Complex classes like vCard 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 vCard, and based on these observations, apply Extract Interface, too.
1 | <?php |
||
10 | class vCard implements Countable, Iterator |
||
11 | { |
||
12 | const MODE_ERROR = 'error'; |
||
13 | const MODE_SINGLE = 'single'; |
||
14 | const MODE_MULTIPLE = 'multiple'; |
||
15 | |||
16 | const endl = "\n"; |
||
17 | |||
18 | /** |
||
19 | * @var string Current object mode - error, single or multiple (for a single vCard within a file and multiple combined vCards) |
||
20 | */ |
||
21 | private $Mode; //single, multiple, error |
||
22 | |||
23 | private $Path = ''; |
||
24 | private $RawData = ''; |
||
25 | |||
26 | /** |
||
27 | * @var array Internal options container. Options: |
||
28 | * bool Collapse: If true, elements that can have multiple values but have only a single value are returned as that value instead of an array |
||
29 | * If false, an array is returned even if it has only one value. |
||
30 | */ |
||
31 | private $Options = array( |
||
32 | 'Collapse' => false |
||
33 | ); |
||
34 | |||
35 | /** |
||
36 | * @var array Internal data container. Contains vCard objects for multiple vCards and just the data for single vCards. |
||
37 | */ |
||
38 | private $Data = array(); |
||
39 | |||
40 | /** |
||
41 | * @static Parts of structured elements according to the spec. |
||
42 | */ |
||
43 | private static $Spec_StructuredElements = array( |
||
44 | 'n' => array('lastname', 'firstname', 'additionalnames', 'prefixes', 'suffixes'), |
||
45 | 'adr' => array('pobox', 'extendedaddress', 'streetaddress', 'locality', 'region', 'postalcode', 'country'), |
||
46 | 'geo' => array('latitude', 'longitude'), |
||
47 | 'org' => array('name', 'unit1', 'unit2') |
||
48 | ); |
||
49 | private static $Spec_MultipleValueElements = array('nickname', 'categories'); |
||
50 | |||
51 | private static $Spec_ElementTypes = array( |
||
52 | 'email' => array('internet', 'x400', 'pref', 'home', 'work'), |
||
53 | 'adr' => array('dom', 'intl', 'postal', 'parcel', 'home', 'work', 'pref'), |
||
54 | 'label' => array('dom', 'intl', 'postal', 'parcel', 'home', 'work', 'pref'), |
||
55 | 'tel' => array('home', 'msg', 'work', 'pref', 'voice', 'fax', 'cell', 'video', 'pager', 'bbs', 'modem', 'car', 'isdn', 'pcs'), |
||
56 | 'impp' => array('personal', 'business', 'home', 'work', 'mobile', 'pref') |
||
57 | ); |
||
58 | |||
59 | private static $Spec_FileElements = array('photo', 'logo', 'sound'); |
||
60 | |||
61 | /** |
||
62 | * vCard constructor |
||
63 | * |
||
64 | * @param string Path to file, optional. |
||
65 | * @param string Raw data, optional. |
||
66 | * @param array Additional options, optional. Currently supported options: |
||
67 | * bool Collapse: If true, elements that can have multiple values but have only a single value are returned as that value instead of an array |
||
68 | * If false, an array is returned even if it has only one value. |
||
69 | * |
||
70 | * One of these parameters must be provided, otherwise an exception is thrown. |
||
71 | */ |
||
72 | public function __construct($Path = false, $RawData = false, array $Options = null) |
||
73 | { |
||
74 | // Checking preconditions for the parser. |
||
75 | // If path is given, the file should be accessible. |
