From 9e56293931457eb70d03a949729ebd2b394a5b03 Mon Sep 17 00:00:00 2001
From: Sam Minnee <sam@silverstripe.com>
Date: Wed, 12 May 2010 05:28:12 +0000
Subject: [PATCH] ENHANCEMENT: Macrons, umlauts, etc, are now transliterated
 when inserted into URLS. API CHANGE: Added Transliterator class, which uses
 iconv() or strtr() to convert characters with diacritical marks to their
 ASCII equivalents. API CHANGE: Added Extension hook updateURLSegment for
 SiteeTree.

git-svn-id: svn://svn.silverstripe.com/silverstripe/open/modules/sapphire/branches/2.4@104671 467b73ca-7a2a-4603-9d3b-597d59a354a9
---
 core/model/SiteTree.php       | 16 ++++++++--
 core/model/Transliterator.php | 56 +++++++++++++++++++++++++++++++++++
 javascript/UpdateURL.js       | 12 ++------
 3 files changed, 73 insertions(+), 11 deletions(-)
 create mode 100644 core/model/Transliterator.php

diff --git a/core/model/SiteTree.php b/core/model/SiteTree.php
index 40fb8b547..518a8fc9d 100755
--- a/core/model/SiteTree.php
+++ b/core/model/SiteTree.php
@@ -1496,11 +1496,18 @@ class SiteTree extends DataObject implements PermissionProvider,i18nEntityProvid
 	
 	/**
 	 * Generate a URL segment based on the title provided.
+	 * 
+	 * If {@link Extension}s wish to alter URL segment generation, they can do so by defining
+	 * updateURLSegment(&$url, $title).  $url will be passed by reference and should be modified.
+	 * $title will contain the title that was originally used as the source of this generated URL.
+	 * This lets decorators either start from scratch, or incrementally modify the generated URL.
+	 * 
 	 * @param string $title Page title.
 	 * @return string Generated url segment
 	 */
 	function generateURLSegment($title){
-		$t = strtolower($title);
+		$t = mb_strtolower($title);
+		$t = Object::create('Transliterator')->toASCII($title);
 		$t = str_replace('&amp;','-and-',$t);
 		$t = str_replace('&','-and-',$t);
 		$t = ereg_replace('[^A-Za-z0-9]+','-',$t);
@@ -1508,7 +1515,12 @@ class SiteTree extends DataObject implements PermissionProvider,i18nEntityProvid
 		if(!$t || $t == '-' || $t == '-1') {
 			$t = "page-$this->ID";
 		}
-		return trim($t, '-');
+		$t = trim($t, '-');
+		
+		// Hook for decorators
+		$this->extend('updateURLSegment', $t, $title);
+		
+		return $t;
 	}
 	
