X-Git-Url: http://git.heureux-cyclage.org/?a=blobdiff_plain;f=languages%2FLanguage.php;h=aa1a65c34abe2cf75d6c24636ddb70a0add5b1cf;hb=0372e9b3f679aa761e3741f1bc4e9270e821e55b;hp=190961971157e7dd9a3b3399b7d9b80956a2b6cb;hpb=38e71e6a2f4fea269f61087373dd22b62de4cd39;p=lhc%2Fweb%2Fwiklou.git
diff --git a/languages/Language.php b/languages/Language.php
index 1909619711..aa1a65c34a 100644
--- a/languages/Language.php
+++ b/languages/Language.php
@@ -37,9 +37,9 @@ class FakeConverter {
function FakeConverter($langobj) {$this->mLang = $langobj;}
function autoConvertToAllVariants($text) {return $text;}
function convert($t, $i) {return $t;}
- function parserConvert($t, $p) {return $t;}
function getVariants() { return array( $this->mLang->getCode() ); }
- function getPreferredVariant() {return $this->mLang->getCode(); }
+ function getPreferredVariant() { return $this->mLang->getCode(); }
+ function getConvRuleTitle() { return false; }
function findVariantLink(&$l, &$n, $ignoreOtherCond = false) {}
function getExtraHashOptions() {return '';}
function getParsedTitle() {return '';}
@@ -59,7 +59,6 @@ class Language {
var $mNamespaceIds, $namespaceNames, $namespaceAliases;
var $dateFormatStrings = array();
- var $minSearchLength;
var $mExtendedSpecialPageAliases;
/**
@@ -116,7 +115,7 @@ class Language {
'hebrew-calendar-m10-gen', 'hebrew-calendar-m11-gen', 'hebrew-calendar-m12-gen',
'hebrew-calendar-m6a-gen', 'hebrew-calendar-m6b-gen'
);
-
+
static public $mHijriCalendarMonthMsgs = array(
'hijri-calendar-m1', 'hijri-calendar-m2', 'hijri-calendar-m3',
'hijri-calendar-m4', 'hijri-calendar-m5', 'hijri-calendar-m6',
@@ -159,7 +158,7 @@ class Language {
if ( $recursionLevel > 5 ) {
throw new MWException( "Language fallback loop detected when creating class $class\n" );
- }
+ }
if( ! class_exists( $class ) ) {
$fallback = Language::getFallbackFor( $code );
@@ -242,7 +241,7 @@ class Language {
function getNamespaces() {
if ( is_null( $this->namespaceNames ) ) {
global $wgExtraNamespaces, $wgMetaNamespace, $wgMetaNamespaceTalk;
-
+
$this->namespaceNames = self::$dataCache->getItem( $this->mCode, 'namespaceNames' );
if ( $wgExtraNamespaces ) {
$this->namespaceNames = $wgExtraNamespaces + $this->namespaceNames;
@@ -256,7 +255,7 @@ class Language {
$this->namespaceNames[NS_PROJECT_TALK] =
$this->fixVariableInNamespace( $talk );
}
-
+
# The above mixing may leave namespaces out of canonical order.
# Re-order by namespace ID number...
ksort( $this->namespaceNames );
@@ -344,7 +343,7 @@ class Language {
if ( is_null( $this->mNamespaceIds ) ) {
global $wgNamespaceAliases;
# Put namespace names and aliases into a hashtable.
- # If this is too slow, then we should arrange it so that it is done
+ # If this is too slow, then we should arrange it so that it is done
# before caching. The catch is that at pre-cache time, the above
# class-specific fixup hasn't been done.
$this->mNamespaceIds = array();
@@ -415,7 +414,7 @@ class Language {
function getDatePreferences() {
return self::$dataCache->getItem( $this->mCode, 'datePreferences' );
}
-
+
function getDateFormats() {
return self::$dataCache->getItem( $this->mCode, 'dateFormats' );
}
@@ -460,7 +459,7 @@ class Language {
if ( !$customisedOnly ) {
return $allNames;
}
-
+
global $IP;
$names = array();
$dir = opendir( "$IP/languages/messages" );
@@ -523,11 +522,11 @@ class Language {
function getHebrewCalendarMonthNameGen( $key ) {
return $this->getMessageFromDB( self::$mHebrewCalendarMonthGenMsgs[$key-1] );
}
-
+
function getHijriCalendarMonthName( $key ) {
return $this->getMessageFromDB( self::$mHijriCalendarMonthMsgs[$key-1] );
}
-
+
/**
* Used by date() and time() to adjust the time output.
