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:  * @package Nette\Utils
  7:  */
  8: 
  9: 
 10: 
 11: /**
 12:  * String tools library.
 13:  *
 14:  * @author     David Grudl
 15:  * @package Nette\Utils
 16:  */
 17: class NStrings
 18: {
 19: 
 20:     /**
 21:      * Static class - cannot be instantiated.
 22:      */
 23:     final public function __construct()
 24:     {
 25:         throw new NStaticClassException;
 26:     }
 27: 
 28: 
 29:     /**
 30:      * Checks if the string is valid for the specified encoding.
 31:      * @param  string  byte stream to check
 32:      * @param  string  expected encoding
 33:      * @return bool
 34:      */
 35:     public static function checkEncoding($s, $encoding = 'UTF-8')
 36:     {
 37:         return $s === self::fixEncoding($s, $encoding);
 38:     }
 39: 
 40: 
 41:     /**
 42:      * Returns correctly encoded string.
 43:      * @param  string  byte stream to fix
 44:      * @param  string  encoding
 45:      * @return string
 46:      */
 47:     public static function fixEncoding($s, $encoding = 'UTF-8')
 48:     {
 49:         // removes xD800-xDFFF, xFEFF, x110000 and higher
 50:         if (strcasecmp($encoding, 'UTF-8') === 0) {
 51:             $s = str_replace("\xEF\xBB\xBF", '', $s); // remove UTF-8 BOM
 52:         }
 53:         if (PHP_VERSION_ID >= 50400) {
 54:             ini_set('mbstring.substitute_character', 'none');
 55:             return mb_convert_encoding($s, $encoding, $encoding);
 56:         }
 57:         return @iconv('UTF-16', $encoding . '//IGNORE', iconv($encoding, 'UTF-16//IGNORE', $s)); // intentionally @
 58:     }
 59: 
 60: 
 61:     /**
 62:      * Returns a specific character.
 63:      * @param  int     codepoint
 64:      * @param  string  encoding
 65:      * @return string
 66:      */
 67:     public static function chr($code, $encoding = 'UTF-8')
 68:     {
 69:         return iconv('UTF-32BE', $encoding . '//IGNORE', pack('N', $code));
 70:     }
 71: 
 72: 
 73:     /**
 74:      * Starts the $haystack string with the prefix $needle?
 75:      * @param  string
 76:      * @param  string
 77:      * @return bool
 78:      */
 79:     public static function startsWith($haystack, $needle)
 80:     {
 81:         return strncmp($haystack, $needle, strlen($needle)) === 0;
 82:     }
 83: 
 84: 
 85:     /**
 86:      * Ends the $haystack string with the suffix $needle?
 87:      * @param  string
 88:      * @param  string
 89:      * @return bool
 90:      */
 91:     public static function endsWith($haystack, $needle)
 92:     {
 93:         return strlen($needle) === 0 || substr($haystack, -strlen($needle)) === $needle;
 94:     }
 95: 
 96: 
 97:     /**
 98:      * Does $haystack contain $needle?
 99:      * @param  string
100:      * @param  string
101:      * @return bool
102:      */
103:     public static function contains($haystack, $needle)
104:     {
105:         return strpos($haystack, $needle) !== FALSE;
106:     }
107: 
108: 
109:     /**
110:      * Returns a part of UTF-8 string.
111:      * @param  string
112:      * @param  int
113:      * @param  int
114:      * @return string
115:      */
116:     public static function substring($s, $start, $length = NULL)
117:     {
118:         if ($length === NULL) {
119:             $length = self::length($s);
120:         }
121:         return function_exists('mb_substr') ? mb_substr($s, $start, $length, 'UTF-8') : iconv_substr($s, $start, $length, 'UTF-8'); // MB is much faster
122:     }
123: 
124: 
125:     /**
126:      * Removes special controls characters and normalizes line endings and spaces.
