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
29 using icu::RuleBasedBreakIterator;
30 using icu::Locale;
31
fetch_rbbi(BreakIterator_object * bio)32 static inline RuleBasedBreakIterator *fetch_rbbi(BreakIterator_object *bio) {
33 return (RuleBasedBreakIterator*)bio->biter;
34 }
35
_php_intlrbbi_constructor_body(INTERNAL_FUNCTION_PARAMETERS)36 static void _php_intlrbbi_constructor_body(INTERNAL_FUNCTION_PARAMETERS)
37 {
38 char *rules;
39 size_t rules_len;
40 zend_bool compiled = 0;
41 UErrorCode status = U_ZERO_ERROR;
42 intl_error_reset(NULL);
43
44 if (zend_parse_parameters_throw(ZEND_NUM_ARGS(), "s|b",
45 &rules, &rules_len, &compiled) == FAILURE) {
46 intl_error_set(NULL, U_ILLEGAL_ARGUMENT_ERROR,
47 "rbbi_create_instance: bad arguments", 0);
48 return;
49 }
50
51 // instantiation of ICU object
52 RuleBasedBreakIterator *rbbi;
53
54 if (!compiled) {
55 UnicodeString rulesStr;
56 UParseError parseError = UParseError();
57 if (intl_stringFromChar(rulesStr, rules, rules_len, &status)
58 == FAILURE) {
59 intl_error_set(NULL, U_ILLEGAL_ARGUMENT_ERROR,
60 "rbbi_create_instance: rules were not a valid UTF-8 string",
61 0);
62 RETURN_NULL();
63 }
64
65 rbbi = new RuleBasedBreakIterator(rulesStr, parseError, status);
66 intl_error_set_code(NULL, status);
67 if (U_FAILURE(status)) {
68 char *msg;
69 smart_str parse_error_str;
70 parse_error_str = intl_parse_error_to_string(&parseError);
71 spprintf(&msg, 0, "rbbi_create_instance: unable to create "
72 "RuleBasedBreakIterator from rules (%s)", parse_error_str.s? ZSTR_VAL(parse_error_str.s) : "");
73 smart_str_free(&parse_error_str);
74 intl_error_set_custom_msg(NULL, msg, 1);
75 efree(msg);
76 delete rbbi;
77 return;
78 }
79 } else { // compiled
80 #if U_ICU_VERSION_MAJOR_NUM * 10 + U_ICU_VERSION_MINOR_NUM >= 48
81 rbbi = new RuleBasedBreakIterator((uint8_t*)rules, rules_len, status);
82 if (U_FAILURE(status)) {
83 intl_error_set(NULL, status, "rbbi_create_instance: unable to "
84 "create instance from compiled rules", 0);
85 delete rbbi;
86 return;
87 }
88 #else
89 intl_error_set(NULL, U_UNSUPPORTED_ERROR, "rbbi_create_instance: "
90 "compiled rules require ICU >= 4.8", 0);
91 return;
92 #endif
93 }
94
95 breakiterator_object_create(return_value, rbbi, 0);
96 }
97
PHP_METHOD(IntlRuleBasedBreakIterator,__construct)98 U_CFUNC PHP_METHOD(IntlRuleBasedBreakIterator, __construct)
99 {
100 zend_error_handling error_handling;
101
102 zend_replace_error_handling(EH_THROW, IntlException_ce_ptr, &error_handling);
103 return_value = getThis();
104 _php_intlrbbi_constructor_body(INTERNAL_FUNCTION_PARAM_PASSTHRU);
105 zend_restore_error_handling(&error_handling);
106 }
107
PHP_FUNCTION(rbbi_get_rules)108 U_CFUNC PHP_FUNCTION(rbbi_get_rules)
109 {
110 BREAKITER_METHOD_INIT_VARS;
111 object = getThis();
112
113 if (zend_parse_parameters_none() == FAILURE) {
114 intl_error_set(NULL, U_ILLEGAL_ARGUMENT_ERROR,
115 "rbbi_get_rules: bad arguments", 0);
116 RETURN_FALSE;
117 }
118
119 BREAKITER_METHOD_FETCH_OBJECT;
120
121 zend_string *u8str;
122 const UnicodeString rules = fetch_rbbi(bio)->getRules();
123
124 u8str = intl_charFromString(rules, BREAKITER_ERROR_CODE_P(bio));
125 if (!u8str)
126 {
127 intl_errors_set(BREAKITER_ERROR_P(bio), BREAKITER_ERROR_CODE(bio),
128 "rbbi_hash_code: Error converting result to UTF-8 string",
129 0);
130 RETURN_FALSE;
131 }
132 RETVAL_STR(u8str);
133 }
134
PHP_FUNCTION(rbbi_get_rule_status)135 U_CFUNC PHP_FUNCTION(rbbi_get_rule_status)
136 {
137 BREAKITER_METHOD_INIT_VARS;
138 object = getThis();
139
140 if (zend_parse_parameters_none() == FAILURE) {
141 intl_error_set(NULL, U_ILLEGAL_ARGUMENT_ERROR,
142 "rbbi_get_rule_status: bad arguments", 0);
143 RETURN_FALSE;
144 }
145
146 BREAKITER_METHOD_FETCH_OBJECT;
147
148 RETURN_LONG(fetch_rbbi(bio)->getRuleStatus());
149 }
150
PHP_FUNCTION(rbbi_get_rule_status_vec)151 U_CFUNC PHP_FUNCTION(rbbi_get_rule_status_vec)
152 {
153 BREAKITER_METHOD_INIT_VARS;
154 object = getThis();
155
156 if (zend_parse_parameters_none() == FAILURE) {
157 intl_error_set(NULL, U_ILLEGAL_ARGUMENT_ERROR,
158 "rbbi_get_rule_status_vec: bad arguments", 0);
159 RETURN_FALSE;
160 }
161
162 BREAKITER_METHOD_FETCH_OBJECT;
163
164 int32_t num_rules = fetch_rbbi(bio)->getRuleStatusVec(NULL, 0,
165 BREAKITER_ERROR_CODE(bio));
166 if (BREAKITER_ERROR_CODE(bio) == U_BUFFER_OVERFLOW_ERROR) {
167 BREAKITER_ERROR_CODE(bio) = U_ZERO_ERROR;
168 } else {
169 // should not happen
170 INTL_METHOD_CHECK_STATUS(bio, "rbbi_get_rule_status_vec: failed "
171 " determining the number of status values");
172 }
173 int32_t *rules = new int32_t[num_rules];
174 num_rules = fetch_rbbi(bio)->getRuleStatusVec(rules, num_rules,
175 BREAKITER_ERROR_CODE(bio));
176 if (U_FAILURE(BREAKITER_ERROR_CODE(bio))) {
177 delete[] rules;
178 intl_errors_set(BREAKITER_ERROR_P(bio), BREAKITER_ERROR_CODE(bio),
179 "rbbi_get_rule_status_vec: failed obtaining the status values",
180 0);
181 RETURN_FALSE;
182 }
183
184 array_init_size(return_value, num_rules);
185 for (int32_t i = 0; i < num_rules; i++) {
186 add_next_index_long(return_value, rules[i]);
187 }
188 delete[] rules;
189 }
190
191 #if U_ICU_VERSION_MAJOR_NUM * 10 + U_ICU_VERSION_MINOR_NUM >= 48
PHP_FUNCTION(rbbi_get_binary_rules)192 U_CFUNC PHP_FUNCTION(rbbi_get_binary_rules)
193 {
194 BREAKITER_METHOD_INIT_VARS;
195 object = getThis();
196
197 if (zend_parse_parameters_none() == FAILURE) {
198 intl_error_set(NULL, U_ILLEGAL_ARGUMENT_ERROR,
199 "rbbi_get_binary_rules: bad arguments", 0);
200 RETURN_FALSE;
201 }
202
203 BREAKITER_METHOD_FETCH_OBJECT;
204
205 uint32_t rules_len;
206 const uint8_t *rules = fetch_rbbi(bio)->getBinaryRules(rules_len);
207
208 if (rules_len > INT_MAX - 1) {
209 intl_errors_set(BREAKITER_ERROR_P(bio), BREAKITER_ERROR_CODE(bio),
210 "rbbi_get_binary_rules: the rules are too large",
211 0);
212 RETURN_FALSE;
213 }
214
215 zend_string *ret_rules = zend_string_alloc(rules_len, 0);
216 memcpy(ZSTR_VAL(ret_rules), rules, rules_len);
217 ZSTR_VAL(ret_rules)[rules_len] = '\0';
218
219 RETURN_STR(ret_rules);
220 }
221 #endif
222