*
@@ -579,7 +578,7 @@ class Language {
wfSuppressWarnings(); // E_STRICT system time bitching
# Generate an adjusted date; take advantage of the fact that mktime
- # will normalize out-of-range values so we don't have to split $minDiff
+ # will normalize out-of-range values so we don't have to split $minDiff
# into hours and minutes.
$t = mktime( (
(int)substr( $ts, 8, 2) ), # Hours
@@ -588,19 +587,19 @@ class Language {
(int)substr( $ts, 4, 2 ), # Month
(int)substr( $ts, 6, 2 ), # Day
(int)substr( $ts, 0, 4 ) ); #Year
-
+
$date = date( 'YmdHis', $t );
wfRestoreWarnings();
-
+
return $date;
}
/**
* This is a workalike of PHP's date() function, but with better
- * internationalisation, a reduced set of format characters, and a better
+ * internationalisation, a reduced set of format characters, and a better
* escaping format.
*
- * Supported format characters are dDjlNwzWFmMntLoYyaAgGhHiscrU. See the
+ * Supported format characters are dDjlNwzWFmMntLoYyaAgGhHiscrU. See the
* PHP manual for definitions. "o" format character is supported since
* PHP 5.1.0, previous versions return literal o.
* There are a number of extensions, which start with "x":
@@ -648,7 +647,7 @@ class Language {
*
* Input timestamp is assumed to be pre-normalized to the desired local
* time zone, if any.
- *
+ *
* @param $format String
* @param $ts String: 14-character timestamp
* YYYYMMDDHHMMSS
@@ -850,7 +849,7 @@ class Language {
case 'h':
$h = substr( $ts, 8, 2 );
$num = sprintf( '%02d', $h % 12 ? $h % 12 : 12 );
- break;
+ break;
case 'H':
$num = substr( $ts, 8, 2 );
break;
@@ -921,11 +920,11 @@ class Language {
private static $GREG_DAYS = array( 31, 28, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31 );
private static $IRANIAN_DAYS = array( 31, 31, 31, 31, 31, 31, 30, 30, 30, 30, 30, 29 );
/**
- * Algorithm by Roozbeh Pournader and Mohammad Toossi to convert
+ * Algorithm by Roozbeh Pournader and Mohammad Toossi to convert
* Gregorian dates to Iranian dates. Originally written in C, it
* is released under the terms of GNU Lesser General Public
* License. Conversion to PHP was performed by Niklas Laxström.
- *
+ *
* Link: http://www.farsiweb.info/jalali/jalali.c
*/
private static function tsToIranian( $ts ) {
@@ -952,7 +951,7 @@ class Language {
// Days passed in current month
$gDayNo += $gd;
-
+
$jDayNo = $gDayNo - 79;
$jNp = floor($jDayNo / 12053);
@@ -986,7 +985,7 @@ class Language {
$year = substr( $ts, 0, 4 );
$month = substr( $ts, 4, 2 );
$day = substr( $ts, 6, 2 );
-
+
$zyr = $year;
$zd=$day;
$zm=$month;
@@ -996,15 +995,15 @@ class Language {
if (($zy>1582)||(($zy==1582)&&($zm>10))||(($zy==1582)&&($zm==10)&&($zd>14)))
{
-
-
+
+
$zjd=(int)((1461*($zy + 4800 + (int)( ($zm-14) /12) ))/4) + (int)((367*($zm-2-12*((int)(($zm-14)/12))))/12)-(int)((3*(int)(( ($zy+4900+(int)(($zm-14)/12))/100)))/4)+$zd-32075;
}
else
{
$zjd = 367*$zy-(int)((7*($zy+5001+(int)(($zm-9)/7)))/4)+(int)((275*$zm)/9)+$zd+1729777;
}
-
+
$zl=$zjd-1948440+10632;
$zn=(int)(($zl-1)/10631);
$zl=$zl-10631*$zn+354;
@@ -1199,8 +1198,9 @@ class Language {
* http://en.wikipedia.org/wiki/Minguo_calendar
* http://en.wikipedia.org/wiki/Japanese_era_name
*
- * @param $ts String: 14-character timestamp, calender name
- * @return array converted year, month, day
+ * @param $ts String: 14-character timestamp
+ * @param $cName String: calender name
+ * @return Array: converted year, month, day
*/
private static function tsToYear( $ts, $cName ) {
$gy = substr( $ts, 0, 4 );
@@ -1268,7 +1268,7 @@ class Language {