127:      * @param  string  UTF-8 encoding or 8-bit
128:      * @return string
129:      */
130:     public static function normalize($s)
131:     {
132:         // standardize line endings to unix-like
133:         $s = str_replace("\r\n", "\n", $s); // DOS
134:         $s = strtr($s, "\r", "\n"); // Mac
135: 
136:         // remove control characters; leave \t + \n
137:         $s = preg_replace('#[\x00-\x08\x0B-\x1F\x7F]+#', '', $s);
138: 
139:         // right trim
140:         $s = preg_replace('#[\t ]+$#m', '', $s);
141: 
142:         // leading and trailing blank lines
143:         $s = trim($s, "\n");
144: 
145:         return $s;
146:     }
147: 
148: 
149:     /**
150:      * Converts to ASCII.
151:      * @param  string  UTF-8 encoding
152:      * @return string  ASCII
153:      */
154:     public static function toAscii($s)
155:     {
156:         $s = preg_replace('#[^\x09\x0A\x0D\x20-\x7E\xA0-\x{2FF}\x{370}-\x{10FFFF}]#u', '', $s);
157:         $s = strtr($s, '`\'"^~', "\x01\x02\x03\x04\x05");
158:         if (ICONV_IMPL === 'glibc') {
159:             $s = @iconv('UTF-8', 'WINDOWS-1250//TRANSLIT', $s); // intentionally @
160:             $s = strtr($s, "\xa5\xa3\xbc\x8c\xa7\x8a\xaa\x8d\x8f\x8e\xaf\xb9\xb3\xbe\x9c\x9a\xba\x9d\x9f\x9e"
161:                 . "\xbf\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf\xd0\xd1\xd2\xd3"
162:                 . "\xd4\xd5\xd6\xd7\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7\xe8"
163:                 . "\xe9\xea\xeb\xec\xed\xee\xef\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf8\xf9\xfa\xfb\xfc\xfd\xfe\x96",
164:                 "ALLSSSSTZZZallssstzzzRAAAALCCCEEEEIIDDNNOOOOxRUUUUYTsraaaalccceeeeiiddnnooooruuuuyt-");
165:         } else {
166:             $s = @iconv('UTF-8', 'ASCII//TRANSLIT', $s); // intentionally @
167:         }
168:         $s = str_replace(array('`', "'", '"', '^', '~'), '', $s);
169:         return strtr($s, "\x01\x02\x03\x04\x05", '`\'"^~');
170:     }
171: 
172: 
173:     /**
174:      * Converts to web safe characters [a-z0-9-] text.
175:      * @param  string  UTF-8 encoding
176:      * @param  string  allowed characters
177:      * @param  bool
178:      * @return string
179:      */
180:     public static function webalize($s, $charlist = NULL, $lower = TRUE)
181:     {
182:         $s = self::toAscii($s);
183:         if ($lower) {
184:             $s = strtolower($s);
185:         }
186:         $s = preg_replace('#[^a-z0-9' . preg_quote($charlist, '#') . ']+#i', '-', $s);
187:         $s = trim($s, '-');
188:         return $s;
189:     }
190: 
191: 
192:     /**
193:      * Truncates string to maximal length.
194:      * @param  string  UTF-8 encoding
195:      * @param  int
196:      * @param  string  UTF-8 encoding
197:      * @return string
198:      */
199:     public static function truncate($s, $maxLen, $append = "\xE2\x80\xA6")
200:     {
201:         if (self::length($s) > $maxLen) {
202:             $maxLen = $maxLen - self::length($append);
203:             if ($maxLen < 1) {
204:                 return $append;
205: 
206:             } elseif ($matches = self::match($s, '#^.{1,'.$maxLen.'}(?=[\s\x00-/:-@\[-`{-~])#us')) {
207:                 return $matches[0] . $append;
208: 
209:             } else {
210:                 return self::substring($s, 0, $maxLen) . $append;
211:             }
212:         }
213:         return $s;
214:     }
215: 
216: 
217:     /**
218:      * Indents the content from the left.
219:      * @param  string  UTF-8 encoding or 8-bit
220:      * @param  int
221:      * @param  string
222:      * @return string
223:      */
224:     public static function indent($s, $level = 1, $chars = "\t")
225:     {
226:         return $level < 1 ? $s : self::replace($s, '#(?:^|[\r\n]+)(?=[^\r\n])#', '$0' . str_repeat($chars, $level));
227:     }
228: 
229: 
230:     /**
231:      * Convert to lower case.
232:      * @param  string  UTF-8 encoding
233:      * @return string
234:      */
235:     public static function lower($s)
236:     {
237:         return mb_strtolower($s, 'UTF-8');
238:     }
239: 
240: 
241:     /**
242:      * Convert to upper case.
243:      * @param  string  UTF-8 encoding
244:      * @return string
245:      */
246:     public static function upper($s)
247:     {
248:         return mb_strtoupper($s, 'UTF-8');
249:     }
250: 
251: 
252:     /**
253:      * Convert first character to upper case.
254:      * @param  string  UTF-8 encoding
255:      * @return string
256:      */
257:     public static function firstUpper($s)
258:     {
259:         return self::upper(self::substring($s, 0, 1)) . self::substring($s, 1);
260:     }
261: 
262: 
263:     /**
264:      * Capitalize string.
265:      * @param  string  UTF-8 encoding
266:      * @return string
267:      */
268:     public static function capitalize($s)
269:     {
270:         return mb_convert_case($s, MB_CASE_TITLE, 'UTF-8');
271:     }
272: 
273: 
274:     /**
275:      * Case-insensitive compares UTF-8 strings.
276:      * @param  string
277:      * @param  string
278:      * @param  int
279:      * @return bool
280:      */
281:     public static function compare($left, $right, $len = NULL)
282:     {
283:         if ($len < 0) {
284:             $left = self::substring($left, $len, -$len);
285:             $right = self::substring($right, $len, -$len);
286:         } elseif ($len !== NULL) {
287:             $left = self::substring($left, 0, $len);
288:             $right = self::substring($right, 0, $len);
289:         }
290:         return self::lower($left) === self::lower($right);
291:     }
292: 
293: 
294:     /**
295:      * Returns UTF-8 string length.
296:      * @param  string
297:      * @return int
298:      */
299:     public static function length($s)
300:     {
301:         return strlen(utf8_decode($s)); // fastest way
302:     }
303: 
304: 
305:     /**
306:      * Strips whitespace.
307:      * @param  string  UTF-8 encoding
308:      * @param  string
309:      * @return string
310:      */
311:     public static function trim($s, $charlist = " \t\n\r\0\x0B\xC2\xA0")
312:     {
313:         $charlist = preg_quote($charlist, '#');
314:         return self::replace($s, '#^['.$charlist.']+|['.$charlist.']+\z#u', '');
315:     }
316: 
317: 
318:     /**
319:      * Pad a string to a certain length with another string.
320:      * @param  string  UTF-8 encoding
321:      * @param  int
322:      * @param  string
323:      * @return string
324:      */
325:     public static function padLeft($s, $length, $pad = ' ')
326:     {
327:         $length = max(0, $length - self::length($s));
328:         $padLen = self::length($pad);
329:         return str_repeat($pad, $length / $padLen) . self::substring($pad, 0, $length % $padLen) . $s;
330:     }
331: 
332: 
333:     /**
334:      * Pad a string to a certain length with another string.
335:      * @param  string  UTF-8 encoding
336:      * @param  int
337:      * @param  string
338:      * @return string
339:      */
340:     public static function padRight($s, $length, $pad = ' ')
341:     {
342:         $length = max(0, $length - self::length($s));
343:         $padLen = self::length($pad);
344:         return $s . str_repeat($pad, $length / $padLen) . self::substring($pad, 0, $length % $padLen);
345:     }
346: 
347: 
348:     /**
349:      * Reverse string.
350:      * @param  string  UTF-8 encoding
351:      * @return string
352:      */
353:     public static function reverse($s)
354:     {
355:         return @iconv('UTF-32LE', 'UTF-8', strrev(@iconv('UTF-8', 'UTF-32BE', $s)));
356:     }
357: 
358: 
359:     /**
360:      * Generate random string.
