1: <?php
  2: 
  3: /**
  4:  * This file is part of the Nette Framework (https://nette.org)
  5:  * Copyright (c) 2004 David Grudl (http://davidgrudl.com)
  6:  */
  7: 
  8: namespace Nette\Utils;
  9: 
 10: use Nette,
 11:     Nette\Diagnostics\Debugger;
 12: 
 13: 
 14: /**
 15:  * String tools library.
 16:  *
 17:  * @author     David Grudl
 18:  */
 19: class Strings
 20: {
 21: 
 22:     /**
 23:      * Static class - cannot be instantiated.
 24:      */
 25:     final public function __construct()
 26:     {
 27:         throw new Nette\StaticClassException;
 28:     }
 29: 
 30: 
 31:     /**
 32:      * Checks if the string is valid for the specified encoding.
 33:      * @param  string  byte stream to check
 34:      * @param  string  expected encoding
 35:      * @return bool
 36:      */
 37:     public static function checkEncoding($s, $encoding = 'UTF-8')
 38:     {
 39:         return $s === self::fixEncoding($s, $encoding);
 40:     }
 41: 
 42: 
 43:     /**
 44:      * Returns correctly encoded string.
 45:      * @param  string  byte stream to fix
 46:      * @param  string  encoding
 47:      * @return string
 48:      */
 49:     public static function fixEncoding($s, $encoding = 'UTF-8')
 50:     {
 51:         // removes xD800-xDFFF, xFEFF, x110000 and higher
 52:         if (strcasecmp($encoding, 'UTF-8') === 0) {
 53:             $s = str_replace("\xEF\xBB\xBF", '', $s); // remove UTF-8 BOM
 54:         }
 55:         if (PHP_VERSION_ID >= 50400) {
 56:             ini_set('mbstring.substitute_character', 'none');
 57:             return mb_convert_encoding($s, $encoding, $encoding);
 58:         }
 59:         return @iconv('UTF-16', $encoding . '//IGNORE', iconv($encoding, 'UTF-16//IGNORE', $s)); // intentionally @
 60:     }
 61: 
 62: 
 63:     /**
 64:      * Returns a specific character.
 65:      * @param  int     codepoint
 66:      * @param  string  encoding
 67:      * @return string
 68:      */
 69:     public static function chr($code, $encoding = 'UTF-8')
 70:     {
 71:         return iconv('UTF-32BE', $encoding . '//IGNORE', pack('N', $code));
 72:     }
 73: 
 74: 
 75:     /**
 76:      * Starts the $haystack string with the prefix $needle?
 77:      * @param  string
 78:      * @param  string
 79:      * @return bool
 80:      */
 81:     public static function startsWith($haystack, $needle)
 82:     {
 83:         return strncmp($haystack, $needle, strlen($needle)) === 0;
 84:     }
 85: 
 86: 
 87:     /**
 88:      * Ends the $haystack string with the suffix $needle?
 89:      * @param  string
 90:      * @param  string
 91:      * @return bool
 92:      */
 93:     public static function endsWith($haystack, $needle)
 94:     {
 95:         return strlen($needle) === 0 || substr($haystack, -strlen($needle)) === $needle;
 96:     }
 97: 
 98: 
 99:     /**
100:      * Does $haystack contain $needle?
101:      * @param  string
102:      * @param  string
103:      * @return bool
104:      */
105:     public static function contains($haystack, $needle)
106:     {
107:         return strpos($haystack, $needle) !== FALSE;
108:     }
109: 
110: 
111:     /**
112:      * Returns a part of UTF-8 string.
113:      * @param  string
114:      * @param  int
115:      * @param  int
116:      * @return string
117:      */
118:     public static function substring($s, $start, $length = NULL)
119:     {
120:         if ($length === NULL) {
121:             $length = self::length($s);
122:         }
123:         return function_exists('mb_substr') ? mb_substr($s, $start, $length, 'UTF-8') : iconv_substr($s, $start, $length, 'UTF-8'); // MB is much faster
124:     }
125: 
126: 
127:     /**
128:      * Removes special controls characters and normalizes line endings and spaces.