||
76 | // If raw data is given, it is taken as it is. |
||
77 | // In both cases the real content is put in $this -> RawData |
||
78 | if ($Path) |
||
79 | { |
||
80 | if (!is_readable($Path)) |
||
81 | { |
||
82 | throw new Exception('vCard: Path not accessible ('.$Path.')'); |
||
83 | } |
||
84 | |||
85 | $this -> Path = $Path; |
||
|
|||
86 | $this -> RawData = file_get_contents($this -> Path); |
||
87 | } |
||
88 | elseif ($RawData) |
||
89 | { |
||
90 | $this -> RawData = $RawData; |
||
91 | } |
||
92 | else |
||
93 | { |
||
94 | //throw new Exception('vCard: No content provided'); |
||
95 | // Not necessary anymore as possibility to create vCards is added |
||
96 | } |
||
97 | |||
98 | if (!$this -> Path && !$this -> RawData) |
||
99 | { |
||
100 | return true; |
||
101 | } |
||
102 | |||
103 | if ($Options) |
||
104 | { |
||
105 | $this -> Options = array_merge($this -> Options, $Options); |
||
106 | } |
||
107 | |||
108 | // Counting the begin/end separators. If there aren't any or the count doesn't match, there is a problem with the file. |
||
109 | // If there is only one, this is a single vCard, if more, multiple vCards are combined. |
||
110 | $Matches = array(); |
||
111 | $vCardBeginCount = preg_match_all('{^BEGIN\:VCARD}miS', $this -> RawData, $Matches); |
||
112 | $vCardEndCount = preg_match_all('{^END\:VCARD}miS', $this -> RawData, $Matches); |
||
113 | |||
114 | if (($vCardBeginCount != $vCardEndCount) || !$vCardBeginCount) |
||
115 | { |
||
116 | $this -> Mode = vCard::MODE_ERROR; |
||
117 | throw new Exception('vCard: invalid vCard'); |
||
118 | } |
||
119 | |||
120 | $this -> Mode = $vCardBeginCount == 1 ? vCard::MODE_SINGLE : vCard::MODE_MULTIPLE; |
||
121 | |||
122 | // Removing/changing inappropriate newlines, i.e., all CRs or multiple newlines are changed to a single newline |
||
123 | $this -> RawData = str_replace("\r", "\n", $this -> RawData); |
||
124 | $this -> RawData = preg_replace('{(\n+)}', "\n", $this -> RawData); |
||
125 | |||
126 | // In multiple card mode the raw text is split at card beginning markers and each |
||
127 | // fragment is parsed in a separate vCard object. |
||
128 | if ($this -> Mode == self::MODE_MULTIPLE) |
||
129 | { |
||
130 | $this -> RawData = explode('BEGIN:VCARD', $this -> RawData); |
||
131 | $this -> RawData = array_filter($this -> RawData); |
||
132 | |||
133 | foreach ($this -> RawData as $SinglevCardRawData) |
||
134 | { |
||
135 | // Prepending "BEGIN:VCARD" to the raw string because we exploded on that one. |
||
136 | // If there won't be the BEGIN marker in the new object, it will fail. |
||
137 | $SinglevCardRawData = 'BEGIN:VCARD'."\n".$SinglevCardRawData; |
||
138 | |||
139 | $ClassName = get_class($this); |
||
140 | $this -> Data[] = new $ClassName(false, $SinglevCardRawData); |
||
141 | } |
||
142 | } |
||
143 | else |
||
144 | { |
||
145 | // Protect the BASE64 final = sign (detected by the line beginning with whitespace), otherwise the next replace will get rid of it |
||
146 | $this -> RawData = preg_replace('{(\n\s.+)=(\n)}', '$1-base64=-$2', $this -> RawData); |
||
147 | |||
148 | // Joining multiple lines that are split with a hard wrap and indicated by an equals sign at the end of line |
||
149 | // (quoted-printable-encoded values in v2.1 vCards) |
||
150 | $this -> RawData = str_replace("=\n", '', $this -> RawData); |
||
151 | |||
152 | // Joining multiple lines that are split with a soft wrap (space or tab on the beginning of the next line |
||
153 | $this -> RawData = str_replace(array("\n ", "\n\t"), '-wrap-', $this -> RawData); |
||
154 | |||
155 | // Restoring the BASE64 final equals sign (see a few lines above) |
||
156 | $this -> RawData = str_replace("-base64=-\n", "=\n", $this -> RawData); |
||
157 | |||
158 | $Lines = explode("\n", $this -> RawData); |
||
159 | |||
160 | foreach ($Lines as $Line) |
||
161 | { |
||
162 | // Lines without colons are skipped because, most likely, they contain no data. |
||
163 | if (strpos($Line, ':') === false) |
||
164 | { |
||
165 | continue; |
||
166 | } |
||
167 | |||
168 | // Each line is split into two parts. The key contains the element name and additional parameters, if present, |
||
169 | // value is just the value |
||
170 | list($Key, $Value) = explode(':', $Line, 2); |
||
171 | |||
172 | // Key is transformed to lowercase because, even though the element and parameter names are written in uppercase, |
||
173 | // it is quite possible that they will be in lower- or mixed case. |
||
174 | // The key is trimmed to allow for non-significant WSP characters as allowed by v2.1 |
||
175 | $Key = strtolower(trim(self::Unescape($Key))); |
||
176 | |||
177 | // These two lines can be skipped as they aren't necessary at all. |
||
178 | if ($Key == 'begin' || $Key == 'end') |
||
179 | { |
||
180 | continue; |
||
181 | } |
||
182 | |||
183 | if ((strpos($Key, 'agent') === 0) && (stripos($Value, 'begin:vcard') !== false)) |
||
184 | { |
||
185 | $ClassName = get_class($this); |
||
186 | $Value = new $ClassName(false, str_replace('-wrap-', "\n", $Value)); |
||
187 | if (!isset($this -> Data[$Key])) |
||
188 | { |
||
189 | $this -> Data[$Key] = array(); |
||
190 | } |
||
191 | $this -> Data[$Key][] = $Value; |
||
192 | continue; |
||
193 | } |
||
194 | else |
||
195 | { |
||
196 | $Value = str_replace('-wrap-', '', $Value); |
||
197 | } |
||
198 | |||
199 | $Value = trim(self::Unescape($Value)); |
||
200 | $Type = array(); |
||
201 | |||
202 | // Here additional parameters are parsed |
||
203 | $KeyParts = explode(';', $Key); |
||
204 | $Key = $KeyParts[0]; |
||
205 | $Encoding = false; |
||
206 | |||
207 | if (strpos($Key, 'item') === 0) |
||
208 | { |
||
209 | $TmpKey = explode('.', $Key, 2); |
||
210 | $Key = $TmpKey[1]; |
||
211 | $ItemIndex = (int)str_ireplace('item', '', $TmpKey[0]); |
||
212 | } |
||
213 | |||
214 | if (count($KeyParts) > 1) |
||
215 | { |
||
216 | $Parameters = self::ParseParameters($Key, array_slice($KeyParts, 1)); |
||
217 | |||
218 | foreach ($Parameters as $ParamKey => $ParamValue) |
||
219 | { |
||
220 | switch ($ParamKey) |
||
221 | { |
||
222 | case 'encoding': |
||
223 | $Encoding = $ParamValue; |
||
224 | if (in_array($ParamValue, array('b', 'base64'))) |
||
225 | { |
||
226 | //$Value = base64_decode($Value); |
||
227 | } |
||
228 | elseif ($ParamValue == 'quoted-printable') // v2.1 |
||
229 | { |
||
230 | $Value = quoted_printable_decode($Value); |
||
231 | } |
||
232 | break; |
||
233 | case 'charset': // v2.1 |
||
234 | if ($ParamValue != 'utf-8' && $ParamValue != 'utf8') |
||
235 | { |
||
236 | $Value = mb_convert_encoding($Value, 'UTF-8', $ParamValue); |
||
237 | } |
||
238 | break; |
||
239 | case 'type': |
||
240 | $Type = $ParamValue; |
||
241 | break; |
||
242 | } |
||
243 | } |
||
244 | } |
||
245 | |||
246 | // Checking files for colon-separated additional parameters (Apple's Address Book does this), for example, "X-ABCROP-RECTANGLE" for photos |
||
247 | if (in_array($Key, self::$Spec_FileElements) && isset($Parameters['encoding']) && in_array($Parameters['encoding'], array('b', 'base64'))) |
||
248 | { |
||
249 | // If colon is present in the value, it must contain Address Book parameters |
||
250 | // (colon is an invalid character for base64 so it shouldn't appear in valid files) |
||
251 | if (strpos($Value, ':') !== false) |
||
252 | { |
||
253 | $Value = explode(':', $Value); |
||
254 | $Value = array_pop($Value); |
||
255 | } |
||
256 | } |
||
257 | |||
258 | // Values are parsed according to their type |
||
259 | if (isset(self::$Spec_StructuredElements[$Key])) |
||
260 | { |
||
261 | $Value = self::ParseStructuredValue($Value, $Key); |
||
262 | if ($Type) |
||
263 | { |
||
264 | $Value['type'] = $Type; |
||
265 | } |
||
266 | } |
||
267 | else |
||
268 | { |
||
269 | if (in_array($Key, self::$Spec_MultipleValueElements)) |
||
270 | { |
||
271 | $Value = self::ParseMultipleTextValue($Value, $Key); |
||
272 | } |
||
273 | |||
274 | if ($Type) |
||
275 | { |
||
276 | $Value = array( |
||
277 | 'value' => $Value, |
||
278 | 'type' => $Type |
||
279 | ); |
||
280 | } |
||
281 | } |
||
282 | |||
283 | if (is_array($Value) && $Encoding) |
||
284 | { |
||
285 | $Value['encoding'] = $Encoding; |
||
286 | } |
||
287 | |||
288 | if (!isset($this -> Data[$Key])) |
||
289 | { |
||
290 | $this -> Data[$Key] = array(); |
||
291 | } |
||
292 | |||
293 | $this -> Data[$Key][] = $Value; |
||
294 | } |
||
295 | } |
||
296 | } |
||
297 | |||
298 | /** |
||
299 | * Magic method to get the various vCard values as object members, e.g. |
||
300 | * a call to $vCard -> N gets the "N" value |
||
301 | * |
||
302 | * @param string Key |
||
303 | * |
||
304 | * @return mixed Value |
||
305 | */ |
||
306 | public function __get($Key) |
||
307 | { |
||
308 | $Key = strtolower($Key); |
||
309 | if (isset($this -> Data[$Key])) |
||
310 | { |
||
311 | if ($Key == 'agent') |
||
312 | { |
||
313 | return $this -> Data[$Key]; |
||
314 | } |
||
315 | elseif (in_array($Key, self::$Spec_FileElements)) |
||
316 | { |
||
317 | $Value = $this -> Data[$Key]; |
||
318 | foreach ($Value as $K => $V) |
||
319 | { |
||
320 | if (stripos($V['value'], 'uri:') === 0) |
||
321 | { |
||
322 | $Value[$K]['value'] = substr($V, 4); |
||
323 | $Value[$K]['encoding'] = 'uri'; |
||
324 | } |
||
325 | } |
||
326 | return $Value; |
||
327 | } |
||
328 | |||
329 | if ($this -> Options['Collapse'] && is_array($this -> Data[$Key]) && (count($this -> Data[$Key]) == 1)) |
||
330 | { |
||
331 | return $this -> Data[$Key][0]; |
||
332 | } |
||
333 | return $this -> Data[$Key]; |
||
334 | } |
||
335 | elseif ($Key == 'Mode') |
||
336 | { |
||
337 | return $this -> Mode; |
||
338 | } |
||
339 | return array(); |
||
340 | } |
||
341 | |||
342 | /** |
||
343 | * Saves an embedded file |
||
344 | * |
||
345 | * @param string Key |
||
346 | * @param int Index of the file, defaults to 0 |
||
347 | * @param string Target path where the file should be saved, including the filename |
||
348 | * |
||
349 | * @return bool Operation status |
||
350 | */ |
||
351 | public function SaveFile($Key, $Index = 0, $TargetPath = '') |
||
352 | { |
||
353 | if (!isset($this -> Data[$Key])) |
||
354 | { |
||
355 | return false; |
||
356 | } |
||
357 | if (!isset($this -> Data[$Key][$Index])) |
||
358 | { |
||
359 | return false; |
||
360 | } |
||
361 | |||
362 | // Returing false if it is an image URL |
||
363 | if (stripos($this -> Data[$Key][$Index]['value'], 'uri:') === 0) |
||
364 | { |
||
365 | return false; |
||
366 | } |
||
367 | |||
368 | if (is_writable($TargetPath) || (!file_exists($TargetPath) && is_writable(dirname($TargetPath)))) |
||
369 | { |
||
370 | $RawContent = $this -> Data[$Key][$Index]['value']; |
||
371 | if (isset($this -> Data[$Key][$Index]['encoding']) && $this -> Data[$Key][$Index]['encoding'] == 'b') |
||
372 | { |
||
373 | $RawContent = base64_decode($RawContent); |
||
374 | } |
||
375 | $Status = file_put_contents($TargetPath, $RawContent); |
||
376 | return (bool)$Status; |
||
377 | } |
||
378 | else |
||
379 | { |
||
380 | throw new Exception('vCard: Cannot save file ('.$Key.'), target path not writable ('.$TargetPath.')'); |
||
381 | } |
||
382 | return false; |
||
383 | } |
||
384 | |||
385 | /** |
||
386 | * Magic method for adding data to the vCard |
||
387 | * |
||
388 | * @param string Key |
||
389 | * @param string Method call arguments. First element is value. |
||
390 | * |
||
391 | * @return vCard Current object for method chaining |
||
392 | */ |
||
393 | public function __call($Key, $Arguments) |
||
453 | |||
454 | /** |
||
455 | * Magic method for getting vCard content out |
||
456 | * |
||
457 | * @return string Raw vCard content |
||
458 | */ |
||
459 | public function __toString() |
||
508 | |||
509 | // !Helper methods |
||
510 | |||
511 | private static function PrepareTypeStrForOutput($Type) |
||
515 | |||
516 | /** |
||
517 | * Removes the escaping slashes from the text. |
||
518 | * |
||
519 | * @access private |
||
520 | * |
||
521 | * @param string Text to prepare. |
||
522 | * |
||
523 | * @return string Resulting text. |
||
524 | */ |
||
525 | private static function Unescape($Text) |
||
529 | |||
530 | /** |
||
531 | * Separates the various parts of a structured value according to the spec. |
||
532 | * |
||
533 | * @access private |
||
534 | * |
||
535 | * @param string Raw text string |
||
536 | * @param string Key (e.g., N, ADR, ORG, etc.) |
||
537 | * |
||
538 | * @return array Parts in an associative array. |
||
539 | */ |
||
540 | private static function ParseStructuredValue($Text, $Key) |
||
541 | { |
||
542 | $Text = array_map('trim', explode(';', $Text)); |
||
543 | |||
544 | $Result = array(); |
||
545 | $Ctr = 0; |
||
546 | |||
547 | foreach (self::$Spec_StructuredElements[$Key] as $Index => $StructurePart) |
||
548 | { |
||
549 | $Result[$StructurePart] = isset($Text[$Index]) ? $Text[$Index] : null; |
||
550 | } |
||
551 | return $Result; |
||
552 | } |
||
553 | |||
554 | /** |
||
555 | * @access private |
||
556 | */ |
||