361:      * @param  int
362:      * @param  string
363:      * @return string
364:      */
365:     public static function random($length = 10, $charlist = '0-9a-z')
366:     {
367:         $charlist = str_shuffle(preg_replace_callback('#.-.#', create_function('$m', '
368:             return implode(\'\', range($m[0][0], $m[0][2]));
369:         '), $charlist));
370:         $chLen = strlen($charlist);
371: 
372:         if (function_exists('openssl_random_pseudo_bytes')
373:             && (PHP_VERSION_ID >= 50400 || !defined('PHP_WINDOWS_VERSION_BUILD')) // slow in PHP 5.3 & Windows
374:         ) {
375:             $rand3 = openssl_random_pseudo_bytes($length);
376:         }
377:         if (empty($rand3) && function_exists('mcrypt_create_iv') && (PHP_VERSION_ID >= 50307 || !defined('PHP_WINDOWS_VERSION_BUILD'))) { // PHP bug #52523
378:             $rand3 = mcrypt_create_iv($length, MCRYPT_DEV_URANDOM);
379:         }
380:         if (empty($rand3) && @is_readable('/dev/urandom')) {
381:             $rand3 = file_get_contents('/dev/urandom', FALSE, NULL, -1, $length);
382:         }
383:         if (empty($rand3)) {
384:             static $cache;
385:             $rand3 = ($tmp=$cache) ? $tmp : $cache = md5(serialize($_SERVER), TRUE);
386:         }
387: 
388:         $s = '';
389:         for ($i = 0; $i < $length; $i++) {
390:             if ($i % 5 === 0) {
391:                 list($rand, $rand2) = explode(' ', microtime());
392:                 $rand += lcg_value();
393:             }
394:             $rand *= $chLen;
395:             $s .= $charlist[($rand + $rand2 + ord($rand3[$i % strlen($rand3)])) % $chLen];
396:             $rand -= (int) $rand;
397:         }
398:         return $s;
399:     }
400: 
401: 
402:     /**
403:      * Splits string by a regular expression.
404:      * @param  string
405:      * @param  string
406:      * @param  int
407:      * @return array
408:      */
409:     public static function split($subject, $pattern, $flags = 0)
410:     {
411:         set_error_handler(create_function('$severity, $message', 'extract($GLOBALS[0]['.array_push($GLOBALS[0], array('pattern'=>$pattern)).'-1], EXTR_REFS); // preg_last_error does not return compile errors
412:             restore_error_handler();
413:             throw new NRegexpException("$message in pattern: $pattern");
414:         '));
415:         $res = preg_split($pattern, $subject, -1, $flags | PREG_SPLIT_DELIM_CAPTURE);
416:         restore_error_handler();
417:         if (preg_last_error()) { // run-time error
418:             throw new NRegexpException(NULL, preg_last_error(), $pattern);
419:         }
420:         return $res;
421:     }
422: 
423: 
424:     /**
425:      * Performs a regular expression match.
426:      * @param  string
427:      * @param  string
428:      * @param  int  can be PREG_OFFSET_CAPTURE (returned in bytes)
429:      * @param  int  offset in bytes
430:      * @return mixed
431:      */
432:     public static function match($subject, $pattern, $flags = 0, $offset = 0)
433:     {
434:         if ($offset > strlen($subject)) {
435:             return NULL;
436:         }
437:         set_error_handler(create_function('$severity, $message', 'extract($GLOBALS[0]['.array_push($GLOBALS[0], array('pattern'=>$pattern)).'-1], EXTR_REFS); // preg_last_error does not return compile errors
438:             restore_error_handler();
439:             throw new NRegexpException("$message in pattern: $pattern");
440:         '));
441:         $res = preg_match($pattern, $subject, $m, $flags, $offset);
442:         restore_error_handler();
443:         if (preg_last_error()) { // run-time error
444:             throw new NRegexpException(NULL, preg_last_error(), $pattern);
445:         }
446:         if ($res) {
447:             return $m;
448:         }
449:     }
450: 
451: 
452:     /**
453:      * Performs a global regular expression match.
