RGBCube/serenity
1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-07-25 06:37:43 +00:00
Code Issues Activity Actions

LibRegex: Do not treat repeats followed by fallthroughs as atomic

Browse source
This commit is contained in:
Ali Mohammad Pur 2023-06-14 01:02:40 +03:30 • committed by Andreas Kling
parent 31c9fe5c32
commit b1ca2e5e39
2 changed files with 18 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

3
Tests/LibRegex/Regex.cpp
View file

@ -995,6 +995,9 @@ TEST_CASE(optimizer_atomic_groups)
Tuple { "[^x]+y"sv, "ay"sv, true }, Tuple { "[^x]+y"sv, "ay"sv, true },
// .+ should not be rewritten here, as it's followed by something that would be matched by `.`. // .+ should not be rewritten here, as it's followed by something that would be matched by `.`.
Tuple { ".+(a|b|c)"sv, "xxa"sv, true }, Tuple { ".+(a|b|c)"sv, "xxa"sv, true },
// (b+)(b+) produces an intermediate block with no matching ops, the optimiser should ignore that block when looking for following matches and correctly detect the overlap between (b+) and (b+).
// note that the second loop may be rewritten to a ForkReplace, but the first loop should not be rewritten.
Tuple { "(b+)(b+)"sv, "bbb"sv, true },
}; };
for (auto& test : tests) { for (auto& test : tests) {

16
Userland/Libraries/LibRegex/RegexOptimizer.cpp
View file

@ -351,7 +351,9 @@ static AtomicRewritePreconditionResult block_satisfies_atomic_rewrite_preconditi
bool following_block_has_at_least_one_compare = false; bool following_block_has_at_least_one_compare = false;
// Find the first compare in the following block, it must NOT match any of the values in `repeated_values'. // Find the first compare in the following block, it must NOT match any of the values in `repeated_values'.
auto final_instruction = following_block.start;
for (state.instruction_position = following_block.start; state.instruction_position < following_block.end;) { for (state.instruction_position = following_block.start; state.instruction_position < following_block.end;) {
final_instruction = state.instruction_position;
auto& opcode = bytecode.get_opcode(state); auto& opcode = bytecode.get_opcode(state);
switch (opcode.opcode_id()) { switch (opcode.opcode_id()) {
// Note: These have to exist since we're effectively repeating the following block as well // Note: These have to exist since we're effectively repeating the following block as well
@ -399,6 +401,18 @@ static AtomicRewritePreconditionResult block_satisfies_atomic_rewrite_preconditi
state.instruction_position += opcode.size(); state.instruction_position += opcode.size();
} }
// If the following block falls through, we can't rewrite it.
state.instruction_position = final_instruction;
switch (bytecode.get_opcode(state).opcode_id()) {
case OpCodeId::Jump:
case OpCodeId::JumpNonEmpty:
case OpCodeId::ForkJump:
case OpCodeId::ForkReplaceJump:
break;
default:
return AtomicRewritePreconditionResult::NotSatisfied;
}
if (following_block_has_at_least_one_compare) if (following_block_has_at_least_one_compare)
return AtomicRewritePreconditionResult::SatisfiedWithProperHeader; return AtomicRewritePreconditionResult::SatisfiedWithProperHeader;
return AtomicRewritePreconditionResult::SatisfiedWithEmptyHeader; return AtomicRewritePreconditionResult::SatisfiedWithEmptyHeader;
@ -717,7 +731,7 @@ void Optimizer::append_alternation(ByteCode& target, Span<ByteCode> alternatives
size_t i = 0; size_t i = 0;
for (auto& entry : alternatives) { for (auto& entry : alternatives) {
auto& blocks = basic_blocks[i]; auto& blocks = basic_blocks[i++];
auto& block = blocks[block_index]; auto& block = blocks[block_index];
auto end = block_index + 1 == blocks.size() ? block.end : blocks[block_index + 1].start; auto end = block_index + 1 == blocks.size() ? block.end : blocks[block_index + 1].start;
state.instruction_position = block.start; state.instruction_position = block.start;