<?php
/**
- * @defgroup Language Language
+ * Internationalisation code
*
* @file
* @ingroup Language
*/
+/**
+ * @defgroup Language Language
+ */
+
if ( !defined( 'MEDIAWIKI' ) ) {
echo "This file is part of MediaWiki, it is not a valid entry point.\n";
exit( 1 );
*/
class FakeConverter {
var $mLang;
- function FakeConverter( $langobj ) { $this->mLang = $langobj; }
- function autoConvertToAllVariants( $text ) { return $text; }
+ function __construct( $langobj ) { $this->mLang = $langobj; }
+ function autoConvertToAllVariants( $text ) { return array( $this->mLang->getCode() => $text ); }
function convert( $t ) { return $t; }
function convertTitle( $t ) { return $t->getPrefixedText(); }
function getVariants() { return array( $this->mLang->getCode() ); }
function getPreferredVariant() { return $this->mLang->getCode(); }
+ function getDefaultVariant() { return $this->mLang->getCode(); }
+ function getURLVariant() { return ''; }
function getConvRuleTitle() { return false; }
function findVariantLink( &$l, &$n, $ignoreOtherCond = false ) { }
function getExtraHashOptions() { return ''; }
function getParsedTitle() { return ''; }
function markNoConversion( $text, $noParse = false ) { return $text; }
function convertCategoryKey( $key ) { return $key; }
- function convertLinkToAllVariants( $text ) { return array( $this->mLang->getCode() => $text ); }
+ function convertLinkToAllVariants( $text ) { return $this->autoConvertToAllVariants( $text ); }
function armourMath( $text ) { return $text; }
}
/**
* Get a cached language object for a given language code
+ * @param $code String
+ * @return Language
*/
static function factory( $code ) {
if ( !isset( self::$mLangObjCache[$code] ) ) {
/**
* Create a language object for a given language code
+ * @param $code String
+ * @return Language
*/
protected static function newFromCode( $code ) {
global $IP;
static $recursionLevel = 0;
+
+ // Protect against path traversal below
+ if ( !Language::isValidCode( $code ) ) {
+ throw new MWException( "Invalid language code \"$code\"" );
+ }
+
if ( $code == 'en' ) {
$class = 'Language';
} else {
return $lang;
}
+ /**
+ * Returns true if a language code string is of a valid form, whether or
+ * not it exists.
+ */
+ public static function isValidCode( $code ) {
+ return strcspn( $code, "/\\\000" ) === strlen( $code );
+ }
+
/**
* Get the LocalisationCache instance
*/
*/
function getNamespaces() {
if ( is_null( $this->namespaceNames ) ) {
- global $wgExtraNamespaces, $wgMetaNamespace, $wgMetaNamespaceTalk;
+ global $wgMetaNamespace, $wgMetaNamespaceTalk, $wgExtraNamespaces;
$this->namespaceNames = self::$dataCache->getItem( $this->mCode, 'namespaceNames' );
- if ( $wgExtraNamespaces ) {
- $this->namespaceNames = $wgExtraNamespaces + $this->namespaceNames;
- }
+ $validNamespaces = MWNamespace::getCanonicalNamespaces();
+
+ $this->namespaceNames = $wgExtraNamespaces + $this->namespaceNames + $validNamespaces;
$this->namespaceNames[NS_PROJECT] = $wgMetaNamespace;
if ( $wgMetaNamespaceTalk ) {
$this->namespaceNames[NS_PROJECT_TALK] =
$this->fixVariableInNamespace( $talk );
}
-
+
# Sometimes a language will be localised but not actually exist on this wiki.
- $validNamespaces = MWNamespace::getValidNamespaces();
foreach( $this->namespaceNames as $key => $text ) {
- if ( ! in_array( $key, $validNamespaces ) ) {
- unset( $this->namespaceNames[$key] );
- }
+ if ( !isset( $validNamespaces[$key] ) ) {
+ unset( $this->namespaceNames[$key] );
+ }
}
# The above mixing may leave namespaces out of canonical order.
return strtr( $ns, '_', ' ' );
}
+ /**
+ * Returns gender-dependent namespace alias if available.
