]> git.sesse.net Git - vlc/blobdiff - src/text/strings.c
Win32: canonicalize path separator, avoid isalpha(), remove dead code
[vlc] / src / text / strings.c
index 82e29db6302214ea7f38a26207b9fb008bac1998..8a0229ce20ccb5e7b56546ffe5e87efd2841cd2a 100644 (file)
@@ -235,130 +235,143 @@ char *encode_URI_component( const char *psz_uri )
 static const struct xml_entity_s
 {
     char    psz_entity[8];
-    uint8_t i_length;
     char    psz_char[4];
-} p_xml_entities[] = {
+} xml_entities[] = {
     /* Important: this list has to be in alphabetical order (psz_entity-wise) */
-    { "AElig;", 6, "Æ" },
-    { "Aacute;", 7, "Á" },
-    { "Acirc;", 6, "Â" },
-    { "Agrave;", 7, "À" },
-    { "Aring;", 6, "Å" },
-    { "Atilde;", 7, "Ã" },
-    { "Auml;", 5, "Ä" },
-    { "Ccedil;", 7, "Ç" },
-    { "Dagger;", 7, "‡" },
-    { "ETH;", 4, "Ð" },
-    { "Eacute;", 7, "É" },
-    { "Ecirc;", 6, "Ê" },
-    { "Egrave;", 7, "È" },
-    { "Euml;", 5, "Ë" },
-    { "Iacute;", 7, "Í" },
-    { "Icirc;", 6, "Î" },
-    { "Igrave;", 7, "Ì" },
-    { "Iuml;", 5, "Ï" },
-    { "Ntilde;", 7, "Ñ" },
-    { "OElig;", 6, "Œ" },
-    { "Oacute;", 7, "Ó" },
-    { "Ocirc;", 6, "Ô" },
-    { "Ograve;", 7, "Ò" },
-    { "Oslash;", 7, "Ø" },
-    { "Otilde;", 7, "Õ" },
-    { "Ouml;", 5, "Ö" },
-    { "Scaron;", 7, "Š" },
-    { "THORN;", 6, "Þ" },
-    { "Uacute;", 7, "Ú" },
-    { "Ucirc;", 6, "Û" },
-    { "Ugrave;", 7, "Ù" },
-    { "Uuml;", 5, "Ü" },
-    { "Yacute;", 7, "Ý" },
-    { "Yuml;", 5, "Ÿ" },
-    { "aacute;", 7, "á" },
-    { "acirc;", 6, "â" },
-    { "acute;", 6, "´" },
-    { "aelig;", 6, "æ" },
-    { "agrave;", 7, "à" },
-    { "aring;", 6, "å" },
-    { "atilde;", 7, "ã" },
-    { "auml;", 5, "ä" },
-    { "bdquo;", 6, "„" },
-    { "brvbar;", 7, "¦" },
-    { "ccedil;", 7, "ç" },
-    { "cedil;", 6, "¸" },
-    { "cent;", 5, "¢" },
-    { "circ;", 5, "ˆ" },
-    { "copy;", 5, "©" },
-    { "curren;", 7, "¤" },
-    { "dagger;", 7, "†" },
-    { "deg;", 4, "°" },
-    { "divide;", 7, "÷" },
-    { "eacute;", 7, "é" },
-    { "ecirc;", 6, "ê" },
-    { "egrave;", 7, "è" },
-    { "eth;", 4, "ð" },
-    { "euml;", 5, "ë" },
-    { "euro;", 5, "€" },
-    { "frac12;", 7, "½" },
-    { "frac14;", 7, "¼" },
-    { "frac34;", 7, "¾" },
-    { "hellip;", 7, "…" },
-    { "iacute;", 7, "í" },
-    { "icirc;", 6, "î" },
-    { "iexcl;", 6, "¡" },
-    { "igrave;", 7, "ì" },
-    { "iquest;", 7, "¿" },
-    { "iuml;", 5, "ï" },
-    { "laquo;", 6, "«" },
-    { "ldquo;", 6, "“" },
-    { "lsaquo;", 7, "‹" },
-    { "lsquo;", 6, "‘" },
-    { "macr;", 5, "¯" },
-    { "mdash;", 6, "—" },
-    { "micro;", 6, "µ" },
-    { "middot;", 7, "·" },
-    { "ndash;", 6, "–" },
-    { "not;", 4, "¬" },
-    { "ntilde;", 7, "ñ" },
-    { "oacute;", 7, "ó" },
-    { "ocirc;", 6, "ô" },
-    { "oelig;", 6, "œ" },
-    { "ograve;", 7, "ò" },
-    { "ordf;", 5, "ª" },
-    { "ordm;", 5, "º" },
-    { "oslash;", 7, "ø" },
-    { "otilde;", 7, "õ" },
-    { "ouml;", 5, "ö" },
-    { "para;", 5, "¶" },
-    { "permil;", 7, "‰" },
-    { "plusmn;", 7, "±" },
-    { "pound;", 6, "£" },
-    { "raquo;", 6, "»" },
-    { "rdquo;", 6, "”" },
-    { "reg;", 4, "®" },
-    { "rsaquo;", 7, "›" },
-    { "rsquo;", 6, "’" },
-    { "sbquo;", 6, "‚" },
-    { "scaron;", 7, "š" },
-    { "sect;", 5, "§" },
-    { "shy;", 4, "­" },
-    { "sup1;", 5, "¹" },
-    { "sup2;", 5, "²" },
-    { "sup3;", 5, "³" },
-    { "szlig;", 6, "ß" },
-    { "thorn;", 6, "þ" },
-    { "tilde;", 6, "˜" },
-    { "times;", 6, "×" },
-    { "trade;", 6, "™" },
-    { "uacute;", 7, "ú" },
-    { "ucirc;", 6, "û" },
-    { "ugrave;", 7, "ù" },
-    { "uml;", 4, "¨" },
-    { "uuml;", 5, "ü" },
-    { "yacute;", 7, "ý" },
-    { "yen;", 4, "¥" },
-    { "yuml;", 5, "ÿ" },
+    { "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 strncmp (name, ent->psz_entity, strlen (ent->psz_entity));
+}
+
 /**
  * Converts "&lt;", "&gt;" and "&amp;" to "<", ">" and "&"
  * \param string to convert
@@ -371,35 +384,42 @@ void resolve_xml_special_chars( char *psz_value )
     {
         if( *psz_value == '&' )
         {
-            char *psz_value1 = psz_value + 1;
-#define TRY_CHAR( src, len, dst )                     \
-            if( !strncmp( psz_value1, src, len ) )   \
-            {                                         \
-                *p_pos = dst;                         \
-                psz_value += len + 1;                 \
-            }
-            TRY_CHAR( "lt;", 3, '<' )
-            else TRY_CHAR( "amp;", 4, '&' )
-            else TRY_CHAR( "apos;", 5, '\'' )
-            else TRY_CHAR( "gt;", 3, '>' )
-            else TRY_CHAR( "quot;", 5, '"' )
-#undef TRY_CHAR
-            else if( *psz_value1 == '#' )
-            {
+            if( psz_value[1] == '#' )
+            {   /* &#xxx; Unicode code point */
                 char *psz_end;
-                int i = strtol( psz_value+2, &psz_end, 10 );
+                unsigned long cp = strtoul( psz_value+2, &psz_end, 10 );
                 if( *psz_end == ';' )
                 {
-                    if( i >= 32 && i <= 126 )
+                    psz_value = psz_end + 1;
+                    if( cp == 0 )
+                        (void)0; /* skip nuls */
+                    else
+                    if( cp <= 0x7F )
                     {
-                        *p_pos = (char)i;
-                        psz_value = psz_end+1;
+                        *p_pos =            cp;
                     }
                     else
+                    /* Unicode code point outside ASCII.
+                     * &#xxx; representation is longer than UTF-8 :) */
+                    if( cp <= 0x7FF )
                     {
-                        /* Unhandled code, FIXME */
-                        *p_pos = *psz_value;
-                        psz_value++;
+                        *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
@@ -410,37 +430,21 @@ void resolve_xml_special_chars( char *psz_value )
                 }
             }
             else
-            {
-                const size_t i_entities = sizeof( p_xml_entities ) /
-                                          sizeof( p_xml_entities[0] );
-                assert( i_entities < 128 );
-                size_t step = 128>>1;
-                size_t i = step-1;
-                int cmp = -1;
-                while( step )
+            {   /* 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)
                 {
-                    step >>= 1;
-                    if( i >= i_entities )
-                        cmp = -1;
-                    else
-                        cmp = strncmp( psz_value1, /* Skip the & */
-                                       p_xml_entities[i].psz_entity,
-                                       p_xml_entities[i].i_length );
-                    if( cmp == 0 )
-                    {
-                        size_t i_len = strlen( p_xml_entities[i].psz_char );
-                        strncpy( p_pos, p_xml_entities[i].psz_char, i_len );
-                        p_pos += i_len - 1;
-                        psz_value += p_xml_entities[i].i_length+1;
-                        break;
-                    }
-                    else if( cmp < 0 )
-                        i -= step;
-                    else
-                        i += step;
+                    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;
                 }
-                if( cmp != 0 )
-                {
+                else
+                {   /* No match */
                     *p_pos = *psz_value;
                     psz_value++;
                 }
@@ -757,8 +761,10 @@ char *__str_format_meta( vlc_object_t *p_object, const char *string )
                 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
                     {
@@ -885,7 +891,7 @@ char *__str_format_meta( vlc_object_t *p_object, const char *string )
                     if( p_item && p_input )
                     {
                         mtime_t i_duration = input_item_GetDuration( p_item );
-                        int64_t i_time = p_input->i_time;
+                        int64_t i_time = var_GetInteger( p_input, "time" );
                         sprintf( buf, "%02d:%02d:%02d",
                      (int)( ( i_duration - i_time ) / 3600000000 ),
                      (int)( ( ( i_duration - i_time ) / 60000000 ) % 60 ),
@@ -953,10 +959,11 @@ char *__str_format_meta( vlc_object_t *p_object, const char *string )
                 case 'T':
                     if( p_input )
                     {
+                        int64_t i_time = var_GetInteger( p_input, "time" );
                         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 ) );
+                            (int)( i_time / ( 3600000000 ) ),
+                            (int)( ( i_time / ( 60000000 ) ) % 60 ),
+                            (int)( ( i_time / 1000000 ) % 60 ) );
                     }
                     else
                     {
@@ -1032,8 +1039,10 @@ 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 )
@@ -1041,9 +1050,15 @@ void filename_sanitize( char *str )
             *str = '_';
             str++;
         }
-        return;
+        return str_base;
     }
 
+#if defined( WIN32 )
+    // Change leading spaces into underscores
+    while( *str && *str == ' ' )
+        *str++ = '_';
+#endif
+
     while( *str )
     {
         switch( *str )
@@ -1065,6 +1080,19 @@ 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;
 }
 
 /**
@@ -1072,16 +1100,10 @@ void filename_sanitize( char *str )
  */
 void path_sanitize( char *str )
 {
-#if 0
-    /*
-     * Uncomment the two blocks to prevent /../ or /./, i'm not sure that we
-     * want to.
-     */
-    char *prev = str - 1;
-#endif
 #ifdef WIN32
     /* check drive prefix if path is absolute */
-    if( isalpha(*str) && (':' == *(str+1)) )
+    if( (((unsigned char)(str[0] - 'A') < 26)
+      || ((unsigned char)(str[0] - 'a') < 26)) && (':' == str[1]) )
         str += 2;
 #endif
     while( *str )
@@ -1090,36 +1112,10 @@ void path_sanitize( char *str )
         if( *str == ':' )
             *str = '_';
 #elif defined( WIN32 )
-        switch( *str )
-        {
-            case '*':
-            case '"':
-            case '?':
-            case ':':
-            case '|':
-            case '<':
-            case '>':
-                *str = '_';
-        }
-#endif
-#if 0
-        if( *str == '/'
-#ifdef WIN32
-            || *str == '\\'
-#endif
-            )
-        {
-            if( str - prev == 2 && prev[1] == '.' )
-            {
-                prev[1] = '.';
-            }
-            else if( str - prev == 3 && prev[1] == '.' && prev[2] == '.' )
-            {
-                prev[1] = '_';
-                prev[2] = '_';
-            }
-            prev = str;
-        }
+        if( strchr( "*\"?:|<>", *str ) )
+            *str = '_';
+        if( *str == '/' )
+            *str = DIR_SEP_CHAR;
 #endif
         str++;
     }