summaryrefslogtreecommitdiff
path: root/ext/standard/string.c
diff options
context:
space:
mode:
authorAndrea Faulds <ajf@ajf.me>2016-10-13 23:33:33 +0100
committerAndrea Faulds <ajf@ajf.me>2016-10-17 15:39:02 +0100
commit1a512eed449128334edf0329b72e53c5caaaa95b (patch)
treec095b8ccdb00a6d364a1b1f642c5a5a218a675d6 /ext/standard/string.c
parenta5251f78f8c3204c04d93c02b63d2e3967d3e1c1 (diff)
downloadphp-git-1a512eed449128334edf0329b72e53c5caaaa95b.tar.gz
Move utf8_encode and utf8_decode to ext/standard
Diffstat (limited to 'ext/standard/string.c')
-rw-r--r--ext/standard/string.c94
1 files changed, 94 insertions, 0 deletions
diff --git a/ext/standard/string.c b/ext/standard/string.c
index fa59ddd06f..4389e10702 100644
--- a/ext/standard/string.c
+++ b/ext/standard/string.c
@@ -64,6 +64,8 @@
/* For str_getcsv() support */
#include "ext/standard/file.h"
+/* For php_next_utf8_char() */
+#include "ext/standard/html.h"
#define STR_PAD_LEFT 0
#define STR_PAD_RIGHT 1
@@ -5653,6 +5655,98 @@ PHP_FUNCTION(substr_compare)
}
/* }}} */
+/* {{{ */
+static zend_string *php_utf8_encode(const char *s, size_t len)
+{
+ size_t pos = len;
+ zend_string *str;
+ unsigned char c;
+
+ str = zend_string_safe_alloc(len, 2, 0, 0);
+ ZSTR_LEN(str) = 0;
+ while (pos > 0) {
+ /* The lower 256 codepoints of Unicode are identical to Latin-1,
+ * so we don't need to do any mapping here. */
+ c = (unsigned char)(*s);
+ if (c < 0x80) {
+ ZSTR_VAL(str)[ZSTR_LEN(str)++] = (char) c;
+ /* We only account for the single-byte and two-byte cases because
+ * we're only dealing with the first 256 Unicode codepoints. */
+ } else {
+ ZSTR_VAL(str)[ZSTR_LEN(str)++] = (0xc0 | (c >> 6));
+ ZSTR_VAL(str)[ZSTR_LEN(str)++] = (0x80 | (c & 0x3f));
+ }
+ pos--;
+ s++;
+ }
+ ZSTR_VAL(str)[ZSTR_LEN(str)] = '\0';
+ str = zend_string_truncate(str, ZSTR_LEN(str), 0);
+ return str;
+}
+/* }}} */
+
+/* {{{ */
+static zend_string *php_utf8_decode(const char *s, size_t len)
+{
+ size_t pos = 0;
+ unsigned int c;
+ zend_string *str;
+
+ str = zend_string_alloc(len, 0);
+ ZSTR_LEN(str) = 0;
+ while (pos < len) {
+ int status = FAILURE;
+ c = php_next_utf8_char((const unsigned char*)s, (size_t) len, &pos, &status);
+
+ /* The lower 256 codepoints of Unicode are identical to Latin-1,
+ * so we don't need to do any mapping here beyond replacing non-Latin-1
+ * characters. */
+ if (status == FAILURE || c > 0xFFU) {
+ c = '?';
+ }
+
+ ZSTR_VAL(str)[ZSTR_LEN(str)++] = c;
+ }
+ ZSTR_VAL(str)[ZSTR_LEN(str)] = '\0';
+ if (ZSTR_LEN(str) < len) {
+ str = zend_string_truncate(str, ZSTR_LEN(str), 0);
+ }
+
+ return str;
+}
+/* }}} */
+
+
+/* {{{ proto string utf8_encode(string data)
+ Encodes an ISO-8859-1 string to UTF-8 */
+PHP_FUNCTION(utf8_encode)
+{
+ char *arg;
+ size_t arg_len;
+
+ if (zend_parse_parameters(ZEND_NUM_ARGS(), "s", &arg, &arg_len) == FAILURE) {
+ return;
+ }
+
+ RETURN_STR(php_utf8_encode(arg, arg_len));
+}
+/* }}} */
+
+/* {{{ proto string utf8_decode(string data)
+ Converts a UTF-8 encoded string to ISO-8859-1 */
+PHP_FUNCTION(utf8_decode)
+{
+ char *arg;
+ size_t arg_len;
+
+ if (zend_parse_parameters(ZEND_NUM_ARGS(), "s", &arg, &arg_len) == FAILURE) {
+ return;
+ }
+
+ RETURN_STR(php_utf8_decode(arg, arg_len));
+}
+/* }}} */
+
/*
* Local variables:
* tab-width: 4