+ * @param $index Int: namespace index
+ * @param $gender String: gender key (male, female... )
+ * @return String
+ * @since 1.18
+ */
+ function getGenderNsText( $index, $gender ) {
+ $ns = self::$dataCache->getItem( $this->mCode, 'namespaceGenderAliases' );
+ return isset( $ns[$index][$gender] ) ? $ns[$index][$gender] : $this->getNsText( $index );
+ }
+
+ /**
+ * Whether this language makes distinguishes genders for example in
+ * namespaces.
+ * @return bool
+ * @since 1.18
+ */
+ function needsGenderDistinction() {
+ $aliases = self::$dataCache->getItem( $this->mCode, 'namespaceGenderAliases' );
+ return count( $aliases ) > 0;
+ }
+
/**
* Get a namespace key by value, case insensitive.
* Only matches namespace names for the current language, not the
}
}
}
+
+ $genders = self::$dataCache->getItem( $this->mCode, 'namespaceGenderAliases' );
+ foreach ( $genders as $index => $forms ) {
+ foreach ( $forms as $alias ) {
+ $aliases[$alias] = $index;
+ }
+ }
+
$this->namespaceAliases = $aliases;
}
return $this->namespaceAliases;
return $names;
}
+ /**
+ * Get translated language names. This is done on best effort and
+ * by default this is exactly the same as Language::getLanguageNames.
+ * The CLDR extension provides translated names.
+ * @param $code String Language code.
+ * @return Array language code => language name
+ * @since 1.18.0
+ */
+ public static function getTranslatedLanguageNames( $code ) {
+ $names = array();
+ wfRunHooks( 'LanguageGetTranslatedLanguageNames', array( &$names, $code ) );
+
+ foreach ( self::getLanguageNames() as $code => $name ) {
+ if ( !isset( $names[$code] ) ) $names[$code] = $name;
+ }
+
+ return $names;
+ }
+
/**
* Get a message from the MediaWiki namespace.
*
* escaping format.
*
* 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":
+ * PHP manual for definitions. There are a number of extensions, which
+ * start with "x":
*
* xn Do not translate digits of the next numeric format character
* xN Toggle raw digit (xn) flag, stays set until explicitly unset
* @param $ts String: 14-character timestamp
* YYYYMMDDHHMMSS
* 01234567890123
- * @todo emulation of "o" format character for PHP pre 5.1.0
* @todo handling of "o" format character for Iranian, Hebrew, Hijri & Thai?
*/
function sprintfDate( $format, $ts ) {
}
$num = gmdate( 'L', $unix );
break;
- # 'o' is supported since PHP 5.1.0
- # return literal if not supported
- # TODO: emulation for pre 5.1.0 versions
case 'o':
if ( !$unix ) {
$unix = wfTimestamp( TS_UNIX, $ts );
}
- if ( version_compare( PHP_VERSION, '5.1.0' ) === 1 ) {
- $num = date( 'o', $unix );
- } else {
- $s .= 'o';
- }
+ $num = date( 'o', $unix );
break;
case 'Y':
$num = substr( $ts, 0, 4 );
} else {
$s .= $this->formatNum( $num, true );
}
- $num = false;
}
}
return $s;
return $s;
}
- /**
+ /**
* Hebrew Gematria number formatting up to 9999
*/
static function hebrewNumeral( $num ) {
* @return string
*/
function date( $ts, $adj = false, $format = true, $timecorrection = false ) {
+ $ts = wfTimestamp( TS_MW, $ts );
if ( $adj ) {
$ts = $this->userAdjust( $ts, $timecorrection );
}
* @return string
*/
function time( $ts, $adj = false, $format = true, $timecorrection = false ) {
+ $ts = wfTimestamp( TS_MW, $ts );
if ( $adj ) {
$ts = $this->userAdjust( $ts, $timecorrection );
}
return strtr( $matches[0], $wikiUpperChars );
}
+ /**
+ * Make a string's first character uppercase
+ */
function ucfirst( $str ) {
$o = ord( $str );
- if ( $o < 96 ) {
+ if ( $o < 96 ) { // if already uppercase...
return $str;
} elseif ( $o < 128 ) {
- return ucfirst( $str );
+ return ucfirst( $str ); // use PHP's ucfirst()
} else {
// fall back to more complex logic in case of multibyte strings
return $this->uc( $str, true );
}
}
+ /**
+ * Convert a string to uppercase
+ */
function uc( $str, $first = false ) {
if ( function_exists( 'mb_strtoupper' ) ) {
if ( $first ) {
}
$isutf8 = preg_match( '/^([\x00-\x7f]|[\xc0-\xdf][\x80-\xbf]|' .
- '[\xe0-\xef][\x80-\xbf]{2}|[\xf0-\xf7][\x80-\xbf]{3})+$/', $s );
+ '[\xe0-\xef][\x80-\xbf]{2}|[\xf0-\xf7][\x80-\xbf]{3})+$/', $s );
if ( $isutf8 ) {
return $s;
}
return self::$dataCache->getItem( $this->mCode, 'magicWords' );
}
+ protected function doMagicHook() {
+ if ( $this->mMagicHookDone ) {
+ return;
+ }
+ $this->mMagicHookDone = true;
+ wfProfileIn( 'LanguageGetMagic' );
+ wfRunHooks( 'LanguageGetMagic', array( &$this->mMagicExtensions, $this->getCode() ) );
+ wfProfileOut( 'LanguageGetMagic' );
+ }
+
# Fill a MagicWord object with data from here
function getMagic( $mw ) {
- if ( !$this->mMagicHookDone ) {
- $this->mMagicHookDone = true;
- wfProfileIn( 'LanguageGetMagic' );
- wfRunHooks( 'LanguageGetMagic', array( &$this->mMagicExtensions, $this->getCode() ) );
- wfProfileOut( 'LanguageGetMagic' );
- }
+ $this->doMagicHook();
+
if ( isset( $this->mMagicExtensions[$mw->mId] ) ) {
$rawEntry = $this->mMagicExtensions[$mw->mId];
} else {
# 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, $m ) )
+ preg_match( '/^(.*)(?:[\xe0-\xef][\x80-\xbf]|' .
+ '[\xf0-\xf7][\x80-\xbf]{1,2})$/', $string, $m ) )
{
# We chopped in the middle of a character; remove it
$string = $m[1];
*
* Note: tries to fix broken HTML with MWTidy
*
- * @param string $text String to truncate
+ * @param string $text HTML string to truncate
* @param int $length (zero/positive) Maximum length (excluding ellipses)
* @param string $ellipsis String to append to the truncated text
* @returns string
$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();
+ $tag = $ret = '';
+ $openTags = array(); // open tag stack
$textLen = strlen( $text );
for ( $pos = 0; $pos < $textLen; ++$pos ) {
$ch = $text[$pos];
if ( $displayLen == 0 ) {
return ''; // no text shown, nothing to format
}
- $this->truncate_endBracket( $tag, $text[$textLen - 1], $tagType, $openTags ); // for bad HTML
+ // Close the last tag if left unclosed by bad HTML
+ $this->truncate_endBracket( $tag, $text[$textLen - 1], $tagType, $openTags );
while ( count( $openTags ) > 0 ) {
$ret .= '</' . array_pop( $openTags ) . '>'; // close open tags
}
return $skipCount;
}
- // truncateHtml() helper function
- // (a) push or pop $tag from $openTags as needed
- // (b) clear $tag value
+ /*
+ * truncateHtml() helper function
+ * (a) push or pop $tag from $openTags as needed
+ * (b) clear $tag value
+ * @param String &$tag Current HTML tag name we are looking at
+ * @param int $tagType (0-open tag, 1-close tag)
+ * @param char $lastCh Character before the '>' that ended this tag
+ * @param array &$openTags Open tag stack (not accounting for $tag)
+ */
private function truncate_endBracket( &$tag, $tagType, $lastCh, &$openTags ) {
$tag = ltrim( $tag );
if ( $tag != '' ) {
}
/**
- * Get the list of variants supported by this langauge
+ * Get the list of variants supported by this language
* see sample implementation in LanguageZh.php
*
* @return array an array of language codes
return $this->mConverter->getVariants();
}
- function getPreferredVariant( $fromUser = true, $fromHeader = false ) {
- return $this->mConverter->getPreferredVariant( $fromUser, $fromHeader );
+ function getPreferredVariant() {
+ return $this->mConverter->getPreferredVariant();
+ }
+
+ function getDefaultVariant() {
+ return $this->mConverter->getDefaultVariant();
+ }
+
+ function getURLVariant() {
+ return $this->mConverter->getURLVariant();
}
/**
* If a language supports multiple variants, converts text
* into an array of all possible variants of the text:
* 'variant' => text in that variant
+ *
+ * @deprecated Use autoConvertToAllVariants()
*/
function convertLinkToAllVariants( $text ) {
return $this->mConverter->convertLinkToAllVariants( $text );
* @return string $prefix . $mangledCode . $suffix
*/
static function getFileName( $prefix = 'Language', $code, $suffix = '.php' ) {
+ // Protect against path traversal
+ if ( !Language::isValidCode( $code ) ) {
+ throw new MWException( "Invalid language code \"$code\"" );
+ }
+
return $prefix . str_replace( '-', '_', ucfirst( $code ) ) . $suffix;
}
throw new MWException(
"Utf8Case.ser is missing, please run \"make\" in the serialized directory\n" );
}
- extract( $arr );
+ $wikiUpperChars = $arr['wikiUpperChars'];
+ $wikiLowerChars = $arr['wikiLowerChars'];
wfProfileOut( __METHOD__ );
return array( $wikiUpperChars, $wikiLowerChars );
}
function formatTimePeriod( $seconds ) {
- if ( $seconds < 10 ) {
- return $this->formatNum( sprintf( "%.1f", $seconds ) ) . $this->getMessageFromDB( 'seconds-abbrev' );
- } elseif ( $seconds < 60 ) {
+ if ( round( $seconds * 10 ) < 100 ) {
+ return $this->formatNum( sprintf( "%.1f", round( $seconds * 10 ) / 10 ) ) . $this->getMessageFromDB( 'seconds-abbrev' );
+ } elseif ( round( $seconds ) < 60 ) {
return $this->formatNum( round( $seconds ) ) . $this->getMessageFromDB( 'seconds-abbrev' );
- } elseif ( $seconds < 3600 ) {
+ } elseif ( round( $seconds ) < 3600 ) {
$minutes = floor( $seconds / 60 );
$secondsPart = round( fmod( $seconds, 60 ) );
if ( $secondsPart == 60 ) {
function getConvRuleTitle() {
return $this->mConverter->getConvRuleTitle();
}
-
- /**
- * Given a string, convert it to a (hopefully short) key that can be used
- * for efficient sorting. A binary sort according to the sortkeys
- * corresponds to a logical sort of the corresponding strings. Current
- * code expects that a null character should sort before all others, but
- * has no other particular expectations (and that one can be changed if
- * necessary).
- *
- * @param string $string UTF-8 string
- * @return string Binary sortkey
- */
- public function convertToSortkey( $string ) {
- # Fake function for now
- return strtoupper( $string );
- }
-
- /**
- * Does it make sense for lists to be split up into sections based on their
- * first letter? Logogram-based scripts probably want to return false.
- *
- * TODO: Use this in CategoryPage.php.
- *
- * @return boolean
- */
- public function usesFirstLettersInLists() {
- return true;
- }
-
- /**
- * Given a string, return the logical "first letter" to be used for
- * grouping on category pages and so on. This has to be coordinated
- * carefully with convertToSortkey(), or else the sorted list might jump
- * back and forth between the same "initial letters" or other pathological
- * behavior. For instance, if you just return the first character, but "a"
- * sorts the same as "A" based on convertToSortkey(), then you might get a
- * list like
- *
- * == A ==
- * * [[Aardvark]]
- *
- * == a ==
- * * [[antelope]]
- *
- * == A ==
- * * [[Ape]]
- *
- * etc., assuming for the sake of argument that $wgCapitalLinks is false.
- * Obviously, this is ignored if usesFirstLettersInLists() is false.
- *
- * @param string $string UTF-8 string
- * @return string UTF-8 string corresponding to the first letter of input
- */
- public function firstLetterForLists( $string ) {
- if ( $string[0] == "\0" ) {
- $string = substr( $string, 1 );
- }
- return strtoupper( $this->firstChar( $string ) );
- }
}
dépôts / lhc / web / wiklou.git / blobdiff