array( '', 'C', 'CC', 'CCC', 'CD', 'D', 'DC', 'DCC', 'DCCC', 'CM', 'M' ),
array( '', 'M', 'MM', 'MMM' )
);
-
+
$num = intval( $num );
if ( $num > 3000 || $num <= 0 ) {
return $num;
@@ -1367,8 +1367,7 @@ class Language {
if( $usePrefs ) {
$datePreference = $wgUser->getDatePreference();
} else {
- $options = User::getDefaultOptions();
- $datePreference = (string)$options['date'];
+ $datePreference = (string)User::getDefaultOption( 'date' );
}
} else {
$datePreference = (string)$usePrefs;
@@ -1378,7 +1377,7 @@ class Language {
if( $datePreference == '' ) {
return 'default';
}
-
+
return $datePreference;
}
@@ -1415,8 +1414,8 @@ class Language {
* @return string
*/
function date( $ts, $adj = false, $format = true, $timecorrection = false ) {
- if ( $adj ) {
- $ts = $this->userAdjust( $ts, $timecorrection );
+ if ( $adj ) {
+ $ts = $this->userAdjust( $ts, $timecorrection );
}
$df = $this->getDateFormatString( 'date', $this->dateFormat( $format ) );
return $this->sprintfDate( $df, $ts );
@@ -1433,8 +1432,8 @@ class Language {
* @return string
*/
function time( $ts, $adj = false, $format = true, $timecorrection = false ) {
- if ( $adj ) {
- $ts = $this->userAdjust( $ts, $timecorrection );
+ if ( $adj ) {
+ $ts = $this->userAdjust( $ts, $timecorrection );
}
$df = $this->getDateFormatString( 'time', $this->dateFormat( $format ) );
return $this->sprintfDate( $df, $ts );
@@ -1453,8 +1452,8 @@ class Language {
*/
function timeanddate( $ts, $adj = false, $format = true, $timecorrection = false) {
$ts = wfTimestamp( TS_MW, $ts );
- if ( $adj ) {
- $ts = $this->userAdjust( $ts, $timecorrection );
+ if ( $adj ) {
+ $ts = $this->userAdjust( $ts, $timecorrection );
}
$df = $this->getDateFormatString( 'both', $this->dateFormat( $format ) );
return $this->sprintfDate( $df, $ts );
@@ -1486,25 +1485,25 @@ class Language {
function ucwordbreaksCallbackAscii($matches){
return $this->ucfirst($matches[1]);
}
-
+
function ucwordbreaksCallbackMB($matches){
return mb_strtoupper($matches[0]);
}
-
+
function ucCallback($matches){
list( $wikiUpperChars ) = self::getCaseMaps();
return strtr( $matches[1], $wikiUpperChars );
}
-
+
function lcCallback($matches){
list( , $wikiLowerChars ) = self::getCaseMaps();
return strtr( $matches[1], $wikiLowerChars );
}
-
+
function ucwordsCallbackMB($matches){
return mb_strtoupper($matches[0]);
}
-
+
function ucwordsCallbackWiki($matches){
list( $wikiUpperChars ) = self::getCaseMaps();
return strtr( $matches[0], $wikiUpperChars );
@@ -1518,7 +1517,7 @@ class Language {
return ucfirst($str);
} else {
// fall back to more complex logic in case of multibyte strings
- return self::uc($str,true);
+ return self::uc($str,true);
}
}
@@ -1547,7 +1546,7 @@ class Language {
}
}
}
-
+
function lcfirst( $str ) {
$o = ord( $str );
if ( !$o ) {
@@ -1602,7 +1601,7 @@ class Language {
array($this,"ucwordsCallbackMB"),
$str
);
- else
+ else
return preg_replace_callback(
$replaceRegexp,
array($this,"ucwordsCallbackWiki"),
@@ -1630,7 +1629,7 @@ class Language {
array($this,"ucwordbreaksCallbackMB"),
$str
);
- else
+ else
return preg_replace_callback(
$replaceRegexp,
array($this,"ucwordsCallbackWiki"),
@@ -1647,7 +1646,7 @@ class Language {
/**
* Return a case-folded representation of $s
*
- * This is a representation such that caseFold($s1)==caseFold($s2) if $s1
+ * This is a representation such that caseFold($s1)==caseFold($s2) if $s1
* and $s2 are the same except for the case of their characters. It is not
* necessary for the value returned to make sense when displayed.
