xenia/home
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Regex: Support non greedy quantifiers

Browse Source
This commit is contained in:
Maxime Coste 2017-09-28 17:50:04 +08:00
parent e4004a7b7f
commit e96cd29f0e
1 changed files with 31 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

43
src/regex_impl.cc
View File

@ -42,6 +42,7 @@ struct ParsedRegex
RepeatMinMax, RepeatMinMax,
}; };
Type type = One; Type type = One;
bool greedy = true;
int min = -1, max = -1; int min = -1, max = -1;
bool allows_none() const bool allows_none() const
@ -346,11 +347,18 @@ private:
return res; return res;
}; };
auto check_greedy = [&]() {
if (at_end() or *m_pos != '?')
return true;
++m_pos;
return false;
};
switch (*m_pos) switch (*m_pos)
{ {
case '*': ++m_pos; return {ParsedRegex::Quantifier::RepeatZeroOrMore}; case '*': ++m_pos; return {ParsedRegex::Quantifier::RepeatZeroOrMore, check_greedy()};
case '+': ++m_pos; return {ParsedRegex::Quantifier::RepeatOneOrMore}; case '+': ++m_pos; return {ParsedRegex::Quantifier::RepeatOneOrMore, check_greedy()};
case '?': ++m_pos; return {ParsedRegex::Quantifier::Optional}; case '?': ++m_pos; return {ParsedRegex::Quantifier::Optional, check_greedy()};
case '{': case '{':
{ {
auto it = m_pos+1; auto it = m_pos+1;
@ -364,7 +372,7 @@ private:
if (*it++ != '}') if (*it++ != '}')
parse_error("expected closing bracket"); parse_error("expected closing bracket");
m_pos = it; m_pos = it;
return {ParsedRegex::Quantifier::RepeatMinMax, min, max}; return {ParsedRegex::Quantifier::RepeatMinMax, true, min, max};
} }
default: return {ParsedRegex::Quantifier::One}; default: return {ParsedRegex::Quantifier::One};
} }
@ -549,27 +557,32 @@ private:
Offset pos = m_program.bytecode.size(); Offset pos = m_program.bytecode.size();
Vector<Offset> goto_end_offsets; Vector<Offset> goto_end_offsets;
if (node->quantifier.allows_none()) auto& quantifier = node->quantifier;
if (quantifier.allows_none())
{ {
push_op(CompiledRegex::Split_PrioritizeParent); push_op(quantifier.greedy ? CompiledRegex::Split_PrioritizeParent
: CompiledRegex::Split_PrioritizeChild);
goto_end_offsets.push_back(alloc_offset()); goto_end_offsets.push_back(alloc_offset());
} }
auto inner_pos = compile_node_inner(node); auto inner_pos = compile_node_inner(node);
// Write the node multiple times when we have a min count quantifier // Write the node multiple times when we have a min count quantifier
for (int i = 1; i < node->quantifier.min; ++i) for (int i = 1; i < quantifier.min; ++i)
inner_pos = compile_node_inner(node); inner_pos = compile_node_inner(node);
if (node->quantifier.allows_infinite_repeat()) if (quantifier.allows_infinite_repeat())
{ {
push_op(CompiledRegex::Split_PrioritizeChild); push_op(quantifier.greedy ? CompiledRegex::Split_PrioritizeChild
: CompiledRegex::Split_PrioritizeParent);
get_offset(alloc_offset()) = inner_pos; get_offset(alloc_offset()) = inner_pos;
} }
// Write the node as an optional match for the min -> max counts // Write the node as an optional match for the min -> max counts
else for (int i = std::max(1, node->quantifier.min); // STILL UGLY ! else for (int i = std::max(1, quantifier.min); // STILL UGLY !
i < node->quantifier.max; ++i) i < quantifier.max; ++i)
{ {
push_op(CompiledRegex::Split_PrioritizeParent); push_op(quantifier.greedy ? CompiledRegex::Split_PrioritizeParent
: CompiledRegex::Split_PrioritizeChild);
goto_end_offsets.push_back(alloc_offset()); goto_end_offsets.push_back(alloc_offset());
compile_node_inner(node); compile_node_inner(node);
} }
@ -1015,6 +1028,12 @@ auto test_regex = UnitTest{[]{
kak_assert(StringView{vm.m_captures[0], vm.m_captures[1]} == "bar"); kak_assert(StringView{vm.m_captures[0], vm.m_captures[1]} == "bar");
kak_assert(not vm.exec("bar", true, true)); kak_assert(not vm.exec("bar", true, true));
} }
{
TestVM vm{R"((fo+?).*)"};
kak_assert(vm.exec("foooo", true, true));
kak_assert(StringView{vm.m_captures[2], vm.m_captures[3]} == "fo");
}
}}; }};
} }