129:      * @param  string  UTF-8 encoding or 8-bit
130:      * @return string
131:      */
132:     public static function normalize($s)
133:     {
134:         // standardize line endings to unix-like
135:         $s = str_replace("\r\n", "\n", $s); // DOS
136:         $s = strtr($s, "\r", "\n"); // Mac
137: 
138:         // remove control characters; leave \t + \n
139:         $s = preg_replace('#[\x00-\x08\x0B-\x1F\x7F]+#', '', $s);
140: 
141:         // right trim
142:         $s = preg_replace('#[\t ]+$#m', '', $s);
143: 
144:         // leading and trailing blank lines
145:         $s = trim($s, "\n");
146: 
147:         return $s;
148:     }
149: 
150: 
151:     /**
152:      * Converts to ASCII.
153:      * @param  string  UTF-8 encoding
154:      * @return string  ASCII
155:      */
156:     public static function toAscii($s)
157:     {
158:         $s = preg_replace('#[^\x09\x0A\x0D\x20-\x7E\xA0-\x{2FF}\x{370}-\x{10FFFF}]#u', '', $s);
159:         $s = strtr($s, '`\'"^~', "\x01\x02\x03\x04\x05");
160:         if (ICONV_IMPL === 'glibc') {
161:             $s = @iconv('UTF-8', 'WINDOWS-1250//TRANSLIT', $s); // intentionally @
162:             $s = strtr($s, "\xa5\xa3\xbc\x8c\xa7\x8a\xaa\x8d\x8f\x8e\xaf\xb9\xb3\xbe\x9c\x9a\xba\x9d\x9f\x9e"
163:                 . "\xbf\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf\xd0\xd1\xd2\xd3"
164:                 . "\xd4\xd5\xd6\xd7\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7\xe8"
165:                 . "\xe9\xea\xeb\xec\xed\xee\xef\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf8\xf9\xfa\xfb\xfc\xfd\xfe\x96",
166:                 "ALLSSSSTZZZallssstzzzRAAAALCCCEEEEIIDDNNOOOOxRUUUUYTsraaaalccceeeeiiddnnooooruuuuyt-");
167:         } else {
168:             $s = @iconv('UTF-8', 'ASCII//TRANSLIT', $s); // intentionally @
169:         }
170:         $s = str_replace(array('`', "'", '"', '^', '~'), '', $s);
171:         return strtr($s, "\x01\x02\x03\x04\x05", '`\'"^~');
172:     }
173: 
174: 
175:     /**
176:      * Converts to web safe characters [a-z0-9-] text.
177:      * @param  string  UTF-8 encoding
178:      * @param  string  allowed characters
179:      * @param  bool
180:      * @return string
181:      */
182:     public static function webalize($s, $charlist = NULL, $lower = TRUE)
183:     {
184:         $s = self::toAscii($s);
185:         if ($lower) {
186:             $s = strtolower($s);
187:         }
188:         $s = preg_replace('#[^a-z0-9' . preg_quote($charlist, '#') . ']+#i', '-', $s);
189:         $s = trim($s, '-');
190:         return $s;
191:     }
192: 
193: 
194:     /**
195:      * Truncates string to maximal length.
196:      * @param  string  UTF-8 encoding
197:      * @param  int
198:      * @param  string  UTF-8 encoding
199:      * @return string
200:      */
201:     public static function truncate($s, $maxLen, $append = "\xE2\x80\xA6")
202:     {
203:         if (self::length($s) > $maxLen) {
204:             $maxLen = $maxLen - self::length($append);
205:             if ($maxLen < 1) {
206:                 return $append;
207: 
208:             } elseif ($matches = self::match($s, '#^.{1,'.$maxLen.'}(?=[\s\x00-/:-@\[-`{-~])#us')) {
209:                 return $matches[0] . $append;
210: 
211:             } else {
212:                 return self::substring($s, 0, $maxLen) . $append;
213:             }
214:         }
215:         return $s;
216:     }
217: 
218: 
219:     /**
220:      * Indents the content from the left.
