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:  * @author     David Grudl
 17:  */
 18: class Strings
 19: {
 20: 
 21:     /**
 22:      * Static class - cannot be instantiated.
 23:      */
 24:     final public function __construct()
 25:     {
 26:         throw new Nette\StaticClassException;
 27:     }
 28: 
 29: 
 30:     /**
 31:      * Checks if the string is valid for the specified encoding.
 32:      * @param  string  byte stream to check
 33:      * @param  string  expected encoding
 34:      * @return bool
 35:      */
 36:     public static function checkEncoding($s, $encoding = 'UTF-8')
 37:     {
 38:         return $s === self::fixEncoding($s, $encoding);
 39:     }
 40: 
 41: 
 42:     /**
 43:      * Returns correctly encoded string.
 44:      * @param  string  byte stream to fix
 45:      * @param  string  encoding
 46:      * @return string
 47:      */
 48:     public static function fixEncoding($s, $encoding = 'UTF-8')
 49:     {
 50:         // removes xD800-xDFFF, x110000 and higher
 51:         if (PHP_VERSION_ID < 50400 || strcasecmp($encoding, 'UTF-8')) {
 52:             return @iconv('UTF-16', $encoding . '//IGNORE', iconv($encoding, 'UTF-16//IGNORE', $s)); // intentionally @
 53:         } else {
 54:             return htmlspecialchars_decode(htmlspecialchars($s, ENT_NOQUOTES | ENT_IGNORE, 'UTF-8'), ENT_NOQUOTES);
 55:         }
 56:     }
 57: 
 58: 
 59:     /**
 60:      * Returns a specific character.
 61:      * @param  int     codepoint
 62:      * @param  string  encoding
 63:      * @return string
 64:      */
 65:     public static function chr($code, $encoding = 'UTF-8')
 66:     {
 67:         return iconv('UTF-32BE', $encoding . '//IGNORE', pack('N', $code));
 68:     }
 69: 
 70: 
 71:     /**
 72:      * Starts the $haystack string with the prefix $needle?
 73:      * @param  string
 74:      * @param  string
 75:      * @return bool
 76:      */
 77:     public static function startsWith($haystack, $needle)
 78:     {
 79:         return strncmp($haystack, $needle, strlen($needle)) === 0;
 80:     }
 81: 
 82: 
 83:     /**
 84:      * Ends the $haystack string with the suffix $needle?
 85:      * @param  string
 86:      * @param  string
 87:      * @return bool
 88:      */
 89:     public static function endsWith($haystack, $needle)
 90:     {
 91:         return strlen($needle) === 0 || substr($haystack, -strlen($needle)) === $needle;
 92:     }
 93: 
 94: 
 95:     /**
 96:      * Does $haystack contain $needle?
 97:      * @param  string
 98:      * @param  string
 99:      * @return bool
100:      */
101:     public static function contains($haystack, $needle)
102:     {
103:         return strpos($haystack, $needle) !== FALSE;
104:     }
105: 
106: 
107:     /**
108:      * Returns a part of UTF-8 string.
109:      * @param  string
110:      * @param  int
111:      * @param  int
112:      * @return string
113:      */
114:     public static function substring($s, $start, $length = NULL)
115:     {
116:         if (function_exists('mb_substr')) {
117:             if ($length === NULL && PHP_VERSION_ID < 50408) {
118:                 $length = self::length($s);
119:             }
120:             return mb_substr($s, $start, $length, 'UTF-8'); // MB is much faster
121:         } elseif ($length === NULL) {
122:             $length = self::length($s);
123:         } elseif ($start < 0 && $length < 0) {
124:             $start += self::length($s); // unifies iconv_substr behavior with mb_substr
125:         }
126:         return iconv_substr($s, $start, $length, 'UTF-8');
127:     }
128: 
129: 
130:     /**
131:      * Removes special controls characters and normalizes line endings and spaces.
