From 5aa4893dc6f74af72949ea7a2cb299a4e2dbd506 Mon Sep 17 00:00:00 2001 From: StarCells Date: Sat, 29 Aug 2026 00:10:20 +0800 Subject: [PATCH] 202608290010 --- include/r8asm/r8asm.hh | 19 +++--- src/r8asm.cc | 144 ++++++++++++++++++++++++++++++++++++----- tests/r8_test.cc | 1 + 3 files changed, 140 insertions(+), 24 deletions(-) diff --git a/include/r8asm/r8asm.hh b/include/r8asm/r8asm.hh index d6d43cb..ddd84a6 100644 --- a/include/r8asm/r8asm.hh +++ b/include/r8asm/r8asm.hh @@ -9,6 +9,8 @@ constexpr size_t CELL_SIZE = 8; +namespace r8asm_maps {} + class r8asm { using data = int16_t; using parameter_index = uint8_t; @@ -29,7 +31,6 @@ private: DIRECTIVE, PARAMETER, BUILTIN, - PREFIX_REMARK, LPAREN, RPAREN, @@ -50,8 +51,11 @@ private: std::string filename; std::vector lines; + const std::set raw_rot8_ins = {"stp", "btp", "ror", "flb", + "biz", "rnz", "out", "inp"}; + token_stream(const std::string &filename); - std::vector lex(const std::string &in_chars); + std::vector lex(const std::string &char_stream); }; enum class ast_node_t { @@ -124,7 +128,6 @@ private: AND, OR, NOT, - NEG, }; const std::map operator_map = { @@ -132,11 +135,11 @@ private: {"op_mul", operator_t::MUL}, {"op_div", operator_t::DIV}, {"op_xor", operator_t::XOR}, {"op_and", operator_t::AND}, {"op_or", operator_t::OR}, {"op_not", operator_t::NOT}, - {"op_neg", operator_t::NEG}}; + }; const std::map op_precedence_map = { - {operator_t::NEG, 0}, {operator_t::NOT, 0}, {operator_t::MUL, 1}, - {operator_t::DIV, 1}, {operator_t::ADD, 2}, {operator_t::SUB, 2}, - {operator_t::AND, 3}, {operator_t::XOR, 4}, {operator_t::OR, 5}, + {operator_t::NOT, 0}, {operator_t::MUL, 1}, {operator_t::DIV, 1}, + {operator_t::ADD, 2}, {operator_t::SUB, 2}, {operator_t::AND, 3}, + {operator_t::XOR, 4}, {operator_t::OR, 5}, }; ast_node expand_node_to_ins(const ast_node &node); std::string expand_root(); @@ -162,8 +165,6 @@ private: ast_node parse(const token_stream &stream); ast_node parse_args(const token_line &line); ast_node parse_single_arg(const token_line &line); - token_stream lex(const std::string &char_stream); - bool is_integer(const std::string &str); public: std::string export_rot8(); diff --git a/src/r8asm.cc b/src/r8asm.cc index 665f6d1..ca9b5cc 100644 --- a/src/r8asm.cc +++ b/src/r8asm.cc @@ -1,13 +1,64 @@ #include "r8asm/r8asm.hh" #include -#include -#include #include #include +#include +#include #include #include #include +#define EMPTY_CHARS (" \t\n\r\f\v") +#define OPERATORS ("+-*/^~\"()") +namespace r8asm_strtools { + +bool is_integer(const std::string &str) { + return !str.empty() && + std::ranges::all_of(str, [](char c) { return c >= '0' && c <= '9'; }); +} + +void trim(std::string &str) { // Clean empty chars in the string. + auto start = str.find_first_not_of(EMPTY_CHARS); + + if (!(start == + std::string::npos)) { // The string isn't filled with empty chars. + auto tmp_end = str.find_last_not_of(EMPTY_CHARS); + str = str.substr(start, tmp_end - start + 1); + } else + str = ""; +} + +std::string peel(std::string &str) { // Peel the first word from string. + for (size_t i = 1; i < str.size(); + i++) { // insert space before and after operators appear + if (std::string(OPERATORS).find(str[i]) != std::string::npos) { + if (str[i - 1] != ' ') { + str = str.substr(0, i) + ' ' + str.substr(i, str.size()); + i++; + } + if (str[i + 1] != ' ') { + str = str.substr(0, i + 1) + ' ' + str.substr(i + 1, str.size()); + } + } + } + trim(str); + auto end = str.find_first_of(EMPTY_CHARS); + + if (end != std::string::npos) { + std::string tmp_str = str.substr(0, end); + str = str.substr(end); + trim(str); + return (tmp_str); + } else { + std::string tmp_str = str; + str = ""; + return (tmp_str); + } +} +} // namespace r8asm_strtools +#undef EMPTY_CHARS +#undef OPERATORS + r8asm::general_data r8asm::ast::expand_node_to_data(const ast_node &node) { general_data to_return = 0; switch (node.type) { @@ -60,9 +111,6 @@ r8asm::general_data r8asm::ast::expand_node_to_data(const ast_node &node) { case operator_t::NOT: to_return = ~expand_node_to_data(node.children[0]); break; - case operator_t::NEG: - to_return = -expand_node_to_data(node.children[0]); - break; default:; break; } @@ -410,9 +458,6 @@ r8asm::ast_node r8asm::parse(const token_stream &stream) { to_return.children.push_back( {r8asm::ast_node_t::ROT8, it->tokens[0].value, {}}); } break; - case token_t::PREFIX_REMARK: - // ignore it. - break; default: break; @@ -422,11 +467,6 @@ r8asm::ast_node r8asm::parse(const token_stream &stream) { } #undef POINTS_AT_DIRECTIVE -bool r8asm::is_integer(const std::string &str) { - return !str.empty() && - std::ranges::all_of(str, [](char c) { return c >= '0' && c <= '9'; }); -} - #define UNRELATED_TO_LEFT(present_token) \ (it != line.tokens.begin() && \ ((present_token->type == token_t::LPAREN || \ @@ -502,7 +542,7 @@ r8asm::ast_node r8asm::parse_single_arg(const token_line &line) { } break; case token_t::STRING: { { - if (is_integer(it->value)) + if (r8asm_strtools::is_integer(it->value)) data_stack.push({r8asm::ast_node_t::CONSTANT, it->value, {}}); else data_stack.push({r8asm::ast_node_t::VARIABLE, it->value, {}}); @@ -524,7 +564,6 @@ r8asm::ast_node r8asm::parse_single_arg(const token_line &line) { #define GET_PRECEDENCE(op_key) \ (main_tree.op_precedence_map.at(main_tree.operator_map.at(op_key))) switch (main_tree.operator_map.at(it->value)) { - case ast::operator_t::NEG: case ast::operator_t::NOT: // unary operators { right_assoc_depth++; @@ -571,3 +610,78 @@ r8asm::ast_node r8asm::parse_single_arg(const token_line &line) { } #undef EMPTY_CODEBLOCK + +#define OPERATORS ("+-*/^~") +std::vector +r8asm::token_stream::lex(const std::string &char_stream) { + std::vector to_return; + + std::istringstream iss(char_stream); + std::string str_buffer; + line_number num = 0; + bool next_will_be_remark = false; + + const std::map operator_map = { + // without "-" + {"+", "op_add"}, {"*", "op_mul"}, {"/", "op_div"}, {"^", "op_xor"}, + {"&", "op_and"}, {"|", "op_or"}, {"~", "op_not"}, {"-", "op_sub"}}; + + while (std::getline(iss, str_buffer)) { + std::vector token_buffer = {}; + bool start_of_line = true; + bool lquote_appeared = false; + + for (std::string token_val = r8asm_strtools::peel(str_buffer); + token_val != "";) { + if (lquote_appeared && token_val[0] != '\"') + token_buffer.push_back({r8asm::token_t::STRING, token_val}); + else { + if (start_of_line) { // it's the first token in the line + start_of_line = false; + if (token_val[0] == '.') // builtin + token_buffer.push_back( + {r8asm::token_t::BUILTIN, "builtin_" + token_val.substr(0, 1)}); + // e.g. .var -> builtin_var + else if (token_val[0] == '%') // directive + token_buffer.push_back( + {r8asm::token_t::DIRECTIVE, token_val.substr(0, 1)}); + // e.g. %macro -> macro + else if (token_val[0] == ';') // remark + next_will_be_remark = true; + else if (raw_rot8_ins.find(token_val) != raw_rot8_ins.end()) // rot8 + token_buffer.push_back({r8asm::token_t::ROT8, "rot8_" + token_val}); + else // macro + token_buffer.push_back({r8asm::token_t::MACROCALL, token_val}); + } else { // it's a part of argument + if (!next_will_be_remark) { + if (token_val[0] == ';') // remark + next_will_be_remark = true; + else if (token_val[0] == '%') // parameter + token_buffer.push_back({r8asm::token_t::PARAMETER, token_val}); + else if (token_val[0] == '"') { // quote(filename) + if (!lquote_appeared) { + lquote_appeared = true; // only %include will use quotes, they + // won't appear again + token_buffer.push_back({r8asm::token_t::LQUOTE}); + } else + token_buffer.push_back({r8asm::token_t::RQUOTE}); + } else if (token_val[0] == '(') + token_buffer.push_back({r8asm::token_t::LPAREN}); + else if (token_val[0] == ')') + token_buffer.push_back({r8asm::token_t::RPAREN}); + else if (std::string(OPERATORS).find(token_val[0]) != + std::string::npos) // operator + token_buffer.push_back({r8asm::token_t::OPERATOR, token_val}); + else + token_buffer.push_back({r8asm::token_t::STRING, token_val}); + } + } + } + } + num++; + if (!token_buffer.empty()) + to_return.push_back({num, token_buffer}); + next_will_be_remark = false; + } + return to_return; +} diff --git a/tests/r8_test.cc b/tests/r8_test.cc index 9b4b6ae..ff1253d 100644 --- a/tests/r8_test.cc +++ b/tests/r8_test.cc @@ -1 +1,2 @@ + int main() { return (0); } \ No newline at end of file