Compare commits
4 Commits
V0.0.3-bet
...
V0.0.5
| Author | SHA1 | Date | |
|---|---|---|---|
| 5a24fa900e | |||
|
|
9bbf8435c6 | ||
|
|
e250b2f5fb | ||
|
|
aa6d16bc52 |
@@ -22,12 +22,14 @@
|
||||
<ClCompile Include="codegen.cpp" />
|
||||
<ClCompile Include="main.cpp" />
|
||||
<ClCompile Include="parser.cpp" />
|
||||
<ClCompile Include="preprocessor.cpp" />
|
||||
<ClCompile Include="utils.cpp" />
|
||||
</ItemGroup>
|
||||
<ItemGroup>
|
||||
<ClInclude Include="codegen.h" />
|
||||
<ClInclude Include="compiler_types.h" />
|
||||
<ClInclude Include="parser.h" />
|
||||
<ClInclude Include="preprocessor.h" />
|
||||
<ClInclude Include="utils.h" />
|
||||
</ItemGroup>
|
||||
<PropertyGroup Label="Globals">
|
||||
|
||||
@@ -27,6 +27,9 @@
|
||||
<ClCompile Include="codegen.cpp">
|
||||
<Filter>Pliki źródłowe</Filter>
|
||||
</ClCompile>
|
||||
<ClCompile Include="preprocessor.cpp">
|
||||
<Filter>Pliki źródłowe</Filter>
|
||||
</ClCompile>
|
||||
</ItemGroup>
|
||||
<ItemGroup>
|
||||
<ClInclude Include="utils.h">
|
||||
@@ -41,5 +44,8 @@
|
||||
<ClInclude Include="compiler_types.h">
|
||||
<Filter>Pliki nagłówkowe</Filter>
|
||||
</ClInclude>
|
||||
<ClInclude Include="preprocessor.h">
|
||||
<Filter>Pliki nagłówkowe</Filter>
|
||||
</ClInclude>
|
||||
</ItemGroup>
|
||||
</Project>
|
||||
@@ -40,9 +40,19 @@ std::string generateAssembly(const CompilerState& state) {
|
||||
std::string result;
|
||||
result += "global main\n";
|
||||
result += "extern printf\n";
|
||||
result += "extern getchar\n"; // <--- DODAJ TO
|
||||
result += "extern getchar\n";
|
||||
result += "section .data\n";
|
||||
result += " fmt db '%d', 10, 0\n";
|
||||
result += "section .data\n";
|
||||
result += " fmt_int db '%d', 10, 0\n"; // Format dla liczb
|
||||
result += " fmt_str db '%s', 10, 0\n"; // NOWOŒÆ: Format dla stringów
|
||||
|
||||
// --- WYPISYWANIE STRINGÓW ---
|
||||
for (const auto& pair : state.stringLiterals) {
|
||||
// Nazwa etykiety: db 'Tresc', 0
|
||||
// Uwaga: ASM nie lubi pewnych znaków, ale zak³adamy proste litery
|
||||
result += " " + pair.second + " db '" + pair.first + "', 0\n";
|
||||
}
|
||||
result += "section .text\n\n";
|
||||
|
||||
for (const auto& pair : state.functions) {
|
||||
@@ -71,19 +81,33 @@ std::string generateAssembly(const CompilerState& state) {
|
||||
// 2. INSTRUKCJE
|
||||
for (const auto& instr : func.instructions) {
|
||||
|
||||
// Rezerwacja miejsca dla nowych zmiennych
|
||||
if ((instr.type == OpType::ASSIGN || instr.type == OpType::ADD || instr.type == OpType::EQ)
|
||||
&& stackMap.find(instr.arg1) == stackMap.end() && instr.arg1 != "RAX") {
|
||||
// Rezerwacja miejsca dla nowych zmiennych (wynikowych)
|
||||
// Dodajemy tu OpType::SUB i OpType::MUL
|
||||
bool isWriteOp = (instr.type == OpType::ASSIGN ||
|
||||
instr.type == OpType::ADD ||
|
||||
instr.type == OpType::EQ ||
|
||||
instr.type == OpType::SUB ||
|
||||
instr.type == OpType::MUL);
|
||||
|
||||
if (isWriteOp && stackMap.find(instr.arg1) == stackMap.end() && instr.arg1 != "RAX") {
