Complex classes like URI often do a lot of different things. To break such a class down, we need to identify a cohesive component within that class. A common approach to find such a component is to look for fields/methods that share the same prefixes, or suffixes. You can also have a look at the cohesion graph to spot any un-connected, or weakly-connected components.
Once you have determined the fields that belong together, you can apply the Extract Class refactoring. If the component makes sense as a sub-class, Extract Subclass is also a candidate, and is often faster.
While breaking up the class, it is a good idea to analyze how other classes use URI, and based on these observations, apply Extract Interface, too.
1 | <?php |
||
4 | class URI { |
||
5 | |||
6 | /** |
||
7 | * List of cached uri segments |
||
8 | * |
||
9 | * @var array |
||
10 | * @access public |
||
11 | * |
||
12 | */ |
||
13 | var $keyval=array (); |
||
14 | /** |
||
15 | * Current uri string |
||
16 | * |
||
17 | * @var string |
||
18 | * @access public |
||
19 | * |
||
20 | */ |
||
21 | var $uri_string; |
||
22 | /** |
||
23 | * List of uri segments |
||
24 | * |
||
25 | * @var array |
||
26 | * @access public |
||
27 | * |
||
28 | */ |
||
29 | var $segments=array (); |
||
30 | /** |
||
31 | * Re-indexed list of uri segments |
||
32 | * Starts at 1 instead of 0 |
||
33 | * |
||
34 | * @var array |
||
35 | * @access public |
||
36 | * |
||
37 | */ |
||
38 | var $rsegments=array (); |
||
39 | var $permitted_uri_chars="a-z 0-9~%.:_\-"; |
||
40 | var $enable_query_strings=false; |
||
41 | |||
42 | /** |
||
43 | * | 'AUTO' Default - auto detects |
||
44 | * | 'PATH_INFO' Uses the PATH_INFO |
||
45 | * | 'QUERY_STRING' Uses the QUERY_STRING |
||
46 | * | 'REQUEST_URI' Uses the REQUEST_URI |
||
47 | * | 'ORIG_PATH_INFO' Uses the ORIG_PATH_INFO |
||
48 | */ |
||
49 | var $uri_protocol="AUTO"; |
||
50 | // |
||
51 | |||
52 | /** |
||
53 | * Constructor |
||
54 | * |
||
55 | * Simply globalizes the $RTR object. The front |
||
56 | * loads the Router class early on so it's not available |
||
57 | * normally as other classes are. |
||
58 | * |
||
59 | * @access public |
||
60 | * |
||
61 | */ |
||
62 | function __construct() { |
||
63 | $this->uri_string=$this->_detect_uri(); |
||
64 | $this->_explode_segments(); |
||
65 | } |
||
66 | |||
67 | // -------------------------------------------------------------------- |
||
68 | |||
69 | /** |
||
70 | * Get the URI String |
||
71 | * |
||
72 | * @access private |
||
73 | * @return string |
||
74 | * |
||
75 | */ |
||
76 | function _fetch_uri_string() { |
||
77 | $uri=""; |
||
78 | if ($this->uri_protocol == 'AUTO') { |
||
79 | // Is the request coming from the command line? |
||
80 | if (php_sapi_name() == 'cli' or defined('STDIN')) { |
||
81 | $this->_set_uri_string($this->_parse_cli_args()); |
||
82 | return; |
||
83 | } |
||
84 | |||
85 | // Let's try the REQUEST_URI first, this will work in most situations |
||
86 | if ($uri=$this->_detect_uri()) { |
||
87 | $this->_set_uri_string($uri); |
||
88 | return; |
||
89 | } |
||
90 | |||
91 | // Is there a PATH_INFO variable? |
||
92 | // Note: some servers seem to have trouble with getenv() so we'll test it two ways |
||
93 | $path=(isset($_SERVER['PATH_INFO'])) ? $_SERVER['PATH_INFO'] : @getenv('PATH_INFO'); |
||
94 | if (trim($path, '/') != '' && $path != "/" . SELF) { |
||
95 | $this->_set_uri_string($path); |
||
96 | return; |
||
97 | } |
||
98 | |||
99 | // No PATH_INFO?... What about QUERY_STRING? |
||
100 | $path=(isset($_SERVER['QUERY_STRING'])) ? $_SERVER['QUERY_STRING'] : @getenv('QUERY_STRING'); |
||
101 | if (trim($path, '/') != '') { |
||
102 | $this->_set_uri_string($path); |
||
103 | return; |
||
104 | } |
||
105 | |||
