1: <?php
  2: 
  3: /**
  4:  * This file is part of the Nette Framework (https://nette.org)
  5:  * Copyright (c) 2004 David Grudl (https://davidgrudl.com)
  6:  */
  7: 
  8: namespace Nette\Utils;
  9: 
 10: use Nette;
 11: 
 12: 
 13: /**
 14:  * String tools library.
 15:  */
 16: class Strings
 17: {
 18: 
 19:     const TRIM_CHARACTERS = " \t\n\r\0\x0B\xC2\xA0";
 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 UTF-8 encoding.
 33:      * @param  string  byte stream to check
 34:      * @return bool
 35:      */
 36:     public static function checkEncoding($s)
 37:     {
 38:         return $s === self::fixEncoding($s);
 39:     }
 40: 
 41: 
 42:     /**
 43:      * Removes invalid code unit sequences from UTF-8 string.
 44:      * @param  string  byte stream to fix
 45:      * @return string
 46:      */
 47:     public static function fixEncoding($s)
 48:     {
 49:         // removes xD800-xDFFF, x110000 and higher
 50:         if (PHP_VERSION_ID < 50400) {
 51:             return @iconv('UTF-16', 'UTF-8//IGNORE', iconv('UTF-8', 'UTF-16//IGNORE', $s)); // @ - ignore encoding errors
 52:         } else {
 53:             return htmlspecialchars_decode(htmlspecialchars($s, ENT_NOQUOTES | ENT_IGNORE, 'UTF-8'), ENT_NOQUOTES);
 54:         }
 55:     }
 56: 
 57: 
 58:     /**
 59:      * Returns a specific character in UTF-8.
 60:      * @param  int     code point (0x0 to 0xD7FF or 0xE000 to 0x10FFFF)
 61:      * @return string
 62:      * @throws Nette\InvalidArgumentException if code point is not in valid range
 63:      */
 64:     public static function chr($code)
 65:     {
 66:         if ($code < 0 || ($code >= 0xD800 && $code <= 0xDFFF) || $code > 0x10FFFF) {
 67:             throw new Nette\InvalidArgumentException('Code point must be in range 0x0 to 0xD7FF or 0xE000 to 0x10FFFF.');
 68:         }
 69:         return iconv('UTF-32BE', 'UTF-8//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, $needle) !== FALSE;
106:     }
107: 
108: 
109:     /**
110:      * Returns a part of UTF-8 string.
111:      * @param  string
112:      * @param  int in characters (code points)
113:      * @param  int in characters (code points)
114:      * @return string
115:      */
116:     public static function substring($s, $start, $length = NULL)
117:     {
118:         if (function_exists('mb_substr')) {
119:             if ($length === NULL && PHP_VERSION_ID < 50408) {
120:                 $length = self::length($s);
121:             }
122:             return mb_substr($s, $start, $length, 'UTF-8'); // MB is much faster
123:         } elseif ($length === NULL) {
124:             $length = self::length($s);
125:         } elseif ($start < 0 && $length < 0) {
126:             $start += self::length($s); // unifies iconv_substr behavior with mb_substr
127:         }
128:         return iconv_substr($s, $start, $length, 'UTF-8');
129:     }
130: 
131: 
132:     /**
133:      * Removes special controls characters and normalizes line endings and spaces.
134:      * @param  string  UTF-8 encoding
135:      * @return string
136:      */
137:     public static function normalize($s)
138:     {
139:         $s = self::normalizeNewLines($s);
140: 
141:         // remove control characters; leave \t + \n
142:         $s = preg_replace('#[\x00-\x08\x0B-\x1F\x7F-\x9F]+#u', '', $s);
143: 
144:         // right trim
145:         $s = preg_replace('#[\t ]+$#m', '', $s);
146: 
147:         // leading and trailing blank lines
148:         $s = trim($s, "\n");
149: 
150:         return $s;
151:     }
152: 
153: 
154:     /**
155:      * Standardize line endings to unix-like.
156:      * @param  string  UTF-8 encoding or 8-bit
157:      * @return string
158:      */
159:     public static function normalizeNewLines($s)
160:     {
161:         return str_replace(array("\r\n", "\r"), "\n", $s);
162:     }
163: 
164: 
165:     /**
166:      * Converts to ASCII.
167:      * @param  string  UTF-8 encoding
168:      * @return string  ASCII
169:      */
170:     public static function toAscii($s)
171:     {
172:         static $transliterator = NULL;
173:         if ($transliterator === NULL && class_exists('Transliterator', FALSE)) {
174:             $transliterator = \Transliterator::create('Any-Latin; Latin-ASCII');
175:         }
176: 
177:         $s = preg_replace('#[^\x09\x0A\x0D\x20-\x7E\xA0-\x{2FF}\x{370}-\x{10FFFF}]#u', '', $s);
178:         $s = strtr($s, '`\'"^~?', "\x01\x02\x03\x04\x05\x06");
179:         $s = str_replace(
180:             array("\xE2\x80\x9E", "\xE2\x80\x9C", "\xE2\x80\x9D", "\xE2\x80\x9A", "\xE2\x80\x98", "\xE2\x80\x99", "\xC2\xB0"),
181:             array("\x03", "\x03", "\x03", "\x02", "\x02", "\x02", "\x04"), $s
182:         );
183:         if ($transliterator !== NULL) {
184:             $s = $transliterator->transliterate($s);
185:         }
186:         if (ICONV_IMPL === 'glibc') {
187:             $s = str_replace(
188:                 array("\xC2\xBB", "\xC2\xAB", "\xE2\x80\xA6", "\xE2\x84\xA2", "\xC2\xA9", "\xC2\xAE"),
189:                 array('>>', '<<', '...', 'TM', '(c)', '(R)'), $s
190:             );
191:             $s = @iconv('UTF-8', 'WINDOWS-1250//TRANSLIT//IGNORE', $s); // intentionally @
192:             $s = strtr($s, "\xa5\xa3\xbc\x8c\xa7\x8a\xaa\x8d\x8f\x8e\xaf\xb9\xb3\xbe\x9c\x9a\xba\x9d\x9f\x9e"
193:                 . "\xbf\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf\xd0\xd1\xd2\xd3"
194:                 . "\xd4\xd5\xd6\xd7\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7\xe8"
195:                 . "\xe9\xea\xeb\xec\xed\xee\xef\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf8\xf9\xfa\xfb\xfc\xfd\xfe"
196:                 . "\x96\xa0\x8b\x97\x9b\xa6\xad\xb7",
197:                 'ALLSSSSTZZZallssstzzzRAAAALCCCEEEEIIDDNNOOOOxRUUUUYTsraaaalccceeeeiiddnnooooruuuuyt- <->|-.');
198:             $s = preg_replace('#[^\x00-\x7F]++#', '', $s);
199:         } else {
200:             $s = @iconv('UTF-8', 'ASCII//TRANSLIT//IGNORE', $s); // intentionally @
201:         }
202:         $s = str_replace(array('`', "'", '"', '^', '~', '?'), '', $s);
203:         return strtr($s, "\x01\x02\x03\x04\x05\x06", '`\'"^~?');
204:     }
205: 
206: 
207:     /**
208:      * Converts to web safe characters [a-z0-9-] text.
209:      * @param  string  UTF-8 encoding
210:      * @param  string  allowed characters
211:      * @param  bool
212:      * @return string
213:      */
214:     public static function webalize($s, $charlist = NULL, $lower = TRUE)
215:     {
216:         $s = self::toAscii($s);
217:         if ($lower) {
218:             $s = strtolower($s);
219:         }
220:         $s = preg_replace('#[^a-z0-9' . preg_quote($charlist, '#') . ']+#i', '-', $s);
221:         $s = trim($s, '-');
222:         return $s;
223:     }
224: 
225: 
226:     /**
227:      * Truncates string to maximal length.
228:      * @param  string  UTF-8 encoding
229:      * @param  int
230:      * @param  string  UTF-8 encoding
231:      * @return string
232:      */
233:     public static function truncate($s, $maxLen, $append = "\xE2\x80\xA6")
234:     {
235:         if (self::length($s) > $maxLen) {
236:             $maxLen = $maxLen - self::length($append);
237:             if ($maxLen < 1) {
238:                 return $append;
239: 
240:             } elseif ($matches = self::match($s, '#^.{1,'.$maxLen.'}(?=[\s\x00-/:-@\[-`{-~])#us')) {
241:                 return $matches[0] . $append;
242: 
243:             } else {
244:                 return self::substring($s, 0, $maxLen) . $append;
245:             }
246:         }
247:         return $s;
248:     }
249: 
250: 
251:     /**
252:      * Indents the content from the left.
