| Index: trunk/phase3/maintenance/parserTests.txt |
| — | — | @@ -2575,6 +2575,64 @@ |
| 2576 | 2576 | </p> |
| 2577 | 2577 | !! end |
| 2578 | 2578 | |
| | 2579 | +!! article |
| | 2580 | +Template:Includeonly section |
| | 2581 | +!! text |
| | 2582 | +<includeonly> |
| | 2583 | +==Includeonly section== |
| | 2584 | +</includeonly> |
| | 2585 | +==Section T-1== |
| | 2586 | +!!endarticle |
| | 2587 | + |
| | 2588 | +!! test |
| | 2589 | +Bug 6563: Edit link generation for section shown by <includeonly> |
| | 2590 | +!! input |
| | 2591 | +{{includeonly section}} |
| | 2592 | +!! result |
| | 2593 | +<a name="Includeonly_section"></a><h2><span class="editsection">[<a href="https://www.mediawiki.org/index.php?title=Template:Includeonly_section&action=edit&section=T-1" title="Template:Includeonly section">edit</a>]</span> <span class="mw-headline">Includeonly section</span></h2> |
| | 2594 | +<a name="Section_T-1"></a><h2><span class="editsection">[<a href="https://www.mediawiki.org/index.php?title=Template:Includeonly_section&action=edit&section=T-2" title="Template:Includeonly section">edit</a>]</span> <span class="mw-headline">Section T-1</span></h2> |
| | 2595 | + |
| | 2596 | +!! end |
| | 2597 | + |
| | 2598 | +# Uses same input as the contents of [[Template:Includeonly section]] |
| | 2599 | +!! test |
| | 2600 | +Bug 6563: Section extraction for section shown by <includeonly> |
| | 2601 | +!! options |
| | 2602 | +section=T-2 |
| | 2603 | +!! input |
| | 2604 | +<includeonly> |
| | 2605 | +==Includeonly section== |
| | 2606 | +</includeonly> |
| | 2607 | +==Section T-2== |
| | 2608 | +!! result |
| | 2609 | +==Section T-2== |
| | 2610 | +!! end |
| | 2611 | + |
| | 2612 | +!! test |
| | 2613 | +Bug 6563: Edit link generation for section suppressed by <includeonly> |
| | 2614 | +!! input |
| | 2615 | +<includeonly> |
| | 2616 | +==Includeonly section== |
| | 2617 | +</includeonly> |
| | 2618 | +==Section 1== |
| | 2619 | +!! result |
| | 2620 | +<a name="Section_1"></a><h2><span class="editsection">[<a href="https://www.mediawiki.org/index.php?title=Parser_test&action=edit&section=1" title="Edit section: Section 1">edit</a>]</span> <span class="mw-headline">Section 1</span></h2> |
| | 2621 | + |
| | 2622 | +!! end |
| | 2623 | + |
| | 2624 | +!! test |
| | 2625 | +Bug 6563: Section extraction for section suppressed by <includeonly> |
| | 2626 | +!! options |
| | 2627 | +section=1 |
| | 2628 | +!! input |
| | 2629 | +<includeonly> |
| | 2630 | +==Includeonly section== |
| | 2631 | +</includeonly> |
| | 2632 | +==Section 1== |
| | 2633 | +!! result |
| | 2634 | +==Section 1== |
| | 2635 | +!! end |
| | 2636 | + |
| 2579 | 2637 | ### |
| 2580 | 2638 | ### Pre-save transform tests |
| 2581 | 2639 | ### |
| — | — | @@ -3504,8 +3562,8 @@ |
| 3505 | 3563 | ==Section 4== |
| 3506 | 3564 | !! result |
| 3507 | 3565 | <a name="Section_0"></a><h2><span class="editsection">[<a href="https://www.mediawiki.org/index.php?title=Parser_test&action=edit&section=1" title="Edit section: Section 0">edit</a>]</span> <span class="mw-headline">Section 0</span></h2> |
| 3508 | | -<a name="Section_1"></a><h3><span class="editsection">[<a href="https://www.mediawiki.org/index.php?title=Template:Sections&action=edit&section=1" title="Template:Sections">edit</a>]</span> <span class="mw-headline">Section 1</span></h3> |