221:      * @param  string  UTF-8 encoding or 8-bit
222:      * @param  int
223:      * @param  string
224:      * @return string
225:      */
226:     public static function indent($s, $level = 1, $chars = "\t")
227:     {
228:         return $level < 1 ? $s : self::replace($s, '#(?:^|[\r\n]+)(?=[^\r\n])#', '$0' . str_repeat($chars, $level));
229:     }
230: 
231: 
232:     /**
233:      * Convert to lower case.
234:      * @param  string  UTF-8 encoding
235:      * @return string
236:      */
237:     public static function lower($s)
238:     {
239:         return mb_strtolower($s, 'UTF-8');
240:     }
241: 
242: 
243:     /**
244:      * Convert to upper case.
245:      * @param  string  UTF-8 encoding
246:      * @return string
247:      */
248:     public static function upper($s)
249:     {
250:         return mb_strtoupper($s, 'UTF-8');
251:     }
252: 
253: 
254:     /**
255:      * Convert first character to upper case.
256:      * @param  string  UTF-8 encoding
257:      * @return string
258:      */
259:     public static function firstUpper($s)
260:     {
261:         return self::upper(self::substring($s, 0, 1)) . self::substring($s, 1);
262:     }
263: 
264: 
265:     /**
266:      * Capitalize string.
267:      * @param  string  UTF-8 encoding
268:      * @return string
269:      */
270:     public static function capitalize($s)
271:     {
272:         return mb_convert_case($s, MB_CASE_TITLE, 'UTF-8');
273:     }
274: 
275: 
276:     /**
277:      * Case-insensitive compares UTF-8 strings.
278:      * @param  string
279:      * @param  string
280:      * @param  int
281:      * @return bool
282:      */
283:     public static function compare($left, $right, $len = NULL)
284:     {
285:         if ($len < 0) {
286:             $left = self::substring($left, $len, -$len);
287:             $right = self::substring($right, $len, -$len);
288:         } elseif ($len !== NULL) {
289:             $left = self::substring($left, 0, $len);
290:             $right = self::substring($right, 0, $len);
291:         }
292:         return self::lower($left) === self::lower($right);
293:     }
294: 
295: 
296:     /**
297:      * Returns UTF-8 string length.
298:      * @param  string
299:      * @return int
300:      */
301:     public static function length($s)
302:     {
303:         return strlen(utf8_decode($s)); // fastest way
304:     }
305: 
306: 
307:     /**
308:      * Strips whitespace.
309:      * @param  string  UTF-8 encoding
310:      * @param  string
311:      * @return string
312:      */
313:     public static function trim($s, $charlist = " \t\n\r\0\x0B\xC2\xA0")
314:     {
315:         $charlist = preg_quote($charlist, '#');
316:         return self::replace($s, '#^['.$charlist.']+|['.$charlist.']+\z#u', '');
317:     }
318: 
319: 
320:     /**
321:      * Pad a string to a certain length with another string.
322:      * @param  string  UTF-8 encoding
323:      * @param  int
324:      * @param  string
325:      * @return string
326:      */
327:     public static function padLeft($s, $length, $pad = ' ')
328:     {
329:         $length = max(0, $length - self::length($s));
330:         $padLen = self::length($pad);
331:         return str_repeat($pad, $length / $padLen) . self::substring($pad, 0, $length % $padLen) . $s;
332:     }
333: 
334: 
335:     /**
336:      * Pad a string to a certain length with another string.
337:      * @param  string  UTF-8 encoding
338:      * @param  int
339:      * @param  string
340:      * @return string
341:      */
342:     public static function padRight($s, $length, $pad = ' ')
343:     {
344:         $length = max(0, $length - self::length($s));
345:         $padLen = self::length($pad);
346:         return $s . str_repeat($pad, $length / $padLen) . self::substring($pad, 0, $length % $padLen);
347:     }
348: 
349: 
350:     /**
351:      * Reverse string.
352:      * @param  string  UTF-8 encoding
353:      * @return string
354:      */
355:     public static function reverse($s)
356:     {
357:         return @iconv('UTF-32LE', 'UTF-8', strrev(@iconv('UTF-8', 'UTF-32BE', $s)));
358:     }
359: 
360: 
361:     /**
362:      * Generate random string.
363:      * @param  int
364:      * @param  string
365:      * @return string
366:      */
367:     public static function random($length = 10, $charlist = '0-9a-z')
368:     {
369:         $charlist = str_shuffle(preg_replace_callback('#.-.#', function($m) {
370:             return implode('', range($m[0][0], $m[0][2]));
371:         }, $charlist));
372:         $chLen = strlen($charlist);
373: 
374:         if (function_exists('openssl_random_pseudo_bytes')
375:             && (PHP_VERSION_ID >= 50400 || !defined('PHP_WINDOWS_VERSION_BUILD')) // slow in PHP 5.3 & Windows
376:         ) {
377:             $rand3 = openssl_random_pseudo_bytes($length);
378:         }
379:         if (empty($rand3) && function_exists('mcrypt_create_iv') && (PHP_VERSION_ID >= 50307 || !defined('PHP_WINDOWS_VERSION_BUILD'))) { // PHP bug #52523
380:             $rand3 = mcrypt_create_iv($length, MCRYPT_DEV_URANDOM);
381:         }
382:         if (empty($rand3) && @is_readable('/dev/urandom')) {
383:             $rand3 = file_get_contents('/dev/urandom', FALSE, NULL, -1, $length);
384:         }
385:         if (empty($rand3)) {
386:             static $cache;
387:             $rand3 = $cache ?: $cache = md5(serialize($_SERVER), TRUE);
388:         }
389: 
390:         $s = '';
391:         for ($i = 0; $i < $length; $i++) {
392:             if ($i % 5 === 0) {
393:                 list($rand, $rand2) = explode(' ', microtime());
394:                 $rand += lcg_value();
395:             }
396:             $rand *= $chLen;
397:             $s .= $charlist[($rand + $rand2 + ord($rand3[$i % strlen($rand3)])) % $chLen];
398:             $rand -= (int) $rand;
399:         }
400:         return $s;
401:     }
402: 
403: 
404:     /**
405:      * Splits string by a regular expression.
406:      * @param  string
407:      * @param  string
408:      * @param  int
409:      * @return array
410:      */
411:     public static function split($subject, $pattern, $flags = 0)
412:     {
413:         set_error_handler(function($severity, $message) use ($pattern) { // preg_last_error does not return compile errors
414:             restore_error_handler();
415:             throw new RegexpException("$message in pattern: $pattern");
416:         });
417:         $res = preg_split($pattern, $subject, -1, $flags | PREG_SPLIT_DELIM_CAPTURE);
418:         restore_error_handler();
419:         if (preg_last_error()) { // run-time error
420:             throw new RegexpException(NULL, preg_last_error(), $pattern);
421:         }
422:         return $res;
423:     }
424: 
425: 
426:     /**
427:      * Performs a regular expression match.
428:      * @param  string
429:      * @param  string
430:      * @param  int  can be PREG_OFFSET_CAPTURE (returned in bytes)
431:      * @param  int  offset in bytes
432:      * @return mixed
433:      */
434:     public static function match($subject, $pattern, $flags = 0, $offset = 0)
435:     {
436:         if ($offset > strlen($subject)) {
437:             return NULL;
438:         }
439:         set_error_handler(function($severity, $message) use ($pattern) { // preg_last_error does not return compile errors
440:             restore_error_handler();
441:             throw new RegexpException("$message in pattern: $pattern");
442:         });
443:         $res = preg_match($pattern, $subject, $m, $flags, $offset);
444:         restore_error_handler();
445:         if (preg_last_error()) { // run-time error
446:             throw new RegexpException(NULL, preg_last_error(), $pattern);
447:         }
448:         if ($res) {
449:             return $m;
450:         }
451:     }
452: 
453: 
454:     /**
455:      * Performs a global regular expression match.
