]>
Commit | Line | Data |
---|---|---|
1 | <?php | |
2 | /** | |
3 | * Converts an array-represented URL to a string | |
4 | * | |
5 | * Source: http://php.net/manual/en/function.parse-url.php#106731 | |
6 | * | |
7 | * @see http://php.net/manual/en/function.parse-url.php | |
8 | * | |
9 | * @param array $parsedUrl an array-represented URL | |
10 | * | |
11 | * @return string the string representation of the URL | |
12 | */ | |
13 | function unparse_url($parsedUrl) | |
14 | { | |
15 | $scheme = isset($parsedUrl['scheme']) ? $parsedUrl['scheme'].'://' : ''; | |
16 | $host = isset($parsedUrl['host']) ? $parsedUrl['host'] : ''; | |
17 | $port = isset($parsedUrl['port']) ? ':'.$parsedUrl['port'] : ''; | |
18 | $user = isset($parsedUrl['user']) ? $parsedUrl['user'] : ''; | |
19 | $pass = isset($parsedUrl['pass']) ? ':'.$parsedUrl['pass'] : ''; | |
20 | $pass = ($user || $pass) ? "$pass@" : ''; | |
21 | $path = isset($parsedUrl['path']) ? $parsedUrl['path'] : ''; | |
22 | $query = isset($parsedUrl['query']) ? '?'.$parsedUrl['query'] : ''; | |
23 | $fragment = isset($parsedUrl['fragment']) ? '#'.$parsedUrl['fragment'] : ''; | |
24 | ||
25 | return "$scheme$user$pass$host$port$path$query$fragment"; | |
26 | } | |
27 | ||
28 | /** | |
29 | * Removes undesired query parameters and fragments | |
30 | * | |
31 | * @param string url Url to be cleaned | |
32 | * | |
33 | * @return string the string representation of this URL after cleanup | |
34 | */ | |
35 | function cleanup_url($url) | |
36 | { | |
37 | $obj_url = new Url($url); | |
38 | return $obj_url->cleanup(); | |
39 | } | |
40 | ||
41 | /** | |
42 | * Get URL scheme. | |
43 | * | |
44 | * @param string url Url for which the scheme is requested | |
45 | * | |
46 | * @return mixed the URL scheme or false if none is provided. | |
47 | */ | |
48 | function get_url_scheme($url) | |
49 | { | |
50 | $obj_url = new Url($url); | |
51 | return $obj_url->getScheme(); | |
52 | } | |
53 | ||
54 | /** | |
55 | * Adds a trailing slash at the end of URL if necessary. | |
56 | * | |
57 | * @param string $url URL to check/edit. | |
58 | * | |
59 | * @return string $url URL with a end trailing slash. | |
60 | */ | |
61 | function add_trailing_slash($url) | |
62 | { | |
63 | return $url . (!endsWith($url, '/') ? '/' : ''); | |
64 | } | |
65 | /** | |
66 | * Converts an URL with an IDN host to a ASCII one. | |
67 | * | |
68 | * @param string $url Input URL. | |
69 | * | |
70 | * @return string converted URL. | |
71 | */ | |
72 | function url_with_idn_to_ascii($url) | |
73 | { | |
74 | $parts = parse_url($url); | |
75 | $parts['host'] = idn_to_ascii($parts['host']); | |
76 | ||
77 | $httpUrl = new \http\Url($parts); | |
78 | return $httpUrl->toString(); | |
79 | } | |
80 | /** | |
81 | * URL representation and cleanup utilities | |
82 | * | |
83 | * Form | |
84 | * scheme://[username:password@]host[:port][/path][?query][#fragment] | |
85 | * | |
86 | * Examples | |
87 | * http://username:password@hostname:9090/path?arg1=value1&arg2=value2#anchor | |
88 | * https://host.name.tld | |
89 | * https://h2.g2/faq/?vendor=hitchhiker&item=guide&dest=galaxy#answer | |
90 | * | |
91 | * @see http://www.faqs.org/rfcs/rfc3986.html | |
92 | */ | |
93 | class Url | |
94 | { | |
95 | private static $annoyingQueryParams = array( | |
96 | ||
97 | 'action_object_map=', | |
98 | 'action_ref_map=', | |
99 | 'action_type_map=', | |
100 | 'fb_', | |
101 | 'fb=', | |
102 | 'PHPSESSID=', | |
103 | ||
104 | // Scoop.it | |
105 | '__scoop', | |
106 | ||
107 | // Google Analytics & FeedProxy | |
108 | 'utm_', | |
109 | ||
110 | // ATInternet | |
111 | 'xtor=' | |
112 | ); | |
113 | ||
114 | private static $annoyingFragments = array( | |
115 | // ATInternet | |
116 | 'xtor=RSS-', | |
117 | ||
118 | // Misc. | |
119 | 'tk.rss_all' | |
120 | ); | |
121 | ||
122 | /* | |
123 | * URL parts represented as an array | |
124 | * | |
125 | * @see http://php.net/parse_url | |
126 | */ | |
127 | protected $parts; | |
128 | ||
129 | /** | |
130 | * Parses a string containing a URL | |
131 | * | |
132 | * @param string $url a string containing a URL | |
133 | */ | |
134 | public function __construct($url) | |
135 | { | |
136 | $url = self::cleanupUnparsedUrl(trim($url)); | |
137 | $this->parts = parse_url($url); | |
138 | ||
139 | if (!empty($url) && empty($this->parts['scheme'])) { | |
140 | $this->parts['scheme'] = 'http'; | |
141 | } | |
142 | } | |
143 | ||
144 | /** | |
145 | * Clean up URL before it's parsed. | |
146 | * ie. handle urlencode, url prefixes, etc. | |
147 | * | |
148 | * @param string $url URL to clean. | |
149 | * | |
150 | * @return string cleaned URL. | |
151 | */ | |
152 | protected static function cleanupUnparsedUrl($url) | |
153 | { | |
154 | return self::removeFirefoxAboutReader($url); | |
155 | } | |
156 | ||
157 | /** | |
158 | * Remove Firefox Reader prefix if it's present. | |
159 | * | |
160 | * @param string $input url | |
161 | * | |
162 | * @return string cleaned url | |
163 | */ | |
164 | protected static function removeFirefoxAboutReader($input) | |
165 | { | |
166 | $firefoxPrefix = 'about://reader?url='; | |
167 | if (startsWith($input, $firefoxPrefix)) { | |
168 | return urldecode(ltrim($input, $firefoxPrefix)); | |
169 | } | |
170 | return $input; | |
171 | } | |
172 | ||
173 | /** | |
174 | * Returns a string representation of this URL | |
175 | */ | |
176 | public function toString() | |
177 | { | |
178 | return unparse_url($this->parts); | |
179 | } | |
180 | ||
181 | /** | |
182 | * Removes undesired query parameters | |
183 | */ | |
184 | protected function cleanupQuery() | |
185 | { | |
186 | if (! isset($this->parts['query'])) { | |
187 | return; | |
188 | } | |
189 | ||
190 | $queryParams = explode('&', $this->parts['query']); | |
191 | ||
192 | foreach (self::$annoyingQueryParams as $annoying) { | |
193 | foreach ($queryParams as $param) { | |
194 | if (startsWith($param, $annoying)) { | |
195 | $queryParams = array_diff($queryParams, array($param)); | |
196 | continue; | |
197 | } | |
198 | } | |
199 | } | |
200 | ||
201 | if (count($queryParams) == 0) { | |
202 | unset($this->parts['query']); | |
203 | return; | |
204 | } | |
205 | ||
206 | $this->parts['query'] = implode('&', $queryParams); | |
207 | } | |
208 | ||
209 | /** | |
210 | * Removes undesired fragments | |
211 | */ | |
212 | protected function cleanupFragment() | |
213 | { | |
214 | if (! isset($this->parts['fragment'])) { | |
215 | return; | |
216 | } | |
217 | ||
218 | foreach (self::$annoyingFragments as $annoying) { | |
219 | if (startsWith($this->parts['fragment'], $annoying)) { | |
220 | unset($this->parts['fragment']); | |
221 | break; | |
222 | } | |
223 | } | |
224 | } | |
225 | ||
226 | /** | |
227 | * Removes undesired query parameters and fragments | |
228 | * | |
229 | * @return string the string representation of this URL after cleanup | |
230 | */ | |
231 | public function cleanup() | |
232 | { | |
233 | $this->cleanupQuery(); | |
234 | $this->cleanupFragment(); | |
235 | return $this->toString(); | |
236 | } | |
237 | ||
238 | /** | |
239 | * Converts an URL with an International Domain Name host to a ASCII one. | |
240 | * This requires PHP-intl. If it's not available, just returns this->cleanup(). | |
241 | * | |
242 | * @return string converted cleaned up URL. | |
243 | */ | |
244 | public function idnToAscii() | |
245 | { | |
246 | $out = $this->cleanup(); | |
247 | if (! function_exists('idn_to_ascii') || ! isset($this->parts['host'])) { | |
248 | return $out; | |
249 | } | |
250 | $asciiHost = idn_to_ascii($this->parts['host']); | |
251 | return str_replace($this->parts['host'], $asciiHost, $out); | |
252 | } | |
253 | ||
254 | /** | |
255 | * Get URL scheme. | |
256 | * | |
257 | * @return string the URL scheme or false if none is provided. | |
258 | */ | |
259 | public function getScheme() { | |
260 | if (!isset($this->parts['scheme'])) { | |
261 | return false; | |
262 | } | |
263 | return $this->parts['scheme']; | |
264 | } | |
265 | ||
266 | /** | |
267 | * Get URL host. | |
268 | * | |
269 | * @return string the URL host or false if none is provided. | |
270 | */ | |
271 | public function getHost() { | |
272 | if (empty($this->parts['host'])) { | |
273 | return false; | |
274 | } | |
275 | return $this->parts['host']; | |
276 | } | |
277 | ||
278 | /** | |
279 | * Test if the Url is an HTTP one. | |
280 | * | |
281 | * @return true is HTTP, false otherwise. | |
282 | */ | |
283 | public function isHttp() { | |
284 | return strpos(strtolower($this->parts['scheme']), 'http') !== false; | |
285 | } | |
286 | } |