* language, script or variant codes actually exist in the repositories.
*
* Based on regexes by Mark Davis of the Unicode Consortium:
- * http://unicode.org/repos/cldr/trunk/tools/java/org/unicode/cldr/util/data/langtagRegex.txt
+ * https://www.unicode.org/repos/cldr/trunk/tools/java/org/unicode/cldr/util/data/langtagRegex.txt
*
* @param string $code
* @param bool $lenient Whether to allow '_' as separator. The default is only '-'.
/**
* Get an array of language names, indexed by code.
* @param null|string $inLanguage Code of language in which to return the names
- * Use null for autonyms (native names)
+ * Use null for autonyms (native names)
* @param string $include One of:
- * 'all' all available languages
- * 'mw' only if the language is defined in MediaWiki or wgExtraLanguageNames (default)
- * 'mwfile' only if the language is in 'mw' *and* has a message file
- * @return array Language code => language name
+ * 'all' all available languages
+ * 'mw' only if the language is defined in MediaWiki or wgExtraLanguageNames (default)
+ * 'mwfile' only if the language is in 'mw' *and* has a message file
+ * @return array Language code => language name (sorted by key)
* @since 1.20
*/
public static function fetchLanguageNames( $inLanguage = null, $include = 'mw' ) {
* 'all' all available languages
* 'mw' only if the language is defined in MediaWiki or wgExtraLanguageNames (default)
* 'mwfile' only if the language is in 'mw' *and* has a message file
- * @return array Language code => language name
+ * @return array Language code => language name (sorted by key)
*/
private static function fetchLanguageNamesUncached( $inLanguage = null, $include = 'mw' ) {
global $wgExtraLanguageNames, $wgUsePigLatinVariant;
* Gets directionality of the first strongly directional codepoint, for embedBidi()
*
* This is the rule the BIDI algorithm uses to determine the directionality of
- * paragraphs ( http://unicode.org/reports/tr9/#The_Paragraph_Level ) and
- * FSI isolates ( http://unicode.org/reports/tr9/#Explicit_Directional_Isolates ).
+ * paragraphs ( https://www.unicode.org/reports/tr9/#The_Paragraph_Level ) and
+ * FSI isolates ( https://www.unicode.org/reports/tr9/#Explicit_Directional_Isolates ).
*
* TODO: Does not handle BIDI control characters inside the text.
* TODO: Does not handle unallocated characters.
global $wgAllUnicodeFixes;
$s = UtfNormal\Validator::cleanUp( $s );
if ( $wgAllUnicodeFixes ) {
- $s = $this->transformUsingPairFile( 'normalize-ar.ser', $s );
- $s = $this->transformUsingPairFile( 'normalize-ml.ser', $s );
+ $s = $this->transformUsingPairFile( 'normalize-ar.php', $s );
+ $s = $this->transformUsingPairFile( 'normalize-ml.php', $s );
}
return $s;
* @throws MWException
* @return string
*/
- function transformUsingPairFile( $file, $string ) {
+ protected function transformUsingPairFile( $file, $string ) {
if ( !isset( $this->transformData[$file] ) ) {
- $data = wfGetPrecompiledData( $file );
- if ( $data === false ) {
- throw new MWException( __METHOD__ . ": The transformation file $file is missing" );
- }
+ global $IP;
+ $data = require "$IP/languages/data/{$file}";
$this->transformData[$file] = new ReplacementArray( $data );
}
return $this->transformData[$file]->replace( $string );
}
/**
- * Truncate a string to a specified length in bytes, appending an optional
- * string (e.g. for ellipses)
+ * This method is deprecated since 1.31 and kept as alias for truncateForDatabase, which
+ * has replaced it. This method provides truncation suitable for DB.
*
* The database offers limited byte lengths for some columns in the database;
* multi-byte character sets mean we need to ensure that only whole characters
- * are included, otherwise broken characters can be passed to the user
+ * are included, otherwise broken characters can be passed to the user.
*
- * If $length is negative, the string will be truncated from the beginning
+ * @deprecated since 1.31, use truncateForDatabase or truncateForVisual as appropriate.
*
* @param string $string String to truncate
- * @param int $length Maximum length (including ellipses)
+ * @param int $length Maximum length (including ellipsis)
* @param string $ellipsis String to append to the truncated text
* @param bool $adjustLength Subtract length of ellipsis from $length.
- * $adjustLength was introduced in 1.18, before that behaved as if false.
+ * $adjustLength was introduced in 1.18, before that behaved as if false.
* @return string
*/
function truncate( $string, $length, $ellipsis = '...', $adjustLength = true ) {
+ return $this->truncateForDatabase( $string, $length, $ellipsis, $adjustLength );
+ }
+
+ /**
+ * Truncate a string to a specified length in bytes, appending an optional
+ * string (e.g. for ellipsis)
+ *
+ * If $length is negative, the string will be truncated from the beginning
+ *
+ * @since 1.31
+ *
+ * @param string $string String to truncate
+ * @param int $length Maximum length in bytes
+ * @param string $ellipsis String to append to the end of truncated text
+ * @param bool $adjustLength Subtract length of ellipsis from $length
+ *
+ * @return string
+ */
+ function truncateForDatabase( $string, $length, $ellipsis = '...', $adjustLength = true ) {
+ return $this->truncateInternal(
+ $string, $length, $ellipsis, $adjustLength, 'strlen', 'substr'
+ );
+ }
+
+ /**
+ * Truncate a string to a specified number of characters, appending an optional
+ * string (e.g. for ellipsis).
+ *
+ * This provides multibyte version of truncate() method of this class, suitable for truncation
+ * based on number of characters, instead of number of bytes.
+ *
+ * If $length is negative, the string will be truncated from the beginning.
+ *
+ * @since 1.31
+ *
+ * @param string $string String to truncate
+ * @param int $length Maximum number of characters
+ * @param string $ellipsis String to append to the end of truncated text
+ * @param bool $adjustLength Subtract length of ellipsis from $length
+ *
+ * @return string
+ */
+ function truncateForVisual( $string, $length, $ellipsis = '...', $adjustLength = true ) {
+ // Passing encoding to mb_strlen and mb_substr is optional.
+ // Encoding defaults to mb_internal_encoding(), which is set to UTF-8 in Setup.php, so
+ // explicit specification of encoding is skipped.
+ // Note: Both multibyte methods are callables invoked in truncateInternal.
+ return $this->truncateInternal(
+ $string, $length, $ellipsis, $adjustLength, 'mb_strlen', 'mb_substr'
+ );
+ }
+
+ /**
+ * Internal method used for truncation. This method abstracts text truncation into
+ * one common method, allowing users to provide length measurement function and
+ * function for finding substring.
+ *
+ * For usages, see truncateForDatabase and truncateForVisual.
+ *
+ * @param string $string String to truncate
+ * @param int $length Maximum length of final text
+ * @param string $ellipsis String to append to the end of truncated text
+ * @param bool $adjustLength Subtract length of ellipsis from $length
+ * @param callable $measureLength Callable function used for determining the length of text
+ * @param callable $getSubstring Callable function used for getting the substrings
+ *
+ * @return string
+ */
+ private function truncateInternal(
+ $string, $length, $ellipsis = '...', $adjustLength = true, $measureLength, $getSubstring
+ ) {
+ if ( !is_callable( $measureLength ) || !is_callable( $getSubstring ) ) {
+ throw new InvalidArgumentException( 'Invalid callback provided' );
+ }
+
# Check if there is no need to truncate
- if ( strlen( $string ) <= abs( $length ) ) {
+ if ( $measureLength( $string ) <= abs( $length ) ) {
return $string; // no need to truncate
}
+
# Use the localized ellipsis character
if ( $ellipsis == '...' ) {
$ellipsis = wfMessage( 'ellipsis' )->inLanguage( $this )->escaped();
if ( $length == 0 ) {
return $ellipsis; // convention
}
+
$stringOriginal = $string;
# If ellipsis length is >= $length then we can't apply $adjustLength
- if ( $adjustLength && strlen( $ellipsis ) >= abs( $length ) ) {
+ if ( $adjustLength && $measureLength( $ellipsis ) >= abs( $length ) ) {
$string = $ellipsis; // this can be slightly unexpected
# Otherwise, truncate and add ellipsis...
} else {
- $eLength = $adjustLength ? strlen( $ellipsis ) : 0;
+ $ellipsisLength = $adjustLength ? $measureLength( $ellipsis ) : 0;
if ( $length > 0 ) {
- $length -= $eLength;
- $string = substr( $string, 0, $length ); // xyz...
+ $length -= $ellipsisLength;
+ $string = $getSubstring( $string, 0, $length ); // xyz...
$string = $this->removeBadCharLast( $string );
$string = rtrim( $string );
$string = $string . $ellipsis;
} else {
- $length += $eLength;
- $string = substr( $string, $length ); // ...xyz
+ $length += $ellipsisLength;
+ $string = $getSubstring( $string, $length ); // ...xyz
$string = $this->removeBadCharFirst( $string );
$string = ltrim( $string );
$string = $ellipsis . $string;
}
}
+
# Do not truncate if the ellipsis makes the string longer/equal (T24181).
# This check is *not* redundant if $adjustLength, due to the single case where
# LEN($ellipsis) > ABS($limit arg); $stringOriginal could be shorter than $string.
- if ( strlen( $string ) < strlen( $stringOriginal ) ) {
+ if ( $measureLength( $string ) < $measureLength( $stringOriginal ) ) {
return $string;
} else {
return $stringOriginal;
return $this->mConverter;
}
+ /**
+ * convert text to a variant
+ *
+ * @param string $text text to convert
+ * @param string|bool $variant variant to convert to, or false to use the user's preferred
+ * variant (if logged in), or the project default variant
+ * @return string the converted string
+ */
+ public function autoConvert( $text, $variant = false ) {
+ return $this->mConverter->autoConvert( $text, $variant );
+ }
+
/**
* convert text to all supported variants
*
/**
* Convert a namespace index to a string in the preferred variant
*
- * @param int $ns
- * @return string
+ * @param int $ns namespace index (https://www.mediawiki.org/wiki/Manual:Namespace)
+ * @param string|null $variant variant to convert to, or null to use the user's preferred
+ * variant (if logged in), or the project default variant
+ * @return string a string representation of the namespace
*/
- public function convertNamespace( $ns ) {
- return $this->mConverter->convertNamespace( $ns );
+ public function convertNamespace( $ns, $variant = null ) {
+ return $this->mConverter->convertNamespace( $ns, $variant );
}
/**
* @param string $details HTML safe text between brackets
* @param bool $oppositedm Add the direction mark opposite to your
* language, to display text properly
- * @return HTML escaped string
+ * @return string HTML escaped
*/
function specialList( $page, $details, $oppositedm = true ) {
if ( !$details ) {
dépôts / lhc / web / wiklou.git / blobdiff