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: 
11: /**
12:  * Simple lexical analyser.
13:  * @deprecated use similar Nette\Tokenizer\Tokenizer
14:  */
15: class Tokenizer
16: {
17:     const VALUE = 0,
18:         OFFSET = 1,
19:         TYPE = 2;
20: 
21:     /** @var string */
22:     private $re;
23: 
24:     /** @var array|false */
25:     private $types;
26: 
27: 
28:     /**
29:      * @param  array of [(int|string) token type => (string) pattern]
30:      * @param  string  regular expression flags
31:      */
32:     public function __construct(array $patterns, $flags = '')
33:     {
34:         $this->re = '~(' . implode(')|(', $patterns) . ')~A' . $flags;
35:         $keys = array_keys($patterns);
36:         $this->types = $keys === range(0, count($patterns) - 1) ? false : $keys;
37:     }
38: 
39: 
40:     /**
41:      * Tokenizes string.
42:      * @param  string
43:      * @return array
44:      * @throws TokenizerException
45:      */
46:     public function tokenize($input)
47:     {
48:         if ($this->types) {
49:             preg_match_all($this->re, $input, $tokens, PREG_SET_ORDER);
50:             $len = 0;
51:             $count = count($this->types);
52:             foreach ($tokens as &$match) {
53:                 $type = null;
54:                 for ($i = 1; $i <= $count; $i++) {
55:                     if (!isset($match[$i])) {
56:                         break;
57:                     } elseif ($match[$i] != null) {
58:                         $type = $this->types[$i - 1];
59:                         break;
60:                     }
61:                 }
62:                 $match = [self::VALUE => $match[0], self::OFFSET => $len, self::TYPE => $type];
63:                 $len += strlen($match[self::VALUE]);
64:             }
65:             if ($len !== strlen($input)) {
66:                 $errorOffset = $len;
67:             }
68: 
69:         } else {
70:             $tokens = preg_split($this->re, $input, -1, PREG_SPLIT_NO_EMPTY | PREG_SPLIT_OFFSET_CAPTURE | PREG_SPLIT_DELIM_CAPTURE);
71:             $last = end($tokens);
72:             if ($tokens && !preg_match($this->re, $last[0])) {
73:                 $errorOffset = $last[1];
74:             }
75:         }
76: 
77:         if (isset($errorOffset)) {
78:             list($line, $col) = $this->getCoordinates($input, $errorOffset);
79:             $token = str_replace("\n", '\n', substr($input, $errorOffset, 10));
80:             throw new TokenizerException("Unexpected '$token' on line $line, column $col.");
81:         }
82:         return $tokens;
83:     }
84: 
85: 
86:     /**
87:      * Returns position of token in input string.
88:      * @param  string
89:      * @param  int
90:      * @return array of [line, column]
91:      */
92:     public static function getCoordinates($text, $offset)
93:     {
94:         $text = substr($text, 0, $offset);
95:         return [substr_count($text, "\n") + 1, $offset - strrpos("\n" . $text, "\n") + 1];
96:     }
97: }
98: