AST Struct Generator

.gitignore

@@ -1,2 +1,3 @@
 .vscode/
 build/
+.idea

Makefile

@@ -7,22 +7,26 @@ CC = g++
 CFLAGS = -Wall -Wextra -std=c++11
 
 # Source directory
-SRC_DIR = ./src
+SRC_DIR = ./source
 
 # Output directory
 BUILD_DIR = ./build
 
-# Get all CPP files in the source directory
-CPP_FILES := $(wildcard $(SRC_DIR)/*.cpp)
+# Find all C++ files recursively in the source directory
+CPP_FILES := $(shell find $(SRC_DIR) -type f -name '*.cpp')
 
 # Generate object file names by replacing the source directory with the build directory
 OBJ_FILES := $(patsubst $(SRC_DIR)/%.cpp,$(BUILD_DIR)/%.o,$(CPP_FILES))
 
+# Create directories for object files
+$(shell mkdir -p $(dir $(OBJ_FILES)))
+
 # Default target
 all: $(BUILD_DIR)/bob
 
-# Rule to create the build directory if it doesn't exist
-$(shell mkdir -p $(BUILD_DIR))
+# Rule to create necessary directories
+$(DIRS):
+	mkdir -p $(patsubst $(SRC_DIR)/%, $(OUTPUT_DIR)/%, $@)
 
 # Rule to compile object files
 $(BUILD_DIR)/%.o: $(SRC_DIR)/%.cpp

headers/Expression.h

@@ -0,0 +1,49 @@
+//
+// Created by Bobby Lucero on 5/21/23.
+//
+
+#pragma once
+#include "Lexer.h"
+#include <iostream>
+struct Expr{
+    virtual ~Expr()
+    {
+
+    }
+};
+
+struct BinaryExpr : Expr
+{
+    const Expr left;
+    const Token oper;
+    const Expr right;
+
+    BinaryExpr(Expr left, Token oper, Expr right) : left(left), oper(oper), right(right)
+    {
+    }
+};
+struct GroupingExpr : Expr
+{
+    const Expr expression;
+
+    GroupingExpr(Expr expression) : expression(expression)
+    {
+    }
+};
+struct LiteralExpr : Expr
+{
+    const std::string value;
+
+    LiteralExpr(std::string value) : value(value)
+    {
+    }
+};
+struct UnaryExpr : Expr
+{
+    const Token oper;
+    const Expr right;
+
+    UnaryExpr(Token oper, Expr right) : oper(oper), right(right)
+    {
+    }
+};

headers/Lexer.h

@@ -24,7 +24,21 @@ enum TokenType{
 };
 
 const std::map<std::string, TokenType> KEYWORDS {
-    
+        {"and", AND},
+        {"or", OR},
+        {"true", TRUE},
+        {"false", FALSE},
+        {"if", IF},
+        {"else", ELSE},
+        {"func", FUNCTION},
+        {"for", FOR},
+        {"while", WHILE},
+        {"var", VAR},
+        {"class", CLASS},
+        {"super", SUPER},
+        {"this", THIS},
+        {"none", NONE},
+        {"return", RETURN}
 };
 
 struct Token
@@ -41,6 +55,8 @@ public:
     std::vector<Token> Tokenize(std::string source);
 private:
     int line;
+    std::vector<char> src;
 private:
-    bool matchOn(char expected, std::vector<char>& src);
+    bool matchOn(char expected);
+    void advance();
 };

headers/bob.h

@@ -0,0 +1,31 @@
+#pragma once
+
+#include <iostream>
+#include <fstream>
+#include <string>
+#include "../headers/Lexer.h"
+
+#define VERSION "0.0.1"
+
+class Bob
+{
+public:
+    Lexer lexer;
+
+public:
+    void runFile(std::string path);
+
+    void runPrompt();
+
+    void error(int line, std::string message);
+
+
+private:
+    bool hadError = false;
+
+private:
+    void run(std::string source);
+
+    void report(int line, std::string where, std::string message);
+};
+

