1 /*
2 +----------------------------------------------------------------------+
3 | PHP Version 5 |
4 +----------------------------------------------------------------------+
5 | Copyright (c) 1997-2013 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: Andrei Zmievski <andrei@php.net> |
16 +----------------------------------------------------------------------+
17 */
18
19 /* $Id$ */
20
21 #ifdef HAVE_CONFIG_H
22 #include "config.h"
23 #endif
24
25 #include "php.h"
26 #include "php_ini.h"
27 #include "ext/standard/info.h"
28 #include "php_tokenizer.h"
29
30 #include "zend.h"
31 #include "zend_language_scanner.h"
32 #include "zend_language_scanner_defs.h"
33 #include <zend_language_parser.h>
34
35 #define zendtext LANG_SCNG(yy_text)
36 #define zendleng LANG_SCNG(yy_leng)
37
38 /* {{{ arginfo */
39 ZEND_BEGIN_ARG_INFO_EX(arginfo_token_get_all, 0, 0, 1)
40 ZEND_ARG_INFO(0, source)
41 ZEND_END_ARG_INFO()
42
43 ZEND_BEGIN_ARG_INFO_EX(arginfo_token_name, 0, 0, 1)
44 ZEND_ARG_INFO(0, token)
45 ZEND_END_ARG_INFO()
46 /* }}} */
47
48 /* {{{ tokenizer_functions[]
49 *
50 * Every user visible function must have an entry in tokenizer_functions[].
51 */
52 const zend_function_entry tokenizer_functions[] = {
53 PHP_FE(token_get_all, arginfo_token_get_all)
54 PHP_FE(token_name, arginfo_token_name)
55 PHP_FE_END
56 };
57 /* }}} */
58
59 /* {{{ tokenizer_module_entry
60 */
61 zend_module_entry tokenizer_module_entry = {
62 #if ZEND_MODULE_API_NO >= 20010901
63 STANDARD_MODULE_HEADER,
64 #endif
65 "tokenizer",
66 tokenizer_functions,
67 PHP_MINIT(tokenizer),
68 NULL,
69 NULL,
70 NULL,
71 PHP_MINFO(tokenizer),
72 #if ZEND_MODULE_API_NO >= 20010901
73 "0.1", /* Replace with version number for your extension */
74 #endif
75 STANDARD_MODULE_PROPERTIES
76 };
77 /* }}} */
78
79 #ifdef COMPILE_DL_TOKENIZER
80 ZEND_GET_MODULE(tokenizer)
81 #endif
82
83 /* {{{ PHP_MINIT_FUNCTION
84 */
PHP_MINIT_FUNCTION(tokenizer)85 PHP_MINIT_FUNCTION(tokenizer)
86 {
87 tokenizer_register_constants(INIT_FUNC_ARGS_PASSTHRU);
88 return SUCCESS;
89 }
90 /* }}} */
91
92 /* {{{ PHP_MINFO_FUNCTION
93 */
PHP_MINFO_FUNCTION(tokenizer)94 PHP_MINFO_FUNCTION(tokenizer)
95 {
96 php_info_print_table_start();
97 php_info_print_table_row(2, "Tokenizer Support", "enabled");
98 php_info_print_table_end();
99 }
100 /* }}} */
101
tokenize(zval * return_value TSRMLS_DC)102 static void tokenize(zval *return_value TSRMLS_DC)
103 {
104 zval token;
105 zval *keyword;
106 int token_type;
107 zend_bool destroy;
108 int token_line = 1;
109
110 array_init(return_value);
111
112 ZVAL_NULL(&token);
113 while ((token_type = lex_scan(&token TSRMLS_CC))) {
114 destroy = 1;
115 switch (token_type) {
116 case T_CLOSE_TAG:
117 if (zendtext[zendleng - 1] != '>') {
118 CG(zend_lineno)++;
119 }
120 case T_OPEN_TAG:
121 case T_OPEN_TAG_WITH_ECHO:
122 case T_WHITESPACE:
123 case T_COMMENT:
124 case T_DOC_COMMENT:
125 destroy = 0;
126 break;
127 }
128
129 if (token_type >= 256) {
130 MAKE_STD_ZVAL(keyword);
131 array_init(keyword);
132 add_next_index_long(keyword, token_type);
133 if (token_type == T_END_HEREDOC) {
134 if (CG(increment_lineno)) {
135 token_line = ++CG(zend_lineno);
136 CG(increment_lineno) = 0;
137 }
138 add_next_index_stringl(keyword, Z_STRVAL(token), Z_STRLEN(token), 1);
139 efree(Z_STRVAL(token));
140 } else {
141 add_next_index_stringl(keyword, (char *)zendtext, zendleng, 1);
142 }
143 add_next_index_long(keyword, token_line);
144 add_next_index_zval(return_value, keyword);
145 } else {
146 add_next_index_stringl(return_value, (char *)zendtext, zendleng, 1);
147 }
148 if (destroy && Z_TYPE(token) != IS_NULL) {
149 zval_dtor(&token);
150 }
151 ZVAL_NULL(&token);
152
153 token_line = CG(zend_lineno);
154
155 if (token_type == T_HALT_COMPILER) {
156 break;
157 }
158 }
159 }
160
161 /* {{{ proto array token_get_all(string source)
162 */
PHP_FUNCTION(token_get_all)163 PHP_FUNCTION(token_get_all)
164 {
165 char *source = NULL;
166 int argc = ZEND_NUM_ARGS();
167 int source_len;
168 zval source_z;
169 zend_lex_state original_lex_state;
170
171 if (zend_parse_parameters(argc TSRMLS_CC, "s", &source, &source_len) == FAILURE)
172 return;
173
174 ZVAL_STRINGL(&source_z, source, source_len, 1);
175 zend_save_lexical_state(&original_lex_state TSRMLS_CC);
176
177 if (zend_prepare_string_for_scanning(&source_z, "" TSRMLS_CC) == FAILURE) {
178 zend_restore_lexical_state(&original_lex_state TSRMLS_CC);
179 RETURN_EMPTY_STRING();
180 }
181
182 LANG_SCNG(yy_state) = yycINITIAL;
183
184 tokenize(return_value TSRMLS_CC);
185
186 zend_restore_lexical_state(&original_lex_state TSRMLS_CC);
187 zval_dtor(&source_z);
188 }
189 /* }}} */
190
191 /* {{{ proto string token_name(int type)
192 */
PHP_FUNCTION(token_name)193 PHP_FUNCTION(token_name)
194 {
195 int argc = ZEND_NUM_ARGS();
196 long type;
197
198 if (zend_parse_parameters(argc TSRMLS_CC, "l", &type) == FAILURE) {
199 return;
200 }
201 RETVAL_STRING(get_token_type_name(type), 1);
202 }
203 /* }}} */
204
205 /*
206 * Local variables:
207 * tab-width: 4
208 * c-basic-offset: 4
209 * End:
210 * vim600: noet sw=4 ts=4 fdm=marker
211 * vim<600: noet sw=4 ts=4
212 */
213
