Implement extended named capture group identifiers (#90)

Perfectly reasonable and not at all uncommon regular expressions like
/(?<𝑓𝑜𝑥>fox).*(?<𝓓𝓸𝓰>dog)/ are now accepted.
This commit is contained in:
Ben Noordhuis 2023-11-19 11:01:38 +01:00 committed by GitHub
parent 20b3aca5a2
commit b56cbb143c
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
2 changed files with 14 additions and 11 deletions

View file

@ -1071,11 +1071,10 @@ static int re_is_simple_quantifier(const uint8_t *bc_buf, int bc_buf_len)
}
/* '*pp' is the first char after '<' */
static int re_parse_group_name(char *buf, int buf_size,
const uint8_t **pp, BOOL is_utf16)
static int re_parse_group_name(char *buf, int buf_size, const uint8_t **pp)
{
const uint8_t *p;
uint32_t c;
const uint8_t *p, *p1;
uint32_t c, d;
char *q;
p = *pp;
@ -1086,11 +1085,18 @@ static int re_parse_group_name(char *buf, int buf_size,
p++;
if (*p != 'u')
return -1;
c = lre_parse_escape(&p, is_utf16 * 2);
c = lre_parse_escape(&p, 2); // accept surrogate pairs
} else if (c == '>') {
break;
} else if (c >= 128) {
c = unicode_from_utf8(p, UTF8_CHAR_LEN_MAX, &p);
if (c >= 0xD800 && c <= 0xDBFF) {
d = unicode_from_utf8(p, UTF8_CHAR_LEN_MAX, &p1);
if (d >= 0xDC00 && d <= 0xDFFF) {
c = 0x10000 + 0x400 * (c - 0xD800) + (d - 0xDC00);
p = p1;
}
}
} else {
p++;
}
@ -1140,8 +1146,7 @@ static int re_parse_captures(REParseState *s, int *phas_named_captures,
/* potential named capture */
if (capture_name) {
p += 3;
if (re_parse_group_name(name, sizeof(name), &p,
s->is_utf16) == 0) {
if (re_parse_group_name(name, sizeof(name), &p) == 0) {
if (!strcmp(name, capture_name))
return capture_index;
}
@ -1314,7 +1319,7 @@ static int re_parse_term(REParseState *s, BOOL is_backward_dir)
} else if (p[2] == '<') {
p += 3;
if (re_parse_group_name(s->u.tmp_buf, sizeof(s->u.tmp_buf),
&p, s->is_utf16)) {
&p)) {
return re_parse_error(s, "invalid group name");
}
if (find_group_name(s, s->u.tmp_buf) > 0) {
@ -1378,7 +1383,7 @@ static int re_parse_term(REParseState *s, BOOL is_backward_dir)
}
p1 += 3;
if (re_parse_group_name(s->u.tmp_buf, sizeof(s->u.tmp_buf),
&p1, s->is_utf16)) {
&p1)) {
if (s->is_utf16 || re_has_named_captures(s))
return re_parse_error(s, "invalid group name");
else

View file

@ -37,8 +37,6 @@ test262/test/built-ins/Function/internals/Construct/derived-this-uninitialized-r
test262/test/built-ins/Function/internals/Construct/derived-this-uninitialized-realm.js:20: strict mode: Test262Error: Expected a ReferenceError but got a different error constructor with the same name
test262/test/built-ins/RegExp/lookahead-quantifier-match-groups.js:27: Test262Error: Expected [a, abc] and [a, undefined] to have the same contents. ? quantifier
test262/test/built-ins/RegExp/lookahead-quantifier-match-groups.js:27: strict mode: Test262Error: Expected [a, abc] and [a, undefined] to have the same contents. ? quantifier
test262/test/built-ins/RegExp/named-groups/non-unicode-property-names-valid.js:46: SyntaxError: invalid group name
test262/test/built-ins/RegExp/named-groups/non-unicode-property-names-valid.js:46: strict mode: SyntaxError: invalid group name
test262/test/built-ins/RegExp/prototype/Symbol.match/flags-tostring-error.js:22: Test262Error: Expected a CustomError but got a Test262Error
test262/test/built-ins/RegExp/prototype/Symbol.match/flags-tostring-error.js:22: strict mode: Test262Error: Expected a CustomError but got a Test262Error
test262/test/built-ins/RegExp/prototype/Symbol.match/get-flags-err.js:23: Test262Error: Expected a CustomError but got a Test262Error