summaryrefslogtreecommitdiff
path: root/include/utf8/utils/patterns.php
diff options
context:
space:
mode:
authorAndreas Baumann <mail@andreasbaumann.cc>2019-11-17 20:57:39 +0100
committerAndreas Baumann <mail@andreasbaumann.cc>2019-11-17 20:57:39 +0100
commit3b06ee0d381dc1be5f40ca98ad4278046d869d21 (patch)
treed31e79fc57d882b8267f40c3434480bb58a3ca73 /include/utf8/utils/patterns.php
downloadfluxbb-3b06ee0d381dc1be5f40ca98ad4278046d869d21.tar.xz
checked in initial customized verison for Archlinux32HEADmaster
Diffstat (limited to 'include/utf8/utils/patterns.php')
-rw-r--r--include/utf8/utils/patterns.php67
1 files changed, 67 insertions, 0 deletions
diff --git a/include/utf8/utils/patterns.php b/include/utf8/utils/patterns.php
new file mode 100644
index 0000000..5a85a4f
--- /dev/null
+++ b/include/utf8/utils/patterns.php
@@ -0,0 +1,67 @@
+<?php
+
+/**
+* PCRE Regular expressions for UTF-8. Note this file is not actually used by
+* the rest of the library but these regular expressions can be useful to have
+* available.
+* @version $Id: patterns.php,v 1.1 2006/02/25 14:20:02 harryf Exp $
+* @see http://www.w3.org/International/questions/qa-forms-utf-8
+* @package utf8
+* @subpackage patterns
+*/
+
+/**
+* PCRE Pattern to check a UTF-8 string is valid
+* Comes from W3 FAQ: Multilingual Forms
+* Note: modified to include full ASCII range including control chars
+* @see http://www.w3.org/International/questions/qa-forms-utf-8
+* @package utf8
+* @subpackage patterns
+*/
+$UTF8_VALID = '^('.
+ '[\x00-\x7F]'. # ASCII (including control chars)
+ '|[\xC2-\xDF][\x80-\xBF]'. # Non-overlong 2-byte
+ '|\xE0[\xA0-\xBF][\x80-\xBF]'. # Excluding overlongs
+ '|[\xE1-\xEC\xEE\xEF][\x80-\xBF]{2}'. # Straight 3-byte
+ '|\xED[\x80-\x9F][\x80-\xBF]'. # Excluding surrogates
+ '|\xF0[\x90-\xBF][\x80-\xBF]{2}'. # Planes 1-3
+ '|[\xF1-\xF3][\x80-\xBF]{3}'. # Planes 4-15
+ '|\xF4[\x80-\x8F][\x80-\xBF]{2}'. # Plane 16
+ ')*$';
+
+/**
+* PCRE Pattern to match single UTF-8 characters
+* Comes from W3 FAQ: Multilingual Forms
+* Note: modified to include full ASCII range including control chars
+* @see http://www.w3.org/International/questions/qa-forms-utf-8
+* @package utf8
+* @subpackage patterns
+*/
+$UTF8_MATCH =
+ '([\x00-\x7F])'. # ASCII (including control chars)
+ '|([\xC2-\xDF][\x80-\xBF])'. # Non-overlong 2-byte
+ '|(\xE0[\xA0-\xBF][\x80-\xBF])'. # Excluding overlongs
+ '|([\xE1-\xEC\xEE\xEF][\x80-\xBF]{2})'. # Straight 3-byte
+ '|(\xED[\x80-\x9F][\x80-\xBF])'. # Excluding surrogates
+ '|(\xF0[\x90-\xBF][\x80-\xBF]{2})'. # Planes 1-3
+ '|([\xF1-\xF3][\x80-\xBF]{3})'. # Planes 4-15
+ '|(\xF4[\x80-\x8F][\x80-\xBF]{2})'; # Plane 16
+
+/**
+* PCRE Pattern to locate bad bytes in a UTF-8 string
+* Comes from W3 FAQ: Multilingual Forms
+* Note: modified to include full ASCII range including control chars
+* @see http://www.w3.org/International/questions/qa-forms-utf-8
+* @package utf8
+* @subpackage patterns
+*/
+$UTF8_BAD =
+ '([\x00-\x7F]'. # ASCII (including control chars)
+ '|[\xC2-\xDF][\x80-\xBF]'. # Non-overlong 2-byte
+ '|\xE0[\xA0-\xBF][\x80-\xBF]'. # Excluding overlongs
+ '|[\xE1-\xEC\xEE\xEF][\x80-\xBF]{2}'. # Straight 3-byte
+ '|\xED[\x80-\x9F][\x80-\xBF]'. # Excluding surrogates
+ '|\xF0[\x90-\xBF][\x80-\xBF]{2}'. # Planes 1-3
+ '|[\xF1-\xF3][\x80-\xBF]{3}'. # Planes 4-15
+ '|\xF4[\x80-\x8F][\x80-\xBF]{2}'. # Plane 16
+ '|(.{1}))'; # Invalid byte