:|<>", *str ) ) + *str = '_'; + if( *str == '/' ) + *str = DIR_SEP_CHAR; +#endif + str++; + } +} + +#include + +/** + * Convert a file path to an URI. + * If already an URI, return a copy of the string. + * @path path path to convert (or URI to copy) + * @return a nul-terminated URI string (use free() to release it), + * or NULL in case of error + */ +char *make_URI (const char *path) +{ + if (path == NULL) + return NULL; + if (!strcmp (path, "-")) + return strdup ("fd://0"); // standard input + if (strstr (path, "://") != NULL) + return strdup (path); /* Already an URI */ + /* Note: VLC cannot handle URI schemes without double slash after the + * scheme name (such as mailto: or news:). */ + + char *buf; +#ifdef WIN32 + if (isalpha (path[0]) && (path[1] == ':')) + { + if (asprintf (&buf, "file:///%c:", path[0]) == -1) + buf = NULL; + path += 2; + } + else +#endif + if (!strncmp (path, "\\\\", 2)) + { /* Windows UNC paths */ +#ifndef WIN32 + /* \\host\share\path -> smb://host/share/path */ + if (strchr (path + 2, '\\') != NULL) + { /* Convert antislashes to slashes */ + char *dup = strdup (path); + if (dup == NULL) + return NULL; + for (size_t i = 2; dup[i]; i++) + if (dup[i] == '\\') + dup[i] = DIR_SEP_CHAR; + + char *ret = make_URI (dup); + free (dup); + return ret; + } +# define SMB_SCHEME "smb" +#else + /* \\host\share\path -> file://host/share/path */ +# define SMB_SCHEME "file" +#endif + size_t hostlen = strcspn (path + 2, DIR_SEP); + + buf = malloc (sizeof (SMB_SCHEME) + 3 + hostlen); + if (buf != NULL) + snprintf (buf, sizeof (SMB_SCHEME) + 3 + hostlen, + SMB_SCHEME"://%s", path + 2); + path += 2 + hostlen; + } + else + if (path[0] != DIR_SEP_CHAR) + { /* Relative path: prepend the current working directory */ + char cwd[PATH_MAX]; + + if (getcwd (cwd, sizeof (cwd)) == NULL) /* FIXME: UTF8? */ + return NULL; + if (asprintf (&buf, "%s/%s", cwd, path) == -1) + return NULL; + char *ret = make_URI (buf); + free (buf); + return ret; + } + else + buf = strdup ("file://"); + if (buf == NULL) + return NULL; + + assert (path[0] == DIR_SEP_CHAR); + + /* Absolute file path */ + for (const char *ptr = path + 1;; ptr++) + { + size_t len = strcspn (ptr, DIR_SEP); + char *component = encode_URI_bytes (ptr, len); + if (component == NULL) { - case '*': - case '"': - case '?': - case ':': - case '|': - case '<': - case '>': - *str = '_'; + free (buf); + return NULL; } - str++; + char *uri; + int val = asprintf (&uri, "%s/%s", buf, component); + free (component); + free (buf); + if (val == -1) + return NULL; + buf = uri; + ptr += len; + if (*ptr == '\0') + return buf; } +} + +/** + * Tries to convert an URI to a local (UTF-8-encoded) file path. + * @param url URI to convert + * @return NULL on error, a nul-terminated string otherwise + * (use free() to release it) + */ +char *make_path (const char *url) +{ + char *ret = NULL; + char *end; + + char *path = strstr (url, "://"); + if (path == NULL) + return NULL; /* unsupported scheme or invalid syntax */ + + end = memchr (url, '/', path - url); + size_t schemelen = ((end != NULL) ? end : path) - url; + path += 3; /* skip "://" */ + + /* Remove HTML anchor if present */ + end = strchr (path, '#'); + if (end) + path = strndup (path, end - path); + else + path = strdup (path); + if (unlikely(path == NULL)) + return NULL; /* boom! */ + + /* Decode path */ + decode_URI (path); + + if (schemelen == 4 && !strncasecmp (url, "file", 4)) + { +#if (DIR_SEP_CHAR != '/') + for (char *p = strchr (path, '/'); p; p = strchr (p, '/')) + *p == DIR_SEP_CHAR; #endif + if (*path == DIR_SEP_CHAR) + return path; + + /* Local path disguised as a remote one (MacOS X) */ + if (!strncasecmp (path, "localhost"DIR_SEP, 10)) + { + memmove (path, path + 9, strlen (path + 9) + 1); + return path; + } + +#ifdef WIN32 + if (*path && asprintf (&ret, "\\\\%s", path) == -1) + ret = NULL; +#endif + /* non-local path :-( */ + } + else + if (schemelen == 2 && !strncasecmp (url, "fd", 2)) + { + int fd = strtol (path, &end, 0); + + if (*end) + goto out; + +#ifndef WIN32 + switch (fd) + { + case 0: + ret = strdup ("/dev/stdin"); + break; + case 1: + ret = strdup ("/dev/stdout"); + break; + case 2: + ret = strdup ("/dev/strerr"); + break; + default: + if (asprintf (&ret, "/dev/fd/%d", fd) == -1) + ret = NULL; + } +#else + if (fd < 2) + ret = strdup ("CON"); + else + ret = NULL; +#endif + } + +out: + free (path); + return ret; /* unknown scheme */ }