headers/helperFunctions/HelperFunctions.h

@@ -0,0 +1,40 @@
+#pragma once
+
+#include <iostream>
+#include <vector>
+
+std::vector<std::string> splitString(const std::string& input, std::string delimiter) {
+    std::vector<std::string> tokens;
+    std::string token;
+    size_t start = 0;
+    size_t end = input.find(delimiter);
+
+    while (end != std::string::npos) {
+        token = input.substr(start, end - start);
+        tokens.push_back(token);
+        start = end + 1;
+        end = input.find(delimiter, start);
+    }
+
+    // Add the last token (after the last delimiter)
+    token = input.substr(start, end);
+    tokens.push_back(token);
+
+    return tokens;
+}
+
+std::string trim(const std::string& str) {
+    // Find the first non-whitespace character
+    size_t start = str.find_first_not_of(" \t\n\r");
+
+    // If the string is all whitespace, return an empty string
+    if (start == std::string::npos) {
+        return "";
+    }
+
+    // Find the last non-whitespace character
+    size_t end = str.find_last_not_of(" \t\n\r");
+
+    // Extract the trimmed substring
+    return str.substr(start, end - start + 1);
+}

source.bob

@@ -2,7 +2,15 @@ bob.test
 10
 11.1
 
-test = (11 + 2 "xs")
+test = (11 + 2 "xs
+hello
+end")
 
-//"
-//11.
+//
+//11.
+
+12//11
+11.
+11.69 + 66.735293857293875 + 235982735987235.0 + 1
+
+123a

source/Expression.cpp

@@ -0,0 +1,5 @@
+//
+// Created by Bobby Lucero on 5/21/23.
+//
+
+#include "../headers/Expression.h"

source/Lexer.cpp

@@ -1,135 +1,142 @@
 #include "../headers/Lexer.h"
-#include <iostream>
 #include <cctype>
 
-
 using namespace std;
 
 std::vector<Token> Lexer::Tokenize(std::string source){
     std::vector<Token> tokens;
-    std::vector<char> src{source.begin(), source.end()};
+    src = std::vector<char>{source.begin(), source.end()};
+    line = 0;
 
-    while(src.size() > 0)
+    while(!src.empty())
     {
         char t = src[0];
         if(t == '(')
         {
-            tokens.push_back(Token{OPEN_PAREN, std::string(1, t)}); //brace initialization in case you forget
-            src.erase(src.begin());
+            tokens.push_back(Token{OPEN_PAREN, std::string(1, t), line}); //brace initialization in case you forget
+            advance();
         }
         else if(t == ')')
         {
-            tokens.push_back(Token{CLOSE_PAREN, std::string(1, t)});
-            src.erase(src.begin());
+            tokens.push_back(Token{CLOSE_PAREN, std::string(1, t), line});
+            advance();
         }
         else if(t == '{')
         {
-            tokens.push_back(Token{OPEN_BRACE, std::string(1, t)});
-            src.erase(src.begin());
+            tokens.push_back(Token{OPEN_BRACE, std::string(1, t), line});
+            advance();
         }
         else if(t == '}')
         {
-            tokens.push_back(Token{CLOSE_BRACE, std::string(1, t)});
-            src.erase(src.begin());
+            tokens.push_back(Token{CLOSE_BRACE, std::string(1, t), line});
+            advance();
         }
         else if(t == ',')
         {
-            tokens.push_back(Token{COMMA, std::string(1, t)});
-            src.erase(src.begin());
+            tokens.push_back(Token{COMMA, std::string(1, t), line});
+            advance();
         }
         else if(t == '.')
         {
-            tokens.push_back(Token{DOT, std::string(1, t)});
-            src.erase(src.begin());
+            tokens.push_back(Token{DOT, std::string(1, t), line});
+            advance();
         }
         else if(t == ';')
         {
-            tokens.push_back(Token{SEMICOLON, std::string(1, t)});
-            src.erase(src.begin());
+            tokens.push_back(Token{SEMICOLON, std::string(1, t), line});
+            advance();
         }
         else if(t == '+')
         {
-            tokens.push_back(Token{PLUS, std::string(1, t)});
-            src.erase(src.begin());
+            tokens.push_back(Token{PLUS, std::string(1, t), line});
+            advance();
         }
         else if(t == '-')
         {
-            tokens.push_back(Token{MINUS, std::string(1, t)});
-            src.erase(src.begin());
+            tokens.push_back(Token{MINUS, std::string(1, t), line});
+            advance();
         }
-         else if(t == '*')
+        else if(t == '*')
         {
-            tokens.push_back(Token{STAR, std::string(1, t)});
-            src.erase(src.begin());
+            tokens.push_back(Token{STAR, std::string(1, t), line});
+            advance();
         }
         else if(t == '=')
         {
             std::string token = std::string(1, t);
-            src.erase(src.begin());
-            bool match = matchOn('=', src);
+            advance();
+            bool match = matchOn('=');
             token += match ? "=" : "";
-            tokens.push_back(Token{match ? DOUBLE_EQUAL : EQUAL, token});
+            tokens.push_back(Token{match ? DOUBLE_EQUAL : EQUAL, token, line});
         }
         else if(t == '!')
         {
             std::string token = std::string(1, t);
-            src.erase(src.begin());
-            bool match = matchOn('=', src);
+            advance();
+            bool match = matchOn('=');
             token += match ? "=" : "";
-            tokens.push_back(Token{match ? BANG_EQUAL : BANG, token});
+            tokens.push_back(Token{match ? BANG_EQUAL : BANG, token, line});
         }
         else if(t == '<')
         {
             std::string token = std::string(1, t);
-            src.erase(src.begin());
-            bool match = matchOn('=', src);
+            advance();
+            bool match = matchOn('=');
             token += match ? "=" : "";
-            tokens.push_back(Token{match ? LESS_EQUAL : LESS, token});
+            tokens.push_back(Token{match ? LESS_EQUAL : LESS, token, line});
         }
         else if(t == '>')
         {
             std::string token = std::string(1, t);
-            src.erase(src.begin());
-            bool match = matchOn('=', src);
+            advance();
+            bool match = matchOn('=');
             token += match ? "=" : "";
-            tokens.push_back(Token{match ? GREATER_EQUAL : GREATER, token});
+            tokens.push_back(Token{match ? GREATER_EQUAL : GREATER, token, line});
+        }
+        else if(t == '&')
+        {
+            std::string token = std::string(1, t);
+            advance();
+            bool match = matchOn('&');
+            token += match ? "&" : "";
+            if(match) tokens.push_back(Token{AND, token, line});
         }
         else if(t == '/')
         {
             std::string token = std::string(1, t);
-            src.erase(src.begin());
-            bool match = matchOn('/', src);
+            advance();
+            bool match = matchOn('/');
             if(match)
             {
-                while(src.size() > 0 && src[0] != '\n')
+                while(!src.empty() && src[0] != '\n')
                 {
-                    src.erase(src.begin());
+                    advance();
                 }
             }
             else
             {
-                tokens.push_back(Token{SLASH, std::string(1, t)});
+                tokens.push_back(Token{SLASH, std::string(1, t), line});
             }
         }
         else if(t == '"')
         {
             std::string str = std::string(1, src[0]);
-            src.erase(src.begin());
-            while(src.size() > 0 && src[0] != '"')
+            advance();
+            while(!src.empty() && src[0] != '"')
             {
                 if(src[0] == '\n') line++;
                 str += src[0];
-                src.erase(src.begin());
+                advance();
             }
-            if(src.size() == 0)
+            if(src.empty())
             {
                 throw std::runtime_error("Unterminated string at line: " + std::to_string(this->line));
             }
             else if(src[0] == '"')
             {
                 str += '"';
-                src.erase(src.begin());
-                tokens.push_back(Token{STRING, str});
+                advance();
+                tokens.push_back(Token{STRING, str, line});
             }
             
 
@@ -137,30 +144,30 @@ std::vector<Token> Lexer::Tokenize(std::string source){
         else if(t == '\n')
         {
             line++;
-            src.erase(src.begin());
+            advance();
         }
         else
         {
             //Multi char tokens
             if(std::isdigit(t))
             {
-                std::string num = "";
-                while(src.size() > 0 && std::isdigit(src[0]))
+                std::string num;
+                while(!src.empty() && std::isdigit(src[0]))
                 {
                     num += src[0];
-                    src.erase(src.begin());
+                    advance();
                 }
 
-                if(src.size() > 0 && src[0] == '.')
+                if(!src.empty() && src[0] == '.')
                 {
-                    src.erase(src.begin());
-                    if(src.size() > 0 && std::isdigit(src[0]))
+                    advance();
+                    if(!src.empty() && std::isdigit(src[0]))
                     {
                         num += '.';
-                        while(src.size() > 0 && std::isdigit(src[0]))
+                        while(!src.empty() && std::isdigit(src[0]))
                         {
                             num += src[0];
-                            src.erase(src.begin());
+                            advance();
                         }
                     }
                     else
@@ -170,30 +177,30 @@ std::vector<Token> Lexer::Tokenize(std::string source){
                     
