Merge pull request #281 from mrclay/2276-friendly-title

Fixes #2276: Better friendly titles

2 files changed, 32 insertions, 5 deletions

diff --git a/engine/tests/api/metadata.php b/engine/tests/api/metadata.php
index 244036f80..c63b0cbec 100644
--- a/engine/tests/api/metadata.php
+++ b/engine/tests/api/metadata.php
@@ -28,6 +28,8 @@ class ElggCoreMetadataAPITest extends ElggCoreUnitTest {
 
 	public function testGetMetastringById() {
 		foreach (array('metaUnitTest', 'metaunittest', 'METAUNITTEST') as $string) {
+			// in case previous tests failed to cleanup after themselves
+			$this->delete_metastrings($string);
 			$this->create_metastring($string);
 		}
 
@@ -194,11 +196,19 @@ class ElggCoreMetadataAPITest extends ElggCoreUnitTest {
 		$u2->delete();
 	}
 
+	protected function delete_metastrings($string) {
+		global $CONFIG, $METASTRINGS_CACHE, $METASTRINGS_DEADNAME_CACHE;
+		$METASTRINGS_CACHE = $METASTRINGS_DEADNAME_CACHE = array();
+
+		$string = sanitise_string($string);
+		mysql_query("DELETE FROM {$CONFIG->dbprefix}metastrings WHERE string = BINARY '$string'");
+	}
 
 	protected function create_metastring($string) {
 		global $CONFIG, $METASTRINGS_CACHE, $METASTRINGS_DEADNAME_CACHE;
 		$METASTRINGS_CACHE = $METASTRINGS_DEADNAME_CACHE = array();
 
+		$string = sanitise_string($string);
 		mysql_query("INSERT INTO {$CONFIG->dbprefix}metastrings (string) VALUES ('$string')");
 		$this->metastrings[$string] = mysql_insert_id();
 	}
diff --git a/engine/tests/regression/trac_bugs.php b/engine/tests/regression/trac_bugs.php
index 26a45ab6a..691433a41 100644
--- a/engine/tests/regression/trac_bugs.php
+++ b/engine/tests/regression/trac_bugs.php
@@ -202,16 +202,33 @@ class ElggCoreRegressionBugsTest extends ElggCoreUnitTest {
 
 	/**
 	 * http://trac.elgg.org/ticket/3210 - Don't remove -s in friendly titles
-	 * @todo: http://trac.elgg.org/ticket/2276 - improve char encoding
+	 * http://trac.elgg.org/ticket/2276 - improve char encoding
 	 */
 	public function test_friendly_title() {
 		$cases = array(
-			'Simple Test' => 'simple-test',
-			'Test top-level page' => 'test-top-level-page',
-//			'éclair' => 'éclair',
-//			'English, Español, and 日本語' => 'english-español-and-日本語'
+			// hyphen, underscore and ASCII whitespace replaced by separator,
+			// other non-alphanumeric ASCII removed
+			"a-a_a a\na\ra\ta\va!a\"a#a\$a%a&a'a(a)a*a+a,a.a/a:a;a<a=a>a?a@a[a\\a]a^a`a{a|a}a~a"
+			=> "a-a-a-a-a-a-aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa",
+
+			// separators trimmed
+			"-_ hello _-" => "hello",
+
+			// accents removed, lower case, other multibyte chars are URL encoded
+			"I\xC3\xB1t\xC3\xABrn\xC3\xA2ti\xC3\xB4n\xC3\xA0liz\xC3\xA6ti\xC3\xB8n, AND \xE6\x97\xA5\xE6\x9C\xAC\xE8\xAA\x9E"
+				// Iñtërnâtiônàlizætiøn, AND 日本語
+			=> 'internationalizaetion-and-%E6%97%A5%E6%9C%AC%E8%AA%9E',
+
+			// some HTML entity replacements
+			"Me &amp; You" => 'me-and-you',
 		);
 
+		// where available, string is converted to NFC before transliteration
+		if (ElggTranslit::hasNormalizerSupport()) {
+			$form_d = "A\xCC\x8A"; // A followed by 'COMBINING RING ABOVE' (U+030A)
+			$cases[$form_d] = "a";
+		}
+
 		foreach ($cases as $case => $expected) {
 			$friendly_title = elgg_get_friendly_title($case);
 			$this->assertIdentical($expected, $friendly_title);