Move everything into it's own file

This commit is contained in:
Maxwell 2025-04-24 19:43:15 +10:00
parent 21064d7ce7
commit 73939b3668
14 changed files with 707 additions and 622 deletions

View File

@ -1,4 +1,4 @@
compiler "g++";
binary "mx";
source "src/main.cpp";
source "src/*.cpp";
compile;

BIN
mx

Binary file not shown.

38
src/ArgParser.cpp Normal file
View File

@ -0,0 +1,38 @@
#include "ArgParser.h"
bool debugMode = false;
ArgParser::ArgParser(int argc, char* argv[]) {
// First, collect all arguments
for (int i = 0; i < argc; i++) {
args.push_back(argv[i]);
}
// Then process them
for (int i = 0; i < args.size(); i++) {
if (args[i] == "--debug") {
debugMode = true;
args.erase(args.begin() + i);
} else if (args[i] == "--help") {
cout << "mxlang interpreter" << endl;
cout << "Usage: mx [file]" << endl;
cout << "Options:" << endl;
cout << " --debug Enable debug mode" << endl;
cout << " --help Show this help message" << endl;
cout << "Issues? Send an email to max@maxwellj.xyz" << endl;
cout << "Report bugs at https://git.maxwellj.xyz/max/mx" << endl;
exit(0);
} else if (args[i] == "--version") {
cout << "mxlang, version 0.0.2" << endl;
exit(0);
}
}
}
string ArgParser::getArg(int index) {
if (index >= 0 && index < args.size()) {
return args[index];
}
return "";
}

11
src/ArgParser.h Normal file
View File

@ -0,0 +1,11 @@
#pragma once
#include "common.h"
class ArgParser {
private:
vector<string> args;
public:
ArgParser(int argc, char* argv[]);
string getArg(int index);
};

263
src/Interpreter.cpp Normal file
View File

@ -0,0 +1,263 @@
#include "Interpreter.h"
optional<Token> Interpreter::consume() {
tokenIndex++;
if (tokenIndex < tokens.size()) return tokens[tokenIndex];
return {};
}
optional<Token> Interpreter::peek(int offset) {
int index = tokenIndex + offset;
if (index >= 0 && index < tokens.size()) return tokens[index];
return {};
}
void Interpreter::convertToTokens(vector<Token> tokenList) {
if (debugMode) log.toggleDebugPrint();
tokens = tokenList;
log.debug("Alright we got " + to_string(tokens.size()) + " tokens");
while (tokenIndex < static_cast<int>(tokens.size() - 1)) {
auto currentToken = consume();
if (!currentToken) break;
vector<Token> currentInstruction;
currentInstruction.push_back(currentToken.value());
// Collect tokens until semicolon
while (auto nextToken = peek(1)) {
if (nextToken->keyword == keywords::SEMICOLON) {
consume(); // consume the semicolon
break;
}
consume(); // consume the peeked token
currentInstruction.push_back(nextToken.value());
}
// Apply variables to tokens
// We start at 1 so we can reassign variables in the execution of code
for (int i = 1; i < currentInstruction.size(); i++) {
if (currentInstruction[i].type == valtype::STR) {
string potentialVarName = get<string>(currentInstruction[i].value.value);
auto varIt = variables.find(potentialVarName);
if (varIt != variables.end()) {
// Replace the token with the variable's value
Token newToken;
newToken.keyword = keywords::VALUE;
newToken.type = varIt->second.type;
newToken.value = varIt->second;
currentInstruction[i] = newToken;
}
}
}
// Do math
for (int i = 0; i < currentInstruction.size(); i++) {
if (currentInstruction[i].keyword == keywords::ADD || currentInstruction[i].keyword == keywords::SUBTRACT || currentInstruction[i].keyword == keywords::MULTIPLY || currentInstruction[i].keyword == keywords::DIVIDE) {
Token newToken;
newToken.keyword = keywords::VALUE;
if (currentInstruction.size() < i + 1) syntaxError.mathTooFewArgs();
Token before = currentInstruction[i - 1];
Token after = currentInstruction[i + 1];
if (before.type != after.type) syntaxError.mathTypeMismatch();
newToken.type = before.type;
if (currentInstruction[i].keyword == keywords::ADD) {
if (newToken.type == valtype::INT) {
newToken.value.value = get<int>(before.value.value) + get<int>(after.value.value);
}
else if (newToken.type == valtype::DEC) {
newToken.value.value = get<double>(before.value.value) + get<double>(after.value.value);
}
else if (newToken.type == valtype::STR) {
newToken.value.value = get<string>(before.value.value) + get<string>(after.value.value);
} else {
syntaxError.mathCannotDoOperationOnType("+", "bool");
}
}
else if (currentInstruction[i].keyword == keywords::SUBTRACT) {
if (newToken.type == valtype::INT) {
newToken.value.value = get<int>(before.value.value) - get<int>(after.value.value);
}
else if (newToken.type == valtype::DEC) {
newToken.value.value = get<double>(before.value.value) - get<double>(after.value.value);
} else {
syntaxError.mathCannotDoOperationOnType("-", "bool or string");
}
}
else if (currentInstruction[i].keyword == keywords::MULTIPLY) {
if (newToken.type == valtype::INT) {
newToken.value.value = get<int>(before.value.value) * get<int>(after.value.value);
}
else if (newToken.type == valtype::DEC) {
newToken.value.value = get<double>(before.value.value) * get<double>(after.value.value);
} else {
syntaxError.mathCannotDoOperationOnType("*", "bool or string");
}
}
else if (currentInstruction[i].keyword == keywords::DIVIDE) {
if (newToken.type == valtype::INT) {
newToken.value.value = get<int>(before.value.value) / get<int>(after.value.value);
}
else if (newToken.type == valtype::DEC) {
newToken.value.value = get<double>(before.value.value) / get<double>(after.value.value);
} else {
syntaxError.mathCannotDoOperationOnType("/", "bool or string");
}
} else {
// Something has gone terribly wrong
// We should never reach this point in the code
syntaxError.generalError("The math aint mathing");
}
// Insert our cool new token and get rid of the boring old stuff
currentInstruction[i - 1] = newToken;
currentInstruction.erase(currentInstruction.begin() + i);
currentInstruction.erase(currentInstruction.begin() + i);
i -= 1;
}
}
// Execute the instruction
executeCode(currentInstruction);
}
}
void Interpreter::executeCode(vector<Token> tokens) {
SyntaxError syntaxError;
log.debug("Token length for this expression is " + to_string(tokens.size()));
for (int i = 0; i < tokens.size(); i++) {
if (tokens[i].keyword == keywords::PRINT) {
i++;
if (tokens.size() <= i) break;
Token nextToken = tokens[i];
// Handle different value types
if (nextToken.keyword == keywords::VALUE) {
switch (nextToken.type) {
case valtype::INT:
cout << get<int>(nextToken.value.value) << endl;
break;
case valtype::DEC:
cout << get<double>(nextToken.value.value) << endl;
break;
case valtype::STR:
cout << get<string>(nextToken.value.value) << endl;
break;
case valtype::BOOL:
cout << (get<bool>(nextToken.value.value) ? "true" : "false") << endl;
break;
default:
vector<string> validTypes = {"int", "dec", "str", "bool"};
syntaxError.fnTypeMismatch("print", validTypes, nextToken.type);
}
} else {
syntaxError.fnNotSufficientArgs("print", 1, 1, 0);
}
} else if (tokens[i].keyword == keywords::EXIT) {
i++;
if (tokens.size() <= i) break;
Token nextToken = tokens[i];
if (nextToken.keyword == keywords::VALUE) {
switch (nextToken.type) {
case valtype::INT:
exit(get<int>(nextToken.value.value));
break;
case valtype::DEC:
exit(get<double>(nextToken.value.value));
break;
default:
vector<string> validTypes = {"int", "dec"};
syntaxError.fnTypeMismatch("exit", validTypes, nextToken.type);
}
}
} else if (tokens[i].keyword == keywords::LET) {
i++;
if (tokens.size() <= i + 2) {
syntaxError.fnNotSufficientArgs("let", 3, 3, tokens.size() - i);
break;
}
Token typeToken = tokens[i];
Token nameToken = tokens[i + 1];
Token valueToken = tokens[i + 2];
i += 2;
// Validate that we have a valid variable name
if (nameToken.type != valtype::STR) {
vector<string> validTypes = {"str"};
syntaxError.fnTypeMismatch("let (variable name)", validTypes, nameToken.type);
continue;
}
string varName = get<string>(nameToken.value.value);
Value newValue;
// Check the type declaration matches the value
if (typeToken.keyword == keywords::INT && valueToken.type == valtype::INT) {
newValue.type = valtype::INT;
newValue.value = get<int>(valueToken.value.value);
}
else if (typeToken.keyword == keywords::DEC && valueToken.type == valtype::DEC) {
newValue.type = valtype::DEC;
newValue.value = get<double>(valueToken.value.value);
}
else if (typeToken.keyword == keywords::STR && valueToken.type == valtype::STR) {
newValue.type = valtype::STR;
newValue.value = get<string>(valueToken.value.value);
}
else if (typeToken.keyword == keywords::BOOL && valueToken.type == valtype::BOOL) {
newValue.type = valtype::BOOL;
newValue.value = get<bool>(valueToken.value.value);
}
else {
vector<string> validTypes;
if (typeToken.keyword == keywords::INT) validTypes = {"int"};
else if (typeToken.keyword == keywords::DEC) validTypes = {"dec"};
else if (typeToken.keyword == keywords::STR) validTypes = {"str"};
else if (typeToken.keyword == keywords::BOOL) validTypes = {"bool"};
syntaxError.fnTypeMismatch("let", validTypes, valueToken.type, "Variable name is " + varName);
continue;
}
// Store the variable
variables[varName] = newValue;
} else {
if (tokens[i].keyword == keywords::VALUE && tokens[i].value.type == valtype::STR && variables.find(get<string>(tokens[i].value.value)) != variables.end()) {
log.debug("Manipulating variable...");
if (tokens.size() < i + 1) {
syntaxError.mathTooFewArgs();
}
string varName = get<string>(tokens[i].value.value);
i++;
if (variables[varName].type == valtype::INT) {
if (tokens[i].keyword == keywords::INCREMENT) {
variables[varName].value = get<int>(variables[varName].value) + 1;
} else if (tokens[i].keyword == keywords::DECREMENT) {
variables[varName].value = get<int>(variables[varName].value) - 1;
} else if (tokens[i].keyword == keywords::ADDTO) {
if (tokens.size() < i + 1) syntaxError.mathTooFewArgs();
if (tokens[i + 1].value.type != valtype::INT) syntaxError.mathTypeMismatch("Expected an int when adding to an int");
variables[varName].value = get<int>(variables[varName].value) + get<int>(tokens[i + 1].value.value);
} else if (tokens[i].keyword == keywords::SUBTRACTFROM) {
if (tokens.size() < i + 1) syntaxError.mathTooFewArgs();
if (tokens[i + 1].value.type != valtype::INT) syntaxError.mathTypeMismatch("Expected an int when subtracting from an int");
variables[varName].value = get<int>(variables[varName].value) - get<int>(tokens[i + 1].value.value);
}
} else if (variables[varName].type == valtype::DEC) {
if (tokens[i].keyword == keywords::INCREMENT) {
variables[varName].value = get<double>(variables[varName].value) + 1;
} else if (tokens[i].keyword == keywords::DECREMENT) {
variables[varName].value = get<double>(variables[varName].value) - 1;
} else if (tokens[i].keyword == keywords::ADDTO) {
if (tokens.size() < i + 1) syntaxError.mathTooFewArgs();
if (tokens[i + 1].value.type != valtype::DEC) syntaxError.mathTypeMismatch("Expected an dec when adding to an dec");
variables[varName].value = get<double>(variables[varName].value) + get<double>(tokens[i + 1].value.value);
} else if (tokens[i].keyword == keywords::SUBTRACTFROM) {
if (tokens.size() < i + 1) syntaxError.mathTooFewArgs();
if (tokens[i + 1].value.type != valtype::DEC) syntaxError.mathTypeMismatch("Expected an dec when subtracting from an dec");
variables[varName].value = get<double>(variables[varName].value) - get<double>(tokens[i + 1].value.value);
}
}
} else {
syntaxError.unknownFn();
}
}
}
}

20
src/Interpreter.h Normal file
View File

@ -0,0 +1,20 @@
#pragma once
#include "common.h"
#include "SyntaxError.h"
#include "Logger.h"
class Interpreter {
private:
SyntaxError syntaxError;
vector<Token> tokens;
map<string, Value> variables;
Logger log;
int tokenIndex = -1;
optional<Token> consume();
optional<Token> peek(int offset = 1);
public:
void convertToTokens(vector<Token> tokenList);
void executeCode(vector<Token> tokens);
};

36
src/Logger.cpp Normal file
View File

@ -0,0 +1,36 @@
#include "Logger.h"
string logList;
void Logger::writeToLog(string type, string in) {
logList += type + ": " + in + "\n";
}
string Logger::getLog() {
return logList;
}
void Logger::toggleDebugPrint() {
isDebug = !isDebug;
}
void Logger::error(string in) {
cout << "Error: " + in + "\n";
writeToLog("Error", in);
}
void Logger::fatalError(string in, int code) {
cout << "Error: " + in + "\n";
writeToLog("Error", in);
exit(code);
}
void Logger::info(string in) {
cout << "Info: " + in + "\n";
writeToLog("Info", in);
}
void Logger::debug(string in) {
if (isDebug) cout << "Debug: " + in + "\n";
writeToLog("Debug", in);
}

17
src/Logger.h Normal file
View File

@ -0,0 +1,17 @@
#pragma once
#include "common.h"
class Logger {
private:
bool isDebug = false;
void writeToLog(string type, string in);
public:
string getLog();
void toggleDebugPrint();
void error(string in);
void fatalError(string in, int code);
void info(string in);
void debug(string in);
};

164
src/Parser.cpp Normal file
View File

@ -0,0 +1,164 @@
#include "Parser.h"
bool Parser::canInt(string in) {
try {
stoi(in);
return true;
} catch (...) {
return false;
}
}
bool Parser::canDec(string in) {
try {
stod(in);
return true;
} catch (...) {
return false;
}
}
void Parser::parseLines(string in, Logger log) {
log.debug("Parsing lines...");
buffer.clear();
termBuffer.clear();
terms.clear();
bool isString = false;
bool isEscaped = false;
for (size_t i = 0; i < in.size(); i++) {
char c = in[i];
if (isEscaped) {
buffer += c;
isEscaped = false;
continue;
}
if (c == '\\') {
isEscaped = true;
buffer += c;
continue;
}
if (c == '"') {
buffer += c;
isString = !isString;
} else if (c == '\n' && !isString) {
// Skip newlines outside strings
continue;
} else if (c == ';' && !isString) {
if (!buffer.empty()) {
termBuffer.push_back(buffer);
buffer.clear();
}
terms.push_back(termBuffer);
termBuffer.clear();
} else if ((c == ',' || c == '{' || c == '}' || c == '(' || c == ')') && !isString) {
if (!buffer.empty()) {
termBuffer.push_back(buffer);
buffer.clear();
}
termBuffer.push_back(string(1, c));
} else if (c == ' ' && !isString) {
if (!buffer.empty()) {
termBuffer.push_back(buffer);
buffer.clear();
}
} else {
buffer += c;
}
}
// Handle any remaining buffer content
if (!buffer.empty()) {
termBuffer.push_back(buffer);
}
if (!termBuffer.empty()) {
terms.push_back(termBuffer);
}
// Debug output
string debugString;
log.debug("Lines have been parsed.");
for (const auto& term : terms) {
for (const auto& t : term) {
debugString += t + ", ";
}
debugString += "\n";
}
log.debug(debugString);
}
void Parser::processLines() {
// Process vector<vector<string>> terms
for (int i = 0; i < terms.size(); i++) {
for (int j = 0; j < terms[i].size(); j++) {
Token token;
string ct = terms[i][j];
if (ct == " ") continue;
// Oh boy here we go
else if (ct == "fun") token.keyword = keywords::FUN;
else if (ct == "let") token.keyword = keywords::LET;
else if (ct == "print") token.keyword = keywords::PRINT;
else if (ct == "return") token.keyword = keywords::RETURN;
else if (ct == "exit") token.keyword = keywords::EXIT;
else if (ct == "int") token.keyword = keywords::INT;
else if (ct == "dec") token.keyword = keywords::DEC;
else if (ct == "str") token.keyword = keywords::STR;
else if (ct == "bool") token.keyword = keywords::BOOL;
else if (ct == "{") token.keyword = keywords::OBRAC;
else if (ct == "}") token.keyword = keywords::CBRAC;
else if (ct == "(") token.keyword = keywords::OPARE;
else if (ct == ")") token.keyword = keywords::CPARE;
else if (ct == "+") token.keyword = keywords::ADD;
else if (ct == "-") token.keyword = keywords::SUBTRACT;
else if (ct == "*") token.keyword = keywords::MULTIPLY;
else if (ct == "/") token.keyword = keywords::DIVIDE;
else if (ct == "++") token.keyword = keywords::INCREMENT;
else if (ct == "--") token.keyword = keywords::DECREMENT;
else if (ct == "+=") token.keyword = keywords::ADDTO;
else if (ct == "-=") token.keyword = keywords::SUBTRACTFROM;
else if (ct == "*=") token.keyword = keywords::MULTIPLYTO;
else if (ct == "/=") token.keyword = keywords::DIVIDEFROM;
else {
token.keyword = keywords::VALUE;
// Convert the value based on its type
if (canDec(ct)) {
token.type = valtype::DEC;
token.value.type = valtype::DEC;
token.value.value = stod(ct);
}
else if (canInt(ct)) {
token.type = valtype::INT;
token.value.type = valtype::INT;
token.value.value = stoi(ct);
}
else if (ct == "true" || ct == "false") {
token.type = valtype::BOOL;
token.value.type = valtype::BOOL;
token.value.value = (ct == "true");
}
else {
// Handle strings - remove quotes if present
token.type = valtype::STR;
token.value.type = valtype::STR;
if (ct.size() >= 2 && ct.front() == '"' && ct.back() == '"') {
// Remove the quotes
token.value.value = ct.substr(1, ct.size() - 2);
} else {
token.value.value = ct;
}
}
}
tokens.push_back(token);
}
Token semi;
semi.keyword = keywords::SEMICOLON;
tokens.push_back(semi);
}
}
vector<Token> Parser::getTokens() {
return tokens;
}

21
src/Parser.h Normal file
View File

@ -0,0 +1,21 @@
#pragma once
#include "common.h"
#include "Logger.h"
class Parser {
private:
string buffer;
vector<string> lines;
vector<string> termBuffer;
vector<Token> tokens;
vector<vector<string>> terms;
vector<vector<keywords>> things;
vector<vector<variant<var, keywords>>> stuff;
bool canInt(string in);
bool canDec(string in);
public:
void parseLines(string in, Logger log);
void processLines();
vector<Token> getTokens();
};

54
src/SyntaxError.cpp Normal file
View File

@ -0,0 +1,54 @@
#include "SyntaxError.h"
void SyntaxError::fnTypeMismatch(string function, vector<string> validTypes, valtype typeGiven, string notes) {
cerr << "TypeError: function type mismatch" << endl;
cerr << "Function '" << function << "' expected one of the following types: ";
for (int i = 0; i < validTypes.size(); i++) {
if (i != 0) cerr << ", ";
cerr << validTypes[i];
}
cerr << endl << "Got type '";
if (typeGiven == valtype::INT) cerr << "int";
else if (typeGiven == valtype::DEC) cerr << "dec";
else if (typeGiven == valtype::STR) cerr << "str";
else if (typeGiven == valtype::BOOL) cerr << "bool";
else cerr << "unknown";
cerr << "' instead" << endl;
if (!notes.empty()) cerr << "Notes: " << notes << endl;
exit(1);
}
void SyntaxError::fnNotSufficientArgs(string function, int minArgs, int maxArgs, int argsGiven) {
cerr << "TypeError: function not sufficient arguments" << endl;
cerr << "Function '" << function << "' expected between " << minArgs << " and " << maxArgs << " arguments, got " << argsGiven << endl;
exit(1);
}
void SyntaxError::unknownFn() {
cerr << "TypeError: unknown function" << endl;
exit(1);
}
void SyntaxError::mathTypeMismatch(string notes) {
cerr << "MathError: cannot add two different types together" << endl;
if (!notes.empty()) cerr << "Notes: " << notes << endl;
exit(1);
}
void SyntaxError::mathTooFewArgs(string notes) {
cerr << "MathError: too few things to add together" << endl;
if (!notes.empty()) cerr << "Notes: " << notes << endl;
exit(1);
}
void SyntaxError::mathCannotDoOperationOnType(string operatorUsed, string type, string notes) {
cerr << "MathError: cannot use operator '" + operatorUsed + " on type '" + type + "'" << endl;
if (!notes.empty()) cerr << "Notes: " << notes << endl;
exit(1);
}
void SyntaxError::generalError(string notes) {
cerr << "GeneralError: Something went awfully wrong and we don't know why lmao" << endl;
if (!notes.empty()) cerr << "Notes: " << notes << endl;
exit(1);
}

14
src/SyntaxError.h Normal file
View File

@ -0,0 +1,14 @@
#pragma once
#include "common.h"
class SyntaxError {
public:
void fnTypeMismatch(string function, vector<string> validTypes, valtype typeGiven, string notes = "");
void fnNotSufficientArgs(string function, int minArgs, int maxArgs, int argsGiven);
void unknownFn();
void mathTypeMismatch(string notes = "");
void mathTooFewArgs(string notes = "");
void mathCannotDoOperationOnType(string operatorUsed, string type, string notes = "");
void generalError(string notes = "");
};

62
src/common.h Normal file
View File

@ -0,0 +1,62 @@
#pragma once
#include <iostream>
#include <string>
#include <vector>
#include <variant>
#include <map>
#include <optional>
using namespace std;
enum class valtype {
INT, DEC, STR, BOOL, KEYWORD, UNKNOWN
};
enum class keywords {
IF, ELSE, WHILE, INT, DEC, STR, BOOL, FUN, RETURN,
OPARE, CPARE, OBRAC, CBRAC,
ADDTO, SUBTRACTFROM, MULTIPLYTO, DIVIDEFROM,
ADD, SUBTRACT, MULTIPLY, DIVIDE,
EQUAL, INEQUAL, LESS, GREATER, EQLESS, EQGREATER,
INCREMENT, DECREMENT,
PRINT, LET, INPUT, EXIT,
VALUE, SEMICOLON, VARIABLE
};
struct Value {
valtype type;
variant<int, double, string, bool> value;
};
struct Token {
keywords keyword;
Value value;
valtype type = valtype::KEYWORD;
};
struct var {
valtype type;
variant<int, double, string, bool> value;
string toString() const {
if (type == valtype::INT) {
return to_string(get<int>(value));
}
else if (type == valtype::DEC) {
return to_string(get<double>(value));
}
else if (type == valtype::STR) {
return get<string>(value);
}
else if (type == valtype::BOOL) {
return to_string(get<bool>(value));
} else {
return "unknown";
}
}
};
// Global variable
extern bool debugMode;
extern string logList;

View File

@ -1,626 +1,11 @@
#include <ios>
#include <iostream>
#include <csignal>
#include <fstream>
#include <string>
#include <vector>
#include <variant>
#include <map>
#include <optional>
using namespace std;
enum class valtype {
INT, DEC, STR, BOOL, KEYWORD, UNKNOWN
};
enum class keywords {
IF, ELSE, WHILE, INT, DEC, STR, BOOL, FUN, RETURN,
OPARE, CPARE, OBRAC, CBRAC,
ADDTO, SUBTRACTFROM, MULTIPLYTO, DIVIDEFROM,
ADD, SUBTRACT, MULTIPLY, DIVIDE,
EQUAL, INEQUAL, LESS, GREATER, EQLESS, EQGREATER,
INCREMENT, DECREMENT,
PRINT, LET, INPUT, EXIT,
VALUE, SEMICOLON, VARIABLE
};
struct Value {
valtype type;
variant<int, double, string, bool> value;
};
struct Token {
keywords keyword;
Value value;
valtype type = valtype::KEYWORD;
};
struct var {
valtype type;
variant<int, double, string, bool> value;
string toString() const {
if (type == valtype::INT) {
return to_string(get<int>(value));
}
else if (type == valtype::DEC) {
return to_string(get<double>(value));
}
else if (type == valtype::STR) {
return get<string>(value);
}
else if (type == valtype::BOOL) {
return to_string(get<bool>(value));
} else {
return "unknown";
}
}
};
bool debugMode = false;
class ArgParser {
private:
vector<string> args;
public:
ArgParser(int argc, char* argv[]) {
// First, collect all arguments
for (int i = 0; i < argc; i++) {
args.push_back(argv[i]);
}
// Then process them
for (int i = 0; i < args.size(); i++) {
if (args[i] == "--debug") {
debugMode = true;
args.erase(args.begin() + i);
} else if (args[i] == "--help") {
cout << "mxlang interpreter" << endl;
cout << "Usage: mx [file]" << endl;
cout << "Options:" << endl;
cout << " --debug Enable debug mode" << endl;
cout << " --help Show this help message" << endl;
cout << "Issues? Send an email to max@maxwellj.xyz" << endl;
cout << "Report bugs at https://git.maxwellj.xyz/max/mx" << endl;
exit(0);
}
}
}
string getArg(int index) {
if (index >= 0 && index < args.size()) {
return args[index];
}
return "";
}
};
string logList;
class Logger {
private:
bool isDebug = false;
void writeToLog(string type, string in) {
logList += type + ": " + in + "\n";
}
public:
string getLog() {
return logList;
}
void toggleDebugPrint() {
isDebug = !isDebug;
}
void error(string in) {
cout << "Error: " + in + "\n";
writeToLog("Error", in);
}
void fatalError(string in, int code) {
cout << "Error: " + in + "\n";
writeToLog("Error", in);
exit(code);
}
void info(string in) {
cout << "Info: " + in + "\n";
writeToLog("Info", in);
}
void debug(string in) {
if (isDebug) cout << "Debug: " + in + "\n";
writeToLog("Debug", in);
}
};
class SyntaxError {
private:
public:
void fnTypeMismatch(string function, vector<string> validTypes, valtype typeGiven, string notes = "") {
cerr << "TypeError: function type mismatch" << endl;
cerr << "Function '" << function << "' expected one of the following types: ";
for (int i = 0; i < validTypes.size(); i++) {
if (i != 0) cerr << ", ";
cerr << validTypes[i];
}
cerr << endl << "Got type '";
if (typeGiven == valtype::INT) cerr << "int";
else if (typeGiven == valtype::DEC) cerr << "dec";
else if (typeGiven == valtype::STR) cerr << "str";
else if (typeGiven == valtype::BOOL) cerr << "bool";
else cerr << "unknown";
cerr << "' instead" << endl;
if (!notes.empty()) cerr << "Notes: " << notes << endl;
exit(1);
}
void fnNotSufficientArgs(string function, int minArgs, int maxArgs, int argsGiven) {
cerr << "TypeError: function not sufficient arguments" << endl;
cerr << "Function '" << function << "' expected between " << minArgs << " and " << maxArgs << " arguments, got " << argsGiven << endl;
exit(1);
}
void unknownFn() {
cerr << "TypeError: unknown function" << endl;
exit(1);
}
void mathTypeMismatch(string notes = "") {
cerr << "MathError: cannot add two different types together" << endl;
if (!notes.empty()) cerr << "Notes: " << notes << endl;
exit(1);
}
void mathTooFewArgs(string notes = "") {
cerr << "MathError: too few things to add together" << endl;
if (!notes.empty()) cerr << "Notes: " << notes << endl;
exit(1);
}
void mathCannotDoOperationOnType(string operatorUsed, string type, string notes = "") {
cerr << "MathError: cannot use operator '" + operatorUsed + " on type '" + type + "'" << endl;
if (!notes.empty()) cerr << "Notes: " << notes << endl;
exit(1);
}
void generalError(string notes = "") {
cerr << "GeneralError: Something went awfully wrong and we don't know why lmao" << endl;
if (!notes.empty()) cerr << "Notes: " << notes << endl;
exit(1);
}
};
class Parser {
private:
string buffer;
vector<string> lines;
vector<string> termBuffer;
vector<Token> tokens;
vector<vector<string>> terms;
vector<vector<keywords>> things;
vector<vector<variant<var, keywords>>> stuff;
bool canInt(string in) {
try {
stoi(in);
return true;
} catch (...) {
return false;
}
}
bool canDec(string in) {
try {
stod(in);
return true;
} catch (...) {
return false;
}
}
public:
void parseLines(string in, Logger log) {
log.debug("Parsing lines...");
buffer.clear();
termBuffer.clear();
terms.clear();
bool isString = false;
bool isEscaped = false;
for (size_t i = 0; i < in.size(); i++) {
char c = in[i];
if (isEscaped) {
buffer += c;
isEscaped = false;
continue;
}
if (c == '\\') {
isEscaped = true;
buffer += c;
continue;
}
if (c == '"') {
buffer += c;
isString = !isString;
} else if (c == '\n' && !isString) {
// Skip newlines outside strings
continue;
} else if (c == ';' && !isString) {
if (!buffer.empty()) {
termBuffer.push_back(buffer);
buffer.clear();
}
terms.push_back(termBuffer);
termBuffer.clear();
} else if ((c == ',' || c == '{' || c == '}' || c == '(' || c == ')') && !isString) {
if (!buffer.empty()) {
termBuffer.push_back(buffer);
buffer.clear();
}
termBuffer.push_back(string(1, c));
} else if (c == ' ' && !isString) {
if (!buffer.empty()) {
termBuffer.push_back(buffer);
buffer.clear();
}
} else {
buffer += c;
}
}
// Handle any remaining buffer content
if (!buffer.empty()) {
termBuffer.push_back(buffer);
}
if (!termBuffer.empty()) {
terms.push_back(termBuffer);
}
// Debug output
string debugString;
log.debug("Lines have been parsed.");
for (const auto& term : terms) {
for (const auto& t : term) {
debugString += t + ", ";
}
debugString += "\n";
}
log.debug(debugString);
}
void processLines() {
// Process vector<vector<string>> terms
for (int i = 0; i < terms.size(); i++) {
for (int j = 0; j < terms[i].size(); j++) {
Token token;
string ct = terms[i][j];
if (ct == " ") continue;
// Oh boy here we go
else if (ct == "fun") token.keyword = keywords::FUN;
else if (ct == "let") token.keyword = keywords::LET;
else if (ct == "print") token.keyword = keywords::PRINT;
else if (ct == "return") token.keyword = keywords::RETURN;
else if (ct == "exit") token.keyword = keywords::EXIT;
else if (ct == "int") token.keyword = keywords::INT;
else if (ct == "dec") token.keyword = keywords::DEC;
else if (ct == "str") token.keyword = keywords::STR;
else if (ct == "bool") token.keyword = keywords::BOOL;
else if (ct == "{") token.keyword = keywords::OBRAC;
else if (ct == "}") token.keyword = keywords::CBRAC;
else if (ct == "(") token.keyword = keywords::OPARE;
else if (ct == ")") token.keyword = keywords::CPARE;
else if (ct == "+") token.keyword = keywords::ADD;
else if (ct == "-") token.keyword = keywords::SUBTRACT;
else if (ct == "*") token.keyword = keywords::MULTIPLY;
else if (ct == "/") token.keyword = keywords::DIVIDE;
else if (ct == "++") token.keyword = keywords::INCREMENT;
else if (ct == "--") token.keyword = keywords::DECREMENT;
else if (ct == "+=") token.keyword = keywords::ADDTO;
else if (ct == "-=") token.keyword = keywords::SUBTRACTFROM;
else if (ct == "*=") token.keyword = keywords::MULTIPLYTO;
else if (ct == "/=") token.keyword = keywords::DIVIDEFROM;
else {
token.keyword = keywords::VALUE;
// Convert the value based on its type
if (canDec(ct)) {
token.type = valtype::DEC;
token.value.type = valtype::DEC;
token.value.value = stod(ct);
}
else if (canInt(ct)) {
token.type = valtype::INT;
token.value.type = valtype::INT;
token.value.value = stoi(ct);
}
else if (ct == "true" || ct == "false") {
token.type = valtype::BOOL;
token.value.type = valtype::BOOL;
token.value.value = (ct == "true");
}
else {
// Handle strings - remove quotes if present
token.type = valtype::STR;
token.value.type = valtype::STR;
if (ct.size() >= 2 && ct.front() == '"' && ct.back() == '"') {
// Remove the quotes
token.value.value = ct.substr(1, ct.size() - 2);
} else {
token.value.value = ct;
}
}
}
tokens.push_back(token);
}
Token semi;
semi.keyword = keywords::SEMICOLON;
tokens.push_back(semi);
}
}
vector<Token> getTokens() {
return tokens;
}
};
class Interpreter {
private:
SyntaxError syntaxError;
vector<Token> tokens;
map<string, Value> variables;
Logger log;
int tokenIndex = -1;
optional<Token> consume() {
tokenIndex++;
if (tokenIndex < tokens.size()) return tokens[tokenIndex];
return {};
}
optional<Token> peek(int offset = 1) {
int index = tokenIndex + offset;
if (index >= 0 && index < tokens.size()) return tokens[index];
return {};
}
public:
void convertToTokens(vector<Token> tokenList) {
if (debugMode) log.toggleDebugPrint();
tokens = tokenList;
log.debug("Alright we got " + to_string(tokens.size()) + " tokens");
while (tokenIndex < static_cast<int>(tokens.size() - 1)) {
auto currentToken = consume();
if (!currentToken) break;
vector<Token> currentInstruction;
currentInstruction.push_back(currentToken.value());
// Collect tokens until semicolon
while (auto nextToken = peek(1)) {
if (nextToken->keyword == keywords::SEMICOLON) {
consume(); // consume the semicolon
break;
}
consume(); // consume the peeked token
currentInstruction.push_back(nextToken.value());
}
// Apply variables to tokens
// We start at 1 so we can reassign variables in the execution of code
for (int i = 1; i < currentInstruction.size(); i++) {
if (currentInstruction[i].type == valtype::STR) {
string potentialVarName = get<string>(currentInstruction[i].value.value);
auto varIt = variables.find(potentialVarName);
if (varIt != variables.end()) {
// Replace the token with the variable's value
Token newToken;
newToken.keyword = keywords::VALUE;
newToken.type = varIt->second.type;
newToken.value = varIt->second;
currentInstruction[i] = newToken;
}
}
}
// Do math
for (int i = 0; i < currentInstruction.size(); i++) {
if (currentInstruction[i].keyword == keywords::ADD || currentInstruction[i].keyword == keywords::SUBTRACT || currentInstruction[i].keyword == keywords::MULTIPLY || currentInstruction[i].keyword == keywords::DIVIDE) {
Token newToken;
newToken.keyword = keywords::VALUE;
if (currentInstruction.size() < i + 1) syntaxError.mathTooFewArgs();
Token before = currentInstruction[i - 1];
Token after = currentInstruction[i + 1];
if (before.type != after.type) syntaxError.mathTypeMismatch();
newToken.type = before.type;
if (currentInstruction[i].keyword == keywords::ADD) {
if (newToken.type == valtype::INT) {
newToken.value.value = get<int>(before.value.value) + get<int>(after.value.value);
}
else if (newToken.type == valtype::DEC) {
newToken.value.value = get<double>(before.value.value) + get<double>(after.value.value);
}
else if (newToken.type == valtype::STR) {
newToken.value.value = get<string>(before.value.value) + get<string>(after.value.value);
} else {
syntaxError.mathCannotDoOperationOnType("+", "bool");
}
}
else if (currentInstruction[i].keyword == keywords::SUBTRACT) {
if (newToken.type == valtype::INT) {
newToken.value.value = get<int>(before.value.value) - get<int>(after.value.value);
}
else if (newToken.type == valtype::DEC) {
newToken.value.value = get<double>(before.value.value) - get<double>(after.value.value);
} else {
syntaxError.mathCannotDoOperationOnType("-", "bool or string");
}
}
else if (currentInstruction[i].keyword == keywords::MULTIPLY) {
if (newToken.type == valtype::INT) {
newToken.value.value = get<int>(before.value.value) * get<int>(after.value.value);
}
else if (newToken.type == valtype::DEC) {
newToken.value.value = get<double>(before.value.value) * get<double>(after.value.value);
} else {
syntaxError.mathCannotDoOperationOnType("*", "bool or string");
}
}
else if (currentInstruction[i].keyword == keywords::DIVIDE) {
if (newToken.type == valtype::INT) {
newToken.value.value = get<int>(before.value.value) / get<int>(after.value.value);
}
else if (newToken.type == valtype::DEC) {
