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

Update RegExp unicode mode case folding to conform to the standard (#4004)

Browse Source 
JerryScript-DCO-1.0-Signed-off-by: Dániel Bátyai daniel.batyai@h-lab.eu
This commit is contained in:
Dániel Bátyai
2020-07-20 15:51:43 +02:00
committed by GitHub
parent 33359ac506
commit 321215fdbb
9 changed files with 284 additions and 84 deletions
Download Patch File Download Diff File Expand all files Collapse all files
jerry-core/ecma/operations/ecma-regexp-object.c
+25 -12
View File
@@ -403,30 +403,43 @@ lit_code_point_t
ecma_regexp_canonicalize_char (lit_code_point_t ch, /**< character */
bool unicode) /**< unicode */
{
if (JERRY_LIKELY (ch <= LIT_UTF8_1_BYTE_CODE_POINT_MAX))
#if ENABLED (JERRY_ESNEXT)
if (unicode)
{
if (ch >= LIT_CHAR_LOWERCASE_A && ch <= LIT_CHAR_LOWERCASE_Z)
/* In unicode mode the mappings contained in the CaseFolding.txt file should be used to canonicalize the character.
* These mappings generally correspond to the lowercase variant of the character, however there are some
* differences. In some cases the uppercase variant is used, in others the lowercase of the uppercase character is
* used, and there are also cases where the character has no case folding mapping even though it has upper/lower
* variants. Since lowercasing is the most common this is used as the default behaviour, and characters with
* differing behaviours are encoded in lookup tables. */
if (lit_char_fold_to_upper (ch))
{
return (ecma_char_t) (ch - (LIT_CHAR_LOWERCASE_A - LIT_CHAR_UPPERCASE_A));
ch = lit_char_to_upper_case (ch, NULL);
JERRY_ASSERT (ch != LIT_MULTIPLE_CU);
}
if (lit_char_fold_to_lower (ch))
{
ch = lit_char_to_lower_case (ch, NULL);
JERRY_ASSERT (ch != LIT_MULTIPLE_CU);
}
return ch;
}
#endif /* !ENABLED (JERRY_ESNEXT) */
JERRY_UNUSED (unicode);
lit_code_point_t cu = lit_char_to_upper_case (ch, NULL);
if (cu == LIT_MULTIPLE_CU)
if (ch <= LIT_UTF8_1_BYTE_CODE_POINT_MAX
|| (cu > LIT_UTF8_1_BYTE_CODE_POINT_MAX
&& cu != LIT_MULTIPLE_CU))
{
return ch;
return cu;
}
if (cu <= LIT_UTF8_1_BYTE_CODE_POINT_MAX && !unicode)
{
/* 6. */
return ch;
}
return cu;
return ch;
} /* ecma_regexp_canonicalize_char */
/**