aboutsummaryrefslogtreecommitdiff
path: root/src/parser
diff options
context:
space:
mode:
Diffstat (limited to 'src/parser')
-rw-r--r--src/parser/html_lexer.cc122
-rw-r--r--src/parser/tree_builder.cc87
2 files changed, 101 insertions, 108 deletions
diff --git a/src/parser/html_lexer.cc b/src/parser/html_lexer.cc
index bbf8c57..73591b3 100644
--- a/src/parser/html_lexer.cc
+++ b/src/parser/html_lexer.cc
@@ -21,99 +21,101 @@
#include <cstdint>
#include <string_view>
-
namespace ricochet::parser {
namespace {
-enum class LexerState : std::uint8_t {
- Data, // Reading plain text
- TagOpen, // Just encountered '<'
- TagName, // Reading the name of an opening tag (e.g., "div")
- ClosingTagName, // Reading the name of a closing tag (e.g., "/div")
- SwallowAttributes // Ignore attributes mode (ignore until encountering '>')
+enum class LexerState : std::uint8_t
+{
+ Data, // Reading plain text
+ TagOpen, // Just encountered '<'
+ TagName, // Reading the name of an opening tag (e.g., "div")
+ ClosingTagName, // Reading the name of a closing tag (e.g., "/div")
+ SwallowAttributes // Ignore attributes mode (ignore until encountering '>')
};
-void emit_text(std::string& buffer, std::vector<HtmlToken>& tokens) {
- if (buffer.empty()) {
- return; // 提前返回，减少 if 嵌套层级
- }
+void emit_text( std::string& buffer, std::vector<HtmlToken>& tokens )
+{
+ if ( buffer.empty() ) {
+ return; // 提前返回，减少 if 嵌套层级
+ }
- bool all_whitespace = true;
- for (const char c : buffer) {
- if (!std::isspace(static_cast<unsigned char>(c))) {
- all_whitespace = false;
- break;
- }
- }
-
- if (!all_whitespace) {
- tokens.emplace_back(TextToken{buffer});
+ bool all_whitespace = true;
+ for ( const char c : buffer ) {
+ if ( !std::isspace( static_cast<unsigned char>( c ) ) ) {
+ all_whitespace = false;
+ break;
}
- buffer.clear();
+ }
+
+ if ( !all_whitespace ) {
+ tokens.emplace_back( TextToken { buffer } );
+ }
+ buffer.clear();
}
} // namespace
-std::vector<HtmlToken> HtmlLexer::tokenize( std::string_view html ) const // NOLINT(readability-convert-member-functions-to-static)
+std::vector<HtmlToken> HtmlLexer::tokenize( // NOLINT(readability-convert-member-functions-to-static)
+ std::string_view html ) const
{
std::vector<HtmlToken> tokens;
LexerState state = LexerState::Data;
- std::string buffer{};
+ std::string buffer {};
- for (const char c : html) {
- switch (state) {
- case LexerState::Data:
- if (c == '<') {
- emit_text(buffer, tokens);
- state = LexerState::TagOpen;
+ for ( const char c : html ) {
+ switch ( state ) {
+ case LexerState::Data:
+ if ( c == '<' ) {
+ emit_text( buffer, tokens );
+ state = LexerState::TagOpen;
} else {
- buffer += c;
+ buffer += c;
}
break;
- case LexerState::TagOpen:
- if (c == '/') {
- state = LexerState::ClosingTagName;
- } else if (std::isalpha(static_cast<unsigned char>(c))) {
- buffer.push_back(c);
- state = LexerState::TagName;
+ case LexerState::TagOpen:
+ if ( c == '/' ) {
+ state = LexerState::ClosingTagName;
+ } else if ( std::isalpha( static_cast<unsigned char>( c ) ) ) {
+ buffer.push_back( c );
+ state = LexerState::TagName;
} else {
- buffer.push_back('<');
- buffer.push_back(c);
- state = LexerState::Data;
+ buffer.push_back( '<' );
+ buffer.push_back( c );
+ state = LexerState::Data;
}
break;
- case LexerState::TagName:
- if (c == '>') {
- tokens.emplace_back(TagOpenToken{buffer});
- buffer.clear();
- state = LexerState::Data;
- } else if (std::isspace(static_cast<unsigned char>(c))) {
- tokens.emplace_back(TagOpenToken{buffer});
- buffer.clear();
- state = LexerState::SwallowAttributes;
+ case LexerState::TagName:
+ if ( c == '>' ) {
+ tokens.emplace_back( TagOpenToken { buffer } );
+ buffer.clear();
+ state = LexerState::Data;
+ } else if ( std::isspace( static_cast<unsigned char>( c ) ) ) {
+ tokens.emplace_back( TagOpenToken { buffer } );
+ buffer.clear();
+ state = LexerState::SwallowAttributes;
} else {
- buffer += static_cast<char>(std::tolower(static_cast<unsigned char>(c)));
+ buffer += static_cast<char>( std::tolower( static_cast<unsigned char>( c ) ) );
}
break;
- case LexerState::ClosingTagName:
- if (c == '>') {
- tokens.emplace_back(TagCloseToken{buffer});
- buffer.clear();
- state = LexerState::Data;
+ case LexerState::ClosingTagName:
+ if ( c == '>' ) {
+ tokens.emplace_back( TagCloseToken { buffer } );
+ buffer.clear();
+ state = LexerState::Data;
} else {
- buffer += static_cast<char>(std::tolower(static_cast<unsigned char>(c)));
+ buffer += static_cast<char>( std::tolower( static_cast<unsigned char>( c ) ) );
}
break;
- case LexerState::SwallowAttributes:
- if (c == '>') {
- state = LexerState::Data;
+ case LexerState::SwallowAttributes:
+ if ( c == '>' ) {
+ state = LexerState::Data;
}
break;
}
}
- emit_text(buffer, tokens);
+ emit_text( buffer, tokens );
return tokens;
}
diff --git a/src/parser/tree_builder.cc b/src/parser/tree_builder.cc
index 51a8a51..4f7f87e 100644
--- a/src/parser/tree_builder.cc
+++ b/src/parser/tree_builder.cc
@@ -24,67 +24,58 @@
namespace ricochet::parser {
namespace {
-struct TokenVisitor {
- std::vector<DomNode>* stack;
+struct TokenVisitor
+{
+ std::vector<DomNode>* stack;
- void operator()(const TextToken& token) const {
- stack->back().children.emplace_back(
- DomNode{
- .tag_name = "",
- .text_content = token.content,
- .children = {}
- }
- );
- }
+ void operator()( const TextToken& token ) const
+ {
+ stack->back().children.emplace_back(
+ DomNode { .tag_name = "", .text_content = token.content, .children = {} } );
+ }
+
+ void operator()( const TagOpenToken& token ) const
+ {
+ DomNode new_node { .tag_name = token.name, .text_content = "", .children = {} };
- void operator()(const TagOpenToken& token) const {
- DomNode new_node{
- .tag_name = token.name,
- .text_content = "",
- .children = {}
- };
+ if ( token.name == "meta" || token.name == "link" || token.name == "img" || token.name == "br"
+ || token.name == "hr" || token.name == "input" || token.name == "base" || token.name == "col" ) {
- if (token.name == "meta" || token.name == "link" || token.name == "img" ||
- token.name == "br" || token.name == "hr" || token.name == "input" ||
- token.name == "base" || token.name == "col") {
-
- stack->back().children.push_back(std::move(new_node));
- } else {
- stack->push_back(std::move(new_node));
- }
+ stack->back().children.push_back( std::move( new_node ) );
+ } else {
+ stack->push_back( std::move( new_node ) );
}
+ }
- void operator()(const TagCloseToken& /*token*/) const {
- if (stack->size() > 1) {
- DomNode completed_node = std::move(stack->back());
- stack->pop_back();
- stack->back().children.push_back(std::move(completed_node));
- }
+ void operator()( const TagCloseToken& /*token*/ ) const
+ {
+ if ( stack->size() > 1 ) {
+ DomNode completed_node = std::move( stack->back() );
+ stack->pop_back();
+ stack->back().children.push_back( std::move( completed_node ) );
}
+ }
};
} // namespace
-DomNode TreeBuilder::build(const std::vector<HtmlToken>& tokens) const // NOLINT(readability-convert-member-functions-to-static)
+DomNode TreeBuilder::build( // NOLINT(readability-convert-member-functions-to-static)
+ const std::vector<HtmlToken>& tokens ) const
{
- std::vector<DomNode> stack;
- stack.emplace_back(DomNode{
- .tag_name = "Document",
- .text_content = "",
- .children = {}
- });
+ std::vector<DomNode> stack;
+ stack.emplace_back( DomNode { .tag_name = "Document", .text_content = "", .children = {} } );
- TokenVisitor visitor{&stack};
- for (const auto& token : tokens) {
- std::visit(visitor, token);
- }
+ TokenVisitor visitor { &stack };
+ for ( const auto& token : tokens ) {
+ std::visit( visitor, token );
+ }
- while (stack.size() > 1) {
- DomNode completed_node = std::move(stack.back());
- stack.pop_back();
- stack.back().children.push_back(std::move(completed_node));
- }
+ while ( stack.size() > 1 ) {
+ DomNode completed_node = std::move( stack.back() );
+ stack.pop_back();
+ stack.back().children.push_back( std::move( completed_node ) );
+ }
- return std::move(stack.front());
+ return std::move( stack.front() );
}
} // namespace ricochet::parser \ No newline at end of file
generated by cgit v1.2.3 (git 2.46.0) at 2026-10-29 08:02:41 +0000