*
@@ -1677,7 +1676,7 @@ class Language {
function fallback8bitEncoding() {
return self::$dataCache->getItem( $this->mCode, 'fallback8bitEncoding' );
}
-
+
/**
* Most writing systems use whitespace to break up words.
* Some languages such as Chinese don't conventionally do this,
@@ -1689,83 +1688,50 @@ class Language {
}
/**
- * Some languages have special punctuation to strip out
- * or characters which need to be converted for MySQL's
- * indexing to grok it correctly. Make such changes here.
- *
+ * Some languages such as Chinese require word segmentation,
+ * Specify such segmentation when overridden in derived class.
+ *
* @param $string String
* @return String
*/
- function stripForSearch( $string ) {
- global $wgDBtype, $wgSearchType;
- if ( $wgDBtype != 'mysql' || $wgSearchType == 'LuceneSearch' ) {
- return $string;
- }
-
- wfProfileIn( __METHOD__ );
-
- // MySQL fulltext index doesn't grok utf-8, so we
- // need to fold cases and convert to hex
- $out = preg_replace_callback(
- "/([\\xc0-\\xff][\\x80-\\xbf]*)/",
- array( $this, 'stripForSearchCallback' ),
- $this->lc( $string ) );
-
- // And to add insult to injury, the default indexing
- // ignores short words... Pad them so we can pass them
- // through without reconfiguring the server...
- $minLength = $this->minSearchLength();
- if( $minLength > 1 ) {
- $n = $minLength-1;
- $out = preg_replace(
- "/\b(\w{1,$n})\b/",
- "$1u800",
- $out );
- }
-
- // Periods within things like hostnames and IP addresses
- // are also important -- we want a search for "example.com"
- // or "192.168.1.1" to work sanely.
- //
- // MySQL's search seems to ignore them, so you'd match on
- // "example.wikipedia.com" and "192.168.83.1" as well.
- $out = preg_replace(
- "/(\w)\.(\w|\*)/u",
- "$1u82e$2",
- $out );
-
- wfProfileOut( __METHOD__ );
- return $out;
+ function segmentByWord( $string ) {
+ return $string;
}
-
+
/**
- * Armor a case-folded UTF-8 string to get through MySQL's
- * fulltext search without being mucked up by funny charset
- * settings or anything else of the sort.
+ * Some languages have special punctuation need to be normalized.
+ * Make such changes here.
+ *
+ * @param $string String
+ * @return String
*/
- protected function stripForSearchCallback( $matches ) {
- return 'u8' . bin2hex( $matches[1] );
+ function normalizeForSearch( $string ) {
+ return self::convertDoubleWidth($string);
}
-
+
/**
- * Check MySQL server's ft_min_word_len setting so we know
- * if we need to pad short words...
- */
- protected function minSearchLength() {
- if( is_null( $this->minSearchLength ) ) {
- $sql = "show global variables like 'ft\\_min\\_word\\_len'";
- $dbr = wfGetDB( DB_SLAVE );
- $result = $dbr->query( $sql );
- $row = $result->fetchObject();
- $result->free();
-
- if( $row && $row->Variable_name == 'ft_min_word_len' ) {
- $this->minSearchLength = intval( $row->Value );
- } else {
- $this->minSearchLength = 0;
- }
+ * convert double-width roman characters to single-width.
+ * range: ff00-ff5f ~= 0020-007f
+ */
+ protected static function convertDoubleWidth( $string ) {
+ static $full = null;
+ static $half = null;
+
+ if( $full === null ) {
+ $fullWidth = "ï¼ï¼ï¼ï¼ï¼ï¼ï¼ï¼ï¼ï¼ï¼¡ï¼¢ï¼£ï¼¤ï¼¥ï¼¦ï¼§ï¼¨ï¼©ï¼ªï¼«ï¼¬ï¼ï¼®ï¼¯ï¼°ï¼±ï¼²ï¼³ï¼´ï¼µï¼¶ï¼·ï¼¸ï¼¹ï¼ºï½ï½ï½ï½ï½
ï½ï½ï½ï½ï½ï½ï½ï½ï½ï½ï½ï½ï½ï½ï½ï½ï½ï½ï½ï½ï½";
+ $halfWidth = "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz";
+ $full = str_split( $fullWidth, 3 );
+ $half = str_split( $halfWidth );
}
- return $this->minSearchLength;
+
+ $string = str_replace( $full, $half, $string );
+ return $string;
+ }
+
+ protected static function insertSpace( $string, $pattern ) {
+ $string = preg_replace( $pattern, " $1 ", $string );
+ $string = preg_replace( '/ +/', ' ', $string );
+ return $string;
}
function convertForSearchResult( $termsArray ) {
@@ -1775,7 +1741,7 @@ class Language {
}
/**
- * Get the first character of a string.