|
||||
stackMap[instr.arg1] = currentStack;
|
||||
currentStack += 8;
|
||||
}
|
||||
|
||||
switch (instr.type) {
|
||||
case OpType::ASSIGN: {
|
||||
std::string src = getVarLocation(instr.arg2, stackMap);
|
||||
std::string src = instr.arg2;
|
||||
if (instr.arg3 == "STRING") {
|
||||
result += " lea rax, [rel " + src + "]\n";
|
||||
std::string dst = getVarLocation(instr.arg1, stackMap);
|
||||
result += " mov eax, " + src + "\n";
|
||||
result += " mov qword " + dst + ", rax\n";
|
||||
}
|
||||
else {
|
||||
std::string srcLoc = getVarLocation(instr.arg2, stackMap);
|
||||
std::string dst = getVarLocation(instr.arg1, stackMap);
|
||||
result += " mov eax, " + srcLoc + "\n";
|
||||
result += " mov " + dst + ", eax\n";
|
||||
}
|
||||
break;
|
||||
}
|
||||
case OpType::ADD: {
|
||||
@@ -95,6 +119,30 @@ std::string generateAssembly(const CompilerState& state) {
|
||||
result += " mov " + dst + ", eax\n";
|
||||
break;
|
||||
}
|
||||
case OpType::SUB: {
|
||||
// a = b - c
|
||||
std::string op1 = getVarLocation(instr.arg2, stackMap);
|
||||
std::string op2 = getVarLocation(instr.arg3, stackMap);
|
||||
std::string dst = getVarLocation(instr.arg1, stackMap);
|
||||
|
||||
result += " mov eax, " + op1 + "\n";
|
||||
result += " sub eax, " + op2 + "\n"; // sub = odejmowanie
|
||||
result += " mov " + dst + ", eax\n";
|
||||
break;
|
||||
}
|
||||
case OpType::MUL: {
|
||||
// a = b * c
|
||||
std::string op1 = getVarLocation(instr.arg2, stackMap);
|
||||
std::string op2 = getVarLocation(instr.arg3, stackMap);
|
||||
std::string dst = getVarLocation(instr.arg1, stackMap);
|
||||
|
||||
result += " mov eax, " + op1 + "\n";
|
||||
// Mno¿enie w x86 jest specyficzne: imul eax, operand
|
||||
// Wynik l¹duje w eax (i edx jeœli du¿y, ale ignorujemy nadmiar dla prostoty)
|
||||
result += " imul eax, " + op2 + "\n";
|
||||
result += " mov " + dst + ", eax\n";
|
||||
break;
|
||||
}
|
||||
case OpType::EQ: {
|
||||
std::string op1 = getVarLocation(instr.arg2, stackMap);
|
||||
std::string op2 = getVarLocation(instr.arg3, stackMap);
|
||||
@@ -144,10 +192,17 @@ std::string generateAssembly(const CompilerState& state) {
|
||||
break;
|
||||
}
|
||||
case OpType::PRINT: {
|
||||
if (instr.arg2 == "STRING") {
|
||||
result += " lea rdx, [rel " + instr.arg1 + "]\n";
|
||||
result += " lea rcx, [rel fmt_str]\n";
|
||||
result += " call printf\n";
|
||||
}
|
||||
else {
|
||||
std::string val = getVarLocation(instr.arg1, stackMap);
|
||||
result += " mov edx, " + val + "\n";
|
||||
result += " lea rcx, [rel fmt]\n";
|
||||
result += " lea rcx, [rel fmt_int]\n";
|
||||
result += " call printf\n";
|
||||
}
|
||||
break;
|
||||
}
|
||||
case OpType::CALL: {
|
||||
|
||||
@@ -51,6 +51,9 @@ struct CompilerState {
|
||||
std::stack<std::string> loopStack; // Do break/continue (przysz³oœciowo)
|
||||
|
||||
std::stack<std::string> blockStack;
|
||||
|
||||
std::map<std::string, std::string> stringLiterals;
|
||||
int stringCounter = 0; // Licznik do generowania nazw str_1, str_2...