253:      * @param  string  UTF-8 encoding or 8-bit
254:      * @param  int
255:      * @param  string
256:      * @return string
257:      */
258:     public static function indent($s, $level = 1, $chars = "\t")
259:     {
260:         if ($level > 0) {
261:             $s = self::replace($s, '#(?:^|[\r\n]+)(?=[^\r\n])#', '$0' . str_repeat($chars, $level));
262:         }
263:         return $s;
264:     }
265: 
266: 
267:     /**
268:      * Convert to lower case.
269:      * @param  string  UTF-8 encoding
270:      * @return string
271:      */
272:     public static function lower($s)
273:     {
274:         return mb_strtolower($s, 'UTF-8');
275:     }
276: 
277: 
278:     /**
279:      * Convert first character to lower case.
280:      * @param  string  UTF-8 encoding
281:      * @return string
282:      */
283:     public static function firstLower($s)
284:     {
285:         return self::lower(self::substring($s, 0, 1)) . self::substring($s, 1);
286:     }
287: 
288: 
289:     /**
290:      * Convert to upper case.
291:      * @param  string  UTF-8 encoding
292:      * @return string
293:      */
294:     public static function upper($s)
295:     {
296:         return mb_strtoupper($s, 'UTF-8');
297:     }
298: 
299: 
300:     /**
301:      * Convert first character to upper case.
302:      * @param  string  UTF-8 encoding
303:      * @return string
304:      */
305:     public static function firstUpper($s)
306:     {
307:         return self::upper(self::substring($s, 0, 1)) . self::substring($s, 1);
308:     }
309: 
310: 
311:     /**
312:      * Capitalize string.
313:      * @param  string  UTF-8 encoding
314:      * @return string
315:      */
316:     public static function capitalize($s)
317:     {
318:         return mb_convert_case($s, MB_CASE_TITLE, 'UTF-8');
319:     }
320: 
321: 
322:     /**
323:      * Case-insensitive compares UTF-8 strings.
324:      * @param  string
325:      * @param  string
326:      * @param  int
327:      * @return bool
328:      */
329:     public static function compare($left, $right, $len = NULL)
330:     {
331:         if ($len < 0) {
332:             $left = self::substring($left, $len, -$len);
333:             $right = self::substring($right, $len, -$len);
334:         } elseif ($len !== NULL) {
335:             $left = self::substring($left, 0, $len);
336:             $right = self::substring($right, 0, $len);
337:         }
338:         return self::lower($left) === self::lower($right);
339:     }
340: 
341: 
342:     /**
343:      * Finds the length of common prefix of strings.
344:      * @param  string|array
345:      * @return string
346:      */
347:     public static function findPrefix($strings)
348:     {
349:         if (!is_array($strings)) {
350:             $strings = func_get_args();
351:         }
352:         $first = array_shift($strings);
353:         for ($i = 0; $i < strlen($first); $i++) {
354:             foreach ($strings as $s) {
355:                 if (!isset($s[$i]) || $first[$i] !== $s[$i]) {
356:                     while ($i && $first[$i - 1] >= "\x80" && $first[$i] >= "\x80" && $first[$i] < "\xC0") {
357:                         $i--;
358:                     }
359:                     return substr($first, 0, $i);
360:                 }
361:             }
362:         }
363:         return $first;
364:     }
365: 
366: 
367:     /**
368:      * Returns number of characters (not bytes) in UTF-8 string.
369:      * That is the number of Unicode code points which may differ from the number of graphemes.
370:      * @param  string
371:      * @return int
372:      */
373:     public static function length($s)
374:     {
375:         return function_exists('mb_strlen') ? mb_strlen($s, 'UTF-8') : strlen(utf8_decode($s));
376:     }
377: 
378: 
379:     /**
380:      * Strips whitespace.
381:      * @param  string  UTF-8 encoding
382:      * @param  string
383:      * @return string
384:      */
385:     public static function trim($s, $charlist = self::TRIM_CHARACTERS)
386:     {
387:         $charlist = preg_quote($charlist, '#');
388:         return self::replace($s, '#^['.$charlist.']+|['.$charlist.']+\z#u', '');
389:     }
390: 
391: 
392:     /**
393:      * Pad a string to a certain length with another string.
394:      * @param  string  UTF-8 encoding
395:      * @param  int
396:      * @param  string
397:      * @return string
398:      */
399:     public static function padLeft($s, $length, $pad = ' ')
400:     {
401:         $length = max(0, $length - self::length($s));
402:         $padLen = self::length($pad);
403:         return str_repeat($pad, $length / $padLen) . self::substring($pad, 0, $length % $padLen) . $s;
404:     }
405: 
406: 
407:     /**
408:      * Pad a string to a certain length with another string.
409:      * @param  string  UTF-8 encoding
410:      * @param  int
411:      * @param  string
412:      * @return string
413:      */
414:     public static function padRight($s, $length, $pad = ' ')
415:     {
416:         $length = max(0, $length - self::length($s));
417:         $padLen = self::length($pad);
418:         return $s . str_repeat($pad, $length / $padLen) . self::substring($pad, 0, $length % $padLen);
419:     }
420: 
421: 
422:     /**
423:      * Reverse string.
424:      * @param  string  UTF-8 encoding
425:      * @return string
426:      */
427:     public static function reverse($s)
428:     {
429:         return @iconv('UTF-32LE', 'UTF-8', strrev(@iconv('UTF-8', 'UTF-32BE', $s)));
430:     }
431: 
432: 
433:     /**
434:      * Use Nette\Utils\Random::generate
435:      * @deprecated
436:      */
437:     public static function random($length = 10, $charlist = '0-9a-z')
438:     {
439:         return Random::generate($length, $charlist);
440:     }
441: 
442: 
443:     /**
444:      * Returns part of $haystack before $nth occurence of $needle.
445:      * @param  string
446:      * @param  string
447:      * @param  int  negative value means searching from the end
448:      * @return string|FALSE  returns FALSE if the needle was not found
449:      */
450:     public static function before($haystack, $needle, $nth = 1)
451:     {
452:         $pos = self::pos($haystack, $needle, $nth);
453:         return $pos === FALSE
454:             ? FALSE
455:             : substr($haystack, 0, $pos);
456:     }
457: 
458: 
459:     /**
460:      * Returns part of $haystack after $nth occurence of $needle.
461:      * @param  string
462:      * @param  string
463:      * @param  int  negative value means searching from the end
464:      * @return string|FALSE  returns FALSE if the needle was not found
465:      */
466:     public static function after($haystack, $needle, $nth = 1)
467:     {
468:         $pos = self::pos($haystack, $needle, $nth);
469:         return $pos === FALSE
470:             ? FALSE
471:             : (string) substr($haystack, $pos + strlen($needle));
472:     }
473: 
474: 
475:     /**
476:      * Returns position of $nth occurence of $needle in $haystack.
477:      * @return int|FALSE  offset in bytes or FALSE if the needle was not found
478:      */
479:     private static function pos($haystack, $needle, $nth = 1)
480:     {
481:         if (!$nth) {
482:             return FALSE;
483:         } elseif ($nth > 0) {
484:             if (strlen($needle) === 0) {
485:                 return 0;
486:             }
487:             $pos = 0;
488:             while (FALSE !== ($pos = strpos($haystack, $needle, $pos)) && --$nth) {
489:                 $pos++;
490:             }
491:         } else {
492:             $len = strlen($haystack);
493:             if (strlen($needle) === 0) {
494:                 return $len;
495:             }
496:             $pos = $len - 1;
497:             while (FALSE !== ($pos = strrpos($haystack, $needle, $pos - $len)) && ++$nth) {
498:                 $pos--;
499:             }
500:         }
501:         return $pos;
502:     }
503: 
504: 
505:     /**
506:      * Splits string by a regular expression.