557 | private static function ParseMultipleTextValue($Text) |
||
561 | |||
562 | /** |
||
563 | * @access private |
||
564 | */ |
||
565 | private static function ParseParameters($Key, array $RawParams = null) |
||
566 | { |
||
567 | if (!$RawParams) |
||
568 | { |
||
569 | return array(); |
||
570 | } |
||
571 | |||
572 | // Parameters are split into (key, value) pairs |
||
573 | $Parameters = array(); |
||
574 | foreach ($RawParams as $Item) |
||
575 | { |
||
576 | $Parameters[] = explode('=', strtolower($Item)); |
||
577 | } |
||
578 | |||
579 | $Type = array(); |
||
580 | $Result = array(); |
||
581 | |||
582 | // And each parameter is checked whether anything can/should be done because of it |
||
583 | foreach ($Parameters as $Index => $Parameter) |
||
584 | { |
||
585 | // Skipping empty elements |
||
586 | if (!$Parameter) |
||
587 | { |
||
588 | continue; |
||
589 | } |
||
590 | |||
591 | // Handling type parameters without the explicit TYPE parameter name (2.1 valid) |
||
592 | if (count($Parameter) == 1) |
||
593 | { |
||
594 | // Checks if the type value is allowed for the specific element |
||
595 | // The second part of the "if" statement means that email elements can have non-standard types (see the spec) |
||
596 | if ( |
||
597 | (isset(self::$Spec_ElementTypes[$Key]) && in_array($Parameter[0], self::$Spec_ElementTypes[$Key])) || |
||
598 | ($Key == 'email' && is_scalar($Parameter[0])) |
||
599 | ) |
||
600 | { |
||
601 | $Type[] = $Parameter[0]; |
||
602 | } |
||
603 | } |
||
604 | elseif (count($Parameter) > 2) |
||
605 | { |
||
606 | if(count(explode(',', $RawParams[$Index], -1)) > 0) |
||
607 | { |
||
608 | $TempTypeParams = self::ParseParameters($Key, explode(',', $RawParams[$Index])); |
||
609 | if ($TempTypeParams['type']) |
||
610 | { |
||
611 | $Type = array_merge($Type, $TempTypeParams['type']); |
||
612 | } |
||
613 | } |
||
614 | } |
||
615 | else |
||
616 | { |
||
617 | switch ($Parameter[0]) |
||
618 | { |
||
619 | case 'encoding': |
||
620 | if (in_array($Parameter[1], array('quoted-printable', 'b', 'base64'))) |
||
621 | { |
||
622 | $Result['encoding'] = $Parameter[1] == 'base64' ? 'b' : $Parameter[1]; |
||
623 | } |
||
624 | break; |
||
625 | case 'charset': |
||
626 | $Result['charset'] = $Parameter[1]; |
||
627 | break; |
||
628 | case 'type': |
||
629 | $Type = array_merge($Type, explode(',', $Parameter[1])); |
||
630 | break; |
||
631 | case 'value': |
||
632 | if (strtolower($Parameter[1]) == 'url') |
||
633 | { |
||
634 | $Result['encoding'] = 'uri'; |
||
635 | } |
||
636 | break; |
||
637 | } |
||
638 | } |
||
639 | } |
||
640 | |||
641 | $Result['type'] = $Type; |
||
642 | |||
643 | return $Result; |
||
644 | } |
||
645 | |||
646 | // !Interface methods |
||
647 | |||
648 | // Countable interface |
||
649 | public function count() |
||
665 | |||
666 | // Iterator interface |
||
667 | public function rewind() |
||
671 | |||
672 | public function current() |
||
676 | |||
677 | public function next() |
||
681 | |||
682 | public function valid() |
||
686 | |||
687 | public function key() |
||
691 | } |
||
692 | ?> |
||
693 |
This check looks for assignments to scalar types that may be of the wrong type.
To ensure the code behaves as expected, it may be a good idea to add an explicit type cast.