YourWishes/jerryscript
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Use unicode iterators in RegExp engine and implement \d, \D, \w, \W, \s, \S, \x, \u matching in RegExp.

Browse Source 
JerryScript-DCO-1.0-Signed-off-by: Szilard Ledan szledan.u-szeged@partner.samsung.com
JerryScript-DCO-1.0-Signed-off-by: László Langó llango.u-szeged@partner.samsung.com
This commit is contained in:
László Langó
2015-07-22 09:17:37 +02:00
parent 6567651b6c
commit 3c71daaf84
8 changed files with 559 additions and 402 deletions
Download Patch File Download Diff File Expand all files Collapse all files
jerry-core/parser/regexp/re-parser.h
+5 -12
View File
@@ -59,16 +59,10 @@ typedef enum
*/
#define RE_MAX_RE_DECESC_DIGITS 9
/* FIXME: depends on unicode support */
#define RE_CHAR_UNDEF ((ecma_char_t)-1)
#define RE_CONTROL_CHAR_NUL 0x0000 /* \0 */
#define RE_CONTROL_CHAR_BEL 0x0008 /* \b */
#define RE_CONTROL_CHAR_TAB 0x0009 /* \t */
#define RE_CONTROL_CHAR_EOL 0x000a /* \n */
#define RE_CONTROL_CHAR_VT 0x000b /* \v */
#define RE_CONTROL_CHAR_FF 0x000c /* \f */
#define RE_CONTROL_CHAR_CR 0x000d /* \r */
/**
* Undefined character (out of the range of the codeunit)
*/
#define RE_CHAR_UNDEF 0xFFFFFFFF
/**
* RegExp token type
@@ -87,8 +81,7 @@ typedef struct
*/
typedef struct
{
lit_utf8_byte_t *pattern_start_p; /**< start of input pattern string */
lit_utf8_byte_t *current_char_p; /**< current character in input pattern */
lit_utf8_iterator_t iter; /**< iterator of input pattern */
int num_of_groups; /**< number of groups */
uint32_t num_of_classes; /**< number of character classes */
} re_parser_ctx_t;