                 }
 
-                tokens.push_back(Token{NUMBER, num});
+                tokens.push_back(Token{NUMBER, num, line});
             }
             else if(std::isalpha(t))
             {
-                std::string ident = "";
-                while(src.size() > 0 && std::isalpha(src[0]))
+                std::string ident;
+                while(!src.empty() && std::isalpha(src[0]))
                 {
                     ident += src[0];
-                    src.erase(src.begin());
+                    advance();
                 }
 
                 if(KEYWORDS.find(ident) != KEYWORDS.end()) //identifier is a keyword
                 {
-                    tokens.push_back(Token{KEYWORDS.at(ident), ident});
+                    tokens.push_back(Token{KEYWORDS.at(ident), ident, line});
                 }
                 else
                 {
-                    tokens.push_back(Token{IDENTIFIER, ident});
+                    tokens.push_back(Token{IDENTIFIER, ident, line});
                 }
 
             }
-            else if(t == ' ' || t == '\t' || t == '\n')
+            else if(t == ' ' || t == '\t')
             {
-                    src.erase(src.begin()); //ignore t
+                    advance();
             }
             else
             {
@@ -209,10 +216,15 @@ std::vector<Token> Lexer::Tokenize(std::string source){
     return tokens;
 }
 
-bool Lexer::matchOn(char expected, std::vector<char> &src)
+bool Lexer::matchOn(char expected)
 {
-    if(src.size() == 0) return false;
+    if(src.empty()) return false;
     if(src[0] != expected) return false;
-    src.erase(src.begin());
+    advance();
     return true;
 }
+
+void Lexer::advance()
+{
+    src.erase(src.begin());
+}

source/bob.cpp

@@ -0,0 +1,60 @@
+#include "../headers/bob.h"
+using namespace std;
+
+void Bob::runFile(string path)
+{
+    ifstream file = ifstream(path);
+
+    string source = "";
+
+    if(file.is_open()){
+        source = string(istreambuf_iterator<char>(file), istreambuf_iterator<char>());
+    }
+    else
+    {
+        cout << "File not found" << endl;
+        return;
+    }
+
+    this->run(source);
+}
+
+void Bob::runPrompt()
+{
+    cout << "Bob v" << VERSION << ", 2023" << endl;
+    for(;;)
+    {
+        string line;
+        cout << "-> ";
+        std::getline(std::cin, line);
+
+        if(std::cin.eof())
+        {
+            break;
+        }
+
+        this->run(line);
+        hadError = false;
+    }
+}
+
+void Bob::error(int line, string message)
+{
+
+}
+
+void Bob::run(string source)
+{
+    vector<Token> tokens = lexer.Tokenize(source);
+
+    for(Token t : tokens){
+        cout << "{type: " << t.type << ", value: " << t.lexeme << "}" << endl;
+    }
+}
+
+void Bob::report(int line, string where, string message)
+{
+    hadError = true;
+}
+
+

source/main.cpp

@@ -0,0 +1,37 @@
+//
+// Created by Bobby Lucero on 5/21/23.
+//
+#include "../headers/bob.h"
+#include "../headers/Expression.h"
+#include "../headers/Lexer.h"
+int main(){
+
+    Bob bobLang;
+
+    //bobLang.runFile("source.bob");
+
+
+    Expr a;
+    Expr b;
+    Token t = {PLUS, "+", 1};
+    Token t2 = {MINUS, "-", 1};
+    BinaryExpr e = BinaryExpr(a, t, b);
+
+    std::shared_ptr<Expr> any = std::make_shared<BinaryExpr>(a, t, b);
+    if(std::shared_ptr<BinaryExpr> binexpr = std::dynamic_pointer_cast<BinaryExpr>(any))
+    {
+        std::cout << binexpr->oper.lexeme;
+    }
+
+    any = std::make_shared<BinaryExpr>(a, t2, b);
+    if(std::shared_ptr<BinaryExpr> binexpr = std::dynamic_pointer_cast<BinaryExpr>(any))
+    {
+        std::cout << binexpr->oper.lexeme;
+    }
+
+    std::cout << std::endl;
+
+    bobLang.runPrompt();
+
+    return 0;
+}

src/bob.cpp

@@ -1,96 +0,0 @@
-#include <iostream>
-#include <fstream>
-#include <string>
-#include "../headers/Lexer.h"
-
-#define VERSION "0.0.1"
-
-using namespace std;
-
-class Bob
-{
-public:
-	Lexer lexer;
-
-public:
-	void runFile(string path)
-	{
-		ifstream file = ifstream(path); 
-	
-		string source = "";
-
-		if(file.is_open()){
-			source = string(istreambuf_iterator<char>(file), istreambuf_iterator<char>());
-		}
-		else
-		{
-			cout << "File not found" << endl;
-			return;
-		}
-
-		this->run(source);
-	}
-
-	void runPrompt()
-	{
-		cout << "Bob v" << VERSION << ", 2023" << endl;
-		for(;;)
-		{
-			string line;
-			cout << "-> ";
-			std::getline(std::cin, line);
-
-			if(std::cin.eof())
-			{
-				break;
-			}
-
-			this->run(line);
-			hadError = false;
-		}
-	}
-
-	void error(int line, string message)
-	{
-
-	}
-	
-
-private:
-	bool hadError = false;
-
-private:
-	void run(string source)
-	{
-		vector<Token> tokens = lexer.Tokenize(source);
-
-		for(Token t : tokens){
-			cout << "{type: " << t.type << ", value: " << t.lexeme << "}" << endl;
-		}
-	}
-
-	void report(int line, string where, string message)
-	{
-		hadError = true;
-	}
-};
-int main(){
-
-	// string TokenTypeMappings[] = {
-	// 	"Identifier",
-	// 	"Number",
-	// 	"Equals",
-	// 	"OpenParen",
-	// 	"CloseParen",
-	// 	"BinaryOperator",
-	// 	"TestKeyword"
-	// };
-	
-	Bob bobLang;
-
-	bobLang.runFile("source.bob");
-	//bobLang.runPrompt();
-
-
-	return 0;
-}

testthing

@@ -0,0 +1,35 @@
+struct BinaryExpr : Expr
+{
+    const Expr left;
+    const Token oper;
+    const Expr right;
+
+    BinaryExpr(Expr left, Token oper, Expr right) : left(left), oper(oper), right(right)
+    {
+    }
+};
+struct GroupingExpr : Expr
+{
+    const Expr expression;
+
+    GroupingExpr(Expr expression) : expression(expression)
+    {
+    }
+};
+struct LiteralExpr : Expr
+{
+    const std::string value;
+
+    LiteralExpr(std::string value) : value(value)
+    {
+    }
+};
+struct UnaryExpr : Expr
+{
+    const Token oper;
+    const Expr right;
+
+    UnaryExpr(Token oper, Expr right) : oper(oper), right(right)
+    {
+    }
+};

tools/GenerateAST.cpp

@@ -0,0 +1,72 @@
+//
+// Created by Bobby Lucero on 5/21/23.
+//
+#include <iostream>
+#include <vector>
+#include <fstream>
+#include "../headers/helperFunctions/HelperFunctions.h"
+
+void defineType(std::ofstream &out, std::string baseName, std::string className, std::string fieldList)
+{
+    out << "struct " << className << "Expr : " << baseName << "\n{\n";
+    std::vector<std::string> fields = splitString(fieldList, ", ");
+    for(std::string field : fields)
+    {
+        out << "    const " << trim(field) << ";\n";
+    }
+
+    out << "\n    " << className << "Expr(" << fieldList << ") : ";
+
+    std::string explicitDeclaration;
+    for(std::string field : fields)
+    {
+        std::string name = splitString(trim(field), " ")[1];
+
+        explicitDeclaration += trim(name) + "(" + trim(name) + "), ";
+    }
+    explicitDeclaration = trim(explicitDeclaration);
+    explicitDeclaration.pop_back();
+
+    out << explicitDeclaration;
+
+    out << "\n    {\n";
+    out << "    }\n";
+    out << "};" << std::endl;
+}
+
+void defineAst(std::string outputDir, std::string baseName, const std::vector<std::string>& types)
+{
+    std::ofstream outFile(outputDir);
+
+    if(outFile.is_open())
+    {
+        for(std::string type : types)
+        {
+            std::vector<std::string> type_split = splitString(type, "$");
+            std::string className = trim(type_split[0]);
+            std::string fields = trim(type_split[1]);
+            defineType(outFile, baseName, className, fields);
+        }
+    }
+}
+
+
+
+int main(int argc, char* argv[]){
+    if(argc != 2)
+    {
+        std::cerr << "Usage " << argv[0] << " <output directory>" << std::endl;
+        std::exit(64);
+    }
+    std::string outputDir = argv[1];
+
+    defineAst(outputDir, "Expr", {
+            "Binary   $ Expr left, Token oper, Expr right",
+            "Grouping $ Expr expression",
+            "Literal  $ std::string value",
+            "Unary    $ Token oper, Expr right"
+    });
+}
+
+
+