From 090a9b33316a448f1af9ba865484fa5dafeda4a7 Mon Sep 17 00:00:00 2001
From: Felipe Pena <felipe@php.net>
Date: Sun, 3 Oct 2010 16:01:38 +0000
Subject: [PATCH] - Fixed bug #52971 (PCRE-Meta-Characters not working with
 utf-8) #   In  PCRE,  by  default, \d, \D, \s, \S, \w, and \W recognize only
 ASCII #       characters, even in UTF-8 mode. However, this can be changed by
 setting #       the PCRE_UCP option.

---
 NEWS                         |  1 +
 ext/pcre/php_pcre.c          |  9 +++++++-
 ext/pcre/tests/bug52971.phpt | 43 ++++++++++++++++++++++++++++++++++++
 3 files changed, 52 insertions(+), 1 deletion(-)
 create mode 100644 ext/pcre/tests/bug52971.phpt

diff --git a/NEWS b/NEWS
index bcd560c8db9..9e47361f747 100644
--- a/NEWS
+++ b/NEWS
@@ -22,6 +22,7 @@
 - Fixed possible crash in mssql_fetch_batch(). (Kalle)
 - Fixed inconsistent backlog default value (-1) in FPM on many systems. (fat)
 
+- Fixed bug #52971 (PCRE-Meta-Characters not working with utf-8). (Felipe)
 - Fixed bug #52947 (segfault when ssl stream option capture_peer_cert_chain
   used). (Felipe)
 - Fixed bug #52944 (Invalid write on second and subsequent reads with an
diff --git a/ext/pcre/php_pcre.c b/ext/pcre/php_pcre.c
index dcbc98fb851..ccb0a51c0ec 100644
--- a/ext/pcre/php_pcre.c
+++ b/ext/pcre/php_pcre.c
@@ -350,7 +350,14 @@ PHPAPI pcre_cache_entry* pcre_get_compiled_regex_cache(char *regex, int regex_le
 			case 'S':	do_study  = 1;					break;
 			case 'U':	coptions |= PCRE_UNGREEDY;		break;
 			case 'X':	coptions |= PCRE_EXTRA;			break;
-			case 'u':	coptions |= PCRE_UTF8;			break;
+			case 'u':	coptions |= PCRE_UTF8;
+	/* In  PCRE,  by  default, \d, \D, \s, \S, \w, and \W recognize only ASCII
+       characters, even in UTF-8 mode. However, this can be changed by setting
+       the PCRE_UCP option. */
+#ifdef PCRE_UCP
+						coptions |= PCRE_UCP;
+#endif			
+				break;
 
 			/* Custom preg options */
 			case 'e':	poptions |= PREG_REPLACE_EVAL;	break;
diff --git a/ext/pcre/tests/bug52971.phpt b/ext/pcre/tests/bug52971.phpt
new file mode 100644
index 00000000000..5949cb220c3
--- /dev/null
+++ b/ext/pcre/tests/bug52971.phpt
@@ -0,0 +1,43 @@
+--TEST--
+Bug #52971 (PCRE-Meta-Characters not working with utf-8)
+--SKIPIF--
+<?php if ((double)PCRE_VERSION < 8.1) die('skip PCRE_VERSION >= 8.1 is required!'); ?>
+--FILE--
+<?php
+
+$message = 'Der ist ein Süßwasserpool Süsswasserpool ... verschiedene Wassersportmöglichkeiten bei ...';
+
+$pattern = '/\bwasser/iu';
+preg_match_all($pattern, $message, $match, PREG_OFFSET_CAPTURE);
+var_dump($match);
+
+$pattern = '/[^\w]wasser/iu';
+preg_match_all($pattern, $message, $match, PREG_OFFSET_CAPTURE);
+var_dump($match);
+
+?>
+--EXPECTF--
+array(1) {
+  [0]=>
+  array(1) {
+    [0]=>
+    array(2) {
+      [0]=>
+      string(6) "Wasser"
+      [1]=>
+      int(61)
+    }
+  }
+}
+array(1) {
+  [0]=>
+  array(1) {
+    [0]=>
+    array(2) {
+      [0]=>
+      string(7) " Wasser"
+      [1]=>
+      int(60)
+    }
+  }
+}