Complex classes like Data_and_files 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 Data_and_files, and based on these observations, apply Extract Interface, too.
1 | <?php |
||
14 | trait Data_and_files { |
||
15 | /** |
||
16 | * Data array, similar to `$_POST` |
||
17 | * |
||
18 | * @var array |
||
19 | */ |
||
20 | public $data; |
||
21 | /** |
||
22 | * Normalized files array |
||
23 | * |
||
24 | * Each file item can be either single file or array of files (in contrast with native PHP arrays where each field like `name` become an array) with keys |
||
25 | * `name`, `type`, `size`, `tmp_name`, `stream` and `error` |
||
26 | * |
||
27 | * `name`, `type`, `size` and `error` keys are similar to native PHP fields in `$_FILES`; `tmp_name` might not be temporary file, but file descriptor |
||
28 | * wrapper like `request-file:///file` instead and `stream` is resource like obtained with `fopen('/tmp/xyz', 'rb')` |
||
29 | * |
||
30 | * @var array[] |
||
31 | */ |
||
32 | public $files; |
||
33 | /** |
||
34 | * Data stream resource, similar to `fopen('php://input', 'rb')` |
||
35 | * |
||
36 | * Make sure you're controlling position in stream where you read something, if code in some other place might seek on this stream |
||
37 | * |
||
38 | * Stream is read-only |
||
39 | * |
||
40 | * @var null|resource |
||
41 | */ |
||
42 | public $data_stream; |
||
43 | /** |
||
44 | * `$this->init_server()` assumed to be called already |
||
45 | * |
||
46 | * @param array $data Typically `$_POST` |
||
47 | * @param array[] $files Typically `$_FILES`; might be like native PHP array `$_FILES` or normalized; each file item MUST contain keys |
||
48 | * `name`, `type`, `size`, `error` and at least one of `tmp_name` or `stream` |
||
49 | * @param null|resource|string $data_stream String, like `php://input` or resource, like `fopen('php://input', 'rb')` with request body, will be parsed for |
||
50 | * data and files if necessary |
||
51 | * @param bool $copy_stream Sometimes data stream can only being read once (like most of times with `php://input`), so it is necessary to |
||
52 | * copy it and store its contents for longer period of time |
||
53 | * |
||
54 | * @throws ExitException |
||
55 | */ |
||
56 | 26 | function init_data_and_files ($data = [], $files = [], $data_stream = null, $copy_stream = true) { |
|
57 | 26 | if (is_resource($this->data_stream)) { |
|
58 | fclose($this->data_stream); |
||
59 | } |
||
60 | 26 | $this->data = $data; |
|
61 | 26 | $this->files = $this->normalize_files($files); |
|
62 | 26 | $data_stream = is_string($data_stream) ? fopen($data_stream, 'rb') : $data_stream; |
|
63 | 26 | if ($copy_stream && is_resource($data_stream)) { |
|
64 | 24 | $this->data_stream = fopen('php://temp', 'w+b'); |
|
65 | 24 | stream_copy_to_stream($data_stream, $this->data_stream); |
|
66 | 24 | fclose($data_stream); |
|
67 | } else { |
||
68 | 2 | $this->data_stream = $data_stream; |
|
69 | } |
||
70 | /** |
||
71 | * If we don't appear to have any data or files detected - probably, we need to parse request ourselves |
||
72 | */ |
||
73 | 26 | if (!$this->data && !$this->files && is_resource($this->data_stream)) { |
|
74 | 20 | $this->parse_data_stream(); |
|
75 | } |
||
76 | // Hack: for compatibility we'll override $_POST since it might be filled during parsing |
||
77 | 26 | $_POST = $this->data; |
|
78 | 26 | } |
|
79 | /** |
||
80 | * Get data item by name |
||
81 | * |
||
82 | * @param string[]|string[][] $name |
||
83 | * |
||
84 | * @return mixed|mixed[]|null Data items (or associative array of data items) if exists or `null` otherwise (in case if `$name` is an array even one |
||
85 | * missing key will cause the whole thing to fail) |
||
86 | */ |