 	/**
diff --git a/core/model/Transliterator.php b/core/model/Transliterator.php
new file mode 100644
index 000000000..2e9e413c0
--- /dev/null
+++ b/core/model/Transliterator.php
@@ -0,0 +1,56 @@
+<?php
+
+/**
+ * Support class for converting unicode strings into a suitable 7-bit ASCII equivalent.
+ * 
+ * Usage:
+ * 
+ * $tr = new Transliterator();
+ * $ascii = $tr->toASCII($unicode);
+ */
+class Transliterator {
+	/**
+	 * Allow the use of iconv() to perform transliteration.  Set to false to disable.
+	 * Even if this variable is true, iconv() won't be used if it's not installed.
+	 */
+	static $use_iconv = false;
+	
+	function __construct() {
+		// A constructor is necessary for Object::create() to work
+	}
+	
+	/**
+	 * Convert the given utf8 string to a safe ASCII source
+	 */
+	function toASCII($source) {
+		if(function_exists('iconv') && self::$use_iconv) return $this->useIconv($source);
+		else return $this->useStrTr($source);
+	}
+
+	/**
+	 * Transliteration using strtr() and a lookup table
+	 */
+	protected function useStrTr($source) {
+	    $table = array(
+	        'Š'=>'S', 'š'=>'s', 'Đ'=>'Dj', 'đ'=>'dj', 'Ž'=>'Z', 'ž'=>'z', 'Č'=>'C', 'č'=>'c', 'Ć'=>'C', 'ć'=>'c',
+	        'À'=>'A', 'Á'=>'A', 'Â'=>'A', 'Ã'=>'A', 'Ä'=>'A', 'Å'=>'A', 'Æ'=>'A', 'Ç'=>'C', 'È'=>'E', 'É'=>'E',
+	        'Ê'=>'E', 'Ë'=>'E', 'Ì'=>'I', 'Í'=>'I', 'Î'=>'I', 'Ï'=>'I', 'Ñ'=>'N', 'Ò'=>'O', 'Ó'=>'O', 'Ô'=>'O',
+	        'Õ'=>'O', 'Ö'=>'O', 'Ø'=>'O', 'Ù'=>'U', 'Ú'=>'U', 'Û'=>'U', 'Ü'=>'U', 'Ý'=>'Y', 'Þ'=>'B', 'ß'=>'Ss',
+	        'à'=>'a', 'á'=>'a', 'â'=>'a', 'ã'=>'a', 'ä'=>'a', 'å'=>'a', 'æ'=>'ae', 'ç'=>'c', 'è'=>'e', 'é'=>'e',
+	        'ê'=>'e', 'ë'=>'e', 'ì'=>'i', 'í'=>'i', 'î'=>'i', 'ï'=>'i', 'ð'=>'o', 'ñ'=>'n', 'ò'=>'o', 'ó'=>'o',
+	        'ô'=>'o', 'õ'=>'o', 'ö'=>'o', 'ø'=>'o', 'ù'=>'u', 'ú'=>'u', 'û'=>'u', 'ý'=>'y', 'ý'=>'y', 'þ'=>'b',
+	        'ÿ'=>'y', 'Ŕ'=>'R', 'ŕ'=>'r',
+			'Ā'=>'A', 'ā'=>'a', 'Ē'=>'E', 'ē'=>'e', 'Ī'=>'I', 'ī'=>'i', 'Ō'=>'O', 'ō'=>'o', 'Ū'=>'U', 'ū'=>'u',
+			'œ'=>'oe', 'ß'=>'ss', 'ij'=>'ij', 
+	    );
+
+    	return strtr($source, $table);
+	}
+	
+	/**
+	 * Transliteration using iconv()
+	 */
+	protected function useIconv($source) {
+ 		return iconv("utf-8", "us-ascii//IGNORE//TRANSLIT", $source);
+	}
+}
\ No newline at end of file
diff --git a/javascript/UpdateURL.js b/javascript/UpdateURL.js
index 36ddee2a9..33311266a 100755
--- a/javascript/UpdateURL.js
+++ b/javascript/UpdateURL.js
@@ -8,20 +8,14 @@ Behaviour.register({
 			if(!$('Form_EditForm_URLSegment')) return;
 			
 			var urlSegmentField = $('Form_EditForm_URLSegment');
-			var newSuggestion = urlSegmentField.suggestNewValue( this.value.toLowerCase() );
 			var isNew = urlSegmentField.value.indexOf("new") == 0;
 			var confirmMessage = ss.i18n.sprintf(
-				ss.i18n._t('UPDATEURL.CONFIRM', 'Would you like me to change the URL to:\n\n%s/\n\nClick Ok to change the URL, click Cancel to leave it as:\n\n%s'),
-				newSuggestion,
+				ss.i18n._t('UPDATEURL.CONFIRM_V2', 'Would you like me to change the URL to match this page name?\n\nClick Ok to change the URL, click Cancel to leave it as:\n\n%s'),
 				urlSegmentField.value
 			);
 			
-			if( 
-				newSuggestion == urlSegmentField.value 
-				|| isNew 
-				|| confirm(confirmMessage)
-			) {
-				urlSegmentField.value = newSuggestion;
+			if(isNew  || confirm(confirmMessage)) {
+				urlSegmentField.value = '';
 			}
 			// If you type in Page name, the Navigation Label and Meta Title should automatically update the first time
 			// @todo: Change file name from UpdateURL to something more geneneric since we now do more than update the URL.