From bd9ee4c9f9ad197f049ee1b0a9863a7a3ef86bcf Mon Sep 17 00:00:00 2001
From: ozh <ozh@ozh.org>
Date: Fri, 5 Jun 2015 16:01:51 +0200
Subject: [PATCH] Fix mb_ fallbacks and PCRE/u compat check

Fixes 1915
---
 includes/functions-compat.php     |  8 ++++----
 includes/functions-formatting.php | 26 ++++++++++++++++++++------
 2 files changed, 24 insertions(+), 10 deletions(-)

diff --git a/includes/functions-compat.php b/includes/functions-compat.php
index 7829eb8..6db1b6b 100644
--- a/includes/functions-compat.php
+++ b/includes/functions-compat.php
@@ -136,14 +136,14 @@ function mb_substr( $str, $start, $length = null, $encoding = null ) {
 endif;
 function yourls_mb_substr( $str, $start, $length = null, $encoding = null ) {
     if ( null === $encoding ) {
-        $encoding = get_option( 'blog_charset' );
+        $encoding = 'UTF-8';
     }
     // The solution below works only for UTF-8,
     // so in case of a different charset just use built-in substr()
     if ( ! in_array( $encoding, array( 'utf8', 'utf-8', 'UTF8', 'UTF-8' ) ) ) {
         return is_null( $length ) ? substr( $str, $start ) : substr( $str, $start, $length );
     }
-    if ( _wp_can_use_pcre_u() ) {
+    if ( yourls_supports_pcre_u() ) {
         // Use the regex unicode support to separate the UTF-8 characters into an array
         preg_match_all( '/./us', $str, $match );
         $chars = is_null( $length ) ? array_slice( $match[0], $start ) : array_slice( $match[0], $start, $length );
@@ -187,14 +187,14 @@ function mb_strlen( $str, $encoding = null ) {
 endif;
 function yourls_mb_strlen( $str, $encoding = null ) {
     if ( null === $encoding ) {
-        $encoding = get_option( 'blog_charset' );
+        $encoding = 'UTF-8';
     }
     // The solution below works only for UTF-8,
     // so in case of a different charset just use built-in strlen()
     if ( ! in_array( $encoding, array( 'utf8', 'utf-8', 'UTF8', 'UTF-8' ) ) ) {
         return strlen( $str );
     }
-    if ( _wp_can_use_pcre_u() ) {
+    if ( yourls_supports_pcre_u() ) {
         // Use the regex unicode support to separate the UTF-8 characters into an array
         preg_match_all( '/./us', $str, $match );
         return count( $match[0] );
diff --git a/includes/functions-formatting.php b/includes/functions-formatting.php
index 2407d91..b288e8f 100644
--- a/includes/functions-formatting.php
+++ b/includes/functions-formatting.php
@@ -261,6 +261,25 @@ function yourls_seems_utf8( $str ) {
 	return true;
 }
 
+
+/**
+ * Check for PCRE /u modifier support. Stolen from WP.
+ *
+ * Just in case "PCRE is not compiled with PCRE_UTF8" which seems to happen
+ * on some distros even for PHP 5.3
+ *
+ * @since 1.7.1
+ *
+ * @return bool whether there's /u support or not
+ */
+function yourls_supports_pcre_u() {
+    static $utf8_pcre;
+    if( !isset( $utf8_pcre ) ) {
+        $utf8_pcre = (bool) @preg_match( '/^./u', 'a' );   
+    }
+    return $utf8_pcre;
+}
+
 /**
  * Checks for invalid UTF8 in a string. Stolen from WP
  *
@@ -277,13 +296,8 @@ function yourls_check_invalid_utf8( $string, $strip = false ) {
 		return '';
 	}
 
-	// Check for support for utf8 in the installed PCRE library once and store the result in a static
-	static $utf8_pcre;
-	if ( !isset( $utf8_pcre ) ) {
-		$utf8_pcre = @preg_match( '/^./u', 'a' );
-	}
 	// We can't demand utf8 in the PCRE installation, so just return the string in those cases
-	if ( !$utf8_pcre ) {
+	if ( !$yourls_supports_pcre_u() ) {
 		return $string;
 	}
 
-- 
2.45.0