|
||||
};
|
||||
|
||||
#endif
|
||||
|
||||
@@ -2,14 +2,24 @@
|
||||
#include <fstream>
|
||||
#include <string>
|
||||
#include <vector>
|
||||
#include <cstdlib> // do std::system
|
||||
#include <cstdlib>
|
||||
#include <filesystem>
|
||||
#include <windows.h>
|
||||
#include "compiler_types.h"
|
||||
#include "parser.h"
|
||||
#include "codegen.h"
|
||||
#include "preprocessor.h"
|
||||
|
||||
std::string getExecutablePath() {
|
||||
char buffer[MAX_PATH];
|
||||
GetModuleFileNameA(NULL, buffer, MAX_PATH);
|
||||
std::string::size_type pos = std::string(buffer).find_last_of("\\/");
|
||||
return std::string(buffer).substr(0, pos);
|
||||
}
|
||||
|
||||
int main(int argc, char* argv[]) {
|
||||
std::string inputFile, outputName;
|
||||
std::string Version = "v1.5.0-modular"; // Zaktualizowałem wersję :)
|
||||
std::string Version = "v0.0.5-beta";
|
||||
bool showHelp = false, showVersion = false, showCredits = false;
|
||||
|
||||
// --- PARSOWANIE ARGUMENTÓW ---
|
||||
@@ -68,10 +78,22 @@ int main(int argc, char* argv[]) {
|
||||
return 1;
|
||||
}
|
||||
|
||||
// ... wczytywanie pliku (to co miałeś) ...
|
||||
std::string src((std::istreambuf_iterator<char>(in)), std::istreambuf_iterator<char>());
|
||||
in.close();
|
||||
|
||||
// --- LOGIKA KOMPILATORA ---
|
||||
// --- PREPROCESSOR START ---
|
||||
std::cout << "[INFO] Preprocessing...\n";
|
||||
|
||||
// 1. Ścieżka projektu (tam gdzie plik wejściowy)
|
||||
std::filesystem::path p(inputFile);
|
||||
std::string projectDir = p.parent_path().string();
|
||||
|
||||
// 2. Ścieżka kompilatora (tam gdzie PCC.exe i folder std)
|
||||
std::string compilerDir = getExecutablePath();
|
||||
|
||||
// Uruchamiamy z obiema ścieżkami
|
||||
src = preprocessSource(src, projectDir, compilerDir);
|
||||
CompilerState state;
|
||||
|
||||
std::cout << "[INFO] Parsing code...\n";
|
||||
|
||||
@@ -92,10 +92,34 @@ void processSource(const std::string& src, CompilerState& state) {
|
||||
size_t start = line.find('(') + 1;
|
||||
size_t end = line.find(')');
|
||||
if (start != std::string::npos && end != std::string::npos) {
|
||||
std::string var = trim(line.substr(start, end - start));
|
||||
f.instructions.push_back({ OpType::PRINT, var, "", "" });
|
||||
std::string arg = trim(line.substr(start, end - start));
|
||||
|
||||
// Czy to bezpośredni napis? np. print("Hello")
|
||||
if (arg.size() >= 2 && arg.front() == '"' && arg.back() == '"') {
|
||||
std::string content = arg.substr(1, arg.size() - 2);
|
||||
|
||||
// Rejestrujemy
|
||||
std::string label;
|
||||
if (state.stringLiterals.count(content)) {
|
||||
label = state.stringLiterals[content];
|
||||
}
|
||||
else {
|
||||
label = "str_" + std::to_string(state.stringCounter++);
|
||||
state.stringLiterals[content] = label;
|
||||
}
|
||||
|
||||
// Dajemy znać generatorowi, że to typ STRING
|
||||
f.instructions.push_back({ OpType::PRINT, label, "STRING", "" });
|
||||
}
|
||||
else {
|
||||
// Zwykła zmienna (int lub string - generator musi zgadnąć lub my musimy wiedzieć)
|
||||
// Na razie załóżmy, że jeśli zmienna ma w nazwie "msg" lub "txt", to string
|
||||
// (To hack, w przyszłości dodamy tabelę typów zmiennych)
|
||||
f.instructions.push_back({ OpType::PRINT, arg, "VAR", "" });
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// C. IF STATEMENT
|
||||
else if (line.substr(0, 2) == "if") {
|
||||
size_t openParen = line.find("(");
|
||||