||
87 | 32 | function data (...$name) { |
|
88 | 32 | if (count($name) === 1) { |
|
89 | 28 | $name = $name[0]; |
|
90 | } |
||
91 | /** |
||
92 | * @var string|string[] $name |
||
93 | */ |
||
94 | 32 | if (is_array($name)) { |
|
95 | 6 | $result = []; |
|
96 | 6 | foreach ($name as &$n) { |
|
97 | 6 | if (!array_key_exists($n, $this->data)) { |
|
98 | 2 | return null; |
|
99 | } |
||
100 | 6 | $result[$n] = $this->data[$n]; |
|
101 | } |
||
102 | 6 | return $result; |
|
103 | } |
||
104 | /** @noinspection OffsetOperationsInspection */ |
||
105 | 28 | return @$this->data[$name]; |
|
106 | } |
||
107 | /** |
||
108 | * Get file item by name |
||
109 | * |
||
110 | * @param string $name |
||
111 | * |
||
112 | * @return array|null File item if exists or `null` otherwise |
||
113 | */ |
||
114 | 2 | function files ($name) { |
|
115 | 2 | return @$this->files[$name]; |
|
116 | } |
||
117 | /** |
||
118 | * @param array[] $files |
||
119 | * @param string $file_path |
||
120 | * |
||
121 | * @return array[] |
||
122 | */ |
||
123 | 26 | protected function normalize_files ($files, $file_path = '') { |
|
124 | 26 | if (!isset($files['name'])) { |
|
125 | 26 | foreach ($files as $field => &$file) { |
|
126 | 2 | $file = $this->normalize_files($file, "$file_path/$field"); |
|
127 | } |
||
128 | 26 | return $files; |
|
129 | } |
||
130 | 2 | if (is_array($files['name'])) { |
|
131 | 2 | $result = []; |
|
132 | 2 | foreach (array_keys($files['name']) as $index) { |
|
133 | 2 | $result[] = $this->normalize_file( |
|
134 | [ |
||
135 | 2 | 'name' => $files['name'][$index], |
|
136 | 2 | 'type' => $files['type'][$index], |
|
137 | 2 | 'size' => $files['size'][$index], |
|
138 | 2 | 'tmp_name' => @$files['tmp_name'][$index], |
|
139 | 2 | 'stream' => @$files['stream'][$index], |
|
140 | 2 | 'error' => $files['error'][$index] |
|
141 | ], |
||
142 | 2 | "$file_path/$index" |
|
143 | ); |
||
144 | } |
||
145 | 2 | return $result; |
|
146 | } else { |
||
147 | 2 | return $this->normalize_file($files, $file_path); |
|
148 | } |
||
149 | } |
||
150 | /** |
||
151 | * @param array $file |
||
152 | * @param string $file_path |
||
153 | * |
||
154 | * @return array |
||
155 | */ |
||
156 | 2 | protected function normalize_file ($file, $file_path) { |
|
157 | $file += [ |
||
158 | 2 | 'tmp_name' => null, |
|
159 | 'stream' => null |
||
160 | ]; |
||
161 | 2 | if (isset($file['tmp_name']) && $file['stream'] === null) { |
|
162 | 2 | $file['stream'] = fopen($file['tmp_name'], 'rb'); |
|
163 | } |
||
164 | 2 | if (isset($file['stream']) && !$file['tmp_name']) { |
|
165 | 2 | $file['tmp_name'] = "request-file://$file_path"; |
|
166 | } |
||
167 | 2 | if ($file['tmp_name'] === null && $file['stream'] === null) { |
|
168 | $file['error'] = UPLOAD_ERR_NO_FILE; |
||
169 | } |
||
170 | 2 | return $file; |
|
171 | } |
||
172 | /** |
||
173 | * Parsing request body for following Content-Type: `application/json`, `application/x-www-form-urlencoded` and `multipart/form-data` |
||
174 | * |
||
175 | * @throws ExitException |
||
176 | */ |
||
177 | 20 | protected function parse_data_stream () { |
|
207 | /** |
||
208 | * Parse content stream |
||
209 | * |
||
210 | * @param resource $stream |
||
211 | * @param string $boundary |
||
212 | * |
||
213 | * @return array[]|false |
||
214 | * |
||
215 | * @throws UnexpectedValueException |
||
216 | * @throws ExitException |
||
217 | */ |
||
218 | protected function parse_multipart_into_parts ($stream, $boundary) { |
||
219 | $parts = []; |
||
220 | $crlf = "\r\n"; |
||
221 | $position = 0; |
||
222 | $body = ''; |
||
223 | list($offset, $body) = $this->parse_multipart_find($stream, $body, "--$boundary$crlf"); |
||
224 | /** |
||