454:      * @param  string
455:      * @param  string
456:      * @param  int  can be PREG_OFFSET_CAPTURE (returned in bytes); PREG_SET_ORDER is default
457:      * @param  int  offset in bytes
458:      * @return array
459:      */
460:     public static function matchAll($subject, $pattern, $flags = 0, $offset = 0)
461:     {
462:         if ($offset > strlen($subject)) {
463:             return array();
464:         }
465:         set_error_handler(create_function('$severity, $message', 'extract($GLOBALS[0]['.array_push($GLOBALS[0], array('pattern'=>$pattern)).'-1], EXTR_REFS); // preg_last_error does not return compile errors
466:             restore_error_handler();
467:             throw new NRegexpException("$message in pattern: $pattern");
468:         '));
469:         preg_match_all(
470:             $pattern, $subject, $m,
471:             ($flags & PREG_PATTERN_ORDER) ? $flags : ($flags | PREG_SET_ORDER),
472:             $offset
473:         );
474:         restore_error_handler();
475:         if (preg_last_error()) { // run-time error
476:             throw new NRegexpException(NULL, preg_last_error(), $pattern);
477:         }
478:         return $m;
479:     }
480: 
481: 
482:     /**
483:      * Perform a regular expression search and replace.
484:      * @param  string
485:      * @param  string|array
486:      * @param  string|callable
487:      * @param  int
488:      * @return string
489:      */
490:     public static function replace($subject, $pattern, $replacement = NULL, $limit = -1)
491:     {
492:         if (is_object($replacement) || is_array($replacement)|| preg_match('#^\x00lambda_\d+\z#', $replacement)) {
493:             if ($replacement instanceof NCallback) {
494:                 $replacement = $replacement->getNative();
495:             }
496:             if (!is_callable($replacement, FALSE, $textual)) {
497:                 throw new InvalidStateException("Callback '$textual' is not callable.");
498:             }
499: 
500:             set_error_handler(create_function('$severity, $message', 'extract($GLOBALS[0]['.array_push($GLOBALS[0], array('tmp'=>& $tmp)).'-1], EXTR_REFS); // preg_last_error does not return compile errors
501:                 restore_error_handler();
502:                 throw new NRegexpException("$message in pattern: $tmp");
503:             '));
504:             foreach ((array) $pattern as $tmp) {
505:                 preg_match($tmp, '');
506:             }
507:             restore_error_handler();
508: 
509:             $res = preg_replace_callback($pattern, $replacement, $subject, $limit);
510:             if ($res === NULL && preg_last_error()) { // run-time error
511:                 throw new NRegexpException(NULL, preg_last_error(), $pattern);
512:             }
513:             return $res;
514: 
515:         } elseif ($replacement === NULL && is_array($pattern)) {
516:             $replacement = array_values($pattern);
517:             $pattern = array_keys($pattern);
518:         }
519: 
520:         set_error_handler(create_function('$severity, $message', 'extract($GLOBALS[0]['.array_push($GLOBALS[0], array('pattern'=>$pattern)).'-1], EXTR_REFS); // preg_last_error does not return compile errors
521:             restore_error_handler();
522:             throw new NRegexpException("$message in pattern: " . implode(\' or \', (array) $pattern));
523:         '));
524:         $res = preg_replace($pattern, $replacement, $subject, $limit);
525:         restore_error_handler();
526:         if (preg_last_error()) { // run-time error
527:             throw new NRegexpException(NULL, preg_last_error(), implode(' or ', (array) $pattern));
528:         }
529:         return $res;
530:     }
531: 
532: }
533: 
534: 
535: /**
536:  * The exception that indicates error of the last Regexp execution.
537:  * @package Nette\Utils
538:  */
539: class NRegexpException extends Exception
540: {
541:     static public $messages = array(
542:         PREG_INTERNAL_ERROR => 'Internal error',
543:         PREG_BACKTRACK_LIMIT_ERROR => 'Backtrack limit was exhausted',
544:         PREG_RECURSION_LIMIT_ERROR => 'Recursion limit was exhausted',
545:         PREG_BAD_UTF8_ERROR => 'Malformed UTF-8 data',
546:         5 => 'Offset didn\'t correspond to the begin of a valid UTF-8 code point', // PREG_BAD_UTF8_OFFSET_ERROR
547:     );
548: 
549:     public function __construct($message, $code = NULL, $pattern = NULL)
550:     {
551:         if (!$message) {
552:             $message = (isset(self::$messages[$code]) ? self::$messages[$code] : 'Unknown error') . ($pattern ? " (pattern: $pattern)" : '');
553:         }
554:         parent::__construct($message, $code);
555:     }
556: 
557: }
558: