Adds mb_parse_url, deprecates Parser::mbParseUrl

Author: jeremykendall

composer.json

@@ -29,6 +29,9 @@
     "autoload": {
         "psr-0": {
             "Pdp\\": "library/"
-        }
+        },
+        "files": [
+            "library/mb-parse-url.php"
+        ]
     }
 }

library/Pdp/Parser.php

@@ -61,7 +61,11 @@ public function parseUrl($url)
             $url = 'http://' . preg_replace('#^//#', '', $url, 1);
         }
 
-        $parts = $this->mbParseUrl($url);
+        $parts = mb_parse_url($url);
+
+        if ($parts === false) {
+            throw new \InvalidArgumentException(sprintf('Invalid url %s', $url));
+        }
 
         $elem = (array) $parts + $elem;
 
@@ -230,10 +234,13 @@ public function getSubdomain($host)
     }
 
     /**
-     * UTF-8 aware parse_url() replacement. 
+     * DEPRECATED: UTF-8 aware parse_url() replacement.
      *
      * Taken from php.net manual comments {@link http://php.net/manual/en/function.parse-url.php#114817}
      *
+     * @deprecated Please use mb_parse_url instead. Will be removed in version 2.0.
+     * @codeCoverageIgnore
+     *
      * @param  string  $url       The URL to parse
      * @param  integer $component Specify one of PHP_URL_SCHEME, PHP_URL_HOST,
      *                            PHP_URL_PORT, PHP_URL_USER, PHP_URL_PASS, PHP_URL_PATH, PHP_URL_QUERY or
@@ -255,7 +262,7 @@ function ($matches) {
         $parts = parse_url($enc_url, $component);
 
         if ($parts === false) {
-            throw new \InvalidArgumentException(sprintf('Invalid url %s', $url));
+            return $parts;
         }
 
         if (is_array($parts)) {

library/mb-parse-url.php

@@ -0,0 +1,45 @@
+<?php
+
+namespace {
+    if (!function_exists('mb_parse_url')) {
+        /**
+         * UTF-8 aware parse_url() replacement.
+         *
+         * Taken from php.net manual comments {@link http://php.net/manual/en/function.parse-url.php#114817}
+         *
+         * @param  string  $url       The URL to parse
+         * @param  integer $component Specify one of PHP_URL_SCHEME, PHP_URL_HOST,
+         *                            PHP_URL_PORT, PHP_URL_USER, PHP_URL_PASS, PHP_URL_PATH, PHP_URL_QUERY or
+         *                            PHP_URL_FRAGMENT to retrieve just a specific URL component as a string
+         *                            (except when PHP_URL_PORT is given, in which case the return value will
+         *                            be an integer).
+         * @return mixed   See parse_url documentation {@link http://us1.php.net/parse_url}
+         */
+        function mb_parse_url($url, $component = -1)
+        {
+            $enc_url = preg_replace_callback(
+                '%[^:/@?&=#]+%usD',
+                function ($matches) {
+                    return urlencode($matches[0]);
+                },
+                    $url
+                );
+
+            $parts = parse_url($enc_url, $component);
+
+            if ($parts === false) {
+                return $parts;
+            }
+
+            if (is_array($parts)) {
+                foreach ($parts as $name => $value) {
+                    $parts[$name] = urldecode($value);
+                }
+            } else {
+                $parts = urldecode($parts);
+            }
+
+            return $parts;
+        }
+    }
+}

tests/library/Pdp/ParserTest.php

@@ -23,7 +23,7 @@ protected function tearDown()
 
     /**
      * @covers Pdp\Parser::parseUrl()
-     * @covers Pdp\Parser::mbParseUrl()
+     * @covers ::mb_parse_url
      */
     public function testParseBadUrlThrowsInvalidArgumentException()
     {
@@ -98,12 +98,13 @@ public function testGetSubdomain($url, $publicSuffix, $registerableDomain, $subd
 
     /**
      * @dataProvider parseDataProvider
+     * @covers ::mb_parse_url
      */
-    public function testMbParseUrlCanReturnCorrectHost($url, $publicSuffix, $registerableDomain, $subdomain, $hostPart)
+    public function testmb_parse_urlCanReturnCorrectHost($url, $publicSuffix, $registerableDomain, $subdomain, $hostPart)
     {
         $this->assertEquals(
             $hostPart,
-            $this->parser->mbParseUrl('http://' . $hostPart, PHP_URL_HOST)
+            mb_parse_url('http://' . $hostPart, PHP_URL_HOST)
         );
     }