132:      * @param  string  UTF-8 encoding or 8-bit
133:      * @return string
134:      */
135:     public static function normalize($s)
136:     {
137:         $s = self::normalizeNewLines($s);
138: 
139:         // remove control characters; leave \t + \n
140:         $s = preg_replace('#[\x00-\x08\x0B-\x1F\x7F]+#', '', $s);
141: 
142:         // right trim
143:         $s = preg_replace('#[\t ]+$#m', '', $s);
144: 
145:         // leading and trailing blank lines
146:         $s = trim($s, "\n");
147: 
148:         return $s;
149:     }
150: 
151: 
152:     /**
153:      * Standardize line endings to unix-like.
154:      * @param  string  UTF-8 encoding or 8-bit
155:      * @return string
156:      */
157:     public static function normalizeNewLines($s)
158:     {
159:         return str_replace(array("\r\n", "\r"), "\n", $s);
160:     }
161: 
162: 
163:     /**
164:      * Converts to ASCII.
165:      * @param  string  UTF-8 encoding
166:      * @return string  ASCII
167:      */
168:     public static function toAscii($s)
169:     {
170:         $s = preg_replace('#[^\x09\x0A\x0D\x20-\x7E\xA0-\x{2FF}\x{370}-\x{10FFFF}]#u', '', $s);
171:         $s = strtr($s, '`\'"^~?', "\x01\x02\x03\x04\x05\x06");
172:         $s = str_replace(array("\xE2\x80\x9E", "\xE2\x80\x9C", "\xE2\x80\x9D", "\xE2\x80\x9A",
173:             "\xE2\x80\x98", "\xE2\x80\x99", "\xC2\xBB", "\xC2\xAB"),
174:             array("\x03", "\x03", "\x03", "\x02", "\x02", "\x02", ">>", "<<"), $s);
175:         if (ICONV_IMPL === 'glibc') {
176:             $s = @iconv('UTF-8', 'WINDOWS-1250//TRANSLIT//IGNORE', $s); // intentionally @
177:             $s = strtr($s, "\xa5\xa3\xbc\x8c\xa7\x8a\xaa\x8d\x8f\x8e\xaf\xb9\xb3\xbe\x9c\x9a\xba\x9d\x9f\x9e"
178:                 . "\xbf\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf\xd0\xd1\xd2\xd3"
179:                 . "\xd4\xd5\xd6\xd7\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7\xe8"
180:                 . "\xe9\xea\xeb\xec\xed\xee\xef\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf8\xf9\xfa\xfb\xfc\xfd\xfe\x96",
181:                 "ALLSSSSTZZZallssstzzzRAAAALCCCEEEEIIDDNNOOOOxRUUUUYTsraaaalccceeeeiiddnnooooruuuuyt-");
182:             $s = preg_replace('#[^\x00-\x7F]++#', '', $s);
183:         } else {
184:             $s = @iconv('UTF-8', 'ASCII//TRANSLIT//IGNORE', $s); // intentionally @
185:         }
186:         $s = str_replace(array('`', "'", '"', '^', '~', '?'), '', $s);
187:         return strtr($s, "\x01\x02\x03\x04\x05\x06", '`\'"^~?');
188:     }
189: 
190: 
191:     /**
192:      * Converts to web safe characters [a-z0-9-] text.
193:      * @param  string  UTF-8 encoding
194:      * @param  string  allowed characters
195:      * @param  bool
196:      * @return string
197:      */
198:     public static function webalize($s, $charlist = NULL, $lower = TRUE)
199:     {
200:         $s = self::toAscii($s);
201:         if ($lower) {
202:             $s = strtolower($s);
203:         }
204:         $s = preg_replace('#[^a-z0-9' . preg_quote($charlist, '#') . ']+#i', '-', $s);
205:         $s = trim($s, '-');
206:         return $s;
207:     }
208: 
209: 
210:     /**
211:      * Truncates string to maximal length.