+ * Get the first character of a string.
*
* @param $s string
* @return string
@@ -1789,7 +1755,7 @@ class Language {
if ( strlen( $matches[1] ) != 3 ) {
return $matches[1];
}
-
+
// Break down Hangul syllables to grab the first jamo
$code = utf8ToCodepoint( $matches[1] );
if ( $code < 0xac00 || 0xd7a4 <= $code) {
@@ -1868,7 +1834,7 @@ class Language {
/**
* Convert a UTF-8 string to normal form C. In Malayalam and Arabic, this
- * also cleans up certain backwards-compatible sequences, converting them
+ * also cleans up certain backwards-compatible sequences, converting them
* to the modern Unicode equivalent.
*
* This is language-specific for performance reasons only.
@@ -1880,9 +1846,9 @@ class Language {
/**
* Transform a string using serialized data stored in the given file (which
* must be in the serialized subdirectory of $IP). The file contains pairs
- * mapping source characters to destination characters.
+ * mapping source characters to destination characters.
*
- * The data is cached in process memory. This will go faster if you have the
+ * The data is cached in process memory. This will go faster if you have the
* FastStringSearch extension.
*/
function transformUsingPairFile( $file, $string ) {
@@ -1901,10 +1867,10 @@ class Language {
*
* @return bool
*/
- function isRTL() {
+ function isRTL() {
return self::$dataCache->getItem( $this->mCode, 'rtl' );
}
-
+
/**
* Return the correct HTML 'dir' attribute value for this language.
* @return String
@@ -1912,7 +1878,7 @@ class Language {
function getDir() {
return $this->isRTL() ? 'rtl' : 'ltr';
}
-
+
/**
* Return 'left' or 'right' as appropriate alignment for line-start
* for this language's text direction.
@@ -1924,7 +1890,7 @@ class Language {
function alignStart() {
return $this->isRTL() ? 'right' : 'left';
}
-
+
/**
* Return 'right' or 'left' as appropriate alignment for line-end
* for this language's text direction.
@@ -2028,7 +1994,7 @@ class Language {
// Cache aliases because it may be slow to load them
if ( is_null( $this->mExtendedSpecialPageAliases ) ) {
// Initialise array
- $this->mExtendedSpecialPageAliases =
+ $this->mExtendedSpecialPageAliases =
self::$dataCache->getItem( $this->mCode, 'specialPageAliases' );
wfRunHooks( 'LanguageGetSpecialPageAliases',
array( &$this->mExtendedSpecialPageAliases, $this->getCode() ) );
@@ -2189,7 +2155,7 @@ class Language {
* are included, otherwise broken characters can be passed to the user
*
* If $length is negative, the string will be truncated from the beginning
- *
+ *
* @param $string String to truncate
* @param $length Int: maximum length (excluding ellipses)
* @param $ellipsis String to append to the truncated text
@@ -2197,38 +2163,205 @@ class Language {
*/
function truncate( $string, $length, $ellipsis = '...' ) {
# Use the localized ellipsis character
- if( $ellipsis == '...' ) {
+ if ( $ellipsis == '...' ) {
$ellipsis = wfMsgExt( 'ellipsis', array( 'escapenoentities', 'language' => $this ) );
}
-
- if( $length == 0 ) {
+ # Check if there is no need to truncate
+ if ( $length == 0 ) {
return $ellipsis;
+ } elseif ( strlen( $string ) <= abs( $length ) ) {
+ return $string;
+ }
+ $stringOriginal = $string;
+ if ( $length > 0 ) {
+ $string = substr( $string, 0, $length ); // xyz...
+ $string = $this->removeBadCharLast( $string );
+ $string = $string . $ellipsis;
+ } else {
+ $string = substr( $string, $length ); // ...xyz
+ $string = $this->removeBadCharFirst( $string );
+ $string = $ellipsis . $string;
}
- if ( strlen( $string ) <= abs( $length ) ) {
+ # Do not truncate if the ellipsis makes the string longer/equal (bug 22181)
+ if ( strlen( $string ) < strlen( $stringOriginal ) ) {
return $string;
+ } else {
+ return $stringOriginal;
+ }
+ }
+
+ /**
+ * Remove bytes that represent an incomplete Unicode character
+ * at the end of string (e.g. bytes of the char are missing)
+ *
+ * @param $string String
+ * @return string
+ */
+ protected function removeBadCharLast( $string ) {
+ $char = ord( $string[strlen( $string ) - 1] );
+ $m = array();
+ if ( $char >= 0xc0 ) {
+ # We got the first byte only of a multibyte char; remove it.
