diff options
| -rw-r--r-- | include/ricochet/render/renderer.hh | 8 | ||||
| -rw-r--r-- | src/core/browser.cc | 104 | ||||
| -rw-r--r-- | src/parser/html_lexer.cc | 6 | ||||
| -rw-r--r-- | src/parser/tree_builder.cc | 8 | ||||
| -rw-r--r-- | src/render/renderer.cc | 69 |
5 files changed, 160 insertions, 35 deletions
diff --git a/include/ricochet/render/renderer.hh b/include/ricochet/render/renderer.hh index f26a788..2a9acae 100644 --- a/include/ricochet/render/renderer.hh +++ b/include/ricochet/render/renderer.hh @@ -22,6 +22,12 @@ namespace ricochet::render { +struct RenderResult +{ + std::string text {}; + std::vector<std::string> links {}; +}; + class Renderer { public: @@ -37,7 +43,7 @@ public: * @brief Traverse the DOM tree and output the formatted result to the terminal using ANSI escape codes * @param node The root node of the DOM tree */ - std::string render( const parser::DomNode& node ) const; + RenderResult render( const parser::DomNode& node ) const; }; } // namespace ricochet::render
\ No newline at end of file diff --git a/src/core/browser.cc b/src/core/browser.cc index 4225986..8bd921c 100644 --- a/src/core/browser.cc +++ b/src/core/browser.cc @@ -23,17 +23,24 @@ #include "ricochet/render/renderer.hh" #include "ricochet/tui/terminal.hh" #include <cstddef> +#include <exception> #include <iostream> namespace ricochet::core { +struct PageData +{ + std::vector<std::string> lines; + std::vector<std::string> links; +}; + namespace { -std::vector<std::string> load_page( std::string_view url ) +PageData load_page( std::string_view url ) { const net::HttpClient client; auto response_result = client.fetch( url ); if ( !response_result.has_value() ) { - return { "[!] Failed to load: " + std::string( url ) }; + return { .lines = { "[!] Failed to load: " + std::string( url ) }, .links = {} }; } const parser::HtmlLexer lexer; @@ -41,20 +48,20 @@ std::vector<std::string> load_page( std::string_view url ) const render::Renderer renderer; const auto dom_root = builder.build( lexer.tokenize( response_result.value().body ) ); - const std::string text = renderer.render( dom_root ); + const render::RenderResult result = renderer.render( dom_root ); std::vector<std::string> lines; std::size_t start = 0; - while ( start < text.length() ) { - const std::size_t end = text.find( '\n', start ); + while ( start < result.text.length() ) { + const std::size_t end = result.text.find( '\n', start ); if ( end == std::string::npos ) { - lines.push_back( text.substr( start ) ); + lines.push_back( result.text.substr( start ) ); break; } - lines.push_back( text.substr( start, end - start ) ); + lines.push_back( result.text.substr( start, end - start ) ); start = end + 1; } - return lines; + return { .lines = lines, .links = result.links }; } void draw_view( const tui::Terminal& terminal, @@ -73,7 +80,7 @@ void draw_view( const tui::Terminal& terminal, std::cout << "~\r\n"; } } - std::cout << "\033[7m URL: " << url << " | [j/k] Scroll [/] Jump [q] Quit \033[0m"; + std::cout << "\033[7m URL: " << url << " | [j/k] Scroll [/] Jump [f] Follow [q] Quit \033[0m"; std::cout.flush(); } @@ -115,6 +122,79 @@ std::string get_url_input( const tui::Terminal& terminal ) return input; } +std::string get_link_input( const tui::Terminal& terminal ) +{ + auto [width, height] = terminal.get_size(); + terminal.move_cursor( height, 1 ); + std::cout << "\x1b[2K\033[7m Follow Link [#]: \033[0m "; + std::cout.flush(); + + std::string num_input; + terminal.show_cursor( true ); + while ( true ) { + const char ch = terminal.read_key(); + if ( ch == '\r' || ch == '\n' ) { + break; + } + if ( ch == 27 ) { + num_input.clear(); + break; + } + if ( ch == 127 || ch == 8 ) { + if ( !num_input.empty() ) { + num_input.pop_back(); + std::cout << "\b \b"; + std::cout.flush(); + } + } else if ( std::isdigit( static_cast<unsigned char>( ch ) ) ) { + num_input += ch; + std::cout << ch; + std::cout.flush(); + } + } + terminal.show_cursor( false ); + return num_input; +} + +void handle_follow_link( const tui::Terminal& terminal, + const std::vector<std::string>& links, + std::string& current_url, + bool& navigate ) +{ + const std::string num_input = get_link_input( terminal ); + if ( !num_input.empty() ) { + try { + const std::size_t index = std::stoul( num_input ); + if ( index > 0 && index <= links.size() ) { + std::string target = links[index - 1]; + + if ( target.starts_with( "/" ) ) { + if ( target.starts_with( "//" ) ) { + target = "https:" + target; // "//example.com" + } else { + const std::size_t slash_pos = current_url.find( '/', 8 ); // 跳过 https:// + const std::string domain + = ( slash_pos == std::string::npos ) ? current_url : current_url.substr( 0, slash_pos ); + target = domain + target; // "/about" -> "https://gnu.org/about" + } + } else if ( !target.starts_with( "http" ) ) { + const std::size_t last_slash = current_url.rfind( '/' ); + if ( last_slash > 7 ) { + target = current_url.substr( 0, last_slash + 1 ) + target; + } else { + target = current_url + "/" + target; + } + } + + current_url = target; + navigate = true; + } + } catch ( const std::exception& e ) { + (void)e; // NOLINT + } + } +} + } // namespace int Browser::run( std::string_view initial_url ) @@ -126,7 +206,9 @@ int Browser::run( std::string_view initial_url ) terminal.clear_screen(); while ( !current_url.empty() ) { - auto lines = load_page( current_url ); + auto page_data = load_page( current_url ); + const auto& lines = page_data.lines; + const auto& links = page_data.links; std::size_t scroll_y = 0; bool navigate = false; @@ -151,6 +233,8 @@ int Browser::run( std::string_view initial_url ) current_url = std::move( next_url ); navigate = true; } + } else if ( c == 'f' ) { + handle_follow_link( terminal, links, current_url, navigate ); } } } diff --git a/src/parser/html_lexer.cc b/src/parser/html_lexer.cc index 73591b3..2aa05b1 100644 --- a/src/parser/html_lexer.cc +++ b/src/parser/html_lexer.cc @@ -90,12 +90,8 @@ std::vector<HtmlToken> HtmlLexer::tokenize( // NOLINT(readability-convert-member 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 += c; } break; case LexerState::ClosingTagName: diff --git a/src/parser/tree_builder.cc b/src/parser/tree_builder.cc index 4f7f87e..64f841b 100644 --- a/src/parser/tree_builder.cc +++ b/src/parser/tree_builder.cc @@ -36,10 +36,14 @@ struct TokenVisitor void operator()( const TagOpenToken& token ) const { + const std::size_t space_pos = token.name.find( ' ' ); + const std::string base_name + = ( space_pos == std::string::npos ) ? token.name : token.name.substr( 0, space_pos ); + 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 ( base_name == "meta" || base_name == "link" || base_name == "img" || base_name == "br" || base_name == "hr" + || base_name == "input" || base_name == "base" || base_name == "col" ) { stack->back().children.push_back( std::move( new_node ) ); } else { diff --git a/src/render/renderer.cc b/src/render/renderer.cc index 2115ee1..1af40f1 100644 --- a/src/render/renderer.cc +++ b/src/render/renderer.cc @@ -17,7 +17,7 @@ */ #include "ricochet/render/renderer.hh" -#include <iostream> +#include <array> namespace ricochet::render { @@ -42,51 +42,86 @@ std::string decode_entities( std::string text ) // NOLINT(readability-convert-me return text; } -void render_node( const parser::DomNode& node, std::string& output ) // NOLINT(misc-no-recursion) +std::string extract_href( const std::string& tag_name ) { - if ( node.tag_name == "head" || node.tag_name == "style" || node.tag_name == "script" || node.tag_name == "meta" - || node.tag_name == "title" || node.tag_name == "link" ) { + std::size_t pos = tag_name.find( "href=\"" ); + if ( pos != std::string::npos ) { + pos += 6; + const std::size_t end = tag_name.find( '\"', pos ); + if ( end != std::string::npos ) { + return tag_name.substr( pos, end - pos ); + } + } + pos = tag_name.find( "href='" ); + if ( pos != std::string::npos ) { + pos += 6; + const std::size_t end = tag_name.find( '\'', pos ); + if ( end != std::string::npos ) { + return tag_name.substr( pos, end - pos ); + } + } + return ""; +} + +void render_node( const parser::DomNode& node, // NOLINT(misc-no-recursion) + std::string& output, + std::vector<std::string>& links ) +{ + const std::size_t space_pos = node.tag_name.find( ' ' ); + const std::string base_tag + = ( space_pos == std::string::npos ) ? node.tag_name : node.tag_name.substr( 0, space_pos ); + + if ( base_tag == "head" || base_tag == "style" || base_tag == "script" || base_tag == "meta" + || base_tag == "title" || base_tag == "link" ) { return; } - if ( node.tag_name.empty() ) { + if ( base_tag.empty() ) { output += decode_entities( node.text_content ); return; } - const bool is_header - = ( node.tag_name == "h1" || node.tag_name == "h2" || node.tag_name == "h3" || node.tag_name == "h4" ); + const bool is_header = ( base_tag == "h1" || base_tag == "h2" || base_tag == "h3" || base_tag == "h4" ); + const bool is_link = ( base_tag == "a" ); if ( is_header ) { output += "\n\n\033[1;31m"; - } else if ( node.tag_name == "a" ) { - std::cout << "\033[4;34m"; - } else if ( node.tag_name == "p" || node.tag_name == "div" || is_header ) { + } else if ( is_link ) { + output += "\033[4;34m"; // 【核心修复】：必须是 output += ，绝对不能用 std::cout + } else if ( base_tag == "p" || base_tag == "div" || is_header ) { if ( !output.empty() && output.back() != '\n' ) { output += "\n"; } } for ( const auto& child : node.children ) { - render_node( child, output ); + render_node( child, output, links ); } - if ( is_header || node.tag_name == "a" ) { + + if ( is_header ) { + output += "\033[0m"; + } else if ( is_link ) { output += "\033[0m"; + const std::string href = extract_href( node.tag_name ); + if ( !href.empty() && !href.starts_with( "javascript:" ) && !href.starts_with( "#" ) ) { + links.push_back( href ); + output += " \033[7m[" + std::to_string( links.size() ) + "]\033[0m"; // 打印数字标签！ + } } - if ( node.tag_name == "h1" || node.tag_name == "p" || node.tag_name == "div" ) { + if ( base_tag == "h1" || base_tag == "p" || base_tag == "div" ) { output += "\n"; } } } // namespace -std::string Renderer::render( // NOLINT(readability-convert-member-functions-to-static) +RenderResult Renderer::render( // NOLINT(readability-convert-member-functions-to-static) const parser::DomNode& node ) const { - std::string output; - render_node( node, output ); - return output; + RenderResult result; + render_node( node, result.text, result.links ); + return result; } } // namespace ricochet::render
\ No newline at end of file |