Code Coverage |
||||||||||
Lines |
Functions and Methods |
Classes and Traits |
||||||||
| Total | |
84.16% |
186 / 221 |
|
25.00% |
3 / 12 |
CRAP | |
0.00% |
0 / 1 |
| LinkFilter | |
84.55% |
186 / 220 |
|
25.00% |
3 / 12 |
113.30 | |
0.00% |
0 / 1 |
| matchEntry | |
0.00% |
0 / 5 |
|
0.00% |
0 / 1 |
6 | |||
| makeRegex | |
0.00% |
0 / 6 |
|
0.00% |
0 / 1 |
6 | |||
| indexifyHost | |
95.65% |
44 / 46 |
|
0.00% |
0 / 1 |
17 | |||
| makeIndexes | |
100.00% |
27 / 27 |
|
100.00% |
1 / 1 |
11 | |||
| getIndexedUrlsNonReversed | |
87.50% |
7 / 8 |
|
0.00% |
0 / 1 |
4.03 | |||
| reverseIndexes | |
85.71% |
12 / 14 |
|
0.00% |
0 / 1 |
6.10 | |||
| reverseDomain | |
100.00% |
9 / 9 |
|
100.00% |
1 / 1 |
5 | |||
| getQueryConditions | |
97.67% |
42 / 43 |
|
0.00% |
0 / 1 |
12 | |||
| getProtocolPrefix | |
0.00% |
0 / 9 |
|
0.00% |
0 / 1 |
30 | |||
| prepareProtocols | |
0.00% |
0 / 7 |
|
0.00% |
0 / 1 |
12 | |||
| makeLikeArray | |
100.00% |
40 / 40 |
|
100.00% |
1 / 1 |
15 | |||
| keepOneWildcard | |
83.33% |
5 / 6 |
|
0.00% |
0 / 1 |
4.07 | |||
| 1 | <?php |
| 2 | /** |
| 3 | * This program is free software; you can redistribute it and/or modify |
| 4 | * it under the terms of the GNU General Public License as published by |
| 5 | * the Free Software Foundation; either version 2 of the License, or |
| 6 | * (at your option) any later version. |
| 7 | * |
| 8 | * This program is distributed in the hope that it will be useful, |
| 9 | * but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 10 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
| 11 | * GNU General Public License for more details. |
| 12 | * |
| 13 | * You should have received a copy of the GNU General Public License along |
| 14 | * with this program; if not, write to the Free Software Foundation, Inc., |
| 15 | * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. |
| 16 | * http://www.gnu.org/copyleft/gpl.html |
| 17 | * |
| 18 | * @file |
| 19 | */ |
| 20 | |
| 21 | namespace MediaWiki\ExternalLinks; |
| 22 | |
| 23 | use Content; |
| 24 | use MediaWiki\MainConfigNames; |
| 25 | use MediaWiki\MediaWikiServices; |
| 26 | use StringUtils; |
| 27 | use TextContent; |
| 28 | use Wikimedia\IPUtils; |
| 29 | use Wikimedia\Rdbms\IExpression; |
| 30 | use Wikimedia\Rdbms\LikeMatch; |
| 31 | use Wikimedia\Rdbms\LikeValue; |
| 32 | use Wikimedia\Rdbms\OrExpressionGroup; |
| 33 | |
| 34 | /** |
| 35 | * Utilities for formatting and querying the externallinks table. |
| 36 | * |
| 37 | * This is primarily used by \MediaWiki\Deferred\LinksUpdate\ExternalLinksTable |
| 38 | * for managing the storage layer, and by SpecialLinkSearch and ApiQueryExtLinksUsage |
| 39 | * as query interface. |
| 40 | * |
| 41 | * For spam removal and anti-spam meausures based on this, see also: |
| 42 | * - maintenance/cleanupSpam.php |
| 43 | * - SpamBlacklist extension |
| 44 | * - AbuseFilter extension (Special:BlockedExternalDomains, T337431) |
| 45 | */ |
| 46 | class LinkFilter { |
| 47 | /** |
| 48 | * Check whether $content contains a link to $filterEntry |
| 49 | * |
| 50 | * @param Content $content Content to check |
| 51 | * @param string $filterEntry Domainparts, see makeRegex() for more details |
| 52 | * @param string $protocol 'http://' or 'https://' |
| 53 | * @return int 0 if no match or 1 if there's at least one match |
| 54 | */ |
| 55 | public static function matchEntry( Content $content, $filterEntry, $protocol = 'http://' ) { |
| 56 | if ( !( $content instanceof TextContent ) ) { |
| 57 | // TODO: handle other types of content too. |
| 58 | // Maybe create ContentHandler::matchFilter( LinkFilter ). |
| 59 | // Think about a common base class for LinkFilter and MagicWord. |
| 60 | return 0; |
| 61 | } |
| 62 | |
| 63 | $text = $content->getText(); |
| 64 | $regex = self::makeRegex( $filterEntry, $protocol ); |
| 65 | return preg_match( $regex, $text ); |
| 66 | } |
| 67 | |
| 68 | /** |
| 69 | * Build a regex pattern for $filterEntry. |
| 70 | * |
| 71 | * @todo This doesn't match the rest of the functionality here. |
| 72 | * @param string $filterEntry URL, if it begins with "*.", it'll be |
| 73 | * replaced to match any subdomain |
| 74 | * @param string $protocol 'http://' or 'https://' |
| 75 | * @return string Regex pattern, for preg_match() |
| 76 | */ |
| 77 | private static function makeRegex( $filterEntry, $protocol ) { |
| 78 | $regex = '!' . preg_quote( $protocol, '!' ); |
| 79 | if ( substr( $filterEntry, 0, 2 ) == '*.' ) { |
| 80 | $regex .= '(?:[A-Za-z0-9.-]+\.|)'; |
| 81 | $filterEntry = substr( $filterEntry, 2 ); |
| 82 | } |
| 83 | $regex .= preg_quote( $filterEntry, '!' ) . '!Si'; |
| 84 | return $regex; |
| 85 | } |
| 86 | |
| 87 | /** |
| 88 | * Canonicalize a hostname for the externallinks table |
| 89 | * |
| 90 | * @param string $host |
| 91 | * @param bool $reverse whether to reverse the domain name or not |
| 92 | * @return string |
| 93 | */ |
| 94 | private static function indexifyHost( $host, $reverse = true ) { |
| 95 | // Canonicalize. |
| 96 | $host = rawurldecode( $host ); |
| 97 | if ( $host !== '' ) { |
| 98 | $tmp = idn_to_utf8( $host ); |
| 99 | if ( $tmp !== false ) { |
| 100 | $host = $tmp; |
| 101 | } |
| 102 | } |
| 103 | $okChars = 'a-zA-Z0-9\\-._~!$&\'()*+,;='; |
| 104 | if ( StringUtils::isUtf8( $host ) ) { |
| 105 | // Save a little space by not percent-encoding valid UTF-8 bytes |
| 106 | $okChars .= '\x80-\xf4'; |
| 107 | } |
| 108 | $host = preg_replace_callback( |
| 109 | '<[^' . $okChars . ']+>', |
| 110 | static fn ( $m ) => rawurlencode( $m[0] ), |
| 111 | strtolower( $host ) |
| 112 | ); |
| 113 | |
| 114 | // IPv6? RFC 3986 syntax. |
| 115 | if ( preg_match( '/^\[([0-9a-f:*]+)\]$/', rawurldecode( $host ), $m ) ) { |
| 116 | $ip = $m[1]; |
| 117 | if ( IPUtils::isValid( $ip ) ) { |
| 118 | if ( !$reverse ) { |
| 119 | return '[' . IPUtils::sanitizeIP( $ip ) . ']'; |
| 120 | } |
| 121 | return 'V6.' . implode( '.', explode( ':', IPUtils::sanitizeIP( $ip ) ) ) . '.'; |
| 122 | } |
| 123 | if ( substr( $ip, -2 ) === ':*' ) { |
| 124 | $cutIp = substr( $ip, 0, -2 ); |
| 125 | if ( IPUtils::isValid( "{$cutIp}::" ) ) { |
| 126 | // Wildcard IP doesn't contain "::", so multiple parts can be wild |
| 127 | $ct = count( explode( ':', $ip ) ) - 1; |
| 128 | if ( !$reverse ) { |
| 129 | return '[' . IPUtils::sanitizeIP( "{$cutIp}::" ) . ']'; |
| 130 | } |
| 131 | return 'V6.' . |
| 132 | implode( '.', array_slice( explode( ':', IPUtils::sanitizeIP( "{$cutIp}::" ) ), 0, $ct ) ) . |
| 133 | '.*.'; |
| 134 | } |
| 135 | if ( IPUtils::isValid( "{$cutIp}:1" ) ) { |
| 136 | // Wildcard IP does contain "::", so only the last part is wild |
| 137 | if ( !$reverse ) { |
| 138 | return '[' . IPUtils::sanitizeIP( "{$cutIp}:1" ) . ']'; |
| 139 | } |
| 140 | return 'V6.' . |
| 141 | substr( implode( '.', explode( ':', IPUtils::sanitizeIP( "{$cutIp}:1" ) ) ), 0, -1 ) . |
| 142 | '*.'; |
| 143 | } |
| 144 | } |
| 145 | } |
| 146 | |
| 147 | // Regularize explicit specification of the DNS root. |
| 148 | // Browsers seem to do this for IPv4 literals too. |
| 149 | if ( substr( $host, -1 ) === '.' ) { |
| 150 | $host = substr( $host, 0, -1 ); |
| 151 | } |
| 152 | |
| 153 | // IPv4? |
| 154 | $b = '(?:0*25[0-5]|0*2[0-4][0-9]|0*1[0-9][0-9]|0*[0-9]?[0-9])'; |
| 155 | if ( preg_match( "/^(?:{$b}\.){3}{$b}$|^(?:{$b}\.){1,3}\*$/", $host ) ) { |
| 156 | if ( !$reverse ) { |
| 157 | return $host; |
| 158 | } |
| 159 | return 'V4.' . implode( '.', array_map( static function ( $v ) { |
| 160 | return $v === '*' ? $v : (int)$v; |
| 161 | }, explode( '.', $host ) ) ) . '.'; |
| 162 | } |
| 163 | |
| 164 | // Must be a host name. |
| 165 | if ( $reverse ) { |
| 166 | return implode( '.', array_reverse( explode( '.', $host ) ) ) . '.'; |
| 167 | } else { |
| 168 | return $host; |
| 169 | } |
| 170 | } |
| 171 | |
| 172 | /** |
| 173 | * Convert given URL to format for the externallinks table |
| 174 | * |
| 175 | * @since 1.33 |
| 176 | * @param string $url |
| 177 | * @param bool $reverseDomain |
| 178 | * @return string[][] One entry. Empty array on error. |
| 179 | * Each entry is an array in form of <host,path> |
| 180 | */ |
| 181 | public static function makeIndexes( $url, $reverseDomain = true ) { |
| 182 | // NOTE: refreshExternallinksIndex.php assumes that only protocol-relative URLs return more |
| 183 | // than one index, and that the indexes for protocol-relative URLs only vary in the "http://" |
| 184 | // versus "https://" prefix. If you change that, you'll likely need to update |
| 185 | // refreshExternallinksIndex.php accordingly. |
| 186 | |
| 187 | $bits = MediaWikiServices::getInstance()->getUrlUtils()->parse( $url ); |
| 188 | if ( !$bits ) { |
| 189 | return []; |
| 190 | } |
| 191 | |
| 192 | // URI RFC identifies the email/server part of mailto or news protocol as 'path', |
| 193 | // while we want to match the email's domain or news server the same way we are |
| 194 | // matching hosts for other URLs. |
| 195 | if ( in_array( $bits['scheme'], [ 'mailto', 'news' ] ) ) { |
| 196 | // (T347574) Only set host if it's not already set (if // is used) |
| 197 | if ( array_key_exists( 'path', $bits ) ) { |
| 198 | $bits['host'] = $bits['path']; |
| 199 | } |
| 200 | $bits['path'] = ''; |
| 201 | } |
| 202 | |
| 203 | // Reverse the labels in the hostname, convert to lower case, unless it's an IP. |
| 204 | // For emails turn it into "domain.reversed@localpart" |
| 205 | if ( $bits['scheme'] == 'mailto' ) { |
| 206 | $mailparts = explode( '@', $bits['host'], 2 ); |
| 207 | if ( count( $mailparts ) === 2 ) { |
| 208 | $domainpart = self::indexifyHost( $mailparts[1], $reverseDomain ); |
| 209 | } else { |
| 210 | // No @, assume it's a local part with no domain |
| 211 | $domainpart = ''; |
| 212 | } |
| 213 | if ( $reverseDomain ) { |
| 214 | $bits['host'] = $domainpart . '@' . $mailparts[0]; |
| 215 | } else { |
| 216 | $bits['host'] = $mailparts[0] . '@' . $domainpart; |
| 217 | } |
| 218 | } else { |
| 219 | $bits['host'] = self::indexifyHost( $bits['host'], $reverseDomain ); |
| 220 | } |
| 221 | |
| 222 | // Reconstruct the pseudo-URL |
| 223 | $index = $bits['scheme'] . $bits['delimiter'] . $bits['host']; |
| 224 | // Leave out user and password. Add the port, path, query and fragment |
| 225 | if ( isset( $bits['port'] ) ) { |
| 226 | $index .= ':' . $bits['port']; |
| 227 | } |
| 228 | $index2 = $bits['path'] ?? '/'; |
| 229 | if ( isset( $bits['query'] ) ) { |
| 230 | $index2 .= '?' . $bits['query']; |
| 231 | } |
| 232 | if ( isset( $bits['fragment'] ) ) { |
| 233 | $index2 .= '#' . $bits['fragment']; |
| 234 | } |
| 235 | |
| 236 | if ( $bits['scheme'] == '' ) { |
| 237 | return [ [ "https:$index", $index2 ] ]; |
| 238 | } else { |
| 239 | return [ [ $index, $index2 ] ]; |
| 240 | } |
| 241 | } |
| 242 | |
| 243 | /** |
| 244 | * Converts a set of URLs to be able to compare them with existing indexes |
| 245 | * @since 1.41 |
| 246 | * @param string[] $urls List of URLs to be indexed |
| 247 | * @return string[] |
| 248 | */ |
| 249 | public static function getIndexedUrlsNonReversed( $urls ) { |
| 250 | $newLinks = []; |
| 251 | foreach ( $urls as $url ) { |
| 252 | $indexes = self::makeIndexes( $url, false ); |
| 253 | if ( !$indexes ) { |
| 254 | continue; |
| 255 | } |
| 256 | foreach ( $indexes as $index ) { |
| 257 | $newLinks[] = $index[0] . $index[1]; |
| 258 | } |
| 259 | } |
| 260 | return $newLinks; |
| 261 | } |
| 262 | |
| 263 | public static function reverseIndexes( $domainIndex ) { |
| 264 | $bits = MediaWikiServices::getInstance()->getUrlUtils()->parse( $domainIndex ); |
| 265 | if ( !$bits ) { |
| 266 | return ''; |
| 267 | } |
| 268 | |
| 269 | // Reverse the labels in the hostname, convert to lower case, unless it's an IP. |
| 270 | // For emails turn it into "domain.reversed@localpart" |
| 271 | if ( $bits['scheme'] == 'mailto' ) { |
| 272 | $mailparts = explode( '@', $bits['path'], 2 ); |
| 273 | if ( count( $mailparts ) === 2 ) { |
| 274 | $domainpart = rtrim( self::reverseDomain( $mailparts[0] ), '.' ); |
| 275 | } else { |
| 276 | // No @, assume it's a local part with no domain |
| 277 | $domainpart = ''; |
| 278 | } |
| 279 | $bits['host'] = $mailparts[1] . '@' . $domainpart; |
| 280 | } else { |
| 281 | $bits['host'] = rtrim( self::reverseDomain( $bits['host'] ), '.' ); |
| 282 | } |
| 283 | |
| 284 | $index = $bits['scheme'] . $bits['delimiter'] . $bits['host']; |
| 285 | if ( isset( $bits['port'] ) && $bits['port'] ) { |
| 286 | $index .= ':' . $bits['port']; |
| 287 | } |
| 288 | return $index; |
| 289 | } |
| 290 | |
| 291 | private static function reverseDomain( $domain ) { |
| 292 | if ( substr( $domain, 0, 3 ) === 'V6.' ) { |
| 293 | $ipv6 = str_replace( '.', ':', trim( substr( $domain, 3 ), '.' ) ); |
| 294 | if ( IPUtils::isValid( $ipv6 ) ) { |
| 295 | return '[' . $ipv6 . ']'; |
