new utf8 safe functions it::grep and it::substr_replace

2 files changed, 29 insertions, 0 deletions

diff --git a/it.class b/it.class
index 924f271..f683568 100644
--- a/it.class
+++ b/it.class
@@ -388,6 +388,22 @@ static function replace($replacements, $string, $p = array())
 }
 
 /**
+ * Returns only the array elements matching the given regex
+ * @param $pattern Regex to match against
+ * @param $array   array to grep
+ * @return New array
+ */
+static function grep($pattern, $array, $p = array())
+{
+	if (!preg_match('/\\\\[wb]|[!\x80-\xff]|\[\[:/i', $pattern) && !$p)
+		$result = preg_grep('!' . $pattern . '!i' . (ini_get('default_charset') == 'utf-8' ? 'u' : ''), $array); # fast path for simple patterns
+	else
+		$result = preg_grep(it::convertregex($pattern, $p), $array);
+
+	return $result;
+}
+
+/**
  * Uppercase first character similar to ucfirst() but for mbstring.internal_encoding
  */
 static function ucfirst($string)
@@ -403,6 +419,11 @@ static function ucwords($string)
 	return preg_replace_callback('/\b\w/u', function($m) { return mb_strtoupper($m[1]); }, mb_strtolower($string));
 }
 
+static function substr_replace($string, $replacement, $start, $length)
+{
+	return grapheme_substr($string, 0, $start) . $replacement . grapheme_substr($string, $start + $length);
+} 
+
 /**
  * Extract key => value pairs from assoc array by key
  * @param $array array to filter
diff --git a/tests/it.t b/tests/it.t
index ec95cc4..317b8a3 100755
--- a/tests/it.t
+++ b/tests/it.t
@@ -305,6 +305,10 @@ is(it::replace(array('#' => "!", '\w' => "x"), "#ö"), "!x");
 is(it::replace(array('ö' => "x"), "Ö"), "x");
 is(it::replace(array('a' => "1"), "aaa", array('limit' => 1)), "1aa");
 
+is(it::grep('ismatch', array('ismatch', 'isnomatch')), array('ismatch'),  'grep with simple regex');
+is(it::grep('!', array('ismatch!', 'isnomatch')),      array('ismatch!'), '! in regex');
+is(it::grep('lower|UPPER', array('lower', 'LOWER', 'upper', 'UPPER'), 'casesensitive' => 1), array(0 => 'lower', 3 => 'UPPER'), 'set casesensitive');
+is(it::grep('match', array('foo' => 'match', 'bar' => 'gna')), array('foo' => 'match'), 'with keys');
 
 setlocale(LC_CTYPE, $oldlocale);
 ini_set('default_charset', $oldcharset);	# end of tests that must run with specific charset
@@ -342,4 +346,8 @@ is(it::date('datetime', 1000000.543), it::date('datetime', "1000000"), '... larg
 is(it::date('time', "10.5"), "10:05", 'interpret string with points with strtotime');
 is(it::date('time', "10.05"), "10:05", 'interpret string with points with strtotime');
 
+# it::substr_replace
+is(it::substr_replace('abcdefgh', 'xyz', 2, 4), substr_replace('abcdefgh', 'xyz', 2, 4), 'it::substr_replace the same as substr_replace for ascii');
+is(it::substr_replace('✔☯♥', '☃☃', 1, 1), '✔☃☃♥', 'it::substr_replace for utf-8');
+
 ?>