| 3509 | | -<a name="Section_2"></a><h2><span class="editsection">[<a href="https://www.mediawiki.org/index.php?title=Template:Sections&action=edit&section=2" title="Template:Sections">edit</a>]</span> <span class="mw-headline">Section 2</span></h2> |
| | 3566 | +<a name="Section_1"></a><h3><span class="editsection">[<a href="https://www.mediawiki.org/index.php?title=Template:Sections&action=edit&section=T-1" title="Template:Sections">edit</a>]</span> <span class="mw-headline">Section 1</span></h3> |
| | 3567 | +<a name="Section_2"></a><h2><span class="editsection">[<a href="https://www.mediawiki.org/index.php?title=Template:Sections&action=edit&section=T-2" title="Template:Sections">edit</a>]</span> <span class="mw-headline">Section 2</span></h2> |
| 3510 | 3568 | <a name="Section_4"></a><h2><span class="editsection">[<a href="https://www.mediawiki.org/index.php?title=Parser_test&action=edit&section=2" title="Edit section: Section 4">edit</a>]</span> <span class="mw-headline">Section 4</span></h2> |
| 3511 | 3569 | |
| 3512 | 3570 | !! end |
| — | — | @@ -6223,7 +6281,7 @@ |
| 6224 | 6282 | !! input |
| 6225 | 6283 | {{MediaWiki:Fake}} |
| 6226 | 6284 | !! result |
| 6227 | | -<a name="header"></a><h2><span class="editsection">[<a href="https://www.mediawiki.org/index.php?title=MediaWiki:Fake&action=edit&section=1" title="MediaWiki:Fake">edit</a>]</span> <span class="mw-headline">header</span></h2> |
| | 6285 | +<a name="header"></a><h2><span class="editsection">[<a href="https://www.mediawiki.org/index.php?title=MediaWiki:Fake&action=edit&section=T-1" title="MediaWiki:Fake">edit</a>]</span> <span class="mw-headline">header</span></h2> |
| 6228 | 6286 | |
| 6229 | 6287 | !! end |
| 6230 | 6288 | |
| Index: trunk/phase3/includes/Parser.php |
| — | — | @@ -71,6 +71,9 @@ |
| 72 | 72 | const COLON_STATE_COMMENTDASH = 6; |
| 73 | 73 | const COLON_STATE_COMMENTDASHDASH = 7; |
| 74 | 74 | |
| | 75 | + // Flags for preprocessToDom |
| | 76 | + const PTD_FOR_INCLUSION = 1; |
| | 77 | + |
| 75 | 78 | /**#@+ |
| 76 | 79 | * @private |
| 77 | 80 | */ |
| — | — | @@ -928,11 +931,6 @@ |
| 929 | 932 | return $text ; |
| 930 | 933 | } |
| 931 | 934 | |
| 932 | | - # Remove <noinclude> tags and <includeonly> sections |
| 933 | | - $text = strtr( $text, array( '<onlyinclude>' => '' , '</onlyinclude>' => '' ) ); |
| 934 | | - $text = strtr( $text, array( '<noinclude>' => '', '</noinclude>' => '') ); |
| 935 | | - $text = StringUtils::delimiterReplace( '<includeonly>', '</includeonly>', '', $text ); |
| 936 | | - |
| 937 | 935 | $text = $this->replaceVariables( $text ); |
| 938 | 936 | $text = Sanitizer::removeHTMLtags( $text, array( &$this, 'attributeStripCallback' ), false, array_keys( $this->mTransparentTagHooks ) ); |
| 939 | 937 | wfRunHooks( 'InternalParseBeforeLinks', array( &$this, &$text, &$this->mStripState ) ); |
| — | — | @@ -2541,17 +2539,32 @@ |
| 2542 | 2540 | } |
| 2543 | 2541 | |
| 2544 | 2542 | /** |
| 2545 | | - * Parse any parentheses in format ((title|part|part)} and return the document tree |
| | 2543 | + * Preprocess some wikitext and return the document tree. |
| 2546 | 2544 | * This is the ghost of replace_variables(). |
| 2547 | 2545 | * |
| 2548 | 2546 | * @param string $text The text to parse |
| | 2547 | + * @param integer flags Bitwise combination of: |
| | 2548 | + * self::PTD_FOR_INCLUSION Handle <noinclude>/<includeonly> as if the text is being |
| | 2549 | + * included. Default is to assume a direct page view. |
| | 2550 | + * |
| | 2551 | + * The generated DOM tree must depend only on the input text, the flags, and $this->ot['msg']. |
| | 2552 | + * The DOM tree must be the same in OT_HTML and OT_WIKI mode, to avoid a regression of bug 4899. |
| | 2553 | + * |
| | 2554 | + * Any flag added to the $flags parameter here, or any other parameter liable to cause a |
| | 2555 | + * change in the DOM tree for a given text, must be passed through the section identifier |
| | 2556 | + * in the section edit link and thus back to extractSections(). |
| | 2557 | + * |
| | 2558 | + * The output of this function is currently only cached in process memory, but a persistent |
| | 2559 | + * cache may be implemented at a later date which takes further advantage of these strict |
| | 2560 | + * dependency requirements. |
| | 2561 | + * |
| 2549 | 2562 | * @private |
| 2550 | 2563 | */ |
| 2551 | | - function preprocessToDom ( $text ) { |
| | 2564 | + function preprocessToDom ( $text, $flags = 0 ) { |
| 2552 | 2565 | wfProfileIn( __METHOD__ ); |
| 2553 | 2566 | wfProfileIn( __METHOD__.'-makexml' ); |
| 2554 | 2567 | |
| 2555 | | - static $msgRules, $normalRules; |
| | 2568 | + static $msgRules, $normalRules, $inclusionSupertags, $nonInclusionSupertags; |
| 2556 | 2569 | if ( !$msgRules ) { |
| 2557 | 2570 | $msgRules = array( |
| 2558 | 2571 | '{' => array( |
| — | — | @@ -2592,19 +2605,32 @@ |
| 2593 | 2606 | } else { |
| 2594 | 2607 | $rules = $normalRules; |
| 2595 | 2608 | } |
| | 2609 | + $forInclusion = $flags & self::PTD_FOR_INCLUSION; |
| 2596 | 2610 | |
| 2597 | | - if ( $this->ot['html'] || ( $this->ot['pre'] && $this->mOptions->getRemoveComments() ) ) { |
| 2598 | | - $text = Sanitizer::removeHTMLcomments( $text ); |
| | 2611 | + $xmlishElements = $this->getStripList(); |
| | 2612 | + $enableOnlyinclude = false; |
| | 2613 | + if ( $forInclusion ) { |
| | 2614 | + $ignoredTags = array( 'includeonly', '/includeonly' ); |
| | 2615 | + $ignoredElements = array( 'noinclude' ); |
| | 2616 | + $xmlishElements[] = 'noinclude'; |
| | 2617 | + if ( strpos( $text, '<onlyinclude>' ) !== false && strpos( $text, '</onlyinclude>' ) !== false ) { |
| | 2618 | + $enableOnlyinclude = true; |
| | 2619 | + } |
| | 2620 | + } else { |
| | 2621 | + $ignoredTags = array( 'noinclude', '/noinclude', 'onlyinclude', '/onlyinclude' ); |
| | 2622 | + $ignoredElements = array( 'includeonly' ); |
| | 2623 | + $xmlishElements[] = 'includeonly'; |
| 2599 | 2624 | } |
| | 2625 | + $xmlishRegex = implode( '|', array_merge( $xmlishElements, $ignoredTags ) ); |
| 2600 | 2626 | |
| 2601 | | - $extElements = implode( '|', $this->getStripList() ); |
| 2602 | 2627 | // Use "A" modifier (anchored) instead of "^", because ^ doesn't work with an offset |
| 2603 | | - $extElementsRegex = "/($extElements)(?:\s|\/>|>)|(!--)/iA"; |
| | 2628 | + $elementsRegex = "~($xmlishRegex)(?:\s|\/>|>)|(!--)~iA"; |
| 2604 | 2629 | |
| 2605 | 2630 | $stack = array(); # Stack of unclosed parentheses |
| 2606 | 2631 | $stackIndex = -1; # Stack read pointer |
| 2607 | 2632 | |
| 2608 | 2633 | $searchBase = implode( '', array_keys( $rules ) ) . '<'; |
| | 2634 | + $revText = strrev( $text ); // For fast reverse searches |
| 2609 | 2635 | |
| 2610 | 2636 | $i = -1; # Input pointer, starts out pointing to a pseudo-newline before the start |
| 2611 | 2637 | $topAccum = '<root>'; # Top level text accumulator |
| — | — | @@ -2614,8 +2640,27 @@ |
| 2615 | 2641 | $findPipe = false; # True to take notice of pipe characters |
| 2616 | 2642 | $headingIndex = 1; |
| 2617 | 2643 | $noMoreGT = false; # True if there are no more greater-than (>) signs right of $i |
| | 2644 | + $findOnlyinclude = $enableOnlyinclude; # True to ignore all input up to the next <onlyinclude> |
| 2618 | 2645 | |
| 2619 | | - while ( $i < strlen( $text ) ) { |
| | 2646 | + if ( $enableOnlyinclude ) { |
| | 2647 | + $i = 0; |
| | 2648 | + } |
| | 2649 | + |
| | 2650 | + while ( true ) { |
| | 2651 | + if ( $findOnlyinclude ) { |
| | 2652 | + // Ignore all input up to the next <onlyinclude> |
| | 2653 | + $startPos = strpos( $text, '<onlyinclude>', $i ); |
| | 2654 | + if ( $startPos === false ) { |
| | 2655 | + // Ignored section runs to the end |
| | 2656 | + $accum .= '<ignore>' . htmlspecialchars( substr( $text, $i ) ) . '</ignore>'; |
| | 2657 | + break; |
| | 2658 | + } |
| | 2659 | + $tagEndPos = $startPos + strlen( '<onlyinclude>' ); // past-the-end |
| | 2660 | + $accum .= '<ignore>' . htmlspecialchars( substr( $text, $i, $tagEndPos - $i ) ) . '</ignore>'; |
| | 2661 | + $i = $tagEndPos; |
| | 2662 | + $findOnlyinclude = false; |
| | 2663 | + } |
| | 2664 | + |
| 2620 | 2665 | if ( $i == -1 ) { |
| 2621 | 2666 | $found = 'line-start'; |
| 2622 | 2667 | $curChar = ''; |
| — | — | @@ -2684,8 +2729,14 @@ |
| 2685 | 2730 | |
| 2686 | 2731 | if ( $found == 'angle' ) { |
| 2687 | 2732 | $matches = false; |
| | 2733 | + // Handle </onlyinclude> |
| | 2734 | + if ( $enableOnlyinclude && substr( $text, $i, strlen( '</onlyinclude>' ) ) == '</onlyinclude>' ) { |
| | 2735 | + $findOnlyinclude = true; |
| | 2736 | + continue; |
| | 2737 | + } |
| | 2738 | + |
| 2688 | 2739 | // Determine element name |
| 2689 | | - if ( !preg_match( $extElementsRegex, $text, $matches, 0, $i + 1 ) ) { |
| | 2740 | + if ( !preg_match( $elementsRegex, $text, $matches, 0, $i + 1 ) ) { |
| 2690 | 2741 | // Element name missing or not listed |
| 2691 | 2742 | $accum .= '<'; |
| 2692 | 2743 | ++$i; |
| — | — | @@ -2693,21 +2744,37 @@ |
| 2694 | 2745 | } |
| 2695 | 2746 | // Handle comments |
| 2696 | 2747 | if ( isset( $matches[2] ) && $matches[2] == '!--' ) { |
| 2697 | | - // HTML comment, scan to end |
| 2698 | | - $endpos = strpos( $text, '-->', $i + 4 ); |
| 2699 | | - if ( $endpos === false ) { |
| | 2748 | + // To avoid leaving blank lines, when a comment is both preceded |
| | 2749 | + // and followed by a newline (ignoring spaces), trim leading and |
| | 2750 | + // trailing spaces and one of the newlines. |
| | 2751 | + |
| | 2752 | + // Find the end |
| | 2753 | + $endPos = strpos( $text, '-->', $i + 4 ); |
| | 2754 | + if ( $endPos === false ) { |
| 2700 | 2755 | // Unclosed comment in input, runs to end |
| 2701 | 2756 | $inner = substr( $text, $i ); |
| 2702 | | - if ( $this->ot['html'] ) { |
| 2703 | | - // Close it so later stripping can remove it |
| 2704 | | - $inner .= '-->'; |
| 2705 | | - } |
| 2706 | 2757 | $accum .= '<comment>' . htmlspecialchars( $inner ) . '</comment>'; |
| 2707 | 2758 | $i = strlen( $text ); |
| 2708 | 2759 | } else { |
| 2709 | | - $inner = substr( $text, $i, $endpos - $i + 3 ); |
| | 2760 | + // Search backwards for leading whitespace |
| | 2761 | + $wsStart = $i ? ( $i - strspn( $revText, ' ', strlen( $text ) - $i - 1 ) ) : 0; |
| | 2762 | + // Search forwards for trailing whitespace |
| | 2763 | + // $wsEnd will be the position of the last space |
| | 2764 | + $wsEnd = $endPos + 2 + strspn( $text, ' ', $endPos + 3 ); |
| | 2765 | + // Eat the line if possible |
| | 2766 | + if ( $wsStart > 0 && substr( $text, $wsStart - 1, 1 ) == "\n" |
| | 2767 | + && substr( $text, $wsEnd + 1, 1 ) == "\n" ) |
| | 2768 | + { |
| | 2769 | + $startPos = $wsStart; |
| | 2770 | + $endPos = $wsEnd + 1; |
| | 2771 | + } else { |
| | 2772 | + // No line to eat, just take the comment itself |
| | 2773 | + $startPos = $i; |
| | 2774 | + $endPos += 2; |
| | 2775 | + } |
| | 2776 | + $inner = substr( $text, $startPos, $endPos - $startPos + 1 ); |
| 2710 | 2777 | $accum .= '<comment>' . htmlspecialchars( $inner ) . '</comment>'; |
| 2711 | | - $i = $endpos + 3; |
| | 2778 | + $i = $endPos + 1; |
| 2712 | 2779 | } |
| 2713 | 2780 | continue; |
| 2714 | 2781 | } |
| — | — | @@ -2724,6 +2791,15 @@ |
| 2725 | 2792 | ++$i; |
| 2726 | 2793 | continue; |
| 2727 | 2794 | } |
| | 2795 | + |
| | 2796 | + // Handle ignored tags |
| | 2797 | + if ( in_array( $name, $ignoredTags ) ) { |
| | 2798 | + $accum .= '<ignore>' . htmlspecialchars( substr( $text, $i, $tagEndPos - $i + 1 ) ) . '</ignore>'; |
| | 2799 | + $i = $tagEndPos + 1; |
| | 2800 | + continue; |
| | 2801 | + } |
| | 2802 | + |
| | 2803 | + $tagStartPos = $i; |
| 2728 | 2804 | if ( $text[$tagEndPos-1] == '/' ) { |
| 2729 | 2805 | $attrEnd = $tagEndPos - 1; |
| 2730 | 2806 | $inner = null; |
| — | — | @@ -2743,6 +2819,13 @@ |
| 2744 | 2820 | $close = ''; |
| 2745 | 2821 | } |
| 2746 | 2822 | } |
| | 2823 | + // <includeonly> and <noinclude> just become <ignore> tags |
| | 2824 | + if ( in_array( $name, $ignoredElements ) ) { |
| | 2825 | + $accum .= '<ignore>' . htmlspecialchars( substr( $text, $tagStartPos, $i - $tagStartPos ) ) |
| | 2826 | + . '</ignore>'; |
| | 2827 | + continue; |
| | 2828 | + } |
| | 2829 | + |
| 2747 | 2830 | $accum .= '<ext>'; |
| 2748 | 2831 | if ( $attrEnd <= $attrStart ) { |
| 2749 | 2832 | $attr = ''; |
| — | — | @@ -2784,13 +2867,11 @@ |
| 2785 | 2868 | // A heading must be open, otherwise \n wouldn't have been in the search list |
| 2786 | 2869 | assert( $piece['open'] == "\n" ); |
| 2787 | 2870 | assert( $stackIndex == 0 ); |
| 2788 | | - // Search back through the accumulator to see if it has a proper close |
| 2789 | | - // No efficient way to do this in PHP AFAICT: strrev, PCRE search with $ anchor |
| 2790 | | - // and rtrim are all O(N) in total size. Optimal would be O(N) in trailing |
| 2791 | | - // whitespace size only. |
| | 2871 | + // Search back through the input to see if it has a proper close |
| | 2872 | + // Do this using the reversed string since the other solutions (end anchor, etc.) are inefficient |
| 2792 | 2873 | $m = false; |
| 2793 | 2874 | $count = $piece['count']; |
| 2794 | | - if ( preg_match( "/(={{$count}})\s*$/", $accum, $m, 0, $count ) ) { |
| | 2875 | + if ( preg_match( "/\s*(={{$count}})/A", $revText, $m, 0, strlen( $text ) - $i ) ) { |
| 2795 | 2876 | // Found match, output <h> |
| 2796 | 2877 | $count = min( strlen( $m[1] ), $count ); |
| 2797 | 2878 | $element = "<h level=\"$count\" i=\"$headingIndex\">$accum</h>"; |
| — | — | @@ -3022,27 +3103,6 @@ |
| 3023 | 3104 | } |
| 3024 | 3105 | |
| 3025 | 3106 | /** |
| 3026 | | - * Convert text to a document tree, like preprocessToDom(), but with some special handling |
| 3027 | | - * assuming the source text is from a template -- specifically noinclude/includeonly behaviour. |
| 3028 | | - */ |
| 3029 | | - function preprocessTplToDom( $text ) { |
| 3030 | | - # If there are any <onlyinclude> tags, only include them |
| 3031 | | - if ( !$this->ot['msg'] ) { |
| 3032 | | - if ( in_string( '<onlyinclude>', $text ) && in_string( '</onlyinclude>', $text ) ) { |
| 3033 | | - $replacer = new OnlyIncludeReplacer; |
| 3034 | | - StringUtils::delimiterReplaceCallback( '<onlyinclude>', '</onlyinclude>', |
| 3035 | | - array( &$replacer, 'replace' ), $text ); |
| 3036 | | - $text = $replacer->output; |
| 3037 | | - } |
| 3038 | | - # Remove <noinclude> sections and <includeonly> tags |
| 3039 | | - $text = StringUtils::delimiterReplace( '<noinclude>', '</noinclude>', '', $text ); |
| 3040 | | - $text = strtr( $text, array( '<includeonly>' => '' , '</includeonly>' => '' ) ); |
| 3041 | | - } |
| 3042 | | - |
| 3043 | | - return $this->preprocessToDom( $text ); |
| 3044 | | - } |
| 3045 | | - |
| 3046 | | - /** |
| 3047 | 3107 | * Replace magic variables, templates, and template arguments |
| 3048 | 3108 | * with the appropriate text. Templates are substituted recursively, |
| 3049 | 3109 | * taking care to avoid infinite loops. |
| — | — | @@ -3311,7 +3371,7 @@ |
| 3312 | 3372 | } else { |
| 3313 | 3373 | $text = $this->interwikiTransclude( $title, 'raw' ); |
| 3314 | 3374 | // Preprocess it like a template |
| 3315 | | - $text = $this->preprocessTplToDom( $text ); |
| | 3375 | + $text = $this->preprocessToDom( $text, self::PTD_FOR_INCLUSION ); |
| 3316 | 3376 | $isDOM = true; |
| 3317 | 3377 | } |
| 3318 | 3378 | $found = true; |
| — | — | @@ -3404,7 +3464,7 @@ |
| 3405 | 3465 | return array( false, $title ); |
| 3406 | 3466 | } |
| 3407 | 3467 | |
| 3408 | | - $dom = $this->preprocessTplToDom( $text ); |
| | 3468 | + $dom = $this->preprocessToDom( $text, self::PTD_FOR_INCLUSION ); |
| 3409 | 3469 | $this->mTplDomCache[ $titleText ] = $dom; |
| 3410 | 3470 | |
| 3411 | 3471 | if (! $title->equals($cacheTitle)) { |
| — | — | @@ -3906,10 +3966,13 @@ |
| 3907 | 3967 | } |
| 3908 | 3968 | # give headline the correct <h#> tag |
| 3909 | 3969 | if( $showEditLink && $sectionIndex !== false ) { |
| 3910 | | - if( $isTemplate ) |
| 3911 | | - $editlink = $sk->editSectionLinkForOther($titleText, $sectionIndex); |
| 3912 | | - else |
| | 3970 | + if( $isTemplate ) { |
| | 3971 | + # Put a T flag in the section identifier, to indicate to extractSections() |
| | 3972 | + # that sections inside <includeonly> should be counted. |
| | 3973 | + $editlink = $sk->editSectionLinkForOther($titleText, "T-$sectionIndex"); |
| | 3974 | + } else { |
| 3913 | 3975 | $editlink = $sk->editSectionLink($this->mTitle, $sectionIndex, $headlineHint); |
| | 3976 | + } |
| 3914 | 3977 | } else { |
| 3915 | 3978 | $editlink = ''; |
| 3916 | 3979 | } |
| — | — | @@ -4910,14 +4973,22 @@ |
| 4911 | 4974 | * |
| 4912 | 4975 | * External callers should use the getSection and replaceSection methods. |
| 4913 | 4976 | * |
| 4914 | | - * @param $text Page wikitext |
| 4915 | | - * @param $section Numbered section. 0 pulls the text before the first |
| 4916 | | - * heading; other numbers will pull the given section |
| 4917 | | - * along with its lower-level subsections. If the section is |
| 4918 | | - * not found, $mode=get will return $newtext, and |
| 4919 | | - * $mode=replace will return $text. |
| 4920 | | - * @param $mode One of "get" or "replace" |
| 4921 | | - * @param $newText Replacement text for section data. |
| | 4977 | + * @param string $text Page wikitext |
| | 4978 | + * @param string $section A section identifier string of the form: |
| | 4979 | + * <flag1> - <flag2> - ... - <section number> |
| | 4980 | + * |
| | 4981 | + * Currently the only recognised flag is "T", which means the target section number |
| | 4982 | + * was derived during a template inclusion parse, in other words this is a template |
| | 4983 | + * section edit link. If no flags are given, it was an ordinary section edit link. |
| | 4984 | + * This flag is required to avoid a section numbering mismatch when a section is |
| | 4985 | + * enclosed by <includeonly> (bug 6563). |
| | 4986 | + * |
| | 4987 | + * The section number 0 pulls the text before the first heading; other numbers will |
| | 4988 | + * pull the given section along with its lower-level subsections. If the section is |
| | 4989 | + * not found, $mode=get will return $newtext, and $mode=replace will return $text. |
| | 4990 | + * |
| | 4991 | + * @param string $mode One of "get" or "replace" |
| | 4992 | + * @param string $newText Replacement text for section data. |
| 4922 | 4993 | * @return string for "get", the extracted section text. |
| 4923 | 4994 | * for "replace", the whole page with the section replaced. |
| 4924 | 4995 | */ |
| — | — | @@ -4931,8 +5002,17 @@ |
| 4932 | 5003 | $outText = ''; |
| 4933 | 5004 | $frame = new PPFrame( $this ); |
| 4934 | 5005 | |
| | 5006 | + // Process section extraction flags |
| | 5007 | + $flags = 0; |
| | 5008 | + $sectionParts = explode( '-', $section ); |
| | 5009 | + $sectionIndex = array_pop( $sectionParts ); |
| | 5010 | + foreach ( $sectionParts as $part ) { |
| | 5011 | + if ( $part == 'T' ) { |
| | 5012 | + $flags |= self::PTD_FOR_INCLUSION; |
| | 5013 | + } |
| | 5014 | + } |
| 4935 | 5015 | // Preprocess the text |
| 4936 | | - $dom = $this->preprocessToDom( $text ); |
| | 5016 | + $dom = $this->preprocessToDom( $text, $flags ); |
| 4937 | 5017 | $root = $dom->documentElement; |
| 4938 | 5018 | |
| 4939 | 5019 | // <h> nodes indicate section breaks |
| — | — | @@ -4940,13 +5020,13 @@ |
| 4941 | 5021 | $node = $root->firstChild; |
| 4942 | 5022 | |
| 4943 | 5023 | // Find the target section |
| 4944 | | - if ( $section == 0 ) { |
| | 5024 | + if ( $sectionIndex == 0 ) { |
| 4945 | 5025 | // Section zero doesn't nest, level=big |
| 4946 | 5026 | $targetLevel = 1000; |
| 4947 | 5027 | } else { |
| 4948 | 5028 | while ( $node ) { |
| 4949 | 5029 | if ( $node->nodeName == 'h' ) { |
| 4950 | | - if ( $curIndex + 1 == $section ) { |
| | 5030 | + if ( $curIndex + 1 == $sectionIndex ) { |
| 4951 | 5031 | break; |
| 4952 | 5032 | } |
| 4953 | 5033 | $curIndex++; |
| — | — | @@ -4975,7 +5055,7 @@ |
| 4976 | 5056 | if ( $node->nodeName == 'h' ) { |
| 4977 | 5057 | $curIndex++; |
| 4978 | 5058 | $curLevel = $node->getAttribute( 'level' ); |
| 4979 | | - if ( $curIndex != $section && $curLevel <= $targetLevel ) { |
| | 5059 | + if ( $curIndex != $sectionIndex && $curLevel <= $targetLevel ) { |
| 4980 | 5060 | break; |
| 4981 | 5061 | } |
| 4982 | 5062 | } |
| — | — | @@ -5012,9 +5092,9 @@ |
| 5013 | 5093 | * |
| 5014 | 5094 | * If a section contains subsections, these are also returned. |
| 5015 | 5095 | * |
| 5016 | | - * @param $text String: text to look in |
| 5017 | | - * @param $section Integer: section number |
| 5018 | | - * @param $deftext: default to return if section is not found |
| | 5096 | + * @param string $text text to look in |
| | 5097 | + * @param string $section section identifier |
| | 5098 | + * @param string $deftext default to return if section is not found |
| 5019 | 5099 | * @return string text of the requested section |
| 5020 | 5100 | */ |
| 5021 | 5101 | public function getSection( $text, $section, $deftext='' ) { |
| — | — | @@ -5217,8 +5297,9 @@ |
| 5218 | 5298 | const NO_ARGS = 1; |
| 5219 | 5299 | const NO_TEMPLATES = 2; |
| 5220 | 5300 | const STRIP_COMMENTS = 4; |
| | 5301 | + const NO_IGNORE = 8; |
| 5221 | 5302 | |
| 5222 | | - const RECOVER_ORIG = 3; |
| | 5303 | + const RECOVER_ORIG = 11; |
| 5223 | 5304 | |
| 5224 | 5305 | /** |
| 5225 | 5306 | * Construct a new preprocessor frame. |
| — | — | @@ -5323,11 +5404,24 @@ |
| 5324 | 5405 | } |
| 5325 | 5406 | } elseif ( $root->nodeName == 'comment' ) { |
| 5326 | 5407 | # HTML-style comment |
| 5327 | | - if ( $flags & self::STRIP_COMMENTS ) { |
| | 5408 | + if ( $this->parser->ot['html'] |
| | 5409 | + || ( $this->parser->ot['pre'] && $this->mOptions->getRemoveComments() ) |
| | 5410 | + || ( $flags & self::STRIP_COMMENTS ) ) |
| | 5411 | + { |
| 5328 | 5412 | $s = ''; |
| 5329 | 5413 | } else { |
| 5330 | 5414 | $s = $root->textContent; |
| 5331 | 5415 | } |
| | 5416 | + } elseif ( $root->nodeName == 'ignore' ) { |
| | 5417 | + # Output suppression used by <includeonly> etc. |
| | 5418 | + # OT_WIKI will only respect <ignore> in substed templates. |
| | 5419 | + # The other output types respect it unless NO_IGNORE is set. |
| | 5420 | + # extractSections() sets NO_IGNORE and so never respects it. |
| | 5421 | + if ( ( !isset( $this->parent ) && $this->parser->ot['wiki'] ) || ( $flags & self::NO_IGNORE ) ) { |
| | 5422 | + $s = $root->textContent; |
| | 5423 | + } else { |
| | 5424 | + $s = ''; |
| | 5425 | + } |
| 5332 | 5426 | } elseif ( $root->nodeName == 'ext' ) { |
| 5333 | 5427 | # Extension tag |
| 5334 | 5428 | $xpath = new DOMXPath( $root->ownerDocument ); |
| — | — | @@ -5417,6 +5511,31 @@ |
| 5418 | 5512 | return array( $name, $index, $values->item( 0 ) ); |
| 5419 | 5513 | } |
| 5420 | 5514 | |
| | 5515 | + /** |
| | 5516 | + * Split an <ext> node into an associative array containing name, attr, inner and close |
| | 5517 | + * All values in the resulting array are DOMNodes. Inner and close are optional. |
| | 5518 | + */ |
| | 5519 | + function splitExtNode( $node ) { |
| | 5520 | + $xpath = new DOMXPath( $node->ownerDocument ); |
| | 5521 | + $names = $xpath->query( 'name', $node ); |
| | 5522 | + $attrs = $xpath->query( 'attr', $node ); |
| | 5523 | + $inners = $xpath->query( 'inner', $node ); |
| | 5524 | + $closes = $xpath->query( 'close', $node ); |
| | 5525 | + if ( !$names->length || !$attrs->length ) { |
| | 5526 | + throw new MWException( 'Invalid ext node passed to ' . __METHOD__ ); |
| | 5527 | + } |
| | 5528 | + $parts = array( |
| | 5529 | + 'name' => $names->item( 0 ), |
| | 5530 | + 'attr' => $attrs->item( 0 ) ); |
| | 5531 | + if ( $inners->length ) { |
| | 5532 | + $parts['inner'] = $inners->item( 0 ); |
| | 5533 | + } |
| | 5534 | + if ( $closes->length ) { |
| | 5535 | + $parts['close'] = $closes->item( 0 ); |
| | 5536 | + } |
| | 5537 | + return $parts; |
| | 5538 | + } |
| | 5539 | + |
| 5421 | 5540 | function __toString() { |
| 5422 | 5541 | return 'frame{}'; |
| 5423 | 5542 | } |