507:      * @param  string
508:      * @param  string
509:      * @param  int
510:      * @return array
511:      */
512:     public static function split($subject, $pattern, $flags = 0)
513:     {
514:         return self::pcre('preg_split', array($pattern, $subject, -1, $flags | PREG_SPLIT_DELIM_CAPTURE));
515:     }
516: 
517: 
518:     /**
519:      * Performs a regular expression match.
520:      * @param  string
521:      * @param  string
522:      * @param  int  can be PREG_OFFSET_CAPTURE (returned in bytes)
523:      * @param  int  offset in bytes
524:      * @return mixed
525:      */
526:     public static function match($subject, $pattern, $flags = 0, $offset = 0)
527:     {
528:         if ($offset > strlen($subject)) {
529:             return NULL;
530:         }
531:         return self::pcre('preg_match', array($pattern, $subject, & $m, $flags, $offset))
532:             ? $m
533:             : NULL;
534:     }
535: 
536: 
537:     /**
538:      * Performs a global regular expression match.
539:      * @param  string
540:      * @param  string
541:      * @param  int  can be PREG_OFFSET_CAPTURE (returned in bytes); PREG_SET_ORDER is default
542:      * @param  int  offset in bytes
543:      * @return array
544:      */
545:     public static function matchAll($subject, $pattern, $flags = 0, $offset = 0)
546:     {
547:         if ($offset > strlen($subject)) {
548:             return array();
549:         }
550:         self::pcre('preg_match_all', array(
551:             $pattern, $subject, & $m,
552:             ($flags & PREG_PATTERN_ORDER) ? $flags : ($flags | PREG_SET_ORDER),
553:             $offset,
554:         ));
555:         return $m;
556:     }
557: 
558: 
559:     /**
560:      * Perform a regular expression search and replace.
561:      * @param  string
562:      * @param  string|array
563:      * @param  string|callable
564:      * @param  int
565:      * @return string
566:      */
567:     public static function replace($subject, $pattern, $replacement = NULL, $limit = -1)
568:     {
569:         if (is_object($replacement) || is_array($replacement)) {
570:             if ($replacement instanceof Nette\Callback) {
571:                 $replacement = $replacement->getNative();
572:             }
573:             if (!is_callable($replacement, FALSE, $textual)) {
574:                 throw new Nette\InvalidStateException("Callback '$textual' is not callable.");
575:             }
576: 
577:             return self::pcre('preg_replace_callback', array($pattern, $replacement, $subject, $limit));
578: 
579:         } elseif ($replacement === NULL && is_array($pattern)) {
580:             $replacement = array_values($pattern);
581:             $pattern = array_keys($pattern);
582:         }
583: 
584:         return self::pcre('preg_replace', array($pattern, $replacement, $subject, $limit));
585:     }
586: 
587: 
588:     /** @internal */
589:     public static function pcre($func, $args)
590:     {
591:         static $messages = array(
592:             PREG_INTERNAL_ERROR => 'Internal error',
593:             PREG_BACKTRACK_LIMIT_ERROR => 'Backtrack limit was exhausted',
594:             PREG_RECURSION_LIMIT_ERROR => 'Recursion limit was exhausted',
595:             PREG_BAD_UTF8_ERROR => 'Malformed UTF-8 data',
596:             PREG_BAD_UTF8_OFFSET_ERROR => 'Offset didn\'t correspond to the begin of a valid UTF-8 code point',
597:             6 => 'Failed due to limited JIT stack space', // PREG_JIT_STACKLIMIT_ERROR
598:         );
599:         $res = Callback::invokeSafe($func, $args, function ($message) use ($args) {
600:             // compile-time error, not detectable by preg_last_error
601:             throw new RegexpException($message . ' in pattern: ' . implode(' or ', (array) $args[0]));
602:         });
603: 
604:         if (($code = preg_last_error()) // run-time error, but preg_last_error & return code are liars
605:             && ($res === NULL || !in_array($func, array('preg_filter', 'preg_replace_callback', 'preg_replace')))
606:         ) {
607:             throw new RegexpException((isset($messages[$code]) ? $messages[$code] : 'Unknown error')
608:                 . ' (pattern: ' . implode(' or ', (array) $args[0]) . ')', $code);
609:         }
610:         return $res;
611:     }
612: 
613: }
614: