X-Git-Url: https://scripts.mit.edu/gitweb/autoinstalls/wordpress.git/blobdiff_plain/899389d1e4043331309c0433543419258b230b60..38ac4bc40322ecdc4052db4263466573e01fa51f:/wp-includes/shortcodes.php

diff --git a/wp-includes/shortcodes.php b/wp-includes/shortcodes.php
index e3c267a4..2661201b 100644
--- a/wp-includes/shortcodes.php
+++ b/wp-includes/shortcodes.php
@@ -184,9 +184,10 @@ function has_shortcode( $content, $tag ) {
  * @global array $shortcode_tags List of shortcode tags and their callback hooks.
  *
  * @param string $content Content to search for shortcodes.
+ * @param bool $ignore_html When true, shortcodes inside HTML elements will be skipped.
  * @return string Content with shortcodes filtered out.
  */
-function do_shortcode($content) {
+function do_shortcode( $content, $ignore_html = false ) {
 	global $shortcode_tags;
 
 	if ( false === strpos( $content, '[' ) ) {
@@ -196,8 +197,24 @@ function do_shortcode($content) {
 	if (empty($shortcode_tags) || !is_array($shortcode_tags))
 		return $content;
 
+	$tagnames = array_keys($shortcode_tags);
+	$tagregexp = join( '|', array_map('preg_quote', $tagnames) );
+	$pattern = "/\\[($tagregexp)/s";
+
+	if ( 1 !== preg_match( $pattern, $content ) ) {
+		// Avoids parsing HTML when there are no shortcodes or embeds anyway.
+		return $content;
+	}
+
+	$content = do_shortcodes_in_html_tags( $content, $ignore_html );
+
 	$pattern = get_shortcode_regex();
-	return preg_replace_callback( "/$pattern/s", 'do_shortcode_tag', $content );
+	$content = preg_replace_callback( "/$pattern/s", 'do_shortcode_tag', $content );
+	
+	// Always restore square braces so we don't break things like <!--[if IE ]>
+	$content = unescape_invalid_shortcodes( $content );
+	
+	return $content;
 }
 
 /**
@@ -290,6 +307,141 @@ function do_shortcode_tag( $m ) {
 	}
 }
 
+/**
+ * Search only inside HTML elements for shortcodes and process them.
+ *
+ * Any [ or ] characters remaining inside elements will be HTML encoded
+ * to prevent interference with shortcodes that are outside the elements.
+ * Assumes $content processed by KSES already.  Users with unfiltered_html
+ * capability may get unexpected output if angle braces are nested in tags.
+ *
+ * @since 4.2.3
+ *
+ * @param string $content Content to search for shortcodes
+ * @param bool $ignore_html When true, all square braces inside elements will be encoded.
+ * @return string Content with shortcodes filtered out.
+ */
+function do_shortcodes_in_html_tags( $content, $ignore_html ) {
+	// Normalize entities in unfiltered HTML before adding placeholders.
+	$trans = array( '&#91;' => '&#091;', '&#93;' => '&#093;' );
+	$content = strtr( $content, $trans );
+	$trans = array( '[' => '&#91;', ']' => '&#93;' );
+	
+	$pattern = get_shortcode_regex();
+
+	$comment_regex =
+		  '!'           // Start of comment, after the <.
+		. '(?:'         // Unroll the loop: Consume everything until --> is found.
+		.     '-(?!->)' // Dash not followed by end of comment.
+		.     '[^\-]*+' // Consume non-dashes.
+		. ')*+'         // Loop possessively.
+		. '(?:-->)?';   // End of comment. If not found, match all input.
+
+	$regex =
+		  '/('                   // Capture the entire match.
+		.     '<'                // Find start of element.
+		.     '(?(?=!--)'        // Is this a comment?
+		.         $comment_regex // Find end of comment.
+		.     '|'
+		.         '[^>]*>?'      // Find end of element. If not found, match all input.
+		.     ')'
+		. ')/s';
+
+	$textarr = preg_split( $regex, $content, -1, PREG_SPLIT_DELIM_CAPTURE | PREG_SPLIT_NO_EMPTY );
+
+	foreach ( $textarr as &$element ) {
+		if ( '<' !== $element[0] ) {
+			continue;
+		}
+
+		$noopen = false === strpos( $element, '[' );
+		$noclose = false === strpos( $element, ']' );
+		if ( $noopen || $noclose ) {
+			// This element does not contain shortcodes.
+			if ( $noopen xor $noclose ) {
+				// Need to encode stray [ or ] chars.
+				$element = strtr( $element, $trans );
+			}
+			continue;
+		}
+
+		if ( $ignore_html || '<!--' === substr( $element, 0, 4 ) ) {
+			// Encode all [ and ] chars.
+			$element = strtr( $element, $trans );
+			continue;
+		}
+
+		$attributes = wp_kses_attr_parse( $element );
+		if ( false === $attributes ) {
+			// Looks like we found some crazy unfiltered HTML.  Skipping it for sanity.
+			$element = strtr( $element, $trans );
+			continue;
+		}
+		
+		// Get element name
+		$front = array_shift( $attributes );
+		$back = array_pop( $attributes );
+		$matches = array();
+		preg_match('%[a-zA-Z0-9]+%', $front, $matches);
+		$elname = $matches[0];
+		
+		// Look for shortcodes in each attribute separately.
+		foreach ( $attributes as &$attr ) {
+			$open = strpos( $attr, '[' );
+			$close = strpos( $attr, ']' );
+			if ( false === $open || false === $close ) {
+				continue; // Go to next attribute.  Square braces will be escaped at end of loop.
+			}
+			$double = strpos( $attr, '"' );
+			$single = strpos( $attr, "'" );
+			if ( ( false === $single || $open < $single ) && ( false === $double || $open < $double ) ) {
+				// $attr like '[shortcode]' or 'name = [shortcode]' implies unfiltered_html.
+				// In this specific situation we assume KSES did not run because the input
+				// was written by an administrator, so we should avoid changing the output
+				// and we do not need to run KSES here.
+				$attr = preg_replace_callback( "/$pattern/s", 'do_shortcode_tag', $attr );
+			} else {
+				// $attr like 'name = "[shortcode]"' or "name = '[shortcode]'"
+				// We do not know if $content was unfiltered. Assume KSES ran before shortcodes.
+				$count = 0;
+				$new_attr = preg_replace_callback( "/$pattern/s", 'do_shortcode_tag', $attr, -1, $count );
+				if ( $count > 0 ) {
+					// Sanitize the shortcode output using KSES.
+					$new_attr = wp_kses_one_attr( $new_attr, $elname );
+					if ( '' !== $new_attr ) {
+						// The shortcode is safe to use now.
+						$attr = $new_attr;
+					}
+				}
+			}
+		}
+		$element = $front . implode( '', $attributes ) . $back;
+		
+		// Now encode any remaining [ or ] chars.
+		$element = strtr( $element, $trans );
+	}
+	
+	$content = implode( '', $textarr );
+	
+	return $content;
+}
+
+/**
+ * Remove placeholders added by do_shortcodes_in_html_tags().
+ *
+ * @since 4.2.3
+ *
+ * @param string $content Content to search for placeholders.
+ * @return string Content with placeholders removed.
+ */
+function unescape_invalid_shortcodes( $content ) {
+        // Clean up entire string, avoids re-parsing HTML.
+        $trans = array( '&#91;' => '[', '&#93;' => ']' );
+        $content = strtr( $content, $trans );
+        
+        return $content;
+}
+
 /**
  * Retrieve all attributes from the shortcodes tag.
  *
@@ -389,9 +541,15 @@ function strip_shortcodes( $content ) {
 	if (empty($shortcode_tags) || !is_array($shortcode_tags))
 		return $content;
 
+	$content = do_shortcodes_in_html_tags( $content, true );
+
 	$pattern = get_shortcode_regex();
+	$content = preg_replace_callback( "/$pattern/s", 'strip_shortcode_tag', $content );
 
-	return preg_replace_callback( "/$pattern/s", 'strip_shortcode_tag', $content );
+	// Always restore square braces so we don't break things like <!--[if IE ]>
+	$content = unescape_invalid_shortcodes( $content );
+	
+	return $content;
 }
 
 function strip_shortcode_tag( $m ) {