Don't allow embedded newlines in magic links, but do allow  

This continues the work started in T67278 to make magic link parsing
more consistent with wiki text parsing in general, and closes two
long-standing bugs.

Bug: T30950
Bug: T31025
Change-Id: I71f8b337543163569c64bbfdec154eb9b69d7264

diff --git a/RELEASE-NOTES-1.25 b/RELEASE-NOTES-1.25
index 1956eb6..9183e44 100644 (file)
--- a/RELEASE-NOTES-1.25
+++ b/RELEASE-NOTES-1.25
@@ -279,6 +279,8 @@ changes to languages because of Bugzilla reports.
    However, this difference is unlikely to arise in practice.
 * (T67278) RFC, PMID, and ISBN "magic links" must be surrounded by non-word
   characters on both sides.
+* (T30950, T31025) RFC, PMID, and ISBN "magic links" can no longer contain
+  newlines; but they can contain   and other non-newline whitespace.
 
 == Compatibility ==
 

diff --git a/includes/parser/Parser.php b/includes/parser/Parser.php
index ecb14ed..e3a4ea5 100644 (file)
--- a/includes/parser/Parser.php
+++ b/includes/parser/Parser.php
@@ -90,6 +90,9 @@ class Parser {
        const EXT_IMAGE_REGEX = '/^(http:\/\/|https:\/\/)([^][<>"\\x00-\\x20\\x7F\p{Zs}]+)
                \\/([A-Za-z0-9_.,~%\\-+&;#*?!=()@\\x80-\\xFF]+)\\.((?i)gif|png|jpg|jpeg)$/Sxu';
 
+       # Regular expression for a non-newline space
+       const SPACE_NOT_NL = '(?:\t|&nbsp;|&\#0*160;|&\#[Xx]0*[Aa]0;|\p{Zs})';
+
        # State constants for the definition list colon extraction
        const COLON_STATE_TEXT = 0;
        const COLON_STATE_TAG = 1;
@@ -1389,18 +1392,22 @@ class Parser {
                wfProfileIn( __METHOD__ );
                $prots = wfUrlProtocolsWithoutProtRel();
                $urlChar = self::EXT_LINK_URL_CLASS;
+               $space = self::SPACE_NOT_NL; #  non-newline space
+               $spdash = "(?:-|$space)"; # a dash or a non-newline space
+               $spaces = "$space++"; # possessive match of 1 or more spaces
                $text = preg_replace_callback(
                        '!(?:                           # Start cases
                                (<a[ \t\r\n>].*?</a>) |     # m[1]: Skip link text
                                (<.*?>) |                   # m[2]: Skip stuff inside HTML elements' . "
-                               (\b(?i:$prots)$urlChar+) |  # m[3]: Free external links" . '
-                               \b(?:RFC|PMID)\s+([0-9]+)\b |# m[4]: RFC or PMID, capture number
-                               \bISBN\s+(                  # m[5]: ISBN, capture number
-                                       (?: 97[89] [\ \-]? )?   # optional 13-digit ISBN prefix
-                                       (?: [0-9]  [\ \-]? ){9} # 9 digits with opt. delimiters
+                               (\b(?i:$prots)$urlChar+) |  # m[3]: Free external links
+                               \b(?:RFC|PMID) $spaces      # m[4]: RFC or PMID, capture number
+                                       ([0-9]+)\b |
+                               \bISBN $spaces (            # m[5]: ISBN, capture number
+                                       (?: 97[89] $spdash? )?   # optional 13-digit ISBN prefix
+                                       (?: [0-9]  $spdash? ){9} # 9 digits with opt. delimiters
                                        [0-9Xx]                 # check digit
-                                       )\b
-                       )!xu', array( &$this, 'magicLinkCallback' ), $text );
+                               )\b
+                       )!xu", array( &$this, 'magicLinkCallback' ), $text );
                wfProfileOut( __METHOD__ );
                return $text;
        }
@@ -1441,6 +1448,8 @@ class Parser {
                } elseif ( isset( $m[5] ) && $m[5] !== '' ) {
                        # ISBN
                        $isbn = $m[5];
+                       $space = self::SPACE_NOT_NL; #  non-newline space
+                       $isbn = preg_replace( "/$space/", ' ', $isbn );
                        $num = strtr( $isbn, array(
                                '-' => '',
                                ' ' => '',

diff --git a/tests/parser/parserTests.txt b/tests/parser/parserTests.txt
index f7dc0a9..cf9d829 100644 (file)
--- a/tests/parser/parserTests.txt
+++ b/tests/parser/parserTests.txt
@@ -8934,6 +8934,19 @@ This is RFC 822 but thisRFC 822 is not RFC 822linked.
 </p>
 !! end
 
+!! test
+Magic links: RFC (w/ non-newline whitespace, bug 28950/29025)
+!! wikitext
+RFC &nbsp;&#160;&#0160;&#xA0;&#Xa0; 822
+RFC
+822
+!! html
+<p><a class="external mw-magiclink-rfc" rel="nofollow" href="//tools.ietf.org/html/rfc822">RFC 822</a>
+RFC
+822
+</p>
+!! end
+
 !! test
 Magic links: ISBN (bug 1937)
 !! wikitext
@@ -8952,6 +8965,23 @@ This is ISBN 978-0-316-09811-3 but thisISBN 978-0-316-09811-3 is not ISBN 978-0-
 </p>
 !! end
 
+!! test
+Magic links: ISBN (w/ non-newline whitespace, bug 28950/29025)
+!! wikitext
+ISBN &nbsp;&#160;&#0160;&#xA0;&#Xa0; 978&nbsp;0&#160;316&#0160;09811&#xA0;3
+ISBN
+9780316098113
+ISBN 978
+0316098113
+!! html
+<p><a href="/wiki/Special:BookSources/9780316098113" class="internal mw-magiclink-isbn">ISBN 978 0 316 09811 3</a>
+ISBN
+9780316098113
+ISBN 978
+0316098113
+</p>
+!! end
+
 !! test
 Magic links: PMID incorrectly converts space to underscore
 !! wikitext
@@ -8970,6 +9000,19 @@ This is PMID 1234 but thisPMID 1234 is not PMID 1234linked.
 </p>
 !! end
 
+!! test
+Magic links: PMID (w/ non-newline whitespace, bug 28950/29025)
+!! wikitext
+PMID &nbsp;&#160;&#0160;&#xA0;&#Xa0; 1234
+PMID
+1234
+!! html
+<p><a class="external mw-magiclink-pmid" rel="nofollow" href="//www.ncbi.nlm.nih.gov/pubmed/1234?dopt=Abstract">PMID 1234</a>
+PMID
+1234
+</p>
+!! end
+
 ###
 ### Templates
 ####