+ $string = substr( $string, 0, -1 );
+ } elseif ( $char >= 0x80 &&
+ preg_match( '/^(.*)(?:[\xe0-\xef][\x80-\xbf]|' .
+ '[\xf0-\xf7][\x80-\xbf]{1,2})$/', $string, $m ) )
+ {
+ # We chopped in the middle of a character; remove it
+ $string = $m[1];
+ }
+ return $string;
+ }
+
+ /**
+ * Remove bytes that represent an incomplete Unicode character
+ * at the start of string (e.g. bytes of the char are missing)
+ *
+ * @param $string String
+ * @return string
+ */
+ protected function removeBadCharFirst( $string ) {
+ $char = ord( $string[0] );
+ if ( $char >= 0x80 && $char < 0xc0 ) {
+ # We chopped in the middle of a character; remove the whole thing
+ $string = preg_replace( '/^[\x80-\xbf]+/', '', $string );
+ }
+ return $string;
+ }
+
+ /*
+ * Truncate a string of valid HTML to a specified length in bytes,
+ * appending an optional string (e.g. for ellipses), and return valid HTML
+ *
+ * This is only intended for styled/linked text, such as HTML with
+ * tags like and , were the tags are self-contained (valid HTML)
+ *
+ * Note: tries to fix broken HTML with MWTidy
+ *
+ * @param string $text String to truncate
+ * @param int $length (zero/positive) Maximum length (excluding ellipses)
+ * @param string $ellipsis String to append to the truncated text
+ * @returns string
+ */
+ function truncateHtml( $text, $length, $ellipsis = '...' ) {
+ # Use the localized ellipsis character
+ if ( $ellipsis == '...' ) {
+ $ellipsis = wfMsgExt( 'ellipsis', array( 'escapenoentities', 'language' => $this ) );
}
- if( $length > 0 ) {
- $string = substr( $string, 0, $length );
- $char = ord( $string[strlen( $string ) - 1] );
- $m = array();
- if ($char >= 0xc0) {
- # We got the first byte only of a multibyte char; remove it.
- $string = substr( $string, 0, -1 );
- } elseif( $char >= 0x80 &&
- preg_match( '/^(.*)(?:[\xe0-\xef][\x80-\xbf]|' .
- '[\xf0-\xf7][\x80-\xbf]{1,2})$/', $string, $m ) ) {
- # We chopped in the middle of a character; remove it
- $string = $m[1];
+ # Check if there is no need to truncate
+ if ( $length <= 0 ) {
+ return $ellipsis; // no text shown, nothing to format
+ } elseif ( strlen($text) <= $length ) {
+ return $text; // string short enough even *with* HTML
+ }
+ $text = MWTidy::tidy( $text ); // fix tags
+ $displayLen = 0; // innerHTML legth so far
+ $testingEllipsis = false; // checking if ellipses will make string longer/equal?
+ $tagType = 0; // 0-open, 1-close
+ $bracketState = 0; // 1-tag start, 2-tag name, 0-neither
+ $entityState = 0; // 0-not entity, 1-entity
+ $tag = $ret = $ch = '';
+ $openTags = array();
+ $textLen = strlen($text);
+ for( $pos = 0; $pos < $textLen; ++$pos ) {
+ $ch = $text[$pos];
+ $lastCh = $pos ? $text[$pos-1] : '';
+ $ret .= $ch; // add to result string
+ if ( $ch == '<' ) {
+ $this->truncate_endBracket( $tag, $tagType, $lastCh, $openTags ); // for bad HTML
+ $entityState = 0; // for bad HTML
+ $bracketState = 1; // tag started (checking for backslash)
+ } elseif ( $ch == '>' ) {
+ $this->truncate_endBracket( $tag, $tagType, $lastCh, $openTags );
+ $entityState = 0; // for bad HTML
+ $bracketState = 0; // out of brackets
+ } elseif ( $bracketState == 1 ) {
+ if ( $ch == '/' ) {
+ $tagType = 1; // close tag (e.g. "")
+ } else {
+ $tagType = 0; // open tag (e.g. "")
+ $tag .= $ch;
+ }
+ $bracketState = 2; // building tag name
+ } elseif ( $bracketState == 2 ) {
+ if ( $ch != ' ' ) {
+ $tag .= $ch;
+ } else {
+ // Name found (e.g. "truncate_skip( $ret, $text, "<>", $pos + 1 );
+ }
+ } elseif ( $bracketState == 0 ) {
+ if ( $entityState ) {
+ if ( $ch == ';' ) {
+ $entityState = 0;
+ $displayLen++; // entity is one displayed char
+ }
+ } else {
+ if ( $ch == '&' ) {
+ $entityState = 1; // entity found, (e.g. " ")
+ } else {
+ $displayLen++; // this char is displayed
+ // Add on the other display text after this...
