Skip to content

<regex>: Clean up parsing logic for quantifiers #5253

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 6 commits into from
Mar 25, 2025
Merged

<regex>: Clean up parsing logic for quantifiers #5253

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
efd32f7
`<regex>`: Clean up parsing logic for quantifiers
muellerj2 Jan 26, 2025
dd0099e
add accidentally missing change in prior commit
muellerj2 Jan 26, 2025
0808bb0
Merge branch 'main' into regex-cleanup-quantifier-parsing
StephanTLavavej Mar 20, 2025
8720e15
We can remove the `_Fl_final` name, leaving a comment about the gap.
StephanTLavavej Mar 20, 2025
8814cc5
Merge branch 'main' into regex-cleanup-quantifier-parsing
StephanTLavavej Mar 25, 2025
d8c6371
Fix stealth merge conflict: `_Add_char` => `_Add_char2`
StephanTLavavej Mar 25, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
33 changes: 13 additions & 20 deletions stl/inc/regex
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1202,8 +1202,7 @@ enum _Node_flags : int { // flags for nfa nodes with special properties
_Fl_none = 0x00,
_Fl_negate = 0x01,
_Fl_greedy = 0x02,
_Fl_final = 0x04,
_Fl_longest = 0x08
_Fl_longest = 0x08 // TRANSITION, ABI: 0x04 is unused; the parser previously marked some nodes with it
};

_BITMASK_OPS(_EMPTY_ARGUMENT, _Node_flags)
Expand Down Expand Up @@ -1521,7 +1520,7 @@ public:
void _Add_eol();
void _Add_wbound();
void _Add_dot();
void _Add_char(_Elem _Ch);
void _Add_char2(_Elem _Ch);
void _Add_class();
void _Add_char_to_class(_Elem _Ch);
void _Add_range2(_Elem, _Elem);
Expand All @@ -1536,9 +1535,8 @@ public:
void _Add_backreference(unsigned int _Idx);
_Node_base* _Begin_if(_Node_base* _Start);
void _Else_if(_Node_base*, _Node_base*);
void _Add_rep(int _Min, int _Max, bool _Greedy);
void _Add_rep2(int _Min, int _Max, bool _Greedy);
void _Negate();
void _Mark_final();
_Root_node* _End_pattern();

private:
Expand Down Expand Up @@ -2773,11 +2771,6 @@ void _Builder<_FwdIt, _Elem, _RxTraits>::_Negate() { // set flag
_Current->_Flags ^= _Fl_negate;
}

template <class _FwdIt, class _Elem, class _RxTraits>
void _Builder<_FwdIt, _Elem, _RxTraits>::_Mark_final() { // set flag
_Current->_Flags |= _Fl_final;
}

template <class _FwdIt, class _Elem, class _RxTraits>
_Node_base* _Builder<_FwdIt, _Elem, _RxTraits>::_Getmark() const {
return _Current;
Expand Down Expand Up @@ -2840,8 +2833,8 @@ void _Builder<_FwdIt, _Elem, _RxTraits>::_Add_str_node() { // add string node
}

template <class _FwdIt, class _Elem, class _RxTraits>
void _Builder<_FwdIt, _Elem, _RxTraits>::_Add_char(_Elem _Ch) { // append character
if (_Current->_Kind != _N_str || (_Current->_Flags & _Fl_final)) {
void _Builder<_FwdIt, _Elem, _RxTraits>::_Add_char2(_Elem _Ch) { // append character
if (_Current->_Kind != _N_str) {
_Add_str_node();
}

Expand Down Expand Up @@ -3090,11 +3083,12 @@ void _Builder<_FwdIt, _Elem, _RxTraits>::_Else_if(_Node_base* _Start, _Node_base
}

template <class _FwdIt, class _Elem, class _RxTraits>
void _Builder<_FwdIt, _Elem, _RxTraits>::_Add_rep(int _Min, int _Max, bool _Greedy) { // add repeat node
void _Builder<_FwdIt, _Elem, _RxTraits>::_Add_rep2(int _Min, int _Max, bool _Greedy) { // add repeat node
if (_Current->_Kind == _N_str
&& static_cast<_Node_str<_Elem>*>(_Current)->_Data._Size() != 1) { // move final character to new string node
_Node_str<_Elem>* _Node = static_cast<_Node_str<_Elem>*>(_Current);
_Add_char(_Node->_Data._Del());
_Add_str_node();
_Add_char2(_Node->_Data._Del());
}

_Node_base* _Pos = _Current;
Expand Down Expand Up @@ -4419,7 +4413,7 @@ void _Parser<_FwdIt, _Elem, _RxTraits>::_AtomEscape() { // check for valid atom
if (!(_L_flags & _L_bzr_chr)) {
_Error(regex_constants::error_escape);
} else {
_Nfa._Add_char(static_cast<_Elem>(_Val));
_Nfa._Add_char2(static_cast<_Elem>(_Val));
}
} else if (_Grp_idx < static_cast<size_t>(_Val) || _Finished_grps.size() <= static_cast<size_t>(_Val)
|| !_Finished_grps[static_cast<size_t>(_Val)]) {
Expand All @@ -4428,7 +4422,7 @@ void _Parser<_FwdIt, _Elem, _RxTraits>::_AtomEscape() { // check for valid atom
_Nfa._Add_backreference(static_cast<size_t>(_Val));
}
} else if (_CharacterEscape()) {
_Nfa._Add_char(static_cast<_Elem>(_Val));
_Nfa._Add_char2(static_cast<_Elem>(_Val));
} else if (!(_L_flags & _L_esc_wsd) || !_CharacterClassEscape(true)) {
_Error(regex_constants::error_escape);
}
Expand Down Expand Up @@ -4471,14 +4465,13 @@ void _Parser<_FwdIt, _Elem, _RxTraits>::_Quantifier() { // check for quantifier
}
}

_Nfa._Mark_final();
_Next();
const bool _Greedy = !(_L_flags & _L_ngr_rep) || _Mchar != _Meta_query;
if (!_Greedy) { // add non-greedy repeat node
_Next();
}

_Nfa._Add_rep(_Min, _Max, _Greedy);
_Nfa._Add_rep2(_Min, _Max, _Greedy);
}

template <class _FwdIt, class _Elem, class _RxTraits>
Expand Down Expand Up @@ -4519,7 +4512,7 @@ bool _Parser<_FwdIt, _Elem, _RxTraits>::_Alternative() { // check for valid alte
_Nfa._Add_bol();
_Next();
if ((_L_flags & _L_star_beg) && _Mchar == _Meta_star && !_Found) {
_Nfa._Add_char(_Char);
_Nfa._Add_char2(_Char);
_Next();
} else {
_Quant = false;
Expand All @@ -4536,7 +4529,7 @@ bool _Parser<_FwdIt, _Elem, _RxTraits>::_Alternative() { // check for valid alte
} else if (_Mchar == _Meta_rsq && !(_L_flags & _L_paren_bal)) {
_Error(regex_constants::error_brack);
} else { // add character
_Nfa._Add_char(_Char);
_Nfa._Add_char2(_Char);
_Next();
}

Expand Down
21 changes: 21 additions & 0 deletions tests/std/tests/VSO_0000000_regex_use/test.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1151,6 +1151,26 @@ void test_gh_5214() {
}
}

void test_gh_5253() {
// GH-5253 cleaned up parsing logic for quantifiers that were applied to single characters
g_regexTester.should_match("abbb", "ab*");
g_regexTester.should_not_match("abab", "ab*");
g_regexTester.should_match("abbb", "(a)b*");
g_regexTester.should_not_match("abab", "(a)b*");
g_regexTester.should_match("abbb", "a(b)*");
g_regexTester.should_not_match("abab", "a(b)*");
g_regexTester.should_match("abbb", "(a)(b)*");
g_regexTester.should_not_match("abab", "(a)(b)*");
g_regexTester.should_not_match("abbb", "(ab)*");
g_regexTester.should_match("abab", "(ab)*");
g_regexTester.should_not_match("abbb", "(?:ab)*");
g_regexTester.should_match("abab", "(?:ab)*");
g_regexTester.should_match("aaaa", "a*");
g_regexTester.should_not_match("b", "a*");
g_regexTester.should_match("", "()*");
g_regexTester.should_not_match("a", "()*");
}

int main() {
test_dev10_449367_case_insensitivity_should_work();
test_dev11_462743_regex_collate_should_not_disable_regex_icase();
Expand Down Expand Up @@ -1187,6 +1207,7 @@ int main() {
test_gh_5167();
test_gh_5192();
test_gh_5214();
test_gh_5253();

return g_regexTester.result();
}