diff options
| -rw-r--r-- | include/error.h | 29 | ||||
| -rw-r--r-- | module-test/lexer_test.cc | 70 | ||||
| -rw-r--r-- | src/lexer/ast.cc | 13 | ||||
| -rw-r--r-- | src/lexer/token.cc | 1 |
4 files changed, 76 insertions, 37 deletions
diff --git a/include/error.h b/include/error.h new file mode 100644 index 0000000..d8be4e5 --- /dev/null +++ b/include/error.h @@ -0,0 +1,29 @@ +#pragma once + +#include <stdexcept> + +namespace ov4 +{ + +enum class error_code +{ + + // tokenizer + TOKEN_UNMATCHED_BRACKET, + + // AST + AST_MISS_LEFT_CHILD, + AST_MISS_RIGHT_CHILD, + AST_EXIST_LEFT_CHILLD, + AST_EXIST_RIGHT_CHILD, + +}; + +class T_error : public std::runtime_error { +public: + error_code code; + T_error(error_code k, std::string msg) + : std::runtime_error(std::move(msg)), code(k) {} +}; + +} diff --git a/module-test/lexer_test.cc b/module-test/lexer_test.cc index d3f5a14..c94a4ef 100644 --- a/module-test/lexer_test.cc +++ b/module-test/lexer_test.cc @@ -16,6 +16,7 @@ g++ \ #include <cstring> #include <vector> #include <print> +#include <stdexcept> #include "magic_enum/magic_enum.hpp" @@ -23,6 +24,7 @@ g++ \ #include "lexer/enum_type.h" #include "lexer/token.h" #include "global.h" +#include "error.h" using namespace std; using namespace ov4; @@ -106,40 +108,50 @@ int main() return 0; } - - T_lexer lexer_instance; - - lexer_instance.tokenizer(s); - cout << "---------- tokenizer ----------" << endl; - for (auto c : lexer_instance.token) + try { - if (c.token_type == LOGIC_AND) - println("{} LOGIC_AND", c.bracket_depth); - else if (c.token_type == LOGIC_OR) - println("{} LOGIC_OR", c.bracket_depth); - else if (c.token_type == ASYNC) - println("{} ASYNC", c.bracket_depth); - else if (c.token_type == PIPE) - println("{} PIPE", c.bracket_depth); - else if (c.token_type == LEFT_BRACKET) - println("{} LEFT_BRACKET", c.bracket_depth); - else if (c.token_type == RIGHT_BRACKET) - println("{} RIGHT_BRACKET", c.bracket_depth); - else if (c.token_type == TEXT) - println("{} TEXT .{}.", c.bracket_depth, c.text); - } + T_lexer lexer_instance; + lexer_instance.tokenizer(s); + cout << "---------- tokenizer ----------" << endl; + for (auto c : lexer_instance.token) + { + if (c.token_type == LOGIC_AND) + println("{} LOGIC_AND", c.bracket_depth); + else if (c.token_type == LOGIC_OR) + println("{} LOGIC_OR", c.bracket_depth); + else if (c.token_type == ASYNC) + println("{} ASYNC", c.bracket_depth); + else if (c.token_type == PIPE) + println("{} PIPE", c.bracket_depth); + else if (c.token_type == LEFT_BRACKET) + println("{} LEFT_BRACKET", c.bracket_depth); + else if (c.token_type == RIGHT_BRACKET) + println("{} RIGHT_BRACKET", c.bracket_depth); + else if (c.token_type == TEXT) + println("{} TEXT .{}.", c.bracket_depth, c.text); + } - cout << "---------- parse ----------" << endl; + cout << "---------- parse ----------" << endl; - lexer_instance.parse(0, lexer_instance.token.size(), lexer_instance.alloc_ast()); + lexer_instance.parse(0, lexer_instance.token.size(), lexer_instance.alloc_ast()); - cout << "---------- AST ----------" << endl; - for (size_t i = 0; i < lexer_instance.ast.size(); i++) + cout << "---------- AST ----------" << endl; + for (size_t i = 0; i < lexer_instance.ast.size(); i++) + { + print("{}: type: {} left: {} right: {}", i, magic_enum::enum_name(lexer_instance.ast[i].token_type), lexer_instance.ast[i].left, lexer_instance.ast[i].right); + if (lexer_instance.ast[i].token_type == TEXT) + print(" command: {}", lexer_instance.ast[i].command_text); + println(); + } + } + catch(const T_error &e) { - print("{}: type: {} left: {} right: {}", i, magic_enum::enum_name(lexer_instance.ast[i].token_type), lexer_instance.ast[i].left, lexer_instance.ast[i].right); - if (lexer_instance.ast[i].token_type == TEXT) - print(" command: {}", lexer_instance.ast[i].command_text); - println(); + std::cerr << e.what() << '\n'; } + + + + + return 0; } diff --git a/src/lexer/ast.cc b/src/lexer/ast.cc index 755e909..4fbb62a 100644 --- a/src/lexer/ast.cc +++ b/src/lexer/ast.cc @@ -9,6 +9,7 @@ #include "lexer/token.h" #include "lexer/symbol_property.h" #include "util/io.h" +#include "error.h" using namespace std; @@ -112,29 +113,25 @@ void T_lexer::parse(size_t cmd_begin, size_t cmd_end, size_t ast_vec) // miss left if (symbol_property[token_str].left == EXIST && cmd_begin == found_pos) { - loggerln("parse: missing left child"); - exit(-1); // TODO + throw T_error{error_code::AST_MISS_LEFT_CHILD, "parse: missing left child"}; } // miss right if (symbol_property[token_str].right == EXIST && cmd_end-1 == found_pos) { - loggerln("parse: missing right child"); - exit(-1); // TODO + throw T_error{error_code::AST_MISS_RIGHT_CHILD, "parse: missing right child"}; } // should not exist left if (symbol_property[token_str].left == NOT_EXIST && cmd_begin != found_pos) { - loggerln("parse: left child should NOT exist"); - exit(-1); // TODO + throw T_error{error_code::AST_EXIST_LEFT_CHILLD, "parse: left child should NOT exist"}; } // should not exist right if (symbol_property[token_str].right == NOT_EXIST && cmd_end-1 != found_pos) { - loggerln("parse: right child should NOT exist"); - exit(-1); // TODO + throw T_error{error_code::AST_EXIST_LEFT_CHILLD, "parse: right child should NOT exist"}; } size_t l = alloc_ast(); diff --git a/src/lexer/token.cc b/src/lexer/token.cc index 071d00c..091a7ab 100644 --- a/src/lexer/token.cc +++ b/src/lexer/token.cc @@ -8,6 +8,7 @@ #include "lexer/enum_type.h" #include "lexer/symbol_property.h" #include "util/io.h" +#include "error.h" using namespace std; // remove this |