225 | * strlen doesn't take into account trailing CRLF since we'll need it in loop below |
||
226 | */ |
||
227 | $position += $offset + strlen("--$boundary"); |
||
228 | $body = substr($body, strlen("--$boundary")); |
||
229 | /** |
||
230 | * Each part always starts with CRLF |
||
231 | */ |
||
232 | while (strpos($body, $crlf) === 0) { |
||
233 | $position += 2; |
||
234 | $body = substr($body, 2); |
||
235 | $part = [ |
||
236 | 'headers' => [ |
||
237 | 'offset' => $position, |
||
238 | 'size' => 0 |
||
239 | ], |
||
240 | 'body' => [ |
||
241 | 'offset' => 0, |
||
242 | 'size' => 0 |
||
243 | ] |
||
244 | ]; |
||
245 | if (strpos($body, $crlf) === 0) { |
||
246 | /** |
||
247 | * No headers |
||
248 | */ |
||
249 | $position += 2; |
||
250 | $body = substr($body, 2); |
||
251 | } else { |
||
252 | /** |
||
253 | * Find headers end in order to determine size |
||
254 | */ |
||
255 | list($offset, $body) = $this->parse_multipart_find($stream, $body, $crlf.$crlf); |
||
256 | $part['headers']['size'] = $offset; |
||
257 | $position += $offset + 4; |
||
258 | $body = substr($body, 4); |
||
259 | } |
||
260 | $part['body']['offset'] = $position; |
||
261 | /** |
||
262 | * Find body end in order to determine its size |
||
263 | */ |
||
264 | list($offset, $body) = $this->parse_multipart_find($stream, $body, "$crlf--$boundary"); |
||
265 | $part['body']['size'] = $offset; |
||
266 | $position += $offset + strlen("$crlf--$boundary"); |
||
267 | $body = substr($body, strlen("$crlf--$boundary")); |
||
268 | if ($part['headers']['size']) { |
||
269 | $parts[] = $part; |
||
270 | } |
||
271 | } |
||
272 | /** |
||
273 | * Last boundary after all parts ends with '--' and we don't care what rubbish happens after it |
||
274 | */ |
||
275 | $post_max_size = $this->post_max_size(); |
||
276 | if (0 !== strpos($body, '--')) { |
||
277 | return false; |
||
278 | } |
||
279 | /** |
||
280 | * Check whether body size is bigger than allowed limit |
||
281 | */ |
||
282 | if ($position + strlen($body) > $post_max_size) { |
||
283 | throw new ExitException(413); |
||
284 | } |
||
285 | return $parts; |
||
286 | } |
||
287 | /** |
||
288 | * @param resource $stream |
||
289 | * @param array[] $parts |
||
290 | * |
||
291 | * @return array[] |
||
292 | */ |
||
293 | protected function parse_multipart_analyze_parts ($stream, $parts) { |
||
294 | $data = []; |
||
295 | $files = []; |
||
296 | foreach ($parts as $part) { |
||
297 | $headers = $this->parse_multipart_headers( |
||
298 | stream_get_contents($stream, $part['headers']['size'], $part['headers']['offset']) |
||
299 | ); |
||
300 | if ( |
||
301 | !isset($headers['content-disposition'][0], $headers['content-disposition']['name']) || |
||
302 | $headers['content-disposition'][0] != 'form-data' |
||
303 | ) { |
||
304 | continue; |
||
305 | } |
||
306 | $name = $headers['content-disposition']['name']; |
||
307 | if (isset($headers['content-disposition']['filename'])) { |
||
308 | $file = [ |
||
309 | 'name' => $headers['content-disposition']['filename'], |
||
310 | 'type' => @$headers['content-type'] ?: 'application/octet-stream', |
||
311 | 'size' => $part['body']['size'], |
||
312 | 'stream' => Stream_slicer::slice($stream, $part['body']['offset'], $part['body']['size']), |
||
313 | 'error' => UPLOAD_ERR_OK |
||
314 | ]; |
||
315 | if ($headers['content-disposition']['filename'] === '') { |
||
316 | $file['type'] = ''; |
||
317 | $file['stream'] = null; |
||
318 | $file['error'] = UPLOAD_ERR_NO_FILE; |
||
319 | } elseif ($file['size'] > $this->upload_max_file_size()) { |
||
320 | $file['stream'] = null; |
||
321 | $file['error'] = UPLOAD_ERR_INI_SIZE; |
||
322 | } |
||
323 | $this->parse_multipart_set_target($files, $name, $file); |
||
324 | } else { |
||
