1 /*
2 +----------------------------------------------------------------------+
3 | PHP Version 7 |
4 +----------------------------------------------------------------------+
5 | This source file is subject to version 3.01 of the PHP license, |
6 | that is bundled with this package in the file LICENSE, and is |
7 | available through the world-wide-web at the following url: |
8 | http://www.php.net/license/3_01.txt |
9 | If you did not receive a copy of the PHP license and are unable to |
10 | obtain it through the world-wide-web, please send a note to |
11 | license@php.net so we can mail you a copy immediately. |
12 +----------------------------------------------------------------------+
13 | Authors: Gustavo Lopes <cataphract@php.net> |
14 +----------------------------------------------------------------------+
15 */
16
17 #include <unicode/rbbi.h>
18
19 extern "C" {
20 #define USE_BREAKITERATOR_POINTER 1
21 #include "breakiterator_class.h"
22 #include <zend_exceptions.h>
23 #include <limits.h>
24 }
25
26 #include "../intl_convertcpp.h"
27 #include "../intl_common.h"
28
fetch_rbbi(BreakIterator_object * bio)29 static inline RuleBasedBreakIterator *fetch_rbbi(BreakIterator_object *bio) {
30 return (RuleBasedBreakIterator*)bio->biter;
31 }
32
_php_intlrbbi_constructor_body(INTERNAL_FUNCTION_PARAMETERS)33 static void _php_intlrbbi_constructor_body(INTERNAL_FUNCTION_PARAMETERS)
34 {
35 char *rules;
36 size_t rules_len;
37 zend_bool compiled = 0;
38 UErrorCode status = U_ZERO_ERROR;
39 intl_error_reset(NULL);
40
41 if (zend_parse_parameters_throw(ZEND_NUM_ARGS(), "s|b",
42 &rules, &rules_len, &compiled) == FAILURE) {
43 intl_error_set(NULL, U_ILLEGAL_ARGUMENT_ERROR,
44 "rbbi_create_instance: bad arguments", 0);
45 return;
46 }
47
48 // instantiation of ICU object
49 RuleBasedBreakIterator *rbbi;
50
51 if (!compiled) {
52 UnicodeString rulesStr;
53 UParseError parseError = UParseError();
54 if (intl_stringFromChar(rulesStr, rules, rules_len, &status)
55 == FAILURE) {
56 intl_error_set(NULL, U_ILLEGAL_ARGUMENT_ERROR,
57 "rbbi_create_instance: rules were not a valid UTF-8 string",
58 0);
59 RETURN_NULL();
60 }
61
62 rbbi = new RuleBasedBreakIterator(rulesStr, parseError, status);
63 intl_error_set_code(NULL, status);
64 if (U_FAILURE(status)) {
65 char *msg;
66 smart_str parse_error_str;
67 parse_error_str = intl_parse_error_to_string(&parseError);
68 spprintf(&msg, 0, "rbbi_create_instance: unable to create "
69 "RuleBasedBreakIterator from rules (%s)", parse_error_str.s? ZSTR_VAL(parse_error_str.s) : "");
70 smart_str_free(&parse_error_str);
71 intl_error_set_custom_msg(NULL, msg, 1);
72 efree(msg);
73 delete rbbi;
74 return;
75 }
76 } else { // compiled
77 #if U_ICU_VERSION_MAJOR_NUM * 10 + U_ICU_VERSION_MINOR_NUM >= 48
78 rbbi = new RuleBasedBreakIterator((uint8_t*)rules, rules_len, status);
79 if (U_FAILURE(status)) {
80 intl_error_set(NULL, status, "rbbi_create_instance: unable to "
81 "create instance from compiled rules", 0);
82 delete rbbi;
83 return;
84 }
85 #else
86 intl_error_set(NULL, U_UNSUPPORTED_ERROR, "rbbi_create_instance: "
87 "compiled rules require ICU >= 4.8", 0);
88 return;
89 #endif
90 }
91
92 breakiterator_object_create(return_value, rbbi, 0);
93 }
94
PHP_METHOD(IntlRuleBasedBreakIterator,__construct)95 U_CFUNC PHP_METHOD(IntlRuleBasedBreakIterator, __construct)
96 {
97 zend_error_handling error_handling;
98
99 zend_replace_error_handling(EH_THROW, IntlException_ce_ptr, &error_handling);
100 return_value = getThis();
101 _php_intlrbbi_constructor_body(INTERNAL_FUNCTION_PARAM_PASSTHRU);
102 zend_restore_error_handling(&error_handling);
103 }
104
PHP_FUNCTION(rbbi_get_rules)105 U_CFUNC PHP_FUNCTION(rbbi_get_rules)
106 {
107 BREAKITER_METHOD_INIT_VARS;
108 object = getThis();
109
110 if (zend_parse_parameters_none() == FAILURE) {
111 intl_error_set(NULL, U_ILLEGAL_ARGUMENT_ERROR,
112 "rbbi_get_rules: bad arguments", 0);
113 RETURN_FALSE;
114 }
115
116 BREAKITER_METHOD_FETCH_OBJECT;
117
118 zend_string *u8str;
119 const UnicodeString rules = fetch_rbbi(bio)->getRules();
120
121 u8str = intl_charFromString(rules, BREAKITER_ERROR_CODE_P(bio));
122 if (!u8str)
123 {
124 intl_errors_set(BREAKITER_ERROR_P(bio), BREAKITER_ERROR_CODE(bio),
125 "rbbi_hash_code: Error converting result to UTF-8 string",
126 0);
127 RETURN_FALSE;
128 }
129 RETVAL_STR(u8str);
130 }
131
PHP_FUNCTION(rbbi_get_rule_status)132 U_CFUNC PHP_FUNCTION(rbbi_get_rule_status)
133 {
134 BREAKITER_METHOD_INIT_VARS;
135 object = getThis();
136
137 if (zend_parse_parameters_none() == FAILURE) {
138 intl_error_set(NULL, U_ILLEGAL_ARGUMENT_ERROR,
139 "rbbi_get_rule_status: bad arguments", 0);
140 RETURN_FALSE;
141 }
142
143 BREAKITER_METHOD_FETCH_OBJECT;
144
145 RETURN_LONG(fetch_rbbi(bio)->getRuleStatus());
146 }
147
PHP_FUNCTION(rbbi_get_rule_status_vec)148 U_CFUNC PHP_FUNCTION(rbbi_get_rule_status_vec)
149 {
150 BREAKITER_METHOD_INIT_VARS;
151 object = getThis();
152
153 if (zend_parse_parameters_none() == FAILURE) {
154 intl_error_set(NULL, U_ILLEGAL_ARGUMENT_ERROR,
155 "rbbi_get_rule_status_vec: bad arguments", 0);
156 RETURN_FALSE;
157 }
158
159 BREAKITER_METHOD_FETCH_OBJECT;
160
161 int32_t num_rules = fetch_rbbi(bio)->getRuleStatusVec(NULL, 0,
162 BREAKITER_ERROR_CODE(bio));
163 if (BREAKITER_ERROR_CODE(bio) == U_BUFFER_OVERFLOW_ERROR) {
164 BREAKITER_ERROR_CODE(bio) = U_ZERO_ERROR;
165 } else {
166 // should not happen
167 INTL_METHOD_CHECK_STATUS(bio, "rbbi_get_rule_status_vec: failed "
168 " determining the number of status values");
169 }
170 int32_t *rules = new int32_t[num_rules];
171 num_rules = fetch_rbbi(bio)->getRuleStatusVec(rules, num_rules,
172 BREAKITER_ERROR_CODE(bio));
173 if (U_FAILURE(BREAKITER_ERROR_CODE(bio))) {
174 delete[] rules;
175 intl_errors_set(BREAKITER_ERROR_P(bio), BREAKITER_ERROR_CODE(bio),
176 "rbbi_get_rule_status_vec: failed obtaining the status values",
177 0);
178 RETURN_FALSE;
179 }
180
181 array_init_size(return_value, num_rules);
182 for (int32_t i = 0; i < num_rules; i++) {
183 add_next_index_long(return_value, rules[i]);
184 }
185 delete[] rules;
186 }
187
188 #if U_ICU_VERSION_MAJOR_NUM * 10 + U_ICU_VERSION_MINOR_NUM >= 48
PHP_FUNCTION(rbbi_get_binary_rules)189 U_CFUNC PHP_FUNCTION(rbbi_get_binary_rules)
190 {
191 BREAKITER_METHOD_INIT_VARS;
192 object = getThis();
193
194 if (zend_parse_parameters_none() == FAILURE) {
195 intl_error_set(NULL, U_ILLEGAL_ARGUMENT_ERROR,
196 "rbbi_get_binary_rules: bad arguments", 0);
197 RETURN_FALSE;
198 }
199
200 BREAKITER_METHOD_FETCH_OBJECT;
201
202 uint32_t rules_len;
203 const uint8_t *rules = fetch_rbbi(bio)->getBinaryRules(rules_len);
204
205 if (rules_len > INT_MAX - 1) {
206 intl_errors_set(BREAKITER_ERROR_P(bio), BREAKITER_ERROR_CODE(bio),
207 "rbbi_get_binary_rules: the rules are too large",
208 0);
209 RETURN_FALSE;
210 }
211
212 zend_string *ret_rules = zend_string_alloc(rules_len, 0);
213 memcpy(ZSTR_VAL(ret_rules), rules, rules_len);
214 ZSTR_VAL(ret_rules)[rules_len] = '\0';
215
216 RETURN_STR(ret_rules);
217 }
218 #endif
219