212:      * @param  string  UTF-8 encoding
213:      * @param  int
214:      * @param  string  UTF-8 encoding
215:      * @return string
216:      */
217:     public static function truncate($s, $maxLen, $append = "\xE2\x80\xA6")
218:     {
219:         if (self::length($s) > $maxLen) {
220:             $maxLen = $maxLen - self::length($append);
221:             if ($maxLen < 1) {
222:                 return $append;
223: 
224:             } elseif ($matches = self::match($s, '#^.{1,'.$maxLen.'}(?=[\s\x00-/:-@\[-`{-~])#us')) {
225:                 return $matches[0] . $append;
226: 
227:             } else {
228:                 return self::substring($s, 0, $maxLen) . $append;
229:             }
230:         }
231:         return $s;
232:     }
233: 
234: 
235:     /**
236:      * Indents the content from the left.
237:      * @param  string  UTF-8 encoding or 8-bit
238:      * @param  int
239:      * @param  string
240:      * @return string
241:      */
242:     public static function indent($s, $level = 1, $chars = "\t")
243:     {
244:         if ($level > 0) {
245:             $s = self::replace($s, '#(?:^|[\r\n]+)(?=[^\r\n])#', '$0' . str_repeat($chars, $level));
246:         }
247:         return $s;
248:     }
249: 
250: 
251:     /**
252:      * Convert to lower case.
253:      * @param  string  UTF-8 encoding
254:      * @return string
255:      */
256:     public static function lower($s)
257:     {
258:         return mb_strtolower($s, 'UTF-8');
259:     }
260: 
261: 
262:     /**
263:      * Convert to upper case.
264:      * @param  string  UTF-8 encoding
265:      * @return string
266:      */
267:     public static function upper($s)
268:     {
269:         return mb_strtoupper($s, 'UTF-8');
270:     }
271: 
272: 
273:     /**
274:      * Convert first character to upper case.
275:      * @param  string  UTF-8 encoding
276:      * @return string
277:      */
278:     public static function firstUpper($s)
279:     {
280:         return self::upper(self::substring($s, 0, 1)) . self::substring($s, 1);
281:     }
282: 
283: 
284:     /**
285:      * Capitalize string.
286:      * @param  string  UTF-8 encoding
287:      * @return string
288:      */
289:     public static function capitalize($s)
290:     {
291:         return mb_convert_case($s, MB_CASE_TITLE, 'UTF-8');
292:     }
293: 
294: 
295:     /**
296:      * Case-insensitive compares UTF-8 strings.
297:      * @param  string
298:      * @param  string
299:      * @param  int
300:      * @return bool
301:      */
302:     public static function compare($left, $right, $len = NULL)
303:     {
304:         if ($len < 0) {
305:             $left = self::substring($left, $len, -$len);
306:             $right = self::substring($right, $len, -$len);
307:         } elseif ($len !== NULL) {
308:             $left = self::substring($left, 0, $len);
309:             $right = self::substring($right, 0, $len);
310:         }
311:         return self::lower($left) === self::lower($right);
312:     }
313: 
314: 
315:     /**
316:      * Finds the length of common prefix of strings.
317:      * @param  string|array
318:      * @return string
319:      */
320:     public static function findPrefix($strings)
321:     {
322:         if (!is_array($strings)) {
323:             $strings = func_get_args();
324:         }
325:         $first = array_shift($strings);
326:         for ($i = 0; $i < strlen($first); $i++) {
327:             foreach ($strings as $s) {
328:                 if (!isset($s[$i]) || $first[$i] !== $s[$i]) {
329:                     while ($i && $first[$i-1] >= "\x80" && $first[$i] >= "\x80" && $first[$i] < "\xC0") {
330:                         $i--;
331:                     }
332:                     return substr($first, 0, $i);
333:                 }
334:             }
335:         }
336:         return $first;
337:     }
338: 
339: 
340:     /**
341:      * Returns UTF-8 string length.
342:      * @param  string
343:      * @return int
344:      */
345:     public static function length($s)
346:     {
347:         return function_exists('mb_strlen') ? mb_strlen($s, 'UTF-8') : strlen(utf8_decode($s));
348:     }
349: 
350: 
351:     /**
352:      * Strips whitespace.
353:      * @param  string  UTF-8 encoding
354:      * @param  string
355:      * @return string
356:      */
357:     public static function trim($s, $charlist = " \t\n\r\0\x0B\xC2\xA0")
358:     {
359:         $charlist = preg_quote($charlist, '#');
360:         return self::replace($s, '#^['.$charlist.']+|['.$charlist.']+\z#u', '');
361:     }
362: 
363: 
364:     /**
365:      * Pad a string to a certain length with another string.
366:      * @param  string  UTF-8 encoding
367:      * @param  int
368:      * @param  string
369:      * @return string
370:      */
371:     public static function padLeft($s, $length, $pad = ' ')
372:     {
373:         $length = max(0, $length - self::length($s));
374:         $padLen = self::length($pad);
375:         return str_repeat($pad, $length / $padLen) . self::substring($pad, 0, $length % $padLen) . $s;
376:     }
377: 
378: 
379:     /**
380:      * Pad a string to a certain length with another string.
381:      * @param  string  UTF-8 encoding
382:      * @param  int
383:      * @param  string
384:      * @return string
385:      */
386:     public static function padRight($s, $length, $pad = ' ')
387:     {
388:         $length = max(0, $length - self::length($s));
389:         $padLen = self::length($pad);
390:         return $s . str_repeat($pad, $length / $padLen) . self::substring($pad, 0, $length % $padLen);
391:     }
392: 
393: 
394:     /**
395:      * Reverse string.
396:      * @param  string  UTF-8 encoding
397:      * @return string
398:      */
399:     public static function reverse($s)
400:     {
401:         return @iconv('UTF-32LE', 'UTF-8', strrev(@iconv('UTF-8', 'UTF-32BE', $s)));
402:     }
403: 
404: 
405:     /**
406:      * Generate random string.
407:      * @param  int
408:      * @param  string
409:      * @return string
410:      */
411:     public static function random($length = 10, $charlist = '0-9a-z')
412:     {
413:         $charlist = str_shuffle(preg_replace_callback('#.-.#', function ($m) {
414:             return implode('', range($m[0][0], $m[0][2]));
415:         }, $charlist));
416:         $chLen = strlen($charlist);
417: 
418:         if (function_exists('openssl_random_pseudo_bytes')
419:             && (PHP_VERSION_ID >= 50400 || !defined('PHP_WINDOWS_VERSION_BUILD')) // slow in PHP 5.3 & Windows
420:         ) {
421:             $rand3 = openssl_random_pseudo_bytes($length);
422:         }
423:         if (empty($rand3) && function_exists('mcrypt_create_iv') && (PHP_VERSION_ID >= 50307 || !defined('PHP_WINDOWS_VERSION_BUILD'))) { // PHP bug #52523
424:             $rand3 = mcrypt_create_iv($length, MCRYPT_DEV_URANDOM);
425:         }
426:         if (empty($rand3) && !defined('PHP_WINDOWS_VERSION_BUILD') && @is_readable('/dev/urandom')) {
427:             $rand3 = file_get_contents('/dev/urandom', FALSE, NULL, -1, $length);
428:         }
429:         if (empty($rand3)) {
430:             static $cache;
431:             $rand3 = $cache ?: $cache = md5(serialize($_SERVER), TRUE);
432:         }
433: 
434:         $s = '';
435:         for ($i = 0; $i < $length; $i++) {
436:             if ($i % 5 === 0) {
437:                 list($rand, $rand2) = explode(' ', microtime());
438:                 $rand += lcg_value();
439:             }
440:             $rand *= $chLen;
441:             $s .= $charlist[($rand + $rand2 + ord($rand3[$i % strlen($rand3)])) % $chLen];
442:             $rand -= (int) $rand;
443:         }
444:         return $s;
445:     }
446: 
447: 
448:     /**
449:      * Splits string by a regular expression.
450:      * @param  string
451:      * @param  string
452:      * @param  int
453:      * @return array
454:      */
455:     public static function split($subject, $pattern, $flags = 0)
456:     {
457:         return self::pcre('preg_split', array($pattern, $subject, -1, $flags | PREG_SPLIT_DELIM_CAPTURE));
458:     }
459: 
460: 
461:     /**
462:      * Performs a regular expression match.
463:      * @param  string
464:      * @param  string
465:      * @param  int  can be PREG_OFFSET_CAPTURE (returned in bytes)
466:      * @param  int  offset in bytes
467:      * @return mixed
468:      */
469:     public static function match($subject, $pattern, $flags = 0, $offset = 0)
470:     {
471:         if ($offset > strlen($subject)) {
472:             return NULL;
473:         }
474:         return self::pcre('preg_match', array($pattern, $subject, & $m, $flags, $offset))
475:             ? $m
476:             : NULL;
477:     }
478: 
479: 
480:     /**
481:      * Performs a global regular expression match.
482:      * @param  string
483:      * @param  string
484:      * @param  int  can be PREG_OFFSET_CAPTURE (returned in bytes); PREG_SET_ORDER is default
485:      * @param  int  offset in bytes
486:      * @return array
487:      */
488:     public static function matchAll($subject, $pattern, $flags = 0, $offset = 0)
489:     {
490:         if ($offset > strlen($subject)) {
491:             return array();
492:         }
493:         self::pcre('preg_match_all', array(
494:             $pattern, $subject, & $m,
495:             ($flags & PREG_PATTERN_ORDER) ? $flags : ($flags | PREG_SET_ORDER),
496:             $offset
497:         ));
498:         return $m;
499:     }
500: 
501: 
502:     /**
503:      * Perform a regular expression search and replace.
504:      * @param  string
505:      * @param  string|array
506:      * @param  string|callable
507:      * @param  int
508:      * @return string
509:      */
510:     public static function replace($subject, $pattern, $replacement = NULL, $limit = -1)
511:     {
512:         if (is_object($replacement) || is_array($replacement)) {
513:             if ($replacement instanceof Nette\Callback) {
514:                 $replacement = $replacement->getNative();
515:             }
516:             if (!is_callable($replacement, FALSE, $textual)) {
517:                 throw new Nette\InvalidStateException("Callback '$textual' is not callable.");
518:             }
519: 
520:             return self::pcre('preg_replace_callback', array($pattern, $replacement, $subject, $limit));
521: 
522:         } elseif ($replacement === NULL && is_array($pattern)) {
523:             $replacement = array_values($pattern);
524:             $pattern = array_keys($pattern);
525:         }
526: 
527:         return self::pcre('preg_replace', array($pattern, $replacement, $subject, $limit));
528:     }
529: 
530: 
531:     /** @internal */
532:     public static function pcre($func, $args)
533:     {
534:         $res = Callback::invokeSafe($func, $args, function ($message) use ($args) {
535:             // compile-time error, not detectable by preg_last_error
536:             throw new RegexpException($message . ' in pattern: ' . implode(' or ', (array) $args[0]));
537:         });
538: 
539:         if (($code = preg_last_error()) // run-time error, but preg_last_error & return code are liars
540:             && ($res === NULL || !in_array($func, array('preg_filter', 'preg_replace_callback', 'preg_replace')))
541:         ) {
542:             throw new RegexpException(NULL, $code, implode(' or ', (array) $args[0]));
543:         }
544:         return $res;
545:     }
546: 
547: }
548: 
549: 
550: /**
551:  * The exception that indicates error of the last Regexp execution.
552:  */
553: class RegexpException extends \Exception
554: {
555:     static public $messages = array(
556:         PREG_INTERNAL_ERROR => 'Internal error',
557:         PREG_BACKTRACK_LIMIT_ERROR => 'Backtrack limit was exhausted',
558:         PREG_RECURSION_LIMIT_ERROR => 'Recursion limit was exhausted',
559:         PREG_BAD_UTF8_ERROR => 'Malformed UTF-8 data',
560:         5 => 'Offset didn\'t correspond to the begin of a valid UTF-8 code point', // PREG_BAD_UTF8_OFFSET_ERROR
561:     );
562: 
563:     public function __construct($message, $code = NULL, $pattern = NULL)
564:     {
565:         if (!$message) {
566:             $message = (isset(self::$messages[$code]) ? self::$messages[$code] : 'Unknown error') . ($pattern ? " (pattern: $pattern)" : '');
567:         }
568:         parent::__construct($message, $code);
569:     }
570: 
571: }
572: