/*****************************************************************************
* strings.c: String related functions
*****************************************************************************
- * Copyright (C) 2006 the VideoLAN team
+ * Copyright (C) 2006 VLC authors and VideoLAN
+ * Copyright (C) 2008-2009 Rémi Denis-Courmont
* $Id$
*
* Authors: Antoine Cellerier <dionoea at videolan dot org>
* Daniel Stranger <vlc at schmaller dot de>
* Rémi Denis-Courmont <rem # videolan org>
*
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU Lesser General Public License as published by
+ * the Free Software Foundation; either version 2.1 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU Lesser General Public License for more details.
*
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston MA 02110-1301, USA.
+ * You should have received a copy of the GNU Lesser General Public License
+ * along with this program; if not, write to the Free Software Foundation,
+ * Inc., 51 Franklin Street, Fifth Floor, Boston MA 02110-1301, USA.
*****************************************************************************/
/*****************************************************************************
# include "config.h"
#endif
-#include <vlc/vlc.h>
+#include <vlc_common.h>
#include <assert.h>
/* Needed by str_format_time */
#include <time.h>
+#include <limits.h>
+#include <math.h>
/* Needed by str_format_meta */
#include <vlc_input.h>
#include <vlc_meta.h>
#include <vlc_playlist.h>
-#include <vlc_aout.h>
+#include <vlc_aout_intf.h>
#include <vlc_strings.h>
-#include <vlc_url.h>
#include <vlc_charset.h>
+#include <vlc_fs.h>
+#include <libvlc.h>
+#include <errno.h>
-/**
- * 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;
- if( psz == NULL )
- return;
-
- 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
- * \return decoded duplicated string
- */
-char *decode_URI_duplicate( const char *psz )
-{
- char *psz_dup = strdup( psz );
- decode_URI( psz_dup );
- return psz_dup;
-}
-
-/**
- * Decode encoded URI string in place
- * \return nothing
- */
-void decode_URI( char *psz )
-{
- unsigned char *in = (unsigned char *)psz, *out = in, c;
- if( psz == NULL )
- return;
-
- while( ( c = *in++ ) != '\0' )
- {
- switch( c )
- {
- case '%':
- {
- char hex[3];
-
- if( ( ( hex[0] = *in++ ) == 0 )
- || ( ( hex[1] = *in++ ) == 0 ) )
- return;
-
- hex[2] = '\0';
- *out++ = (unsigned char)strtoul( hex, NULL, 0x10 );
- break;
- }
-
- case '+':
- *out++ = ' ';
- break;
-
- 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';
- EnsureUTF8( psz );
-}
-
-static inline int isurlsafe( int c )
-{
- 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 );
-}
-
-static inline char url_hexchar( int c )
+static const struct xml_entity_s
{
- return ( c < 10 ) ? c + '0' : c + 'A' - 10;
-}
-
-/**
- * encode_URI_component
- * Encodes an URI component.
- *
- * @param psz_url 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)
- */
-char *encode_URI_component( const char *psz_url )
+ 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)
{
- char psz_enc[3 * strlen( psz_url ) + 1], *out = psz_enc;
- const uint8_t *in;
+ const struct xml_entity_s *ent = elem;
+ const char *name = key;
- 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';
-
- return strdup( psz_enc );
+ return strncmp (name, ent->psz_entity, strlen (ent->psz_entity));
}
/**
{
if( *psz_value == '&' )
{
-#define TRY_CHAR( src, len, dst ) \
- if( !strncmp( psz_value, src, len ) ) \
- { \
- *p_pos = dst; \
- psz_value += len; \
- }
-#define TRY_LONGCHAR( src, len, dst ) \
- if( !strncmp( psz_value, src, len ) ) \
- { \
- strncpy( p_pos, dst, strlen( dst ) ); \
- p_pos += strlen( dst ) - 1; \
- psz_value += len; \
- }
- TRY_CHAR( "<", 4, '<' )
- else TRY_CHAR( ">", 4, '>' )
- else TRY_CHAR( "&", 5, '&' )
- else TRY_CHAR( """, 6, '"' )
- else TRY_CHAR( "'", 6, '\'' )
- else if( psz_value[1] == '#' )
- {
+ 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
psz_value++;
}
}
- else TRY_LONGCHAR( "À", 8, "À" )
- else TRY_LONGCHAR( "Á", 8, "Á" )
- else TRY_LONGCHAR( "Â", 7, "Â" )
- else TRY_LONGCHAR( "Ã", 8, "Ã" )
- else TRY_LONGCHAR( "Ä", 6, "Ä" )
- else TRY_LONGCHAR( "Å", 7, "Å" )
- else TRY_LONGCHAR( "Æ", 7, "Æ" )
- else TRY_LONGCHAR( "Ç", 8, "Ç" )
- else TRY_LONGCHAR( "È", 8, "È" )
- else TRY_LONGCHAR( "É", 8, "É" )
- else TRY_LONGCHAR( "Ê", 7, "Ê" )
- else TRY_LONGCHAR( "Ë", 6, "Ë" )
- else TRY_LONGCHAR( "Ì", 8, "Ì" )
- else TRY_LONGCHAR( "Í", 8, "Í" )
- else TRY_LONGCHAR( "Î", 7, "Î" )
- else TRY_LONGCHAR( "Ï", 6, "Ï" )
- else TRY_LONGCHAR( "Ð", 5, "Ð" )
- else TRY_LONGCHAR( "Ñ", 8, "Ñ" )
- else TRY_LONGCHAR( "Ò", 8, "Ò" )
- else TRY_LONGCHAR( "Ó", 8, "Ó" )
- else TRY_LONGCHAR( "Ô", 7, "Ô" )
- else TRY_LONGCHAR( "Õ", 8, "Õ" )
- else TRY_LONGCHAR( "Ö", 6, "Ö" )
- else TRY_LONGCHAR( "Ø", 8, "Ø" )
- else TRY_LONGCHAR( "Ù", 8, "Ù" )
- else TRY_LONGCHAR( "Ú", 8, "Ú" )
- else TRY_LONGCHAR( "Û", 7, "Û" )
- else TRY_LONGCHAR( "Ü", 6, "Ü" )
- else TRY_LONGCHAR( "Ý", 8, "Ý" )
- else TRY_LONGCHAR( "Þ", 7, "Þ" )
- else TRY_LONGCHAR( "ß", 7, "ß" )
- else TRY_LONGCHAR( "à", 8, "à" )
- else TRY_LONGCHAR( "á", 8, "á" )
- else TRY_LONGCHAR( "â", 7, "â" )
- else TRY_LONGCHAR( "ã", 8, "ã" )
- else TRY_LONGCHAR( "ä", 6, "ä" )
- else TRY_LONGCHAR( "å", 7, "å" )
- else TRY_LONGCHAR( "æ", 7, "æ" )
- else TRY_LONGCHAR( "ç", 8, "ç" )
- else TRY_LONGCHAR( "è", 8, "è" )
- else TRY_LONGCHAR( "é", 8, "é" )
- else TRY_LONGCHAR( "ê", 7, "ê" )
- else TRY_LONGCHAR( "ë", 6, "ë" )
- else TRY_LONGCHAR( "ì", 8, "ì" )
- else TRY_LONGCHAR( "í", 8, "í" )
- else TRY_LONGCHAR( "î", 7, "î" )
- else TRY_LONGCHAR( "ï", 6, "ï" )
- else TRY_LONGCHAR( "ð", 5, "ð" )
- else TRY_LONGCHAR( "ñ", 8, "ñ" )
- else TRY_LONGCHAR( "ò", 8, "ò" )
- else TRY_LONGCHAR( "ó", 8, "ó" )
- else TRY_LONGCHAR( "ô", 7, "ô" )
- else TRY_LONGCHAR( "õ", 8, "õ" )
- else TRY_LONGCHAR( "ö", 6, "ö" )
- else TRY_LONGCHAR( "ø", 8, "ø" )
- else TRY_LONGCHAR( "ù", 8, "ù" )
- else TRY_LONGCHAR( "ú", 8, "ú" )
- else TRY_LONGCHAR( "û", 7, "û" )
- else TRY_LONGCHAR( "ü", 6, "ü" )
- else TRY_LONGCHAR( "ý", 8, "ý" )
- else TRY_LONGCHAR( "þ", 7, "þ" )
- else TRY_LONGCHAR( "ÿ", 6, "ÿ" )
- else TRY_LONGCHAR( "¡", 7, "¡" )
- else TRY_LONGCHAR( "¤", 8, "¤" )
- else TRY_LONGCHAR( "¢", 6, "¢" )
- else TRY_LONGCHAR( "£", 7, "£" )
- else TRY_LONGCHAR( "¥", 5, "¥" )
- else TRY_LONGCHAR( "¦", 8, "¦" )
- else TRY_LONGCHAR( "§", 6, "§" )
- else TRY_LONGCHAR( "¨", 5, "¨" )
- else TRY_LONGCHAR( "©", 6, "©" )
- else TRY_LONGCHAR( "ª", 6, "ª" )
- else TRY_LONGCHAR( "«", 7, "«" )
- else TRY_LONGCHAR( "¬", 5, "¬" )
- else TRY_LONGCHAR( "­", 5, "" )
- else TRY_LONGCHAR( "®", 5, "®" )
- else TRY_LONGCHAR( "™", 7, "™" )
- else TRY_LONGCHAR( "¯", 6, "¯" )
- else TRY_LONGCHAR( "°", 5, "°" )
- else TRY_LONGCHAR( "±", 8, "±" )
- else TRY_LONGCHAR( "²", 6, "²" )
- else TRY_LONGCHAR( "³", 6, "³" )
- else TRY_LONGCHAR( "´", 7, "´" )
- else TRY_LONGCHAR( "µ", 7, "µ" )
- else TRY_LONGCHAR( "¶", 6, "¶" )
- else TRY_LONGCHAR( "·", 8, "·" )
- else TRY_LONGCHAR( "¸", 7, "¸" )
- else TRY_LONGCHAR( "¹", 6, "¹" )
- else TRY_LONGCHAR( "º", 6, "º" )
- else TRY_LONGCHAR( "»", 7, "»" )
- else TRY_LONGCHAR( "¼", 8, "¼" )
- else TRY_LONGCHAR( "½", 8, "½" )
- else TRY_LONGCHAR( "¾", 8, "¾" )
- else TRY_LONGCHAR( "¿", 8, "¿" )
- else TRY_LONGCHAR( "×", 7, "×" )
- else TRY_LONGCHAR( "÷", 8, "÷" )
- else TRY_LONGCHAR( "Œ", 7, "Œ" )
- else TRY_LONGCHAR( "œ", 7, "œ" )
- else TRY_LONGCHAR( "Š", 8, "Š" )
- else TRY_LONGCHAR( "š", 8, "š" )
- else TRY_LONGCHAR( "Ÿ", 6, "Ÿ" )
- else TRY_LONGCHAR( "ˆ", 6, "ˆ" )
- else TRY_LONGCHAR( "˜", 7, "˜" )
- else TRY_LONGCHAR( "–", 7, "–" )
- else TRY_LONGCHAR( "—", 7, "—" )
- else TRY_LONGCHAR( "‘", 7, "‘" )
- else TRY_LONGCHAR( "’", 7, "’" )
- else TRY_LONGCHAR( "‚", 7, "‚" )
- else TRY_LONGCHAR( "“", 7, "“" )
- else TRY_LONGCHAR( "”", 7, "”" )
- else TRY_LONGCHAR( "„", 7, "„" )
- else TRY_LONGCHAR( "†", 8, "†" )
- else TRY_LONGCHAR( "‡", 8, "‡" )
- else TRY_LONGCHAR( "…", 8, "…" )
- else TRY_LONGCHAR( "‰", 8, "‰" )
- else TRY_LONGCHAR( "‹", 8, "‹" )
- else TRY_LONGCHAR( "›", 8, "›" )
- else TRY_LONGCHAR( "€", 6, "€" )
else
- {
- *p_pos = *psz_value;
- psz_value++;
+ { /* 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
}
/**
- * Converts '<', '>', '\"', '\'' and '&' to their html entities
- * \param psz_content simple element content that is to be converted
+ * XML-encode an UTF-8 string
+ * \param str nul-terminated UTF-8 byte sequence to XML-encode
+ * \return XML encoded string or NULL on error
+ * (errno is set to ENOMEM or EILSEQ as appropriate)
*/
-char *convert_xml_special_chars( const char *psz_content )
+char *convert_xml_special_chars (const char *str)
{
- char *psz_temp = malloc( 6 * strlen( psz_content ) + 1 );
- const char *p_from = psz_content;
- char *p_to = psz_temp;
+ assert (str != NULL);
+
+ const size_t len = strlen (str);
+ char *const buf = malloc (6 * len + 1), *ptr = buf;
+ if (unlikely(buf == NULL))
+ return NULL;
+
+ size_t n;
+ uint32_t cp;
- while ( *p_from )
+ while ((n = vlc_towc (str, &cp)) != 0)
{
- 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 == '\"' )
+ if (unlikely(n == (size_t)-1))
{
- strcpy( p_to, """ );
- p_to += 6;
- }
- else if( *p_from == '\'' )
- {
- strcpy( p_to, "'" );
- p_to += 6;
+ free (buf);
+ errno = EILSEQ;
+ return NULL;
}
+
+ if ((cp & ~0x0080) < 32 /* C0/C1 control codes */
+ && memchr ("\x09\x0A\x0D\x85", cp, 4) == NULL)
+ ptr += sprintf (ptr, "&#%"PRIu32";", cp);
else
+ switch (cp)
{
- *p_to = *p_from;
- p_to++;
+ case '\"': strcpy (ptr, """); ptr += 6; break;
+ case '&': strcpy (ptr, "&"); ptr += 5; break;
+ case '\'': strcpy (ptr, "'"); ptr += 5; break;
+ case '<': strcpy (ptr, "<"); ptr += 4; break;
+ case '>': strcpy (ptr, ">"); ptr += 4; break;
+ default: memcpy (ptr, str, n); ptr += n; break;
}
- p_from++;
+ str += n;
}
- *p_to = '\0';
+ *(ptr++) = '\0';
- return psz_temp;
+ ptr = realloc (buf, ptr - buf);
+ return likely(ptr != NULL) ? ptr : buf; /* cannot fail */
}
/* Base64 encoding */
{
const int c = b64[(unsigned int)*p];
if( c == -1 )
- continue;
+ break;
switch( i_level )
{
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;
struct tm loctime;
+ 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. */
localtime_r( &curtime, &loctime );
- strftime( buffer, 255, tformat, &loctime );
- 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);
+}
+
+static void format_duration (char *buf, size_t len, int64_t duration)
+{
+ lldiv_t d;
+ int sec;
+
+ duration /= CLOCK_FREQ;
+ d = lldiv (duration, 60);
+ sec = d.rem;
+ d = lldiv (d.quot, 60);
+ snprintf (buf, len, "%02lld:%02d:%02d", d.quot, (int)d.rem, sec);
}
#define INSERT_STRING( string ) \
if( string != NULL ) \
{ \
int len = strlen( string ); \
- dst = realloc( dst, i_size = i_size + len );\
+ dst = xrealloc( dst, i_size = i_size + len );\
memcpy( (dst+d), string, len ); \
d += len; \
free( string ); \
- } \
- else \
- { \
- *(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 = realloc( dst, i_size = i_size + len );\
+ 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 )
+char *str_format_meta( playlist_t *p_object, const char *string )
{
const char *s = string;
- int b_is_format = 0;
- int b_empty_if_na = 0;
+ bool b_is_format = false;
+ bool b_empty_if_na = false;
char buf[10];
int i_size = strlen( string ) + 1; /* +1 to store '\0' */
- char *dst = malloc( i_size );
+ 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;
+ input_thread_t *p_input = playlist_CurrentInput( p_object );
input_item_t *p_item = NULL;
- pl_Release( p_object );
if( p_input )
{
- vlc_object_yield( p_input );
p_item = input_GetItem(p_input);
}
- sprintf( dst, string );
-
while( *s )
{
if( b_is_format )
case 'f':
if( p_item && p_item->p_stats )
{
- snprintf( buf, 10, "%d",
+ vlc_mutex_lock( &p_item->p_stats->lock );
+ snprintf( buf, 10, "%"PRIi64,
p_item->p_stats->i_displayed_pictures );
+ vlc_mutex_unlock( &p_item->p_stats->lock );
}
else
- {
- sprintf( buf, b_empty_if_na ? "" : "-" );
- }
+ strcpy( buf, b_empty_if_na ? "" : "-" );
INSERT_STRING_NO_FREE( buf );
break;
case 'g':
}
break;
case 's':
- {
- char *lang = NULL;
- if( p_input )
- lang = var_GetNonEmptyString( p_input, "sub-language" );
- if( lang == NULL )
- lang = strdup( b_empty_if_na ? "" : "-" );
- INSERT_STRING( lang );
- break;
- }
+ {
+ char *psz_lang = NULL;
+ if( p_input )
+ psz_lang = var_GetNonEmptyString( p_input, "sub-language" );
+ if( psz_lang == NULL )
+ psz_lang = strdup( b_empty_if_na ? "" : "-" );
+ INSERT_STRING( psz_lang );
+ break;
+ }
case 't':
if( p_item )
{
case 'B':
if( p_input )
{
- snprintf( buf, 10, "%d",
+ snprintf( buf, 10, "%"PRId64,
var_GetInteger( p_input, "bit-rate" )/1000 );
}
else
- {
- sprintf( buf, b_empty_if_na ? "" : "-" );
- }
+ strcpy( buf, b_empty_if_na ? "" : "-" );
INSERT_STRING_NO_FREE( buf );
break;
case 'C':
if( p_input )
{
- snprintf( buf, 10, "%d",
+ snprintf( buf, 10, "%"PRId64,
var_GetInteger( p_input, "chapter" ) );
}
else
- {
- sprintf( buf, b_empty_if_na ? "" : "-" );
- }
+ strcpy( buf, b_empty_if_na ? "" : "-" );
INSERT_STRING_NO_FREE( buf );
break;
case 'D':
if( p_item )
{
mtime_t i_duration = input_item_GetDuration( p_item );
- sprintf( buf, "%02d:%02d:%02d",
- (int)(i_duration/(3600000000)),
- (int)((i_duration/(60000000))%60),
- (int)((i_duration/1000000)%60) );
+ format_duration (buf, sizeof (buf), i_duration);
}
else
- {
- sprintf( buf, b_empty_if_na ? "" : "--:--:--" );
- }
+ strcpy( buf, b_empty_if_na ? "" : "--:--:--" );
INSERT_STRING_NO_FREE( buf );
break;
case 'F':
case 'I':
if( p_input )
{
- snprintf( buf, 10, "%d",
+ snprintf( buf, 10, "%"PRId64,
var_GetInteger( p_input, "title" ) );
}
else
- {
- sprintf( buf, b_empty_if_na ? "" : "-" );
- }
+ strcpy( buf, b_empty_if_na ? "" : "-" );
INSERT_STRING_NO_FREE( buf );
break;
case 'L':
if( p_item && p_input )
{
mtime_t i_duration = input_item_GetDuration( p_item );
- int64_t i_time = p_input->i_time;
- sprintf( buf, "%02d:%02d:%02d",
- (int)( ( i_duration - i_time ) / 3600000000 ),
- (int)( ( ( i_duration - i_time ) / 60000000 ) % 60 ),
- (int)( ( ( i_duration - i_time ) / 1000000 ) % 60 ) );
+ int64_t i_time = var_GetTime( p_input, "time" );
+ format_duration( buf, sizeof(buf),
+ i_duration - i_time );
}
else
- {
- sprintf( buf, b_empty_if_na ? "" : "--:--:--" );
- }
+ strcpy( buf, b_empty_if_na ? "" : "--:--:--" );
INSERT_STRING_NO_FREE( buf );
break;
case 'N':
}
break;
case 'O':
- {
- char *lang = NULL;
- if( p_input )
- lang = var_GetNonEmptyString( p_input,
- "audio-language" );
- if( lang == NULL )
- lang = strdup( b_empty_if_na ? "" : "-" );
- INSERT_STRING( lang );
- break;
- }
+ {
+ char *lang = NULL;
+ if( p_input )
+ lang = var_GetNonEmptyString( p_input,
+ "audio-language" );
+ if( lang == NULL )
+ lang = strdup( b_empty_if_na ? "" : "-" );
+ INSERT_STRING( lang );
+ break;
+ }
case 'P':
if( p_input )
{
}
else
{
- sprintf( buf, b_empty_if_na ? "" : "--.-%%" );
+ snprintf( buf, 10, b_empty_if_na ? "" : "--.-%%" );
}
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, b_empty_if_na ? "" : "-" );
- }
+ strcpy( buf, b_empty_if_na ? "" : "-" );
INSERT_STRING_NO_FREE( buf );
break;
case 'S':
snprintf( buf, 10, "%d.%d", r/1000, (r/100)%10 );
}
else
- {
- sprintf( buf, b_empty_if_na ? "" : "-" );
- }
+ strcpy( buf, b_empty_if_na ? "" : "-" );
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" );
+ format_duration( buf, sizeof(buf), i_time );
}
else
- {
- sprintf( buf, b_empty_if_na ? "" : "--:--:--" );
- }
+ strcpy( buf, b_empty_if_na ? "" : "--:--:--" );
INSERT_STRING_NO_FREE( buf );
break;
case 'U':
break;
case 'V':
{
- audio_volume_t volume;
- aout_VolumeGet( p_object, &volume );
- snprintf( buf, 10, "%d", volume );
- INSERT_STRING_NO_FREE( buf );
+ float vol = aout_VolumeGet( p_object );
+ if( vol >= 0. )
+ {
+ snprintf( buf, 10, "%ld",
+ lroundf(vol * AOUT_VOLUME_DEFAULT ) );
+ INSERT_STRING_NO_FREE( buf );
+ }
+ else
+ INSERT_STRING_NO_FREE( "---" );
break;
}
case '_':
*(dst+d) = '\n';
d++;
break;
+ case 'Z':
+ if( p_item )
+ {
+ char *psz_now_playing = input_item_GetNowPlaying( p_item );
+ if( EMPTY_STR( psz_now_playing ) )
+ {
+ char *psz_temp = input_item_GetTitleFbName( p_item );
+ char *psz_artist = input_item_GetArtist( p_item );
+ if( !EMPTY_STR( psz_artist ) )
+ {
+ INSERT_STRING( psz_artist );
+ if ( !EMPTY_STR( psz_temp ) )
+ INSERT_STRING_NO_FREE( " - " );
+ }
+ INSERT_STRING( psz_temp );
+ }
+ else
+ INSERT_STRING( psz_now_playing );
+ }
+ break;
case ' ':
- b_empty_if_na = 1;
+ b_empty_if_na = true;
break;
default:
break;
}
if( *s != ' ' )
- b_is_format = 0;
+ b_is_format = false;
}
else if( *s == '$' )
{
- b_is_format = 1;
- b_empty_if_na = 0;
+ b_is_format = true;
+ b_empty_if_na = false;
}
else
{
return dst;
}
+#undef INSERT_STRING
+#undef INSERT_STRING_NO_FREE
/**
- * Apply str format time and str format meta
- */
-char *__str_format( vlc_object_t *p_this, const char *psz_src )
-{
- char *psz_buf1, *psz_buf2;
- psz_buf1 = str_format_time( psz_src );
- psz_buf2 = str_format_meta( p_this, psz_buf1 );
- free( psz_buf1 );
- return psz_buf2;
-}
-
-/**
- * Remove forbidden characters from filenames (including slashes)
+ * Remove forbidden, potentially forbidden and otherwise evil characters from
+ * filenames. This includes slashes, and popular characters like colon
+ * (on Unix anyway), so this should only be used for automatically generated
+ * filenames.
+ * \warning Do not use this on full paths,
+ * only single file names without any directory separator!
*/
void filename_sanitize( char *str )
{
- if( *str == '.' && (str[1] == '\0' || (str[1] == '.' && str[2] == '\0' ) ) )
+ unsigned char c;
+
+ /* Special file names, not allowed */
+ if( !strcmp( str, "." ) || !strcmp( str, ".." ) )
{
while( *str )
- {
- *str = '_';
- str++;
- }
+ *(str++) = '_';
return;
}
- while( *str )
+ /* On platforms not using UTF-7, VLC cannot access non-Unicode paths.
+ * Also, some file systems require Unicode file names.
+ * NOTE: This may inserts '?' thus is done replacing '?' with '_'. */
+ EnsureUTF8( str );
+
+ /* Avoid leading spaces to please Windows. */
+ while( (c = *str) != '\0' )
{
- switch( *str )
- {
- case '/':
-#if defined( __APPLE__ )
- case ':':
-#elif defined( WIN32 )
- case '\\':
- case '*':
- case '"':
- case '?':
- case ':':
- case '|':
- case '<':
- case '>':
-#endif
- *str = '_';
- }
+ if( c != ' ' )
+ break;
+ *(str++) = '_';
+ }
+
+ char *start = str;
+
+ while( (c = *str) != '\0' )
+ {
+ /* Non-printable characters are not a good idea */
+ if( c < 32 )
+ *str = '_';
+ /* This is the list of characters not allowed by Microsoft.
+ * We also black-list them on Unix as they may be confusing, and are
+ * not supported by some file system types (notably CIFS). */
+ else if( strchr( "/:\\*\"?|<>", c ) != NULL )
+ *str = '_';
str++;
}
+
+ /* Avoid trailing spaces also to please Windows. */
+ while( str > start )
+ {
+ if( *(--str) != ' ' )
+ break;
+ *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
+#if defined( WIN32 ) || defined( __OS2__ )
/* 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 )
#if defined( __APPLE__ )
if( *str == ':' )
*str = '_';
-#elif defined( WIN32 )
- switch( *str )
- {
- case '*':
- case '"':
- case '?':
- case ':':
- case '|':
- case '<':
- case '>':
- *str = '_';
- }
-#endif
-#if 0
- if( *str == '/'
-#ifdef WIN32
- || *str == '\\'
+#elif defined( WIN32 ) || defined( __OS2__ )
+ if( strchr( "*\"?:|<>", *str ) )
+ *str = '_';
+ if( *str == '/' )
+ *str = DIR_SEP_CHAR;
#endif
- )
+ str++;
+ }
+}
+
+/*
+ Decodes a duration as defined by ISO 8601
+ http://en.wikipedia.org/wiki/ISO_8601#Durations
+ @param str A null-terminated string to convert
+ @return: The duration in seconds. -1 if an error occurred.
+
+ Exemple input string: "PT0H9M56.46S"
+ */
+time_t str_duration( const char *psz_duration )
+{
+ bool timeDesignatorReached = false;
+ time_t res = 0;
+ char* end_ptr;
+
+ if ( psz_duration == NULL )
+ return -1;
+ if ( ( *(psz_duration++) ) != 'P' )
+ return -1;
+ do
+ {
+ double number = strtod( psz_duration, &end_ptr );
+ double mul = 0;
+ if ( psz_duration != end_ptr )
+ psz_duration = end_ptr;
+ switch( *psz_duration )
{
- if( str - prev == 2 && prev[1] == '.' )
+ case 'M':
{
- prev[1] = '.';
+ //M can mean month or minutes, if the 'T' flag has been reached.
+ //We don't handle months though.
+ if ( timeDesignatorReached == true )
+ mul = 60.0;
+ break ;
}
- else if( str - prev == 3 && prev[1] == '.' && prev[2] == '.' )
- {
- prev[1] = '_';
- prev[2] = '_';
- }
- prev = str;
+ case 'Y':
+ case 'W':
+ break ; //Don't handle this duration.
+ case 'D':
+ mul = 86400.0;
+ break ;
+ case 'T':
+ timeDesignatorReached = true;
+ break ;
+ case 'H':
+ mul = 3600.0;
+ break ;
+ case 'S':
+ mul = 1.0;
+ break ;
+ default:
+ break ;
}
-#endif
- str++;
- }
+ res += (time_t)(mul * number);
+ if ( *psz_duration )
+ psz_duration++;
+ } while ( *psz_duration );
+ return res;
}