456:      * @param  string
457:      * @param  string
458:      * @param  int  can be PREG_OFFSET_CAPTURE (returned in bytes); PREG_SET_ORDER is default
459:      * @param  int  offset in bytes
460:      * @return array
461:      */
462:     public static function matchAll($subject, $pattern, $flags = 0, $offset = 0)
463:     {
464:         if ($offset > strlen($subject)) {
465:             return array();
466:         }
467:         set_error_handler(function($severity, $message) use ($pattern) { // preg_last_error does not return compile errors
468:             restore_error_handler();
469:             throw new RegexpException("$message in pattern: $pattern");
470:         });
471:         preg_match_all(
472:             $pattern, $subject, $m,
473:             ($flags & PREG_PATTERN_ORDER) ? $flags : ($flags | PREG_SET_ORDER),
474:             $offset
475:         );
476:         restore_error_handler();
477:         if (preg_last_error()) { // run-time error
478:             throw new RegexpException(NULL, preg_last_error(), $pattern);
479:         }
480:         return $m;
481:     }
482: 
483: 
484:     /**
485:      * Perform a regular expression search and replace.
486:      * @param  string
487:      * @param  string|array
488:      * @param  string|callable
489:      * @param  int
490:      * @return string
491:      */
492:     public static function replace($subject, $pattern, $replacement = NULL, $limit = -1)
493:     {
494:         if (is_object($replacement) || is_array($replacement)) {
495:             if ($replacement instanceof Nette\Callback) {
496:                 $replacement = $replacement->getNative();
497:             }
498:             if (!is_callable($replacement, FALSE, $textual)) {
499:                 throw new Nette\InvalidStateException("Callback '$textual' is not callable.");
500:             }
501: 
502:             set_error_handler(function($severity, $message) use (& $tmp) { // preg_last_error does not return compile errors
503:                 restore_error_handler();
504:                 throw new RegexpException("$message in pattern: $tmp");
505:             });
506:             foreach ((array) $pattern as $tmp) {
507:                 preg_match($tmp, '');
508:             }
509:             restore_error_handler();
510: 
511:             $res = preg_replace_callback($pattern, $replacement, $subject, $limit);
512:             if ($res === NULL && preg_last_error()) { // run-time error
513:                 throw new RegexpException(NULL, preg_last_error(), $pattern);
514:             }
515:             return $res;
516: 
517:         } elseif ($replacement === NULL && is_array($pattern)) {
518:             $replacement = array_values($pattern);
519:             $pattern = array_keys($pattern);
520:         }
521: 
522:         set_error_handler(function($severity, $message) use ($pattern) { // preg_last_error does not return compile errors
523:             restore_error_handler();
524:             throw new RegexpException("$message in pattern: " . implode(' or ', (array) $pattern));
525:         });
526:         $res = preg_replace($pattern, $replacement, $subject, $limit);
527:         restore_error_handler();
528:         if (preg_last_error()) { // run-time error
529:             throw new RegexpException(NULL, preg_last_error(), implode(' or ', (array) $pattern));
530:         }
531:         return $res;
532:     }
533: 
534: }
535: 
536: 
537: /**
538:  * The exception that indicates error of the last Regexp execution.
539:  */
540: class RegexpException extends \Exception
541: {
542:     static public $messages = array(
543:         PREG_INTERNAL_ERROR => 'Internal error',
544:         PREG_BACKTRACK_LIMIT_ERROR => 'Backtrack limit was exhausted',
545:         PREG_RECURSION_LIMIT_ERROR => 'Recursion limit was exhausted',
546:         PREG_BAD_UTF8_ERROR => 'Malformed UTF-8 data',
547:         5 => 'Offset didn\'t correspond to the begin of a valid UTF-8 code point', // PREG_BAD_UTF8_OFFSET_ERROR
548:     );
549: 
550:     public function __construct($message, $code = NULL, $pattern = NULL)
551:     {
552:         if (!$message) {
553:             $message = (isset(self::$messages[$code]) ? self::$messages[$code] : 'Unknown error') . ($pattern ? " (pattern: $pattern)" : '');
554:         }
555:         parent::__construct($message, $code);
556:     }
557: 
558: }
559: