MediaWiki  1.29.1
LinkFilterTest.php
Go to the documentation of this file.
1 <?php
2 
3 use Wikimedia\Rdbms\LikeMatch;
4 
8 class LinkFilterTest extends MediaWikiLangTestCase {
9 
10  protected function setUp() {
11  parent::setUp();
12 
13  $this->setMwGlobals( 'wgUrlProtocols', [
14  'http://',
15  'https://',
16  'ftp://',
17  'irc://',
18  'ircs://',
19  'gopher://',
20  'telnet://',
21  'nntp://',
22  'worldwind://',
23  'mailto:',
24  'news:',
25  'svn://',
26  'git://',
27  'mms://',
28  '//',
29  ] );
30  }
31 
40  function createRegexFromLIKE( $like ) {
41  $regex = '!^';
42 
43  foreach ( $like as $item ) {
44  if ( $item instanceof LikeMatch ) {
45  if ( $item->toString() == '%' ) {
46  $regex .= '.*';
47  } elseif ( $item->toString() == '_' ) {
48  $regex .= '.';
49  }
50  } else {
51  $regex .= preg_quote( $item, '!' );
52  }
53 
54  }
55 
56  $regex .= '$!';
57 
58  return $regex;
59  }
60 
66  public static function provideValidPatterns() {
67  return [
68  // Protocol, Search pattern, URL which matches the pattern
69  [ 'http://', '*.test.com', 'http://www.test.com' ],
70  [ 'http://', 'test.com:8080/dir/file', 'http://name:pass@test.com:8080/dir/file' ],
71  [ 'https://', '*.com', 'https://s.s.test..com:88/dir/file?a=1&b=2' ],
72  [ 'https://', '*.com', 'https://name:pass@secure.com/index.html' ],
73  [ 'http://', 'name:pass@test.com', 'http://test.com' ],
74  [ 'http://', 'test.com', 'http://name:pass@test.com' ],
75  [ 'http://', '*.test.com', 'http://a.b.c.test.com/dir/dir/file?a=6' ],
76  [ null, 'http://*.test.com', 'http://www.test.com' ],
77  [ 'mailto:', 'name@mail.test123.com', 'mailto:name@mail.test123.com' ],
78  [ '',
79  'http://name:pass@www.test.com:12345/dir/dir/file.xyz.php#__se__?arg1=_&arg2[]=4rtg',
80  'http://name:pass@www.test.com:12345/dir/dir/file.xyz.php#__se__?arg1=_&arg2[]=4rtg'
81  ],
82  [ '', 'http://name:pass@*.test.com:12345/dir/dir/file.xyz.php#__se__?arg1=_&arg2[]=4rtg',
83  'http://name:pass@www.test.com:12345/dir/dir/file.xyz.php#__se__?arg1=_&arg2[]=4rtg' ],
84  [ '', 'http://name:wrongpass@*.com:12345/dir/dir/file.xyz.php#__se__?arg1=_&arg2[]',
85  'http://name:pass@www.test.com:12345/dir/dir/file.xyz.php#__se__?arg1=_&arg2[]=4rtg' ],
86  [ 'http://', 'name:pass@*.test.com:12345/dir/dir/file.xyz.php#__se__?arg1=_&arg2[]=4rtg',
87  'http://name:pass@www.test.com:12345/dir/dir/file.xyz.php#__se__?arg1=_&arg2[]=4rtg' ],
88  [ '', 'http://name:pass@www.test.com:12345',
89  'http://name:pass@www.test.com:12345/dir/dir/file.xyz.php#__se__?arg1=_&arg2[]=4rtg' ],
90  [ 'ftp://', 'user:pass@ftp.test.com:1233/home/user/file;type=efw',
91  'ftp://user:pass@ftp.test.com:1233/home/user/file;type=efw' ],
92  [ null, 'ftp://otheruser:otherpass@ftp.test.com:1233/home/user/file;type=',
93  'ftp://user:pass@ftp.test.com:1233/home/user/file;type=efw' ],
94  [ null, 'ftp://@ftp.test.com:1233/home/user/file;type=',
95  'ftp://user:pass@ftp.test.com:1233/home/user/file;type=efw' ],
96  [ null, 'ftp://ftp.test.com/',
97  'ftp://user:pass@ftp.test.com/home/user/file;type=efw' ],
98  [ null, 'ftp://ftp.test.com/',
99  'ftp://user:pass@ftp.test.com/home/user/file;type=efw' ],
100  [ null, 'ftp://*.test.com:222/',
101  'ftp://user:pass@ftp.test.com:222/home' ],
102  [ 'irc://', '*.myserver:6667/', 'irc://test.myserver:6667/' ],
103  [ 'irc://', 'name:pass@*.myserver/', 'irc://test.myserver:6667/' ],
104  [ 'irc://', 'name:pass@*.myserver/', 'irc://other:@test.myserver:6667/' ],
105  [ '', 'irc://test/name,string,abc?msg=t', 'irc://test/name,string,abc?msg=test' ],
106  [ '', 'https://gerrit.wikimedia.org/r/#/q/status:open,n,z',
107  'https://gerrit.wikimedia.org/r/#/q/status:open,n,z' ],
108  [ '', 'https://gerrit.wikimedia.org',
109  'https://gerrit.wikimedia.org/r/#/q/status:open,n,z' ],
110  [ 'mailto:', '*.test.com', 'mailto:name@pop3.test.com' ],
111  [ 'mailto:', 'test.com', 'mailto:name@test.com' ],
112  [ 'news:', 'test.1234afc@news.test.com', 'news:test.1234afc@news.test.com' ],
113  [ 'news:', '*.test.com', 'news:test.1234afc@news.test.com' ],
114  [ '', 'news:4df8kh$iagfewewf(at)newsbf02aaa.news.aol.com',
115  'news:4df8kh$iagfewewf(at)newsbf02aaa.news.aol.com' ],
116  [ '', 'news:*.aol.com',
117  'news:4df8kh$iagfewewf(at)newsbf02aaa.news.aol.com' ],
118  [ '', 'git://github.com/prwef/abc-def.git', 'git://github.com/prwef/abc-def.git' ],
119  [ 'git://', 'github.com/', 'git://github.com/prwef/abc-def.git' ],
120  [ 'git://', '*.github.com/', 'git://a.b.c.d.e.f.github.com/prwef/abc-def.git' ],
121  [ '', 'gopher://*.test.com/', 'gopher://gopher.test.com/0/v2/vstat' ],
122  [ 'telnet://', '*.test.com', 'telnet://shell.test.com/~home/' ],
123  [ '', 'http://test.com', 'http://test.com/index?arg=1' ],
124  [ 'http://', '*.test.com', 'http://www.test.com/index?arg=1' ],
125  [ '' ,
126  'http://xx23124:__ffdfdef__@www.test.com:12345/dir' ,
127  'http://name:pass@www.test.com:12345/dir/dir/file.xyz.php#__se__?arg1=_&arg2[]=4rtg'
128  ],
129 
130  // Tests for false positives
131  [ 'http://', 'test.com', 'http://www.test.com', false ],
132  [ 'http://', 'www1.test.com', 'http://www.test.com', false ],
133  [ 'http://', '*.test.com', 'http://www.test.t.com', false ],
134  [ '', 'http://test.com:8080', 'http://www.test.com:8080', false ],
135  [ '', 'https://test.com', 'http://test.com', false ],
136  [ '', 'http://test.com', 'https://test.com', false ],
137  [ 'http://', 'http://test.com', 'http://test.com', false ],
138  [ null, 'http://www.test.com', 'http://www.test.com:80', false ],
139  [ null, 'http://www.test.com:80', 'http://www.test.com', false ],
140  [ null, 'http://*.test.com:80', 'http://www.test.com', false ],
141  [ '', 'https://gerrit.wikimedia.org/r/#/XXX/status:open,n,z',
142  'https://gerrit.wikimedia.org/r/#/q/status:open,n,z', false ],
143  [ '', 'https://*.wikimedia.org/r/#/q/status:open,n,z',
144  'https://gerrit.wikimedia.org/r/#/XXX/status:open,n,z', false ],
145  [ 'mailto:', '@test.com', '@abc.test.com', false ],
146  [ 'mailto:', 'mail@test.com', 'mail2@test.com', false ],
147  [ '', 'mailto:mail@test.com', 'mail2@test.com', false ],
148  [ '', 'mailto:@test.com', '@abc.test.com', false ],
149  [ 'ftp://', '*.co', 'ftp://www.co.uk', false ],
150  [ 'ftp://', '*.co', 'ftp://www.co.m', false ],
151  [ 'ftp://', '*.co/dir/', 'ftp://www.co/dir2/', false ],
152  [ 'ftp://', 'www.co/dir/', 'ftp://www.co/dir2/', false ],
153  [ 'ftp://', 'test.com/dir/', 'ftp://test.com/', false ],
154  [ '', 'http://test.com:8080/dir/', 'http://test.com:808/dir/', false ],
155  [ '', 'http://test.com/dir/index.html', 'http://test.com/dir/index.php', false ],
156 
157  // These are false positives too and ideally shouldn't match, but that
158  // would require using regexes and RLIKE instead of LIKE
159  // [ null, 'http://*.test.com', 'http://www.test.com:80', false ],
160  // [ '', 'https://*.wikimedia.org/r/#/q/status:open,n,z',
161  // 'https://gerrit.wikimedia.org/XXX/r/#/q/status:open,n,z', false ],
162  ];
163  }
164 
178  function testMakeLikeArrayWithValidPatterns( $protocol, $pattern, $url, $shouldBeFound = true ) {
179  $indexes = wfMakeUrlIndexes( $url );
180  $likeArray = LinkFilter::makeLikeArray( $pattern, $protocol );
181 
182  $this->assertTrue( $likeArray !== false,
183  "LinkFilter::makeLikeArray('$pattern', '$protocol') returned false on a valid pattern"
184  );
185 
186  $regex = $this->createRegexFromLIKE( $likeArray );
187  $debugmsg = "Regex: '" . $regex . "'\n";
188  $debugmsg .= count( $indexes ) . " index(es) created by wfMakeUrlIndexes():\n";
189 
190  $matches = 0;
191 
192  foreach ( $indexes as $index ) {
193  $matches += preg_match( $regex, $index );
194  $debugmsg .= "\t'$index'\n";
195  }
196 
197  if ( $shouldBeFound ) {
198  $this->assertTrue(
199  $matches > 0,
200  "Search pattern '$protocol$pattern' does not find url '$url' \n$debugmsg"
201  );
202  } else {
203  $this->assertFalse(
204  $matches > 0,
205  "Search pattern '$protocol$pattern' should not find url '$url' \n$debugmsg"
206  );
207  }
208  }
209 
215  public static function provideInvalidPatterns() {
216  return [
217  [ '' ],
218  [ '*' ],
219  [ 'http://*' ],
220  [ 'http://*/' ],
221  [ 'http://*/dir/file' ],
222  [ 'test.*.com' ],
223  [ 'http://test.*.com' ],
224  [ 'test.*.com' ],
225  [ 'http://*.test.*' ],
226  [ 'http://*test.com' ],
227  [ 'https://*' ],
228  [ '*://test.com' ],
229  [ 'mailto:name:pass@t*est.com' ],
230  [ 'http://*:888/' ],
231  [ '*http://' ],
232  [ 'test.com/*/index' ],
233  [ 'test.com/dir/index?arg=*' ],
234  ];
235  }
236 
246  function testMakeLikeArrayWithInvalidPatterns( $pattern ) {
247  $this->assertFalse(
248  LinkFilter::makeLikeArray( $pattern ),
249  "'$pattern' is not a valid pattern and should be rejected"
250  );
251  }
252 
253 }
LinkFilterTest\provideValidPatterns
static provideValidPatterns()
provideValidPatterns()
Definition: LinkFilterTest.php:66
false
processing should stop and the error should be shown to the user * false
Definition: hooks.txt:189
captcha-old.count
count
Definition: captcha-old.py:225
wfMakeUrlIndexes
wfMakeUrlIndexes( $url)
Make URL indexes, appropriate for the el_index field of externallinks.
Definition: GlobalFunctions.php:911
use
as see the revision history and available at free of to any person obtaining a copy of this software and associated documentation to deal in the Software without including without limitation the rights to use
Definition: MIT-LICENSE.txt:10
LinkFilterTest\testMakeLikeArrayWithInvalidPatterns
testMakeLikeArrayWithInvalidPatterns( $pattern)
testMakeLikeArrayWithInvalidPatterns()
Definition: LinkFilterTest.php:246
php
injection txt This is an overview of how MediaWiki makes use of dependency injection The design described here grew from the discussion of RFC T384 The term dependency this means that anything an object needs to operate should be injected from the the object itself should only know narrow no concrete implementation of the logic it relies on The requirement to inject everything typically results in an architecture that based on two main types of and essentially stateless service objects that use other service objects to operate on the value objects As of the beginning MediaWiki is only starting to use the DI approach Much of the code still relies on global state or direct resulting in a highly cyclical dependency which acts as the top level factory for services in MediaWiki which can be used to gain access to default instances of various services MediaWikiServices however also allows new services to be defined and default services to be redefined Services are defined or redefined by providing a callback the instantiator that will return a new instance of the service When it will create an instance of MediaWikiServices and populate it with the services defined in the files listed by thereby bootstrapping the DI framework Per $wgServiceWiringFiles lists includes ServiceWiring php
Definition: injection.txt:35
MediaWikiTestCase\setMwGlobals
setMwGlobals( $pairs, $value=null)
Definition: MediaWikiTestCase.php:658
$matches
$matches
Definition: NoLocalSettings.php:24
LinkFilter\makeLikeArray
static makeLikeArray( $filterEntry, $protocol='http://')
Make an array to be used for calls to Database::buildLike(), which will match the specified string.
Definition: LinkFilter.php:95
LinkFilterTest\setUp
setUp()
Definition: LinkFilterTest.php:10
MediaWikiLangTestCase
Base class that store and restore the Language objects.
Definition: MediaWikiLangTestCase.php:6
Wikimedia\Rdbms\LikeMatch
Used by Database::buildLike() to represent characters that have special meaning in SQL LIKE clauses a...
Definition: LikeMatch.php:10
LinkFilterTest\testMakeLikeArrayWithValidPatterns
testMakeLikeArrayWithValidPatterns( $protocol, $pattern, $url, $shouldBeFound=true)
testMakeLikeArrayWithValidPatterns()
Definition: LinkFilterTest.php:178
LinkFilterTest\createRegexFromLIKE
createRegexFromLIKE( $like)
createRegexFromLike($like)
Definition: LinkFilterTest.php:40
as
This document is intended to provide useful advice for parties seeking to redistribute MediaWiki to end users It s targeted particularly at maintainers for Linux since it s been observed that distribution packages of MediaWiki often break We ve consistently had to recommend that users seeking support use official tarballs instead of their distribution s and this often solves whatever problem the user is having It would be nice if this could such as
Definition: distributors.txt:9
LinkFilterTest\provideInvalidPatterns
static provideInvalidPatterns()
provideInvalidPatterns()
Definition: LinkFilterTest.php:215
LinkFilterTest
Database.
Definition: LinkFilterTest.php:8