* with no UTF-8 support.
*
* @param string $string String having html entities
- * @param $quote_style
- * @param string $charset Encoding set to use (default 'ISO-8859-1')
+ * @param $quote_style the quote style to pass as the second argument to
+ * get_html_translation_table()
+ * @param string $charset Encoding set to use (default 'UTF-8')
*/
-function do_html_entity_decode( $string, $quote_style=ENT_COMPAT, $charset='ISO-8859-1' ) {
+function do_html_entity_decode( $string, $quote_style=ENT_COMPAT, $charset='UTF-8' ) {
$fname = 'do_html_entity_decode';
wfProfileIn( $fname );
# The maximum random value is "only" 2^31-1, so get two random
# values to reduce the chance of dupes
$max = mt_getrandmax();
- $rand = number_format( mt_rand() * mt_rand()
+ $rand = number_format( (mt_rand() * $max + mt_rand())
/ $max / $max, 12, '.', '' );
return $rand;
}
/**
* Return the UTF-8 sequence for a given Unicode code point.
- * Currently doesn't work for values outside the Basic Multilingual Plane.
+ * Doesn't work for values outside the Basic Multilingual Plane.
*
* @param string $codepoint UTF-8 code point.
- * @return string HTML UTF-8 Entitie such as 'Ӓ'.
+ * @return string An UTF-8 character if the codepoint is in the BMP and
+ * &#$codepoint if it isn't;
*/
function wfUtf8Sequence( $codepoint ) {
- if($codepoint < 0x80) return chr($codepoint);
- if($codepoint < 0x800) return chr($codepoint >> 6 & 0x3f | 0xc0) .
- chr($codepoint & 0x3f | 0x80);
- if($codepoint < 0x10000) return chr($codepoint >> 12 & 0x0f | 0xe0) .
- chr($codepoint >> 6 & 0x3f | 0x80) .
- chr($codepoint & 0x3f | 0x80);
- if($codepoint < 0x110000) return chr($codepoint >> 18 & 0x07 | 0xf0) .
- chr($codepoint >> 12 & 0x3f | 0x80) .
- chr($codepoint >> 6 & 0x3f | 0x80) .
- chr($codepoint & 0x3f | 0x80);
-
+ if($codepoint < 0x80)
+ return chr($codepoint);
+ if($codepoint < 0x800)
+ return chr($codepoint >> 6 & 0x3f | 0xc0) . chr($codepoint & 0x3f | 0x80);
+ if($codepoint < 0x10000)
+ return chr($codepoint >> 12 & 0x0f | 0xe0) .
+ chr($codepoint >> 6 & 0x3f | 0x80) .
+ chr($codepoint & 0x3f | 0x80);
+ if($codepoint < 0x110000)
+ return chr($codepoint >> 18 & 0x07 | 0xf0) .
+ chr($codepoint >> 12 & 0x3f | 0x80) .
+ chr($codepoint >> 6 & 0x3f | 0x80) .
+ chr($codepoint & 0x3f | 0x80);
# There should be no assigned code points outside this range, but...
return "&#$codepoint;";
}
/**
* Converts numeric character entities to UTF-8
*
+ * @todo Do named entities
+ *
* @param string $string String to convert.
* @return string Converted string.
*/
#$string = iconv($wgInputEncoding, "UTF-8", $string);
$string = preg_replace ( '/�*([0-9]+);/e', 'wfUtf8Sequence($1)', $string );
$string = preg_replace ( '/&#x([0-9a-f]+);/ie', 'wfUtf8Sequence(0x$1)', $string );
- # Should also do named entities here
return $string;
}
$wgOut->debug( $text );
}
if ( '' != $wgDebugLogFile && !$wgProfileOnly ) {
- error_log( $text, 3, $wgDebugLogFile );
+ # Strip unprintables; they can switch terminal modes when binary data
+ # gets dumped, which is pretty annoying.
+ $text = preg_replace( '![\x00-\x08\x0b\x0c\x0e-\x1f]!', ' ', $text );
+ @error_log( $text, 3, $wgDebugLogFile );
}
}
/**
- * Get a message from anywhere, for the UI elements
+ * Get a message from anywhere, for the current user language
+ *
+ * @param string
*/
function wfMsg( $key ) {
$args = func_get_args();
}
/**
- * Get a message from anywhere, for the content
+ * Get a message from anywhere, for the current global language
*/
function wfMsgForContent( $key ) {
global $wgForceUIMsgAsContentMsg;
$message = $wgParser->transformMsg($message, $wgMsgParserOptions);
}
}
+
+ # Fix windows line-endings
+ # Some messages are split with explode("\n", $msg)
+ $message = str_replace( "\r", '', $message );
# Replace arguments
if( count( $args ) ) {
return $out;
}
+/**
+ * Returns an escaped string suitable for inclusion in a string literal
+ * for JavaScript source code.
+ * Illegal control characters are assumed not to be present.
+ *
+ * @param string $string
+ * @return string
+ */
+function wfEscapeJsString( $string ) {
+ // See ECMA 262 section 7.8.4 for string literal format
+ $pairs = array(
+ "\\" => "\\\\",
+ "\"" => "\\\"",
+ "\'" => "\\\'",
+ "\n" => "\\n",
+ "\r" => "\\r",
+
+ # To avoid closing the element or CDATA section
+ "<" => "\\x3c",
+ ">" => "\\x3e",
+ );
+ return strtr( $string, $pairs );
+}
+
/**
* @todo document
* @return float
}
# Make temporary files
- $td = '/tmp/';
+ $td = wfTempDir();
$oldtextFile = fopen( $oldtextName = tempnam( $td, 'merge-old-' ), 'w' );
$mytextFile = fopen( $mytextName = tempnam( $td, 'merge-mine-' ), 'w' );
$yourtextFile = fopen( $yourtextName = tempnam( $td, 'merge-your-' ), 'w' );
header( "Status: $code $label" );
$wgOut->sendCacheControl();
- header( 'Content-type: text/plain' );
- print $desc."\n";
+ header( 'Content-type: text/html' );
+ print "<html><head><title>" .
+ htmlspecialchars( $label ) .
+ "</title></head><body><h1>" .
+ htmlspecialchars( $label ) .
+ "</h1><p>" .
+ htmlspecialchars( $desc ) .
+ "</p></body></html>\n";
}
/**
# Autodetect, convert and provide timestamps of various types
-/** Standard unix timestamp (number of seconds since 1 Jan 1970) */
-define('TS_UNIX',0);
-/** MediaWiki concatenated string timestamp (yyyymmddhhmmss) */
-define('TS_MW',1);
-/** Standard database timestamp (yyyy-mm-dd hh:mm:ss) */
-define('TS_DB',2);
-/** For HTTP and e-mail headers -- output only */
-define('TS_RFC2822', 3 );
+/**
+ * Unix time - the number of seconds since 1970-01-01 00:00:00 UTC
+ */
+define('TS_UNIX', 0);
/**
- * @todo document
+ * MediaWiki concatenated string timestamp (YYYYMMDDHHMMSS)
+ */
+define('TS_MW', 1);
+
+/**
+ * MySQL DATETIME (YYYY-MM-DD HH:MM:SS)
+ */
+define('TS_DB', 2);
+
+/**
+ * RFC 2822 format, for E-mail and HTTP headers
+ */
+define('TS_RFC2822', 3);
+
+/**
+ * An Exif timestamp (YYYY:MM:DD HH:MM:SS)
+ *
+ * @link http://exif.org/Exif2-2.PDF The Exif 2.2 spec, see page 28 for the
+ * DateTime tag and page 36 for the DateTimeOriginal and
+ * DateTimeDigitized tags.
+ */
+define('TS_EXIF', 4);
+
+
+/**
+ * @param mixed $outputtype A timestamp in one of the supported formats, the
+ * function will autodetect which format is supplied
+ and act accordingly.