newToken.value.value = get<double>(before.value.value) / get<double>(after.value.value);
} else {
syntaxError.mathCannotDoOperationOnType("/", "bool or string");
}
} else {
// Something has gone terribly wrong
// We should never reach this point in the code
syntaxError.generalError("The math aint mathing");
}
// Insert our cool new token and get rid of the boring old stuff
currentInstruction[i - 1] = newToken;
currentInstruction.erase(currentInstruction.begin() + i);
currentInstruction.erase(currentInstruction.begin() + i);
i -= 1;
}
}
// Execute the instruction
executeCode(currentInstruction);
}
}
void executeCode(vector<Token> tokens) {
SyntaxError syntaxError;
log.debug("Token length for this expression is " + to_string(tokens.size()));
for (int i = 0; i < tokens.size(); i++) {
if (tokens[i].keyword == keywords::PRINT) {
i++;
if (tokens.size() <= i) break;
Token nextToken = tokens[i];
// Handle different value types
if (nextToken.keyword == keywords::VALUE) {
switch (nextToken.type) {
case valtype::INT:
cout << get<int>(nextToken.value.value) << endl;
break;
case valtype::DEC:
cout << get<double>(nextToken.value.value) << endl;
break;
case valtype::STR:
cout << get<string>(nextToken.value.value) << endl;
break;
case valtype::BOOL:
cout << (get<bool>(nextToken.value.value) ? "true" : "false") << endl;
break;
default:
vector<string> validTypes = {"int", "dec", "str", "bool"};
syntaxError.fnTypeMismatch("print", validTypes, nextToken.type);
}
} else {
syntaxError.fnNotSufficientArgs("print", 1, 1, 0);
}
} else if (tokens[i].keyword == keywords::EXIT) {
i++;
if (tokens.size() <= i) break;
Token nextToken = tokens[i];
if (nextToken.keyword == keywords::VALUE) {
switch (nextToken.type) {
case valtype::INT:
exit(get<int>(nextToken.value.value));
break;
case valtype::DEC:
exit(get<double>(nextToken.value.value));
break;
default:
vector<string> validTypes = {"int", "dec"};
syntaxError.fnTypeMismatch("exit", validTypes, nextToken.type);
}
}
} else if (tokens[i].keyword == keywords::LET) {
i++;
if (tokens.size() <= i + 2) {
syntaxError.fnNotSufficientArgs("let", 3, 3, tokens.size() - i);
break;
}
Token typeToken = tokens[i];
Token nameToken = tokens[i + 1];
Token valueToken = tokens[i + 2];
i += 2;
// Validate that we have a valid variable name
if (nameToken.type != valtype::STR) {
vector<string> validTypes = {"str"};
syntaxError.fnTypeMismatch("let (variable name)", validTypes, nameToken.type);
continue;
}
string varName = get<string>(nameToken.value.value);
Value newValue;
// Check the type declaration matches the value
if (typeToken.keyword == keywords::INT && valueToken.type == valtype::INT) {
newValue.type = valtype::INT;
newValue.value = get<int>(valueToken.value.value);
}
else if (typeToken.keyword == keywords::DEC && valueToken.type == valtype::DEC) {
newValue.type = valtype::DEC;
newValue.value = get<double>(valueToken.value.value);
}
else if (typeToken.keyword == keywords::STR && valueToken.type == valtype::STR) {
newValue.type = valtype::STR;
newValue.value = get<string>(valueToken.value.value);
}
else if (typeToken.keyword == keywords::BOOL && valueToken.type == valtype::BOOL) {
newValue.type = valtype::BOOL;
newValue.value = get<bool>(valueToken.value.value);
}
else {
vector<string> validTypes;
if (typeToken.keyword == keywords::INT) validTypes = {"int"};
else if (typeToken.keyword == keywords::DEC) validTypes = {"dec"};
else if (typeToken.keyword == keywords::STR) validTypes = {"str"};
else if (typeToken.keyword == keywords::BOOL) validTypes = {"bool"};
syntaxError.fnTypeMismatch("let", validTypes, valueToken.type, "Variable name is " + varName);
continue;
}
// Store the variable
variables[varName] = newValue;
} else {
if (tokens[i].keyword == keywords::VALUE && tokens[i].value.type == valtype::STR && variables.find(get<string>(tokens[i].value.value)) != variables.end()) {
log.debug("Manipulating variable...");
if (tokens.size() < i + 1) {
syntaxError.mathTooFewArgs();
}
string varName = get<string>(tokens[i].value.value);
i++;
if (variables[varName].type == valtype::INT) {
if (tokens[i].keyword == keywords::INCREMENT) {
variables[varName].value = get<int>(variables[varName].value) + 1;
} else if (tokens[i].keyword == keywords::DECREMENT) {
variables[varName].value = get<int>(variables[varName].value) - 1;
} else if (tokens[i].keyword == keywords::ADDTO) {
if (tokens.size() < i + 1) syntaxError.mathTooFewArgs();
if (tokens[i + 1].value.type != valtype::INT) syntaxError.mathTypeMismatch("Expected an int when adding to an int");
variables[varName].value = get<int>(variables[varName].value) + get<int>(tokens[i + 1].value.value);
} else if (tokens[i].keyword == keywords::SUBTRACTFROM) {
if (tokens.size() < i + 1) syntaxError.mathTooFewArgs();
if (tokens[i + 1].value.type != valtype::INT) syntaxError.mathTypeMismatch("Expected an int when subtracting from an int");
variables[varName].value = get<int>(variables[varName].value) - get<int>(tokens[i + 1].value.value);
}
} else if (variables[varName].type == valtype::DEC) {
if (tokens[i].keyword == keywords::INCREMENT) {
variables[varName].value = get<double>(variables[varName].value) + 1;
} else if (tokens[i].keyword == keywords::DECREMENT) {
variables[varName].value = get<double>(variables[varName].value) - 1;
} else if (tokens[i].keyword == keywords::ADDTO) {
if (tokens.size() < i + 1) syntaxError.mathTooFewArgs();
if (tokens[i + 1].value.type != valtype::DEC) syntaxError.mathTypeMismatch("Expected an dec when adding to an dec");
variables[varName].value = get<double>(variables[varName].value) + get<double>(tokens[i + 1].value.value);
} else if (tokens[i].keyword == keywords::SUBTRACTFROM) {
if (tokens.size() < i + 1) syntaxError.mathTooFewArgs();
if (tokens[i + 1].value.type != valtype::DEC) syntaxError.mathTypeMismatch("Expected an dec when subtracting from an dec");
variables[varName].value = get<double>(variables[varName].value) - get<double>(tokens[i + 1].value.value);
}
}
} else {
syntaxError.unknownFn();
}
}
}
}
};
#include <csignal>
#include "common.h"
#include "ArgParser.h"
#include "Logger.h"
#include "Parser.h"
#include "Interpreter.h"
int main(int argc, char* argv[]) {
// Parse and act upon command line arguments