325 | if ($part['body']['size'] == 0) { |
||
326 | $this->parse_multipart_set_target($data, $name, ''); |
||
327 | } else { |
||
328 | $this->parse_multipart_set_target( |
||
329 | $data, |
||
330 | $name, |
||
331 | stream_get_contents($stream, $part['body']['size'], $part['body']['offset']) |
||
332 | ); |
||
333 | } |
||
334 | } |
||
335 | } |
||
336 | return [$data, $files]; |
||
337 | } |
||
338 | /** |
||
339 | * @return int |
||
340 | */ |
||
341 | protected function post_max_size () { |
||
342 | $size = ini_get('post_max_size') ?: ini_get('hhvm.server.max_post_size'); |
||
343 | return $this->convert_size_to_bytes($size); |
||
344 | } |
||
345 | /** |
||
346 | * @return int |
||
347 | */ |
||
348 | protected function upload_max_file_size () { |
||
349 | $size = ini_get('upload_max_filesize') ?: ini_get('hhvm.server.upload.upload_max_file_size'); |
||
350 | return $this->convert_size_to_bytes($size); |
||
351 | } |
||
352 | /** |
||
353 | * @param int|string $size |
||
354 | * |
||
355 | * @return int |
||
356 | */ |
||
357 | protected function convert_size_to_bytes ($size) { |
||
358 | switch (strtolower(substr($size, -1))) { |
||
359 | case 'g'; |
||
360 | $size = (int)$size * 1024; |
||
361 | case 'm'; |
||
362 | $size = (int)$size * 1024; |
||
363 | case 'k'; |
||
364 | $size = (int)$size * 1024; |
||
365 | } |
||
366 | return (int)$size ?: PHP_INT_MAX; |
||
367 | } |
||
368 | /** |
||
369 | * @param resource $stream |
||
370 | * @param string $next_data |
||
371 | * @param string $target |
||
372 | * |
||
373 | * @return array |
||
374 | * |
||
375 | * @throws UnexpectedValueException |
||
376 | */ |
||
377 | protected function parse_multipart_find ($stream, $next_data, $target) { |
||
378 | $offset = 0; |
||
379 | $prev_data = ''; |
||
380 | while (($found = strpos($prev_data.$next_data, $target)) === false) { |
||
381 | if (feof($stream)) { |
||
382 | throw new UnexpectedValueException; |
||
383 | } |
||
384 | if ($prev_data) { |
||
385 | $offset += strlen($prev_data); |
||
386 | } |
||
387 | $prev_data = $next_data; |
||
388 | $next_data = fread($stream, 1024); |
||
389 | } |
||
390 | $offset += $found; |
||
391 | /** |
||
392 | * Read some more bytes so that we'll always have some remainder in place, since empty remainder might cause problems with `strpos()` call later |
||
393 | */ |
||
394 | $remainder = substr($prev_data.$next_data, $found).(fread($stream, 1024) ?: ''); |
||
395 | return [$offset, $remainder]; |
||
396 | } |
||
397 | /** |
||
398 | * @param string $content |
||
399 | * |
||
400 | * @return array |
||
401 | */ |
||
402 | protected function parse_multipart_headers ($content) { |
||
403 | $headers = []; |
||
404 | foreach (explode("\r\n", $content) as $header) { |
||
405 | list($name, $value) = explode(':', $header, 2); |
||
406 | if (!preg_match_all('/(.+)(?:="*?(.*)"?)?(?:;\s|$)/U', $value, $matches)) { |
||
407 | continue; |
||
408 | } |
||
409 | $name = strtolower($name); |
||
410 | $headers[$name] = []; |
||
411 | foreach (array_keys($matches[1]) as $index) { |
||
412 | if (isset($headers[$name][0]) || strlen($matches[2][$index])) { |
||
413 | $headers[$name][trim($matches[1][$index])] = urldecode(trim($matches[2][$index])); |
||
414 | } else { |
||
415 | $headers[$name][] = trim($matches[1][$index]); |
||
416 | } |
||
417 | } |
||
418 | if (count($headers[$name]) == 1) { |
||
419 | $headers[$name] = $headers[$name][0]; |
||
420 | } |
||
421 | } |
||
422 | return $headers; |
||
423 | } |
||
424 | /** |
||
425 | * @param array $source |
||
426 | * @param string $name |
||
427 | * @param array|string $value |
||
428 | */ |
||
429 | protected function parse_multipart_set_target (&$source, $name, $value) { |
||
446 | } |
||
447 |