+ $skipped = $this->truncate_skip(
+ $ret, $text, "<>&", $pos + 1, $length - $displayLen );
+ $displayLen += $skipped;
+ $pos += $skipped;
+ }
+ }
}
- return $string . $ellipsis;
- } else {
- $string = substr( $string, $length );
- $char = ord( $string[0] );
- if( $char >= 0x80 && $char < 0xc0 ) {
- # We chopped in the middle of a character; remove the whole thing
- $string = preg_replace( '/^[\x80-\xbf]+/', '', $string );
+ # Consider truncation once the display length has reached the maximim.
+ # Double-check that we're not in the middle of a bracket/entity...
+ if ( $displayLen >= $length && $bracketState == 0 && $entityState == 0 ) {
+ if ( !$testingEllipsis ) {
+ $testingEllipsis = true;
+ # Save where we are; we will truncate here unless
+ # the ellipsis actually makes the string longer.
+ $pOpenTags = $openTags; // save state
+ $pRet = $ret; // save state
+ } elseif ( $displayLen > ($length + strlen($ellipsis)) ) {
+ # Ellipsis won't make string longer/equal, the truncation point was OK.
+ $openTags = $pOpenTags; // reload state
+ $ret = $this->removeBadCharLast( $pRet ); // reload state, multi-byte char fix
+ $ret .= $ellipsis; // add ellipsis
+ break;
+ }
+ }
+ }
+ if ( $displayLen == 0 ) {
+ return ''; // no text shown, nothing to format
+ }
+ $this->truncate_endBracket( $tag, $text[$textLen-1], $tagType, $openTags ); // for bad HTML
+ while ( count( $openTags ) > 0 ) {
+ $ret .= '' . array_pop( $openTags ) . '>'; // close open tags
+ }
+ return $ret;
+ }
+
+ // truncateHtml() helper function
+ // like strcspn() but adds the skipped chars to $ret
+ private function truncate_skip( &$ret, $text, $search, $start, $len = -1 ) {
+ $skipCount = 0;
+ if( $start < strlen($text) ) {
+ $skipCount = strcspn( $text, $search, $start, $len );
+ $ret .= substr( $text, $start, $skipCount );
+ }
+ return $skipCount;
+ }
+
+ // truncateHtml() helper function
+ // (a) push or pop $tag from $openTags as needed
+ // (b) clear $tag value
+ private function truncate_endBracket( &$tag, $tagType, $lastCh, &$openTags ) {
+ $tag = ltrim( $tag );
+ if( $tag != '' ) {
+ if( $tagType == 0 && $lastCh != '/' ) {
+ $openTags[] = $tag; // tag opened (didn't close itself)
+ } else if( $tagType == 1 ) {
+ if( $openTags && $tag == $openTags[count($openTags)-1] ) {
+ array_pop( $openTags ); // tag closed
+ }
}
- return $ellipsis . $string;
+ $tag = '';
}
}
@@ -2358,18 +2491,13 @@ class Language {
return $this->mConverter->convert($text, $isTitle);
}
- # Convert text from within Parser
- function parserConvert( $text, &$parser ) {
- return $this->mConverter->parserConvert( $text, $parser );
- }
-
# Check if this is a language with variants
function hasVariants(){
return sizeof($this->getVariants())>1;
}
# Put custom tags (e.g. -{ }-) around math to prevent conversion
- function armourMath($text){
+ function armourMath($text){
return $this->mConverter->armourMath($text);
}
@@ -2412,7 +2540,7 @@ class Language {
*
* @param $link String: the name of the link
* @param $nt Mixed: the title object of the link
- * @param boolean $ignoreOtherCond: to disable other conditions when
+ * @param $ignoreOtherCond Boolean: to disable other conditions when
* we need to transclude a template or update a category's link
* @return null the input parameters may be modified upon return
*/
@@ -2498,7 +2626,7 @@ class Language {
static function getFileName( $prefix = 'Language', $code, $suffix = '.php' ) {
return $prefix . str_replace( '-', '_', ucfirst( $code ) ) . $suffix;
}
-
+
/**
* Get the language code from a file name. Inverse of getFileName()
* @param $filename string $prefix . $languageCode . $suffix
@@ -2525,7 +2653,7 @@ class Language {
global $IP;
return self::getFileName( "$IP/languages/classes/Language", $code, '.php' );
}
-
+
/**
* Get the fallback for a given language
*/
@@ -2538,7 +2666,7 @@ class Language {
}
}
- /**
+ /**
* Get all messages for a given language
* WARNING: this may take a long time
*/
@@ -2546,7 +2674,7 @@ class Language {
return self::getLocalisationCache()->getItem( $code, 'messages' );
}
- /**
+ /**
* Get a message for a given language
*/
static function getMessageFor( $key, $code ) {
@@ -2560,10 +2688,10 @@ class Language {
$talk = str_replace( '$1', $wgMetaNamespace, $talk );
# Allow grammar transformations
- # Allowing full message-style parsing would make simple requests
- # such as action=raw much more expensive than they need to be.
+ # Allowing full message-style parsing would make simple requests
+ # such as action=raw much more expensive than they need to be.
# This will hopefully cover most cases.
- $talk = preg_replace_callback( '/{{grammar:(.*?)\|(.*?)}}/i',
+ $talk = preg_replace_callback( '/{{grammar:(.*?)\|(.*?)}}/i',
array( &$this, 'replaceGrammarInNamespace' ), $talk );
return str_replace( ' ', '_', $talk );
}
@@ -2581,7 +2709,7 @@ class Language {
wfProfileIn( __METHOD__ );
$arr = wfGetPrecompiledData( 'Utf8Case.ser' );
if ( $arr === false ) {
- throw new MWException(
+ throw new MWException(
"Utf8Case.ser is missing, please run \"make\" in the serialized directory\n" );
}
extract( $arr );
@@ -2591,19 +2719,19 @@ class Language {
function formatTimePeriod( $seconds ) {
if ( $seconds < 10 ) {
- return $this->formatNum( sprintf( "%.1f", $seconds ) ) . wfMsg( 'seconds-abbrev' );
+ return $this->formatNum( sprintf( "%.1f", $seconds ) ) . ' ' . wfMsg( 'seconds-abbrev' );
} elseif ( $seconds < 60 ) {
- return $this->formatNum( round( $seconds ) ) . wfMsg( 'seconds-abbrev' );
+ return $this->formatNum( round( $seconds ) ) . ' ' . wfMsg( 'seconds-abbrev' );
} elseif ( $seconds < 3600 ) {
- return $this->formatNum( floor( $seconds / 60 ) ) . wfMsg( 'minutes-abbrev' ) .
- $this->formatNum( round( fmod( $seconds, 60 ) ) ) . wfMsg( 'seconds-abbrev' );
+ return $this->formatNum( floor( $seconds / 60 ) ) . ' ' . wfMsg( 'minutes-abbrev' ) . ' ' .
+ $this->formatNum( round( fmod( $seconds, 60 ) ) ) . ' ' . wfMsg( 'seconds-abbrev' );
} else {
$hours = floor( $seconds / 3600 );
$minutes = floor( ( $seconds - $hours * 3600 ) / 60 );
$secondsPart = round( $seconds - $hours * 3600 - $minutes * 60 );
- return $this->formatNum( $hours ) . wfMsg( 'hours-abbrev' ) .
- $this->formatNum( $minutes ) . wfMsg( 'minutes-abbrev' ) .
- $this->formatNum( $secondsPart ) . wfMsg( 'seconds-abbrev' );
+ return $this->formatNum( $hours ) . ' ' . wfMsg( 'hours-abbrev' ) . ' ' .
+ $this->formatNum( $minutes ) . ' ' . wfMsg( 'minutes-abbrev' ) . ' ' .
+ $this->formatNum( $secondsPart ) . ' ' . wfMsg( 'seconds-abbrev' );
}
}
@@ -2654,4 +2782,11 @@ class Language {
$text = $this->getMessageFromDB( $msg );
return str_replace( '$1', $this->formatNum( $size ), $text );
}
+
+ /**
+ * Get the conversion rule title, if any.
+ */
+ function getConvRuleTitle() {
+ return $this->mConverter->getConvRuleTitle();
+ }
}