106 | // As a last ditch effort lets try using the $_GET array |
||
107 | if (is_array($_GET) && count($_GET) == 1 && trim(key($_GET), '/') != '') { |
||
108 | $this->_set_uri_string(key($_GET)); |
||
109 | return; |
||
110 | } |
||
111 | |||
112 | // We've exhausted all our options... |
||
113 | $this->uri_string=''; |
||
114 | return; |
||
115 | } |
||
116 | |||
117 | if ($uri == 'REQUEST_URI') { |
||
118 | $this->_set_uri_string($this->_detect_uri()); |
||
119 | return; |
||
120 | } |
||
121 | |||
122 | $path=(isset($_SERVER[$uri])) ? $_SERVER[$uri] : @getenv($uri); |
||
123 | $this->_set_uri_string($path); |
||
124 | } |
||
125 | |||
126 | // -------------------------------------------------------------------- |
||
127 | |||
128 | /** |
||
129 | * Set the URI String |
||
130 | * |
||
131 | * @access public |
||
132 | * @param string |
||
133 | * @return string |
||
134 | * |
||
135 | */ |
||
136 | function _set_uri_string($str) { |
||
137 | // Filter out control characters |
||
138 | // $str = remove_invisible_characters($str, FALSE); |
||
139 | |||
140 | // If the URI contains only a slash we'll kill it |
||
141 | $this->uri_string=($str == '/') ? '' : $str; |
||
142 | } |
||
143 | |||
144 | // -------------------------------------------------------------------- |
||
145 | |||
146 | /** |
||
147 | * Detects the URI |
||
148 | * |
||
149 | * This function will detect the URI automatically and fix the query string |
||
150 | * if necessary. |
||
151 | * |
||
152 | * @access private |
||
153 | * @return string |
||
154 | * |
||
155 | */ |
||
156 | private function _detect_uri() { |
||
157 | if (!isset($_SERVER['REQUEST_URI']) or !isset($_SERVER['SCRIPT_NAME'])) { |
||
158 | return ''; |
||
159 | } |
||
160 | |||
161 | $uri=$_SERVER['REQUEST_URI']; |
||
162 | if (strpos($uri, $_SERVER['SCRIPT_NAME']) === 0) { |
||
163 | $uri=substr($uri, strlen($_SERVER['SCRIPT_NAME'])); |
||
164 | } elseif (strpos($uri, dirname($_SERVER['SCRIPT_NAME'])) === 0) { |
||
165 | $uri=substr($uri, strlen(dirname($_SERVER['SCRIPT_NAME']))); |
||
166 | } |
||
167 | |||
168 | // This section ensures that even on servers that require the URI to be in the query string (Nginx) a correct |
||
169 | // URI is found, and also fixes the QUERY_STRING server var and $_GET array. |
||
170 | if (strncmp($uri, '?/', 2) === 0) { |
||
171 | $uri=substr($uri, 2); |
||
172 | } |
||
173 | $parts=preg_split('#\?#i', $uri, 2); |
||
174 | $uri=$parts[0]; |
||
175 | if (isset($parts[1])) { |
||
176 | $_SERVER['QUERY_STRING']=$parts[1]; |
||
177 | parse_str($_SERVER['QUERY_STRING'], $_GET); |
||
178 | } else { |
||
179 | $_SERVER['QUERY_STRING']=''; |
||
180 | $_GET=array (); |
||
181 | } |
||
182 | |||
183 | if ($uri == '/' || empty($uri)) { |
||
184 | return '/'; |
||
185 | } |
||
186 | |||
187 | $uri=parse_url($uri, PHP_URL_PATH); |
||
188 | |||
189 | // Do some final cleaning of the URI and return it |
||
190 | return str_replace(array ('//','../' ), '/', trim($uri, '/')); |
||
191 | } |
||
192 | |||
193 | // -------------------------------------------------------------------- |
||
194 | |||
195 | /** |
||
196 | * Parse cli arguments |
||
197 | * |
||
198 | * Take each command line argument and assume it is a URI segment. |
||
199 | * |
||
200 | * @access private |
||
201 | * @return string |
||
202 | * |
||
203 | */ |
||
204 | private function _parse_cli_args() { |
||
205 | $args=array_slice($_SERVER['argv'], 1); |
||
206 | |||
207 | return $args ? '/' . implode('/', $args) : ''; |
||
208 | } |
||
209 | |||
210 | // -------------------------------------------------------------------- |
||
211 | |||
212 | /** |
||
213 | * Filter segments for malicious characters |
||
214 | * |
||
215 | * @access private |
||
216 | * @param string |
||
217 | * @return string |
||
218 | * |
||
219 | */ |
