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

Regex: Cleanup character class parsing a bit

Browse Source
This commit is contained in:
Maxime Coste 2017-10-06 19:51:41 +08:00
parent 236751cb84
commit 337e58d4f9
1 changed files with 15 additions and 20 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

25
src/regex_impl.cc
View File

@ -162,7 +162,6 @@ private:
case 'K': m_pos += 2; return new_node(ParsedRegex::ResetStart); case 'K': m_pos += 2; return new_node(ParsedRegex::ResetStart);
} }
break; break;
/* TODO: look ahead, look behind */
} }
return nullptr; return nullptr;
} }
@ -267,19 +266,18 @@ private:
} }
// CharacterClassEscape // CharacterClassEscape
for (auto& character_class : character_class_escapes) auto class_it = find_if(character_class_escapes,
{ [cp = to_lower(cp)](auto& c) { return c.cp == cp; });
if (character_class.cp == cp) if (class_it != std::end(character_class_escapes))
{ {
auto matcher_id = m_parsed_regex.matchers.size(); auto matcher_id = m_parsed_regex.matchers.size();
m_parsed_regex.matchers.push_back( m_parsed_regex.matchers.push_back(
[ctype = character_class.ctype ? wctype(character_class.ctype) : (wctype_t)0, [ctype = class_it->ctype ? wctype(class_it->ctype) : (wctype_t)0,
chars = character_class.additional_chars, neg = character_class.neg] (Codepoint cp) { chars = class_it->additional_chars, neg = is_upper(cp)] (Codepoint cp) {
return ((ctype != 0 and iswctype(cp, ctype)) or contains(chars, cp)) != neg; return ((ctype != 0 and iswctype(cp, ctype)) or contains(chars, cp)) != neg;
}); });
return new_node(ParsedRegex::Matcher, matcher_id); return new_node(ParsedRegex::Matcher, matcher_id);
} }
}
// CharacterEscape // CharacterEscape
struct { Codepoint name; Codepoint value; } control_escapes[] = { struct { Codepoint name; Codepoint value; } control_escapes[] = {
@ -323,14 +321,15 @@ private:
if (cp == '\\') if (cp == '\\')
{ {
auto it = find_if(character_class_escapes, auto it = find_if(character_class_escapes,
[cp = *m_pos](auto& t) { return t.cp == cp; }); [cp = to_lower(*m_pos)](auto& t) { return t.cp == cp; });
if (it != std::end(character_class_escapes)) if (it != std::end(character_class_escapes))
{ {
auto negative = is_upper(*m_pos);
if (it->ctype) if (it->ctype)
ctypes.push_back({wctype(it->ctype), not it->neg}); ctypes.push_back({wctype(it->ctype), not negative});
for (auto& c : it->additional_chars) // TODO: handle negative case for (auto& c : it->additional_chars)
{ {
if (it->neg) if (negative)
excluded.push_back((Codepoint)c); excluded.push_back((Codepoint)c);
else else
ranges.push_back({(Codepoint)c, (Codepoint)c}); ranges.push_back({(Codepoint)c, (Codepoint)c});
@ -497,13 +496,9 @@ private:
// For some reason Gcc fails to link if this is constexpr // For some reason Gcc fails to link if this is constexpr
const RegexParser::CharacterClassEscape RegexParser::character_class_escapes[8] = { const RegexParser::CharacterClassEscape RegexParser::character_class_escapes[8] = {
{ 'd', "digit", "", false }, { 'd', "digit", "", false },
{ 'D', "digit", "", true },
{ 'w', "alnum", "_", false }, { 'w', "alnum", "_", false },
{ 'W', "alnum", "_", true },
{ 's', "space", "", false }, { 's', "space", "", false },
{ 'S', "space", "", true },
{ 'h', nullptr, " \t", false }, { 'h', nullptr, " \t", false },
{ 'H', nullptr, " \t", true },
}; };
struct RegexCompiler struct RegexCompiler