+ * @return string Time in the format specified in $outputtype
*/
function wfTimestamp($outputtype=TS_UNIX,$ts=0) {
if ($ts==0) {
# TS_DB
$uts=gmmktime((int)$da[4],(int)$da[5],(int)$da[6],
(int)$da[2],(int)$da[3],(int)$da[1]);
+ } elseif (preg_match("/^(\d{4}):(\d\d):(\d\d) (\d\d):(\d\d):(\d\d)$/",$ts,$da)) {
+ # TS_EXIF
+ $uts=gmmktime((int)$da[4],(int)$da[5],(int)$da[6],
+ (int)$da[2],(int)$da[3],(int)$da[1]);
} elseif (preg_match("/^(\d{4})(\d\d)(\d\d)(\d\d)(\d\d)(\d\d)$/",$ts,$da)) {
# TS_MW
$uts=gmmktime((int)$da[4],(int)$da[5],(int)$da[6],
return gmdate( 'YmdHis', $uts );
case TS_DB:
return gmdate( 'Y-m-d H:i:s', $uts );
+ // This shouldn't ever be used, but is included for completeness
+ case TS_EXIF:
+ return gmdate( 'Y:m:d H:i:s', $uts );
case TS_RFC2822:
return gmdate( 'D, d M Y H:i:s', $uts ) . ' GMT';
default:
}
}
+/**
+ * Return a formatted timestamp, or null if input is null.
+ * For dealing with nullable timestamp columns in the database.
+ * @param int $outputtype
+ * @param string $ts
+ * @return string
+ */
+function wfTimestampOrNull( $outputtype = TS_UNIX, $ts = null ) {
+ if( is_null( $ts ) ) {
+ return null;
+ } else {
+ return wfTimestamp( $outputtype, $ts );
+ }
+}
+
/**
* Check where as the operating system is Windows
*
return $out;
}
+/**
+ * Format an XML element as with wfElement(), but run text through the
+ * UtfNormal::cleanUp() validator first to ensure that no invalid UTF-8
+ * is passed.
+ *
+ * @param string $element
+ * @param array $attribs Name=>value pairs. Values will be escaped.
+ * @param bool $contents NULL to make an open tag only; '' for a contentless closed tag (default)
+ * @return string
+ */
+function wfElementClean( $element, $attribs = array(), $contents = '') {
+ if( $attribs ) {
+ $attribs = array_map( array( 'UtfNormal', 'cleanUp' ), $attribs );
+ }
+ return wfElement( $element, $attribs, UtfNormal::cleanUp( $contents ) );
+}
+
+/** Global singleton instance of MimeMagic. This is initialized on demand,
+* please always use the wfGetMimeMagic() function to get the instance.
+*
+* @private
+*/
+$wgMimeMagic= NULL;
+
+/** Factory functions for the global MimeMagic object.
+* This function always returns the same singleton instance of MimeMagic.
+* That objects will be instantiated on the first call to this function.
+* If needed, the MimeMagic.php file is automatically included by this function.
+* @return MimeMagic the global MimeMagic objects.
+*/
+function &wfGetMimeMagic() {
+ global $wgMimeMagic;
+
+ if (!is_null($wgMimeMagic)) {
+ return $wgMimeMagic;
+ }
+
+ if (!class_exists("MimeMagic")) {
+ #include on demand
+ require_once("MimeMagic.php");
+ }
+
+ $wgMimeMagic= new MimeMagic();
+
+ return $wgMimeMagic;
+}
+
+
+/**
+ * Tries to get the system directory for temporary files.
+ * The TMPDIR, TMP, and TEMP environment variables are checked in sequence,
+ * and if none are set /tmp is returned as the generic Unix default.
+ *
+ * NOTE: When possible, use the tempfile() function to create temporary
+ * files to avoid race conditions on file creation, etc.
+ *
+ * @return string
+ */
+function wfTempDir() {
+ foreach( array( 'TMPDIR', 'TMP', 'TEMP' ) as $var ) {
+ $tmp = getenv( 'TMPDIR' );
+ if( $tmp && file_exists( $tmp ) && is_dir( $tmp ) && is_writable( $tmp ) ) {
+ return $tmp;
+ }
+ }
+ # Hope this is Unix of some kind!
+ return '/tmp';
+}
+
+/**
+ * Make directory, and make all parent directories if they don't exist
+ */
+function wfMkdirParents( $fullDir, $mode ) {
+ $parts = explode( '/', $fullDir );
+ $path = '';
+ $success = false;
+ foreach ( $parts as $dir ) {
+ $path .= $dir . '/';
+ if ( !is_dir( $path ) ) {
+ if ( !mkdir( $path, $mode ) ) {
+ return false;
+ }
+ }
+ }
+ return true;
+}
+
?>