From 909e195a9b54f38294f217ee8e10b17a62876756 Mon Sep 17 00:00:00 2001
From: Nils Adermann <naderman@naderman.de>
Date: Sun, 15 Jul 2007 20:53:27 +0000
Subject: - search result extract shouldn't end in the middle of a multibyte
 character [Bug #11863] - missing localisation for an imageset shouldn't
 create lots of "imageset refreshed" log messages [Bug #12027] - explain that
 themes which need parsing cannot be stored on the filesystem [Bug #11134] -
 normalize usernames (we really need to make sure we normalize everything) -
 improved utf8_clean_string, more complete list of homographs and NFKC
 normalization, also the resulting string is now trimmed - corrected searching
 subforums explanation [Bug #12209]

git-svn-id: file:///svn/phpbb/trunk@7890 89ea8834-ac86-4346-8a33-228a782c2dd0
---
 phpBB/develop/unicode_testing.php | 120 ++++++++++++++++++++++++++++++++++++++
 1 file changed, 120 insertions(+)
 create mode 100644 phpBB/develop/unicode_testing.php

(limited to 'phpBB/develop/unicode_testing.php')

diff --git a/phpBB/develop/unicode_testing.php b/phpBB/develop/unicode_testing.php
new file mode 100644
index 0000000000..25a13d1325
--- /dev/null
+++ b/phpBB/develop/unicode_testing.php
@@ -0,0 +1,120 @@
+<?php
+//
+// This file provides some useful functions for debugging the unicode/UTF-8 library
+// It requires utf_tools.php to be loaded
+//
+die("Please read the first lines of this script for instructions on how to enable it");
+
+if (!headers_sent())
+{
+	header('Content-type: text/html; charset=UTF-8');
+}
+
+/**
+ * Converts unicode escape sequences (\u0123) into UTF-8 characters
+ *
+ * @param	string	A unicode sequence
+ * @return	string	UTF-8 representation of the given unicode sequence
+ */
+function unicode_to_utf8($string)
+{
+	$utf8 = '';
+	$chars = array();
+	for ($i = 0; $i < strlen($string); $i++)
+	{
+		if (isset($string[$i + 5]) && substr($string, $i, 2) == '\\u' && ctype_xdigit(substr($string, $i + 2, 4)))
+		{
+			$utf8 .= utf8_from_unicode(array(base_convert(substr($string, $i + 2, 4), 16, 10)));
+			$i += 5;
+		}
+		else
+		{
+			$utf8 .= $string[$i];
+		}
+	}
+	return $utf8;
+}
+
+/**
+ * Takes an array of ints representing the Unicode characters and returns
+ * a UTF-8 string.
+ *
+ * @param array $array array of unicode code points representing a string
+ * @return string UTF-8 character string
+ */
+function utf8_from_unicode($array)
+{
+	$str = '';
+	foreach ($array as $value)
+	{
+		$str .= utf8_chr($value);
+	}
+	return $str;
+}
+
+/**
+* Converts a UTF-8 string to unicode code points
+*
+* @param	string	$text		UTF-8 string
+* @return	string				Unicode code points
+*/
+function utf8_to_unicode($text)
+{
+	return preg_replace_callback(
+		'#[\\xC2-\\xF4][\\x80-\\xBF]?[\\x80-\\xBF]?[\\x80-\\xBF]#',
+		'utf8_to_unicode_callback',
+		preg_replace_callback(
+			'#[\\x00-\\x7f]#',
+			'utf8_to_unicode_callback',
+			$text
+		)
+	);
+}
+
+/**
+* Takes a UTF-8 char and replaces it with its unicode escape sequence. Attention, $m is an array
+*
+* @param	array	$m			0-based numerically indexed array passed by preg_replace_callback()
+* @return	string				A unicode escape sequence
+*/
+function utf8_to_unicode_callback($m)
+{
+	return '\u' . str_pad(base_convert(utf8_ord($m[0]), 10, 16), 4, '0', STR_PAD_LEFT) . '';
+}
+
+/**
+* A wrapper function for the normalizer which takes care of including the class if required and modifies the passed strings
+* to be in NFKC
+*
+* @param	mixed	$strings	a string or an array of strings to normalize
+* @return	mixed				the normalized content, preserving array keys if array given.
+*/
+function utf8_normalize_nfkc($strings)
+{
+	if (empty($strings))
+	{
+		return $strings;
+	}
+
+	if (!class_exists('utf_normalizer'))
+	{
+		global $phpbb_root_path, $phpEx;
+		include($phpbb_root_path . 'includes/utf/utf_normalizer.' . $phpEx);
+	}
+
+	if (!is_array($strings))
+	{
+		utf_normalizer::nfkc($strings);
+	}
+	else if (is_array($strings))
+	{
+		foreach ($strings as $key => $string)
+		{
+			utf_normalizer::nfkc($strings[$key]);
+		}
+	}
+
+	return $strings;
+}
+
+?>
\ No newline at end of file
-- 
cgit v1.2.1