Merge "maintenance: Document secondary purpose of --server"

[lhc/web/wiklou.git] / includes / parser / Parser.php

diff --git a/includes/parser/Parser.php b/includes/parser/Parser.php
index 8e5dcbd..b66031c 100644 (file)
--- a/includes/parser/Parser.php
+++ b/includes/parser/Parser.php
@@ -1113,7 +1113,11 @@ class Parser {
                                        $line = "</{$last_tag}>{$line}";
                                }
                                array_pop( $tr_attributes );
-                               $outLine = $line . str_repeat( '</dd></dl>', $indent_level );
+                               if ( $indent_level > 0 ) {
+                                       $outLine = rtrim( $line ) . str_repeat( '</dd></dl>', $indent_level );
+                               } else {
+                                       $outLine = $line;
+                               }
                        } elseif ( $first_two === '|-' ) {
                                # Now we have a table row
                                $line = preg_replace( '#^\|-+#', '', $line );
@@ -1204,13 +1208,15 @@ class Parser {
                                        # be mistaken as delimiting cell parameters
                                        # Bug T153140: Neither should language converter markup.
                                        if ( preg_match( '/\[\[|-\{/', $cell_data[0] ) === 1 ) {
-                                               $cell = "{$previous}<{$last_tag}>{$cell}";
+                                               $cell = "{$previous}<{$last_tag}>" . trim( $cell );
                                        } elseif ( count( $cell_data ) == 1 ) {
-                                               $cell = "{$previous}<{$last_tag}>{$cell_data[0]}";
+                                               // Whitespace in cells is trimmed
+                                               $cell = "{$previous}<{$last_tag}>" . trim( $cell_data[0] );
                                        } else {
                                                $attributes = $this->mStripState->unstripBoth( $cell_data[0] );
                                                $attributes = Sanitizer::fixTagAttributes( $attributes, $last_tag );
-                                               $cell = "{$previous}<{$last_tag}{$attributes}>{$cell_data[1]}";
+                                               // Whitespace in cells is trimmed
+                                               $cell = "{$previous}<{$last_tag}{$attributes}>" . trim( $cell_data[1] );
                                        }
 
                                        $outLine .= $cell;
@@ -1465,7 +1471,7 @@ class Parser {
        /**
         * @throws MWException
         * @param array $m
-        * @return HTML|string
+        * @return string HTML
         */
        public function magicLinkCallback( $m ) {
                if ( isset( $m[1] ) && $m[1] !== '' ) {
@@ -1617,7 +1623,9 @@ class Parser {
        public function doHeadings( $text ) {
                for ( $i = 6; $i >= 1; --$i ) {
                        $h = str_repeat( '=', $i );
-                       $text = preg_replace( "/^$h(.+)$h\\s*$/m", "<h$i>\\1</h$i>", $text );
+                       // Trim non-newline whitespace from headings
+                       // Using \s* will break for: "==\n===\n" and parse as <h2>=</h2>
+                       $text = preg_replace( "/^(?:$h)[ \\t]*(.+?)[ \\t]*(?:$h)\\s*$/m", "<h$i>\\1</h$i>", $text );
                }
                return $text;
        }
@@ -1868,8 +1876,8 @@ class Parser {
 
                        $dtrail = '';
 
-                       # Set linktype for CSS - if URL==text, link is essentially free
-                       $linktype = ( $text === $url ) ? 'free' : 'text';
+                       # Set linktype for CSS
+                       $linktype = 'text';
 
                        # No link text, e.g. [http://domain.tld/some.link]
                        if ( $text == '' ) {
@@ -4050,9 +4058,11 @@ class Parser {
 
                # Get all headlines for numbering them and adding funky stuff like [edit]
                # links - this is for later, but we need the number of headlines right now
+               # NOTE: white space in headings have been trimmed in doHeadings. They shouldn't
+               # be trimmed here since whitespace in HTML headings is significant.
                $matches = [];
                $numMatches = preg_match_all(
-                       '/<H(?P<level>[1-6])(?P<attrib>.*?>)\s*(?P<header>[\s\S]*?)\s*<\/H[1-6] *>/i',
+                       '/<H(?P<level>[1-6])(?P<attrib>.*?>)(?P<header>[\s\S]*?)<\/H[1-6] *>/i',
                        $text,
                        $matches
                );