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

Implement RegExp unicode and sticky flags (#3379)

Browse Source 
JerryScript-DCO-1.0-Signed-off-by: Dániel Bátyai dbatyai@inf.u-szeged.hu
This commit is contained in:
Dániel Bátyai
2019-11-29 14:08:30 +01:00
committed by GitHub
parent 8956eff2bd
commit 35c0a6e299
9 changed files with 484 additions and 77 deletions
Download Patch File Download Diff File Expand all files Collapse all files
jerry-core/parser/js/js-lexer.c
+8
View File
@@ -2317,6 +2317,14 @@ lexer_construct_regexp_object (parser_context_t *context_p, /**< context */
{
flag = RE_FLAG_MULTILINE;
}
else if (source_p[0] == LIT_CHAR_LOWERCASE_U)
{
flag = RE_FLAG_UNICODE;
}
else if (source_p[0] == LIT_CHAR_LOWERCASE_Y)
{
flag = RE_FLAG_STICKY;
}
if (flag == 0)
{
jerry-core/parser/regexp/re-bytecode.c
+11 -2
View File
@@ -15,6 +15,7 @@
#include "ecma-globals.h"
#include "re-bytecode.h"
#include "ecma-regexp-object.h"
#if ENABLED (JERRY_BUILTIN_REGEXP)
@@ -455,8 +456,16 @@ re_dump_bytecode (re_bytecode_ctx_t *bc_ctx_p) /**< RegExp bytecode context */
JERRY_DEBUG_MSG ("%d", num_of_class);
while (num_of_class)
{
JERRY_DEBUG_MSG (" %d", re_get_char (&bytecode_p));
JERRY_DEBUG_MSG ("-%d", re_get_char (&bytecode_p));
if ((compiled_code_p->header.status_flags & RE_FLAG_UNICODE) != 0)
{
JERRY_DEBUG_MSG (" %u", re_get_value (&bytecode_p));
JERRY_DEBUG_MSG ("-%u", re_get_value (&bytecode_p));
}
else
{
JERRY_DEBUG_MSG (" %u", re_get_char (&bytecode_p));
JERRY_DEBUG_MSG ("-%u", re_get_char (&bytecode_p));
}
num_of_class--;
}
JERRY_DEBUG_MSG (", ");
jerry-core/parser/regexp/re-compiler.c
+39 -8
View File
@@ -226,12 +226,29 @@ re_insert_into_group_with_jump (re_compiler_ctx_t *re_ctx_p, /**< RegExp compile
*/
static void
re_append_char_class (re_compiler_ctx_t *re_ctx_p, /**< RegExp compiler context */
ecma_char_t start, /**< character class range from */
ecma_char_t end) /**< character class range to */
lit_code_point_t start, /**< character class range from */
lit_code_point_t end) /**< character class range to */
{
re_append_char (re_ctx_p->bytecode_ctx_p, ecma_regexp_canonicalize (start, re_ctx_p->flags & RE_FLAG_IGNORE_CASE));
re_append_char (re_ctx_p->bytecode_ctx_p, ecma_regexp_canonicalize (end, re_ctx_p->flags & RE_FLAG_IGNORE_CASE));
re_ctx_p->parser_ctx_p->classes_count++;
#if ENABLED (JERRY_ES2015)
if (re_ctx_p->flags & RE_FLAG_UNICODE)
{
re_append_u32 (re_ctx_p->bytecode_ctx_p, ecma_regexp_canonicalize (start, re_ctx_p->flags & RE_FLAG_IGNORE_CASE));
re_append_u32 (re_ctx_p->bytecode_ctx_p, ecma_regexp_canonicalize (end, re_ctx_p->flags & RE_FLAG_IGNORE_CASE));
return;
}
#endif /* ENABLED (JERRY_ES2015) */
JERRY_ASSERT (start <= LIT_UTF16_CODE_UNIT_MAX);
JERRY_ASSERT (end <= LIT_UTF16_CODE_UNIT_MAX);
re_append_char (re_ctx_p->bytecode_ctx_p,
(ecma_char_t) ecma_regexp_canonicalize (start,
re_ctx_p->flags & RE_FLAG_IGNORE_CASE));
re_append_char (re_ctx_p->bytecode_ctx_p,
(ecma_char_t) ecma_regexp_canonicalize (end,
re_ctx_p->flags & RE_FLAG_IGNORE_CASE));
} /* re_append_char_class */
/**
@@ -250,7 +267,7 @@ re_parse_char_class (re_compiler_ctx_t *re_ctx_p, /**< number of classes */
out_token_p->qmax = out_token_p->qmin = 1;
parser_ctx_p->classes_count = 0;
ecma_char_t start = LIT_CHAR_UNDEF;
lit_code_point_t start = LIT_CHAR_UNDEF;
bool is_range = false;
const bool is_char_class = (re_ctx_p->current_token.type == RE_TOK_START_CHAR_CLASS
|| re_ctx_p->current_token.type == RE_TOK_START_INV_CHAR_CLASS);
@@ -269,7 +286,7 @@ re_parse_char_class (re_compiler_ctx_t *re_ctx_p, /**< number of classes */
return ecma_raise_syntax_error (ECMA_ERR_MSG ("invalid character class, end of string"));
}
ecma_char_t ch = lit_utf8_read_next (&parser_ctx_p->input_curr_p);
lit_code_point_t ch = lit_utf8_read_next (&parser_ctx_p->input_curr_p);
if (ch == LIT_CHAR_RIGHT_SQUARE)
{
@@ -459,6 +476,20 @@ re_parse_char_class (re_compiler_ctx_t *re_ctx_p, /**< number of classes */
}
} /* ch == LIT_CHAR_BACKSLASH */
#if ENABLED (JERRY_ES2015)
if (re_ctx_p->flags & RE_FLAG_UNICODE
&& lit_is_code_point_utf16_high_surrogate (ch)
&& parser_ctx_p->input_curr_p < parser_ctx_p->input_end_p)
{
const ecma_char_t next_ch = lit_utf8_peek_next (parser_ctx_p->input_curr_p);
if (lit_is_code_point_utf16_low_surrogate (next_ch))
{
ch = lit_convert_surrogate_pair_to_code_point ((ecma_char_t) ch, next_ch);
lit_utf8_incr (&parser_ctx_p->input_curr_p);
}
}
#endif /* ENABLED (JERRY_ES2015) */
if (start != LIT_CHAR_UNDEF)
{
if (is_range)
@@ -559,8 +590,8 @@ re_parse_alternative (re_compiler_ctx_t *re_ctx_p, /**< RegExp compiler context
(unsigned int) re_ctx_p->current_token.qmax);
re_append_opcode (bc_ctx_p, RE_OP_CHAR);
re_append_char (bc_ctx_p, ecma_regexp_canonicalize ((ecma_char_t) re_ctx_p->current_token.value,
re_ctx_p->flags & RE_FLAG_IGNORE_CASE));
re_append_char (bc_ctx_p, (ecma_char_t) ecma_regexp_canonicalize ((ecma_char_t) re_ctx_p->current_token.value,
re_ctx_p->flags & RE_FLAG_IGNORE_CASE));
ret_value = re_insert_simple_iterator (re_ctx_p, new_atom_start_offset);
break;