| 296 | } |
| 297 | } elseif ( substr( $domain, 0, 3 ) === 'V4.' ) { |
| 298 | $ipv4 = trim( substr( $domain, 3 ), '.' ); |
| 299 | if ( IPUtils::isValid( $ipv4 ) ) { |
| 300 | return $ipv4; |
| 301 | } |
| 302 | } |
| 303 | return self::indexifyHost( $domain ); |
| 304 | } |
| 305 | |
| 306 | /** |
| 307 | * Return conditions for the externallinks table from a given filter entry. |
| 308 | * |
| 309 | * There are several ways you can query: |
| 310 | * |
| 311 | * *.domain.com - Matches domain.com and www.domain.com |
| 312 | * domain.com - Matches domain.com or domain.com/ but not www.domain.com |
| 313 | * *.domain.com/x - Matches domain.com/xy or www.domain.com/xy. Also probably matches |
| 314 | * domain.com/foobar/xy due to limitations of LIKE syntax. |
| 315 | * domain.com/x - Matches domain.com/xy but not www.domain.com/xy |
| 316 | * 192.0.2.* - Matches any IP in 192.0.2.0/24. Can also have a path appended. |
| 317 | * [2001:db8::*] - Matches any IP in 2001:db8::/112. Can also have a path appended. |
| 318 | * [2001:db8:*] - Matches any IP in 2001:db8::/32. Can also have a path appended. |
| 319 | * foo@domain.com - With protocol 'mailto:', matches the email address foo@domain.com. |
| 320 | * *@domain.com - With protocol 'mailto:', matches any email address at domain.com, but |
| 321 | * not subdomains like foo@mail.domain.com |
| 322 | * |
| 323 | * Asterisks in any other location are considered invalid. |
| 324 | * |
| 325 | * @since 1.33 |
| 326 | * @param string $filterEntry Filter entry, as described above |
| 327 | * @param array $options Options are: |
| 328 | * - protocol: (null, string, array) Protocol to query (default: `http://` and `https://`) |
| 329 | * - oneWildcard: (bool) Stop at the first wildcard (default: false) |
| 330 | * - db: (IReadableDatabase|null) Database for building SQL text. |
| 331 | * @return array|false Query conditions (to be ANDed) or false on error. |
| 332 | */ |
| 333 | public static function getQueryConditions( $filterEntry, array $options = [] ) { |
| 334 | $options += [ |
| 335 | 'protocol' => [ 'http://', 'https://' ], |
| 336 | 'oneWildcard' => false, |
| 337 | 'db' => null, |
| 338 | ]; |
| 339 | $domainGaps = MediaWikiServices::getInstance()->getMainConfig()->get( |
| 340 | MainConfigNames::ExternalLinksDomainGaps |
| 341 | ); |
| 342 | |
| 343 | if ( is_string( $options['protocol'] ) ) { |
| 344 | $options['protocol'] = [ $options['protocol'] ]; |
| 345 | } elseif ( $options['protocol'] === null ) { |
| 346 | $options['protocol'] = [ 'http://', 'https://' ]; |
| 347 | } |
| 348 | |
| 349 | $domainConditions = []; |
| 350 | $db = $options['db'] ?: MediaWikiServices::getInstance()->getConnectionProvider()->getReplicaDatabase(); |
| 351 | foreach ( $options['protocol'] as $protocol ) { |
| 352 | $like = self::makeLikeArray( $filterEntry, $protocol ); |
| 353 | if ( $like === false ) { |
| 354 | continue; |
| 355 | } |
| 356 | [ $likeDomain, $likePath ] = $like; |
| 357 | $trimmedlikeDomain = self::keepOneWildcard( $likeDomain ); |
| 358 | if ( $trimmedlikeDomain[count( $trimmedlikeDomain ) - 1] instanceof LikeMatch ) { |
| 359 | array_pop( $trimmedlikeDomain ); |
| 360 | } |
| 361 | $index1 = implode( '', $trimmedlikeDomain ); |
| 362 | if ( $options['oneWildcard'] && $likePath[0] != '/' ) { |
| 363 | $thisDomainExpr = $db->expr( 'el_to_domain_index', '=', $index1 ); |
| 364 | } else { |
| 365 | $thisDomainExpr = $db->expr( |
| 366 | 'el_to_domain_index', |
| 367 | IExpression::LIKE, |
| 368 | new LikeValue( $index1, $db->anyString() ) |
| 369 | ); |
| 370 | } |
| 371 | foreach ( $domainGaps[$index1] ?? [] as $from => $to ) { |
| 372 | $thisDomainExpr = $thisDomainExpr->andExpr( $db->expr( 'el_id', '<', $from )->or( 'el_id', '>', $to ) ); |
| 373 | } |
| 374 | $domainConditions[] = $thisDomainExpr; |
| 375 | } |
| 376 | if ( !$domainConditions ) { |
| 377 | return false; |
| 378 | } |
| 379 | // @phan-suppress-next-line PhanPossiblyUndeclaredVariable |
| 380 | $trimmedlikePath = self::keepOneWildcard( $likePath ); |
| 381 | if ( $trimmedlikePath[count( $trimmedlikePath ) - 1] instanceof LikeMatch ) { |
| 382 | array_pop( $trimmedlikePath ); |
| 383 | } |
| 384 | $index2 = implode( '', $trimmedlikePath ); |
| 385 | |
| 386 | return [ |
| 387 | new OrExpressionGroup( ...$domainConditions ), |
| 388 | $db->expr( 'el_to_path', IExpression::LIKE, new LikeValue( $index2, $db->anyString() ) ), |
| 389 | ]; |
| 390 | } |
| 391 | |
| 392 | public static function getProtocolPrefix( $protocol ) { |
| 393 | // Find the right prefix |
| 394 | $urlProtocols = MediaWikiServices::getInstance()->getMainConfig() |
| 395 | ->get( MainConfigNames::UrlProtocols ); |
| 396 | if ( $protocol && !in_array( $protocol, $urlProtocols ) ) { |
| 397 | foreach ( $urlProtocols as $p ) { |
| 398 | if ( str_starts_with( $p, $protocol ) ) { |
| 399 | $protocol = $p; |
| 400 | break; |
| 401 | } |
| 402 | } |
| 403 | |
| 404 | return $protocol; |
| 405 | } else { |
| 406 | return null; |
| 407 | } |
| 408 | } |
| 409 | |
| 410 | public static function prepareProtocols() { |
| 411 | $urlProtocols = MediaWikiServices::getInstance()->getMainConfig() |
| 412 | ->get( MainConfigNames::UrlProtocols ); |
| 413 | $protocols = [ '' ]; |
| 414 | foreach ( $urlProtocols as $p ) { |
| 415 | if ( $p !== '//' ) { |
| 416 | $protocols[] = substr( $p, 0, strpos( $p, ':' ) ); |
| 417 | } |
| 418 | } |
| 419 | |
| 420 | return $protocols; |
| 421 | } |
| 422 | |
| 423 | /** |
| 424 | * Make an array to be used for calls to Database::buildLike(), which |
| 425 | * will match the specified string. |
| 426 | * |
| 427 | * This function does the same as LinkFilter::makeIndexes(), except it also takes care |
| 428 | * of adding wildcards |
| 429 | * |
| 430 | * @note You probably want self::getQueryConditions() instead |
| 431 | * @param string $filterEntry Filter entry, @see self::getQueryConditions() |
| 432 | * @param string $protocol Protocol (default http://) |
| 433 | * @return array|false Array to be passed to Database::buildLike() or false on error |
| 434 | */ |
| 435 | public static function makeLikeArray( $filterEntry, $protocol = 'http://' ) { |
| 436 | $services = MediaWikiServices::getInstance(); |
| 437 | $db = $services->getConnectionProvider()->getReplicaDatabase(); |
| 438 | $likeDomain = []; |
| 439 | $likePath = []; |
| 440 | |
| 441 | $target = $protocol . $filterEntry; |
| 442 | $bits = $services->getUrlUtils()->parse( $target ); |
| 443 | if ( !$bits ) { |
| 444 | return false; |
| 445 | } |
| 446 | |
| 447 | // URI RFC identifies the email/server part of mailto or news protocol as 'path', |
| 448 | // while we want to match the email's domain or news server the same way we are |
| 449 | // matching hosts for other URLs. |
| 450 | if ( in_array( $bits['scheme'], [ 'mailto', 'news' ] ) ) { |
| 451 | $bits['host'] = $bits['path']; |
| 452 | $bits['path'] = ''; |
| 453 | } |
| 454 | |
| 455 | $subdomains = false; |
| 456 | if ( $bits['scheme'] === 'mailto' && strpos( $bits['host'], '@' ) ) { |
| 457 | // Email address with domain and non-empty local part |
| 458 | $mailparts = explode( '@', $bits['host'], 2 ); |
| 459 | $domainpart = self::indexifyHost( $mailparts[1] ); |
| 460 | if ( $mailparts[0] === '*' ) { |
| 461 | $subdomains = true; |
| 462 | $bits['host'] = $domainpart . '@'; |
| 463 | } else { |
| 464 | $bits['host'] = $domainpart . '@' . $mailparts[0]; |
| 465 | } |
| 466 | } else { |
| 467 | // Non-email, or email with only a domain part. |
| 468 | $bits['host'] = self::indexifyHost( $bits['host'] ); |
| 469 | if ( substr( $bits['host'], -3 ) === '.*.' ) { |
| 470 | $subdomains = true; |
| 471 | $bits['host'] = substr( $bits['host'], 0, -2 ); |
| 472 | } |
| 473 | } |
| 474 | |
| 475 | $likeDomain[] = $bits['scheme'] . $bits['delimiter'] . $bits['host']; |
| 476 | |
| 477 | if ( $subdomains ) { |
| 478 | $likeDomain[] = $db->anyString(); |
| 479 | } |
| 480 | |
| 481 | if ( isset( $bits['port'] ) ) { |
| 482 | $likeDomain[] = ':' . $bits['port']; |
| 483 | } |
| 484 | if ( isset( $bits['path'] ) ) { |
| 485 | $likePath[] = $bits['path']; |
| 486 | } else { |
| 487 | $likePath[] = '/'; |
| 488 | } |
| 489 | if ( isset( $bits['query'] ) ) { |
| 490 | $likePath[] = '?' . $bits['query']; |
| 491 | } |
| 492 | if ( isset( $bits['fragment'] ) ) { |
| 493 | $likePath[] = '#' . $bits['fragment']; |
| 494 | } |
| 495 | $likePath[] = $db->anyString(); |
| 496 | |
| 497 | // Check for stray asterisks: asterisk only allowed at the start of the domain |
| 498 | foreach ( array_merge( $likeDomain, $likePath ) as $likepart ) { |
| 499 | if ( !( $likepart instanceof LikeMatch ) && strpos( $likepart, '*' ) !== false ) { |
| 500 | return false; |
| 501 | } |
| 502 | } |
| 503 | |
| 504 | return [ $likeDomain, $likePath ]; |
| 505 | } |
| 506 | |
| 507 | /** |
| 508 | * Filters an array returned by makeLikeArray(), removing everything past first |
| 509 | * pattern placeholder. |
| 510 | * |
| 511 | * @note You probably want self::getQueryConditions() instead |
| 512 | * @param array $arr Array to filter |
| 513 | * @return array Filtered array |
| 514 | */ |
| 515 | public static function keepOneWildcard( $arr ) { |
| 516 | if ( !is_array( $arr ) ) { |
| 517 | return $arr; |
| 518 | } |
| 519 | |
| 520 | foreach ( $arr as $key => $value ) { |
| 521 | if ( $value instanceof LikeMatch ) { |
| 522 | return array_slice( $arr, 0, $key + 1 ); |
| 523 | } |
| 524 | } |
| 525 | |
| 526 | return $arr; |
| 527 | } |
| 528 | } |
| 529 | |
| 530 | /** @deprecated class alias since 1.40 */ |
| 531 | class_alias( LinkFilter::class, 'LinkFilter' ); |