| Index: trunk/extensions/WikiArticleFeeds/WikiArticleFeeds.php |
| — | — | @@ -2,7 +2,7 @@ |
| 3 | 3 | /* |
| 4 | 4 | * WikiArticleFeeds.php - A MediaWiki extension for converting regular pages into feeds. |
| 5 | 5 | * @author Jim R. Wilson |
| 6 | | - * @version 0.6.3 |
| | 6 | + * @version 0.6.5 |
| 7 | 7 | * @copyright Copyright (C) 2007 Jim R. Wilson |
| 8 | 8 | * @license The MIT License - http://www.opensource.org/licenses/mit-license.php |
| 9 | 9 | * ----------------------------------------------------------------------- |
| — | — | @@ -35,6 +35,8 @@ |
| 36 | 36 | * {{#itemTags:dogs, cats}} |
| 37 | 37 | * {{#itemTags:dogs|cats}} |
| 38 | 38 | * Version Notes: |
| | 39 | + * version 0.6.5: |
| | 40 | + * Simplified many regular expression to get it working on MW 1.16 |
| 39 | 41 | * version 0.6.4: |
| 40 | 42 | * Small fix for MW 1.14 in which section header anchors changed format. |
| 41 | 43 | * First version to be checked into wikimedia SVN. |
| — | — | @@ -454,17 +456,19 @@ |
| 455 | 457 | if ( $match < $lvl ) $lvl = $match; |
| 456 | 458 | } |
| 457 | 459 | |
| | 460 | + $sectionRegExp = '#<h' . $lvl . '>\s*<span.+?id="(.*?)">\s*(.*?)\s*</span>\s*</h' . $lvl . '>#m'; |
| | 461 | + |
| 458 | 462 | # Determine the item titles and default item links |
| 459 | 463 | preg_match_all( |
| 460 | | - '/<a[^>]*\\s+name=([\'"])(.*?)\\1[^>]*><\\/a><h' . $lvl . '>\\s*(.*?)\\s*<\\/h' . $lvl . '>/m', |
| 461 | | - $feedContent, |
| | 464 | + $sectionRegExp, |
| | 465 | + $feedContent, |
| 462 | 466 | $matches |
| 463 | 467 | ); |
| 464 | | - $itemLinks = $matches[2]; |
| 465 | | - $itemTitles = $matches[3]; |
| | 468 | + $itemLinks = $matches[1]; |
| | 469 | + $itemTitles = $matches[2]; |
| 466 | 470 | |
| 467 | 471 | # Split content into segments |
| 468 | | - $segments = preg_split( '/<a name=([\'"]).*?\\1\\s*><\\/a><h' . $lvl . '>.*?<\\/h' . $lvl . '>/m', $feedContent ); |
| | 472 | + $segments = preg_split( $sectionRegExp, $feedContent ); |
| 469 | 473 | $segDesc = trim( strip_tags( array_shift( $segments ) ) ); |
| 470 | 474 | if ( $segDesc ) { |
| 471 | 475 | if ( !$feedDescription ) { |
| — | — | @@ -505,26 +509,13 @@ |
| 506 | 510 | # Determine the item author and date |
| 507 | 511 | $author = null; |
| 508 | 512 | $date = null; |
| 509 | | - |
| | 513 | + $signatureRegExp = '#<a href=".+?User:.+?" title="User:.+?">(.*?)</a> (\d\d):(\d\d), (\d+) ([a-z]+) (\d{4}) \([A-Z]+\)#im'; |
| 510 | 514 | # Look for a regular ~~~~ sig |
| 511 | | - $isAttributable = preg_match( |
| 512 | | - '%<a [^>]*href=([\'"])' . preg_quote( $wgScript ) . '(/|\\?title=)User:.*?\\1[^>]*>(.*?)</a> (\\d\\d):(\\d\\d), (\\d+) ([A-z][a-z]+) (\\d{4}) \\([A-Z]+\\)%m', |
| 513 | | - $seg, |
| 514 | | - $matches |
| 515 | | - ); |
| | 515 | + $isAttributable = preg_match($signatureRegExp, $seg, $matches ); |
| 516 | 516 | |
| 517 | | - # As a fallback - look for a --~~~~ like sig with a user page outside the User NS |
| 518 | | - if ( !$isAttributable ) { |
| 519 | | - $isAttributable = preg_match( |
| 520 | | - '%--<a [^>]*href=([\'"])' . preg_quote( $wgScript ) . '(/|\\?title=).*?\\1[^>]*>(.*?)</a> (\\d\\d):(\\d\\d), (\\d+) ([A-z][a-z]+) (\\d{4}) \\([A-Z]+\\)%m', |
| 521 | | - $seg, |
| 522 | | - $matches |
| 523 | | - ); |
| 524 | | - } |
| 525 | | - |
| 526 | 517 | # Parse it out - if we can |
| 527 | 518 | if ( $isAttributable ) { |
| 528 | | - list( $author, $hour, $min, $day, $monthName, $year ) = array_slice( $matches, 3 ); |
| | 519 | + list( $author, $hour, $min, $day, $monthName, $year ) = array_slice( $matches, 1 ); |
| 529 | 520 | $months = array( |
| 530 | 521 | 'January' => '01', 'February' => '02', 'March' => '03', 'April' => '04', |
| 531 | 522 | 'May' => '05', 'June' => '06', 'July' => '07', 'August' => '08', |
| — | — | @@ -541,22 +532,15 @@ |
| 542 | 533 | # Look for an alternative to the default link (unless default 'section linking' has been forced) |
| 543 | 534 | global $wgForceArticleFeedSectionLinks; |
| 544 | 535 | if ( !$wgForceArticleFeedSectionLinks ) { |
| 545 | | - $strippedSeg = preg_replace( |
| 546 | | - array( |
| 547 | | - '%<a [^>]*href=([\'"])' . preg_quote( $wgScript ) . '(/|\\?title=)User:.*?\\1[^>]*>(.*?)</a> (\\d\\d:\\d\\d, \\d+ [A-z][a-z]+ \\d{4} \\([A-Z]+\\))%m', |
| 548 | | - '%--<a [^>]*href=([\'"])' . preg_quote( $wgScript ) . '(/|\\?title=).*?\\1[^>]*>(.*?)</a> (\\d\\d:\\d\\d, \\d+ [A-z][a-z]+ \\d{4} \\([A-Z]+\\))%m' |
| 549 | | - ), |
| 550 | | - '', |
| 551 | | - $seg |
| 552 | | - ); |
| | 536 | + $strippedSeg = preg_replace($signatureRegExp, '', $seg ); |
| 553 | 537 | preg_match( |
| 554 | | - '%<a [^>]*href=([\'"])(.*?)\\1[^>]*>(.*?)</a>%m', |
| 555 | | - $strippedSeg, |
| 556 | | - $matches |
| 557 | | - ); |
| | 538 | + '#<a [^>]*href=([\'"])(.*?)\\1[^>]*>(.*?)</a>#m', |
| | 539 | + $strippedSeg, |
| | 540 | + $matches |
| | 541 | + ); |
| 558 | 542 | if ( $matches[2] ) { |
| 559 | 543 | $url = $matches[2]; |
| 560 | | - if ( preg_match( '%^/%', $url ) ) { |
| | 544 | + if ( preg_match( '#^/#', $url ) ) { |
| 561 | 545 | $url = $wgServer . $url; |
| 562 | 546 | } |
| 563 | 547 | } |