Formatting: Check the result of preg_split() in convert_smilies().

This aims to avoid a fatal error from `count()` when `preg_split()` fails on large input.

Includes:
* Optimizing the regular expression used to split the input by tags to avoid unlimited backtracking for better performance.
* Adjusting the function logic for better readability.

Follow-up to [340], [4380], [26191].

Props podpirate, nathkrill, rajinsharwar, dmsnell, bjorsch, q0rban, audrasjb, rupw, Ov3rfly, jorbin, nhrrob, chaion07, mcqueen22, azaozz, narenin, roybellingan, SergeyBiryukov.
See #51019.

git-svn-id: https://develop.svn.wordpress.org/trunk@59515 602fd350-edb4-49c9-b593-d223f7449a82

src/wp-includes/formatting.php

@@ -3473,40 +3473,49 @@ function translate_smiley( $matches ) {
  */
 function convert_smilies( $text ) {
 	global $wp_smiliessearch;
-	$output = '';
-	if ( get_option( 'use_smilies' ) && ! empty( $wp_smiliessearch ) ) {
-		// HTML loop taken from texturize function, could possible be consolidated.
-		$textarr = preg_split( '/(<.*>)/U', $text, -1, PREG_SPLIT_DELIM_CAPTURE ); // Capture the tags as well as in between.
-		$stop    = count( $textarr ); // Loop stuff.
 
-		// Ignore processing of specific tags.
-		$tags_to_ignore       = 'code|pre|style|script|textarea';
-		$ignore_block_element = '';
+	if ( ! get_option( 'use_smilies' ) || empty( $wp_smiliessearch ) ) {
+		// Return default text.
+		return $text;
+	}
 
-		for ( $i = 0; $i < $stop; $i++ ) {
-			$content = $textarr[ $i ];
+	// HTML loop taken from texturize function, could possible be consolidated.
+	$textarr = preg_split( '/(<[^>]*>)/U', $text, -1, PREG_SPLIT_DELIM_CAPTURE ); // Capture the tags as well as in between.
 
-			// If we're in an ignore block, wait until we find its closing tag.
-			if ( '' === $ignore_block_element && preg_match( '/^<(' . $tags_to_ignore . ')[^>]*>/', $content, $matches ) ) {
-				$ignore_block_element = $matches[1];
-			}
+	if ( false === $textarr ) {
+		// Return default text.
+		return $text;
+	}
 
-			// If it's not a tag and not in ignore block.
-			if ( '' === $ignore_block_element && strlen( $content ) > 0 && '<' !== $content[0] ) {
-				$content = preg_replace_callback( $wp_smiliessearch, 'translate_smiley', $content );
-			}
+	// Loop stuff.
+	$stop   = count( $textarr );
+	$output = '';
 
-			// Did we exit ignore block?
-			if ( '' !== $ignore_block_element && '</' . $ignore_block_element . '>' === $content ) {
-				$ignore_block_element = '';
-			}
+	// Ignore processing of specific tags.
+	$tags_to_ignore       = 'code|pre|style|script|textarea';
+	$ignore_block_element = '';
+
+	for ( $i = 0; $i < $stop; $i++ ) {
+		$content = $textarr[ $i ];
 
-			$output .= $content;
+		// If we're in an ignore block, wait until we find its closing tag.
+		if ( '' === $ignore_block_element && preg_match( '/^<(' . $tags_to_ignore . ')[^>]*>/', $content, $matches ) ) {
+			$ignore_block_element = $matches[1];
 		}
-	} else {
-		// Return default text.
-		$output = $text;
+
+		// If it's not a tag and not in ignore block.
+		if ( '' === $ignore_block_element && strlen( $content ) > 0 && '<' !== $content[0] ) {
+			$content = preg_replace_callback( $wp_smiliessearch, 'translate_smiley', $content );
+		}
+
+		// Did we exit ignore block?
+		if ( '' !== $ignore_block_element && '</' . $ignore_block_element . '>' === $content ) {
+			$ignore_block_element = '';
+		}
+
+		$output .= $content;
 	}
+
 	return $output;
 }
 

tests/phpunit/tests/formatting/convertSmilies.php

@@ -393,4 +393,16 @@ public function _filter_add_smilies( $wpsmiliestrans ) {
 		$wpsmiliestrans['<3'] = '\xe2\x9d\xa4';
 		return $wpsmiliestrans;
 	}
+
+
+	/**
+	 * Tests that the function does not throw a fatal error from count()
+	 * when preg_split() fails on large input.
+	 *
+	 * @ticket 51019
+	 */
+	public function test_smilies_with_large_text_input() {
+		$text = '<p><img alt="" src="data:image/png;base64,' . str_repeat( 'iVBORw0KGgoAAAAN', 65536 ) . '="></p> :)';
+		$this->assertStringContainsString( "\xf0\x9f\x99\x82", convert_smilies( $text ) );
+	}
 }