1 : /*
2 : +----------------------------------------------------------------------+
3 : | PHP Version 5 |
4 : +----------------------------------------------------------------------+
5 : | Copyright (c) 1997-2009 The PHP Group |
6 : +----------------------------------------------------------------------+
7 : | This source file is subject to version 3.01 of the PHP license, |
8 : | that is bundled with this package in the file LICENSE, and is |
9 : | available through the world-wide-web at the following url: |
10 : | http://www.php.net/license/3_01.txt |
11 : | If you did not receive a copy of the PHP license and are unable to |
12 : | obtain it through the world-wide-web, please send a note to |
13 : | license@php.net so we can mail you a copy immediately. |
14 : +----------------------------------------------------------------------+
15 : | Author: Hartmut Holzgraefe <hholzgra@php.net> |
16 : +----------------------------------------------------------------------+
17 : */
18 : /* $Id: levenshtein.c 272374 2008-12-31 11:17:49Z sebastian $ */
19 :
20 : #include "php.h"
21 : #include <stdlib.h>
22 : #include <errno.h>
23 : #include <ctype.h>
24 : #include "php_string.h"
25 :
26 : #define LEVENSHTEIN_MAX_LENGTH 255
27 :
28 : /* {{{ reference_levdist
29 : * reference implementation, only optimized for memory usage, not speed */
30 : static int reference_levdist(const char *s1, int l1, const char *s2, int l2, int cost_ins, int cost_rep, int cost_del )
31 50 : {
32 : int *p1, *p2, *tmp;
33 : int i1, i2, c0, c1, c2;
34 :
35 50 : if (l1 == 0) {
36 6 : return l2 * cost_ins;
37 : }
38 44 : if (l2 == 0) {
39 2 : return l1 * cost_del;
40 : }
41 :
42 42 : if ((l1 > LEVENSHTEIN_MAX_LENGTH) || (l2 > LEVENSHTEIN_MAX_LENGTH)) {
43 2 : return -1;
44 : }
45 40 : p1 = safe_emalloc((l2 + 1), sizeof(int), 0);
46 40 : p2 = safe_emalloc((l2 + 1), sizeof(int), 0);
47 :
48 476 : for (i2 = 0; i2 <= l2; i2++) {
49 436 : p1[i2] = i2 * cost_ins;
50 : }
51 438 : for (i1 = 0; i1 < l1 ; i1++) {
52 398 : p2[0] = p1[0] + cost_del;
53 :
54 1536 : for (i2 = 0; i2 < l2; i2++) {
55 1138 : c0 = p1[i2] + ((s1[i1] == s2[i2]) ? 0 : cost_rep);
56 1138 : c1 = p1[i2 + 1] + cost_del;
57 1138 : if (c1 < c0) {
58 439 : c0 = c1;
59 : }
60 1138 : c2 = p2[i2] + cost_ins;
61 1138 : if (c2 < c0) {
62 405 : c0 = c2;
63 : }
64 1138 : p2[i2 + 1] = c0;
65 : }
66 398 : tmp = p1;
67 398 : p1 = p2;
68 398 : p2 = tmp;
69 : }
70 40 : c0 = p1[l2];
71 :
72 40 : efree(p1);
73 40 : efree(p2);
74 :
75 40 : return c0;
76 : }
77 : /* }}} */
78 :
79 : /* {{{ custom_levdist
80 : */
81 : static int custom_levdist(char *str1, char *str2, char *callback_name TSRMLS_DC)
82 1 : {
83 1 : php_error_docref(NULL TSRMLS_CC, E_WARNING, "The general Levenshtein support is not there yet");
84 : /* not there yet */
85 :
86 1 : return -1;
87 : }
88 : /* }}} */
89 :
90 : /* {{{ proto int levenshtein(string str1, string str2[, int cost_ins, int cost_rep, int cost_del])
91 : Calculate Levenshtein distance between two strings */
92 : PHP_FUNCTION(levenshtein)
93 53 : {
94 53 : int argc = ZEND_NUM_ARGS();
95 : char *str1, *str2;
96 : char *callback_name;
97 : int str1_len, str2_len, callback_len;
98 : long cost_ins, cost_rep, cost_del;
99 53 : int distance = -1;
100 :
101 53 : switch (argc) {
102 : case 2: /* just two strings: use maximum performance version */
103 37 : if (zend_parse_parameters(2 TSRMLS_CC, "ss", &str1, &str1_len, &str2, &str2_len) == FAILURE) {
104 1 : return;
105 : }
106 36 : distance = reference_levdist(str1, str1_len, str2, str2_len, 1, 1, 1);
107 36 : break;
108 :
109 : case 5: /* more general version: calc cost by ins/rep/del weights */
110 14 : if (zend_parse_parameters(5 TSRMLS_CC, "sslll", &str1, &str1_len, &str2, &str2_len, &cost_ins, &cost_rep, &cost_del) == FAILURE) {
111 0 : return;
112 : }
113 14 : distance = reference_levdist(str1, str1_len, str2, str2_len, cost_ins, cost_rep, cost_del);
114 14 : break;
115 :
116 : case 3: /* most general version: calc cost by user-supplied function */
117 1 : if (zend_parse_parameters(3 TSRMLS_CC, "sss", &str1, &str1_len, &str2, &str2_len, &callback_name, &callback_len) == FAILURE) {
118 0 : return;
119 : }
120 1 : distance = custom_levdist(str1, str2, callback_name TSRMLS_CC);
121 1 : break;
122 :
123 : default:
124 1 : WRONG_PARAM_COUNT;
125 : }
126 :
127 51 : if (distance < 0 && /* TODO */ ZEND_NUM_ARGS() != 3) {
128 2 : php_error_docref(NULL TSRMLS_CC, E_WARNING, "Argument string(s) too long");
129 : }
130 :
131 51 : RETURN_LONG(distance);
132 : }
133 : /* }}} */
134 :
135 : /*
136 : * Local variables:
137 : * tab-width: 4
138 : * c-basic-offset: 4
139 : * End:
140 : * vim600: sw=4 ts=4 fdm=marker
141 : * vim<600: sw=4 ts=4
142 : */
|
