RGBCube/serenity
1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-07-25 05:27:43 +00:00
Code Issues Activity Actions

LibRegex: Make the optimizer understand references and capture groups

Browse source 
Otherwise the fork in patterns like `(1+)\1` would be (incorrectly)
optimized away.
This commit is contained in:
Ali Mohammad Pur 2021-09-15 14:31:55 +04:30 • committed by Ali Mohammad Pur
parent 7de23aede2
commit 741886a4c4
2 changed files with 23 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

4
Tests/LibRegex/Regex.cpp
View file

@ -898,6 +898,10 @@ TEST_CASE(optimizer_atomic_groups)
// Alternative fuse // Alternative fuse
Tuple { "(abcfoo|abcbar|abcbaz).*x"sv, "abcbarx"sv, true }, Tuple { "(abcfoo|abcbar|abcbaz).*x"sv, "abcbarx"sv, true },
Tuple { "(a|a)"sv, "a"sv, true }, Tuple { "(a|a)"sv, "a"sv, true },
// ForkReplace shouldn't be applied where it would change the semantics
Tuple { "(1+)\\1"sv, "11"sv, true },
Tuple { "(1+)1"sv, "11"sv, true },
Tuple { "(1+)0"sv, "10"sv, true },
}; };
for (auto& test : tests) { for (auto& test : tests) {

21
Userland/Libraries/LibRegex/RegexOptimizer.cpp
View file

@ -104,6 +104,7 @@ typename Regex<Parser>::BasicBlockList Regex<Parser>::split_basic_blocks()
static bool block_satisfies_atomic_rewrite_precondition(ByteCode const& bytecode, Block const& repeated_block, Block const& following_block) static bool block_satisfies_atomic_rewrite_precondition(ByteCode const& bytecode, Block const& repeated_block, Block const& following_block)
{ {
Vector<Vector<CompareTypeAndValuePair>> repeated_values; Vector<Vector<CompareTypeAndValuePair>> repeated_values;
HashTable<size_t> active_capture_groups;
MatchState state; MatchState state;
for (state.instruction_position = repeated_block.start; state.instruction_position < repeated_block.end;) { for (state.instruction_position = repeated_block.start; state.instruction_position < repeated_block.end;) {
auto& opcode = bytecode.get_opcode(state); auto& opcode = bytecode.get_opcode(state);
@ -126,6 +127,12 @@ static bool block_satisfies_atomic_rewrite_precondition(ByteCode const& bytecode
case OpCodeId::Restore: case OpCodeId::Restore:
case OpCodeId::GoBack: case OpCodeId::GoBack:
return false; return false;
case OpCodeId::SaveRightCaptureGroup:
active_capture_groups.set(static_cast<OpCode_SaveRightCaptureGroup const&>(opcode).id());
break;
case OpCodeId::SaveLeftCaptureGroup:
active_capture_groups.set(static_cast<OpCode_SaveLeftCaptureGroup const&>(opcode).id());
break;
default: default:
break; break;
} }
@ -133,19 +140,29 @@ static bool block_satisfies_atomic_rewrite_precondition(ByteCode const& bytecode
state.instruction_position += opcode.size(); state.instruction_position += opcode.size();
} }
dbgln_if(REGEX_DEBUG, "Found {} entries in reference", repeated_values.size()); dbgln_if(REGEX_DEBUG, "Found {} entries in reference", repeated_values.size());
dbgln_if(REGEX_DEBUG, "Found {} active capture groups", active_capture_groups.size());
// 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'.
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;) {
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
case OpCodeId::SaveRightCaptureGroup:
active_capture_groups.set(static_cast<OpCode_SaveRightCaptureGroup const&>(opcode).id());
break;
case OpCodeId::SaveLeftCaptureGroup:
active_capture_groups.set(static_cast<OpCode_SaveLeftCaptureGroup const&>(opcode).id());
break;
case OpCodeId::Compare: { case OpCodeId::Compare: {
// We found a compare, let's see what it has. // We found a compare, let's see what it has.
auto compares = static_cast<OpCode_Compare const&>(opcode).flat_compares(); auto compares = static_cast<OpCode_Compare const&>(opcode).flat_compares();
if (compares.is_empty()) if (compares.is_empty())
break; break;
// If either side can match _anything_, fail. if (any_of(compares, [&](auto& compare) {
if (any_of(compares, [](auto& compare) { return compare.type == CharacterCompareType::AnyChar; })) return compare.type == CharacterCompareType::AnyChar
|| (compare.type == CharacterCompareType::Reference && active_capture_groups.contains(compare.value));
}))
return false; return false;
for (auto& repeated_value : repeated_values) { for (auto& repeated_value : repeated_values) {