LibRegex: Avoid generating ForkJumps when jumping to the next alt block

Fixes #2398.
This commit is contained in:
Ali Mohammad Pur 2024-11-17 16:01:15 +01:00 committed by Ali Mohammad Pur
parent b264d18ad1
commit 5a4d657a4e
Notes: github-actions[bot] 2024-11-17 19:13:35 +00:00
2 changed files with 11 additions and 3 deletions

View file

@ -1197,6 +1197,8 @@ void Optimizer::append_alternation(ByteCode& target, Span<ByteCode> alternatives
for (auto& patch : patch_locations) {
if (!patch.done && node_is(node, patch.source_ip)) {
auto value = static_cast<ByteCodeValueType>(target.size() - patch.target_ip - 1);
if (value == 0)
target[patch.target_ip - 1] = static_cast<ByteCodeValueType>(OpCodeId::Jump);
target[patch.target_ip] = value;
patch.done = true;
}

View file

@ -1074,7 +1074,7 @@ TEST_CASE(optimizer_char_class_lut)
TEST_CASE(optimizer_alternation)
{
Array tests {
// Pattern, Subject, Expected length
// Pattern, Subject, Expected length [0 == fail]
Tuple { "a|"sv, "a"sv, 1u },
Tuple { "a|a|a|a|a|a|a|a|a|b"sv, "a"sv, 1u },
Tuple { "ab|ac|ad|bc"sv, "bc"sv, 2u },
@ -1084,13 +1084,19 @@ TEST_CASE(optimizer_alternation)
Tuple { "^(\\d+|x)"sv, "42"sv, 2u },
// `Repeat' does not add its insn size to the jump target.
Tuple { "[0-9]{2}|[0-9]"sv, "92"sv, 2u },
// Don't ForkJump to the next instruction, rerunning it would produce the same result. see ladybird#2398.
Tuple { "(xxxxxxxxxxxxxxxxxxxxxxx|xxxxxxxxxxxxxxxxxxxxxxx)?b"sv, "xxxxxxxxxxxxxxxxxxxxxxx"sv, 0u },
};
for (auto& test : tests) {
Regex<ECMA262> re(test.get<0>());
auto result = re.match(test.get<1>());
EXPECT(result.success);
EXPECT_EQ(result.matches.first().view.length(), test.get<2>());
if (test.get<2>() != 0) {
EXPECT(result.success);
EXPECT_EQ(result.matches.first().view.length(), test.get<2>());
} else {
EXPECT(!result.success);
}
}
}