||
220 | function _filter_uri($str) { |
||
221 | if ($str != '' && $this->permitted_uri_chars != '' && $this->enable_query_strings == FALSE) { |
||
222 | // preg_quote() in PHP 5.3 escapes -, so the str_replace() and addition of - to preg_quote() is to maintain backwards |
||
223 | // compatibility as many are unaware of how characters in the permitted_uri_chars will be parsed as a regex pattern |
||
224 | if (!preg_match("|^[" . str_replace(array ('\\-','\-' ), '-', preg_quote($this->permitted_uri_chars, '-')) . "]+$|i", $str)) { |
||
225 | show_error('The URI you submitted has disallowed characters.', 400); |
||
226 | } |
||
227 | } |
||
228 | |||
229 | // Convert programatic characters to entities |
||
230 | $bad=array ('$','(',')','%28','%29' ); |
||
231 | $good=array ('$','(',')','(',')' ); |
||
232 | |||
233 | return str_replace($bad, $good, $str); |
||
234 | } |
||
235 | |||
236 | /** |
||
237 | * Explode the URI Segments. |
||
238 | * The individual segments will |
||
239 | * be stored in the $this->segments array. |
||
240 | * |
||
241 | * @access private |
||
242 | * @return void |
||
243 | * |
||
244 | */ |
||
245 | function _explode_segments() { |
||
246 | foreach ( explode("/", preg_replace("|/*(.+?)/*$|", "\\1", $this->uri_string)) as $val ) { |
||
247 | // Filter segments for security |
||
248 | $val=trim($this->_filter_uri($val)); |
||
249 | |||
250 | if ($val != '') { |
||
251 | $this->segments[]=$val; |
||
252 | } |
||
253 | } |
||
254 | } |
||
255 | |||
256 | // -------------------------------------------------------------------- |
||
257 | /** |
||
258 | * Re-index Segments |
||
259 | * |
||
260 | * This function re-indexes the $this->segment array so that it |
||
261 | * starts at 1 rather than 0. Doing so makes it simpler to |
||
262 | * use functions like $this->uri->segment(n) since there is |
||
263 | * a 1:1 relationship between the segment array and the actual segments. |
||
264 | * |
||
265 | * @access private |
||
266 | * @return void |
||
267 | * |
||
268 | */ |
||
269 | function _reindex_segments() { |
||
270 | array_unshift($this->segments, NULL); |
||
271 | array_unshift($this->rsegments, NULL); |
||
272 | unset($this->segments[0]); |
||
273 | unset($this->rsegments[0]); |
||
274 | } |
||
275 | |||
276 | // -------------------------------------------------------------------- |
||
277 | |||
278 | /** |
||
279 | * Fetch a URI Segment |
||
280 | * |
||
281 | * This function returns the URI segment based on the number provided. |
||
282 | * |
||
283 | * @access public |
||
284 | * @param integer |
||
285 | * @param bool |
||
286 | * @return string |
||
287 | * |
||
288 | */ |
||
289 | function segment($n, $no_result=FALSE) { |
||
290 | return (!isset($this->segments[$n])) ? $no_result : $this->segments[$n]; |
||
291 | } |
||
292 | |||
293 | // -------------------------------------------------------------------- |
||
294 | |||
295 | /** |
||
296 | * Fetch a URI "routed" Segment |
||
297 | * |
||
298 | * This function returns the re-routed URI segment (assuming routing rules are used) |
||
299 | * based on the number provided. If there is no routing this function returns the |
||
300 | * same result as $this->segment() |
||
301 | * |
||
302 | * @access public |
||
303 | * @param integer |
||
304 | * @param bool |
||
305 | * @return string |
||
306 | * |
||
307 | */ |
||
308 | function rsegment($n, $no_result=FALSE) { |
||
309 | return (!isset($this->rsegments[$n])) ? $no_result : $this->rsegments[$n]; |
||
310 | } |
||
311 | |||
312 | // -------------------------------------------------------------------- |
||
313 | |||
314 | /** |
||
315 | * Generate a key value pair from the URI string |
||
316 | * |
||
317 | * This function generates and associative array of URI data starting |
||
318 | * at the supplied segment. For example, if this is your URI: |
||
319 | * |
||
320 | * example.com/user/search/name/joe/location/UK/gender/male |
||
321 | * |
||
322 | * You can use this function to generate an array with this prototype: |
||
323 | * |
||
324 | * array ( |
||
325 | * name => joe |
||
326 | * location => UK |
||
327 | * gender => male |
||
328 | * ) |
||
329 | * |
||
330 | * @access public |
||
331 | * @param integer the starting segment number |
||
332 | * @param array an array of default values |
||
333 | * @return array |
||
334 | * |
||
335 | */ |
||
336 | function uri_to_assoc($n=3, $default=array()) { |
||
337 | return $this->_uri_to_assoc($n, $default, 'segment'); |
||
338 | } |
||
339 | |||
340 | /** |
||
341 | * Identical to above only it uses the re-routed segment array |
||
342 | * |
||
343 | * @access public |
||
344 | * @param integer the starting segment number |
||
345 | * @param array an array of default values |
||
346 | * @return array |
||
347 | * |
||
348 | * |
||
349 | */ |
||
350 | function ruri_to_assoc($n=3, $default=array()) { |
||
351 | return $this->_uri_to_assoc($n, $default, 'rsegment'); |
||
352 | } |
||
353 | |||
354 | // -------------------------------------------------------------------- |
||
355 | |||
356 | /** |
||
357 | * Generate a key value pair from the URI string or Re-routed URI string |
||
358 | * |
||
359 | * @access private |
||
360 | * @param integer the starting segment number |
||
361 | * @param array an array of default values |
||
362 | * @param string which array we should use |
||
363 | * @return array |
||
364 | * |
||
365 | */ |
||
366 | function _uri_to_assoc($n=3, $default=array(), $which='segment') { |
||
367 | if ($which == 'segment') { |
||
368 | $total_segments='total_segments'; |
||
369 | $segment_array='segment_array'; |
||
370 | } else { |
||
371 | $total_segments='total_rsegments'; |
||
372 | $segment_array='rsegment_array'; |
||
373 | } |
||
374 | |||
375 | if (!is_numeric($n)) { |
||
376 | return $default; |
||
377 | } |
||
378 | |||
379 | if (isset($this->keyval[$n])) { |
||
380 | return $this->keyval[$n]; |
||
381 | } |
||
382 | |||
383 | if ($this->$total_segments() < $n) { |
||
384 | if (count($default) == 0) { |
||
385 | return array (); |
||
386 | } |
||
387 | |||
388 | $retval=array (); |
||
389 | foreach ( $default as $val ) { |
||
390 | $retval[$val]=FALSE; |
||
391 | } |
||
392 | return $retval; |
||
393 | } |
||
394 | |||
395 | $segments=array_slice($this->$segment_array(), ($n - 1)); |
||
396 | |||
397 | $i=0; |
||
398 | $lastval=''; |
||
399 | $retval=array (); |
||
400 | foreach ( $segments as $seg ) { |
||
401 | if ($i % 2) { |
||
402 | $retval[$lastval]=$seg; |
||
403 | } else { |
||
404 | $retval[$seg]=FALSE; |
||
405 | $lastval=$seg; |
||
406 | } |
||
407 | |||
408 | $i++; |
||
409 | } |
||
410 | |||
411 | if (count($default) > 0) { |
||
412 | foreach ( $default as $val ) { |
||
413 | if (!array_key_exists($val, $retval)) { |
||
414 | $retval[$val]=FALSE; |
||
415 | } |
||
416 | } |
||
417 | } |
||
418 | |||
419 | // Cache the array for reuse |
||
420 | $this->keyval[$n]=$retval; |
||
421 | return $retval; |
||
422 | } |
||
423 | |||
424 | // -------------------------------------------------------------------- |
||
425 | |||
426 | /** |
||
427 | * Generate a URI string from an associative array |
||
428 | * |
||
429 | * |
||
430 | * @access public |
||
431 | * @param array an associative array of key/values |
||
432 | * @return array |
||
433 | * |
||
434 | */ |
||
435 | function assoc_to_uri($array) { |
||
436 | $temp=array (); |
||
437 | foreach ( ( array ) $array as $key => $val ) { |
||
438 | $temp[]=$key; |
||
439 | $temp[]=$val; |
||
440 | } |
||
441 | |||
442 | return implode('/', $temp); |
||
443 | } |
||
444 | |||
445 | // -------------------------------------------------------------------- |
||
446 | |||
447 | /** |
||
448 | * Fetch a URI Segment and add a trailing slash |
||
449 | * |
||
450 | * @access public |
||
451 | * @param integer |
||
452 | * @param string |
||
453 | * @return string |
||
454 | * |
||
455 | */ |
||
456 | function slash_segment($n, $where='trailing') { |
||
457 | return $this->_slash_segment($n, $where, 'segment'); |
||
458 | } |
||
459 | |||
460 | // -------------------------------------------------------------------- |
||
461 | |||
462 | /** |
||
463 | * Fetch a URI Segment and add a trailing slash |
||
464 | * |
||
465 | * @access public |
||
466 | * @param integer |
||
467 | * @param string |
||
468 | * @return string |
||
469 | * |
||
470 | */ |
||
471 | function slash_rsegment($n, $where='trailing') { |
||
472 | return $this->_slash_segment($n, $where, 'rsegment'); |
||
473 | } |
||
474 | |||
475 | // -------------------------------------------------------------------- |
||
476 | |||
477 | /** |
||
478 | * Fetch a URI Segment and add a trailing slash - helper function |
||
479 | * |
||
480 | * @access private |
||
481 | * @param integer |
||
482 | * @param string |
||
483 | * @param string |
||
484 | * @return string |
||
485 | * |
||
486 | */ |
||
487 | function _slash_segment($n, $where='trailing', $which='segment') { |
||
488 | $leading='/'; |
||
489 | $trailing='/'; |
||
490 | |||
491 | if ($where == 'trailing') { |
||
492 | $leading=''; |
||
493 | } elseif ($where == 'leading') { |
||
494 | $trailing=''; |
||
495 | } |
||
496 | |||
497 | return $leading . $this->$which($n) . $trailing; |
||
498 | } |
||
499 | |||
500 | // -------------------------------------------------------------------- |
||
501 | |||
502 | /** |
||
503 | * Segment Array |
||
504 | * |
||
505 | * @access public |
||
506 | * @return array |
||
507 | * |
||
508 | */ |
||
509 | function segment_array() { |
||
510 | return $this->segments; |
||
511 | } |
||
512 | |||
513 | // -------------------------------------------------------------------- |
||
514 | |||
515 | /** |
||
516 | * Routed Segment Array |
||
517 | * |
||
518 | * @access public |
||
519 | * @return array |
||
520 | * |
||
521 | */ |
||
522 | function rsegment_array() { |
||
523 | return $this->rsegments; |
||
524 | } |
||
525 | |||
526 | // -------------------------------------------------------------------- |
||
527 | |||
528 | /** |
||
529 | * Total number of segments |
||
530 | * |
||
531 | * @access public |
||
532 | * @return integer |
||
533 | * |
||
534 | */ |
||
535 | function total_segments() { |
||
536 | return count($this->segments); |
||
537 | } |
||
538 | |||
539 | // -------------------------------------------------------------------- |
||
540 | |||
541 | /** |
||
542 | * Total number of routed segments |
||
543 | * |
||
544 | * @access public |
||
545 | * @return integer |
||
546 | * |
||
547 | */ |
||
548 | function total_rsegments() { |
||
549 | return count($this->rsegments); |
||
550 | } |
||
551 | |||
552 | // -------------------------------------------------------------------- |
||
553 | |||
554 | /** |
||
555 | * Fetch the entire URI string |
||
556 | * |
||
557 | * @access public |
||
558 | * @return string |
||
559 | * |
||
560 | */ |
||
561 | function uri_string() { |
||
562 | return $this->uri_string; |
||
563 | } |
||
564 | |||
565 | // -------------------------------------------------------------------- |
||
566 | |||
567 | /** |
||
568 | * Fetch the entire Re-routed URI string |
||
569 | * |
||
570 | * @access public |
||
571 | * @return string |
||
572 | * |
||
573 | */ |
||
574 | function ruri_string() { |
||
575 | return '/' . implode('/', $this->rsegment_array()); |
||
576 | } |
||
577 | } |
||
578 | // END URI Class |