(bug 34939) Handle mixed-case URL protocols in wikitext

This patch marks the regex matching url protocol as being case
insensitive. We will from now render links like [HTTP://ww].

Tests added.

Change-Id: I706acb7a0ae194b50d2318763beae4e5e83671f3

diff --git a/RELEASE-NOTES-1.20 b/RELEASE-NOTES-1.20
index 783d231..01c2f9c 100644 (file)
--- a/RELEASE-NOTES-1.20
+++ b/RELEASE-NOTES-1.20
@@ -141,6 +141,7 @@ upgrade PHP if you have not done so prior to upgrading MediaWiki.
   with auto-hide, multi-message support, and message replacement tags.
 * jquery.messageBox which appears to be unused by both core and extensions has
   been removed.
+* (bug 34939) made link parsking insensitive ([HttP://])
 
 === Bug fixes in 1.20 ===
 * (bug 30245) Use the correct way to construct a log page title.

diff --git a/includes/GlobalFunctions.php b/includes/GlobalFunctions.php
index 0a60518..98bc65f 100644 (file)
--- a/includes/GlobalFunctions.php
+++ b/includes/GlobalFunctions.php
@@ -786,6 +786,9 @@ function wfParseUrl( $url ) {
                return false;
        }
 
+       // parse_url() incorrectly handles schemes case-sensitively. Convert it to lowercase.
+       $bits['scheme'] = strtolower( $bits['scheme'] );
+
        // most of the protocols are followed by ://, but mailto: and sometimes news: not, check for it
        if ( in_array( $bits['scheme'] . '://', $wgUrlProtocols ) ) {
                $bits['delimiter'] = '://';

diff --git a/includes/Sanitizer.php b/includes/Sanitizer.php
index a0c77da..734c4ec 100644 (file)
--- a/includes/Sanitizer.php
+++ b/includes/Sanitizer.php
@@ -1026,7 +1026,7 @@ class Sanitizer {
 
                # Stupid hack
                $encValue = preg_replace_callback(
-                       '/(' . wfUrlProtocols() . ')/',
+                       '/((?i)' . wfUrlProtocols() . ')/',
                        array( 'Sanitizer', 'armorLinksCallback' ),
                        $encValue );
                return $encValue;

diff --git a/includes/Skin.php b/includes/Skin.php
index 00eb5e8..968f215 100644 (file)
--- a/includes/Skin.php
+++ b/includes/Skin.php
@@ -1063,7 +1063,7 @@ abstract class Skin extends ContextSource {
         * @return String URL
         */
        static function makeInternalOrExternalUrl( $name ) {
-               if ( preg_match( '/^(?:' . wfUrlProtocols() . ')/', $name ) ) {
+               if ( preg_match( '/^(?i:' . wfUrlProtocols() . ')/', $name ) ) {
                        return $name;
                } else {
                        return self::makeUrl( $name );
@@ -1227,7 +1227,7 @@ abstract class Skin extends ContextSource {
                                                $text = $line[1];
                                        }
 
-                                       if ( preg_match( '/^(?:' . wfUrlProtocols() . ')/', $link ) ) {
+                                       if ( preg_match( '/^(?i:' . wfUrlProtocols() . ')/', $link ) ) {
                                                $href = $link;
 
                                                // Parser::getExternalLinkAttribs won't work here because of the Namespace things

diff --git a/includes/api/ApiFormatBase.php b/includes/api/ApiFormatBase.php
index 54c90a6..8ad9b8c 100644 (file)
--- a/includes/api/ApiFormatBase.php
+++ b/includes/api/ApiFormatBase.php
@@ -271,7 +271,7 @@ See the <a href='https://www.mediawiki.org/wiki/API'>complete documentation</a>,
                // identify URLs
                $protos = wfUrlProtocolsWithoutProtRel();
                // This regex hacks around bug 13218 (&quot; included in the URL)
-               $text = preg_replace( "#(($protos).*?)(&quot;)?([ \\'\"<>\n]|&lt;|&gt;|&quot;)#", '<a href="\\1">\\1</a>\\3\\4', $text );
+               $text = preg_replace( "#(((?i)$protos).*?)(&quot;)?([ \\'\"<>\n]|&lt;|&gt;|&quot;)#", '<a href="\\1">\\1</a>\\3\\4', $text );
                // identify requests to api.php
                $text = preg_replace( "#api\\.php\\?[^ <\n\t]+#", '<a href="\\0">\\0</a>', $text );
                if ( $this->mHelp ) {

diff --git a/includes/parser/Parser.php b/includes/parser/Parser.php
index 8e9444a..de55de0 100644 (file)
--- a/includes/parser/Parser.php
+++ b/includes/parser/Parser.php
@@ -207,7 +207,7 @@ class Parser {
        public function __construct( $conf = array() ) {
                $this->mConf = $conf;
                $this->mUrlProtocols = wfUrlProtocols();
-               $this->mExtLinkBracketedRegex = '/\[((' . $this->mUrlProtocols . ')'.
+               $this->mExtLinkBracketedRegex = '/\[(((?i)' . $this->mUrlProtocols . ')'.
                        self::EXT_LINK_URL_CLASS.'+)\p{Zs}*([^\]\\x00-\\x08\\x0a-\\x1F]*?)\]/Su';
                if ( isset( $conf['preprocessorClass'] ) ) {
                        $this->mPreprocessorClass = $conf['preprocessorClass'];
@@ -1187,7 +1187,7 @@ class Parser {
                        '!(?:                           # Start cases
                                (<a[ \t\r\n>].*?</a>) |     # m[1]: Skip link text
                                (<.*?>) |                   # m[2]: Skip stuff inside HTML elements' . "
-                               (\\b(?:$prots)$urlChar+) |  # m[3]: Free external links" . '
+                               (\\b(?i:$prots)$urlChar+) |  # m[3]: Free external links" . '
                                (?:RFC|PMID)\s+([0-9]+) |   # m[4]: RFC or PMID, capture number
                                ISBN\s+(\b                  # m[5]: ISBN, capture number
                                        (?: 97[89] [\ \-]? )?   # optional 13-digit ISBN prefix
@@ -1853,7 +1853,7 @@ class Parser {
                        # Don't allow internal links to pages containing
                        # PROTO: where PROTO is a valid URL protocol; these
                        # should be external links.
-                       if ( preg_match( '/^(?:' . $this->mUrlProtocols . ')/', $m[1] ) ) {
+                       if ( preg_match( '/^(?i:' . $this->mUrlProtocols . ')/', $m[1] ) ) {
                                $s .= $prefix . '[[' . $line ;
                                wfProfileOut( __METHOD__."-misc" );
                                continue;
@@ -2090,7 +2090,7 @@ class Parser {
         * @return String: less-or-more HTML with NOPARSE bits
         */
        function armorLinks( $text ) {
-               return preg_replace( '/\b(' . $this->mUrlProtocols . ')/',
+               return preg_replace( '/\b((?i)' . $this->mUrlProtocols . ')/',
                        "{$this->mUniqPrefix}NOPARSE$1", $text );
        }
 
@@ -5095,8 +5095,8 @@ class Parser {
                                                                $paramName = 'no-link';
                                                                $value = true;
                                                                $validated = true;
-                                                       } elseif ( preg_match( "/^$prots/", $value ) ) {
-                                                               if ( preg_match( "/^($prots)$chars+$/u", $value, $m ) ) {
+                                                       } elseif ( preg_match( "/^(?i)$prots/", $value ) ) {
+                                                               if ( preg_match( "/^((?i)$prots)$chars+$/u", $value, $m ) ) {
                                                                        $paramName = 'link-url';
                                                                        $this->mOutput->addExternalLink( $value );
                                                                        if ( $this->mOptions->getExternalLinkTarget() ) {
@@ -5622,7 +5622,7 @@ class Parser {
                # @todo FIXME: Not tolerant to blank link text
                # I.E. [http://www.mediawiki.org] will render as [1] or something depending
                # on how many empty links there are on the page - need to figure that out.
-               $text = preg_replace( '/\[(?:' . $this->mUrlProtocols . ')([^ ]+?) ([^[]+)\]/', '$2', $text );
+               $text = preg_replace( '/\[(?i:' . $this->mUrlProtocols . ')([^ ]+?) ([^[]+)\]/', '$2', $text );
 
                # Parse wikitext quotes (italics & bold)
                $text = $this->doQuotes( $text );

diff --git a/includes/parser/Parser_LinkHooks.php b/includes/parser/Parser_LinkHooks.php
index 9555bdb..6bcc324 100644 (file)
--- a/includes/parser/Parser_LinkHooks.php
+++ b/includes/parser/Parser_LinkHooks.php
@@ -226,7 +226,7 @@ class Parser_LinkHooks extends Parser {
                # Don't allow internal links to pages containing
                # PROTO: where PROTO is a valid URL protocol; these
                # should be external links.
-               if( preg_match('/^\b(?:' . wfUrlProtocols() . ')/', $titleText) ) {
+               if( preg_match('/^\b(?i:' . wfUrlProtocols() . ')/', $titleText) ) {
                        wfProfileOut( __METHOD__ );
                        return $wt;
                }

diff --git a/tests/parser/parserTests.txt b/tests/parser/parserTests.txt
index 453e88a..c418116 100644 (file)
--- a/tests/parser/parserTests.txt
+++ b/tests/parser/parserTests.txt
@@ -10343,6 +10343,33 @@ abc
 </p>
 !! end
 
+!!test
+Bug 34939 - Case insensitive link parsing ([HttP://])
+!! input
+[HttP://MediaWiki.Org/]
+!! result
+<p><a rel="nofollow" class="external autonumber" href="HttP://MediaWiki.Org/">[1]</a>
+</p>
+!! end
+
+!!test
+Bug 34939 - Case insensitive link parsing ([HttP:// title])
+!! input
+[HttP://MediaWiki.Org/ MediaWiki]
+!! result
+<p><a rel="nofollow" class="external text" href="HttP://MediaWiki.Org/">MediaWiki</a>
+</p>
+!! end
+
+!!test
+Bug 34939 - Case insensitive link parsing (HttP://)
+!! input
+HttP://MediaWiki.Org/
+!! result
+<p><a rel="nofollow" class="external free" href="HttP://MediaWiki.Org/">HttP://MediaWiki.Org/</a>
+</p>
+!! end
+
 
 TODO:
 more images