@@ -117,12 +141,19 @@ void processSource(const std::string& src, CompilerState& state) {
|
||||
size_t eqPos = line.find('=');
|
||||
std::string leftSide = trim(line.substr(0, eqPos));
|
||||
std::string rightSide = trim(line.substr(eqPos + 1));
|
||||
|
||||
bool isStringDecl = false; // Flaga, czy to string
|
||||
if (!rightSide.empty() && rightSide.back() == ';') rightSide.pop_back();
|
||||
|
||||
// Obsługa nazwy zmiennej (usuwanie "int ", "bool ")
|
||||
std::string varName = leftSide;
|
||||
if (leftSide.rfind("int ", 0) == 0) varName = trim(leftSide.substr(4));
|
||||
else if (leftSide.rfind("bool ", 0) == 0) varName = trim(leftSide.substr(5));
|
||||
else if (leftSide.rfind("string ", 0) == 0) { // NOWOŚĆ
|
||||
varName = trim(leftSide.substr(7));
|
||||
isStringDecl = true;
|
||||
}
|
||||
|
||||
|
||||
// 1. Czy to wywołanie funkcji? int x = func();
|
||||
if (rightSide.find("(") != std::string::npos && rightSide.find(")") != std::string::npos) {
|
||||
@@ -143,6 +174,20 @@ void processSource(const std::string& src, CompilerState& state) {
|
||||
std::string b = trim(rightSide.substr(opPos + 1));
|
||||
f.instructions.push_back({ OpType::ADD, varName, a, b });
|
||||
}
|
||||
// 3. NOWOŚĆ: Czy to odejmowanie? a - b
|
||||
else if (rightSide.find("-") != std::string::npos) {
|
||||
size_t opPos = rightSide.find("-");
|
||||
std::string a = trim(rightSide.substr(0, opPos));
|
||||
std::string b = trim(rightSide.substr(opPos + 1));
|
||||
f.instructions.push_back({ OpType::SUB, varName, a, b }); // <--- Używamy SUB
|
||||
}
|
||||
// 4. NOWOŚĆ: Czy to mnożenie? a * b
|
||||
else if (rightSide.find("*") != std::string::npos) {
|
||||
size_t opPos = rightSide.find("*");
|
||||
std::string a = trim(rightSide.substr(0, opPos));
|
||||
std::string b = trim(rightSide.substr(opPos + 1));
|
||||
f.instructions.push_back({ OpType::MUL, varName, a, b }); // <--- Używamy MUL
|
||||
}
|
||||
// 3. Czy to porównanie? a == b (Ważne: == może być w IFie, ale tu jesteśmy w linii z '=')
|
||||
// UWAGA: To rzadkie w C++ (bool x = a == b), ale obsłużmy proste przypisanie wartości logicznej
|
||||
else if (rightSide.find("==") != std::string::npos) {
|
||||
@@ -151,6 +196,26 @@ void processSource(const std::string& src, CompilerState& state) {
|
||||
std::string b = trim(rightSide.substr(opPos + 2));
|
||||
f.instructions.push_back({ OpType::EQ, varName, a, b });
|
||||
}
|
||||
else if (rightSide.size() >= 2 && rightSide.front() == '"' && rightSide.back() == '"')
|
||||
{
|
||||
{
|
||||
// Wyciągamy treść bez cudzysłowów
|
||||
std::string content = rightSide.substr(1, rightSide.size() - 2);
|
||||
|
||||
// Rejestrujemy stringa w sekcji danych, jeśli jeszcze go nie ma
|
||||
std::string label;
|
||||
if (state.stringLiterals.count(content)) {
|
||||
label = state.stringLiterals[content];
|
||||
}
|
||||
else {
|
||||
label = "str_" + std::to_string(state.stringCounter++);
|
||||
state.stringLiterals[content] = label;
|
||||
}
|
||||
|
||||
// Generujemy instrukcję przypisania ADRESU etykiety do zmiennej
|
||||
f.instructions.push_back({ OpType::ASSIGN, varName, label, "STRING" });
|
||||
}
|
||||
}
|
||||
// 4. Zwykłe przypisanie: a = 5
|
||||
else {
|
||||
f.instructions.push_back({ OpType::ASSIGN, varName, rightSide, "" });
|
||||
|
||||
79
PCCcompiler/preprocessor.cpp
Normal file
79
PCCcompiler/preprocessor.cpp
Normal file
@@ -0,0 +1,79 @@
|
||||
#include "preprocessor.h"
|
||||
#include <iostream>
|
||||
#include <sstream>
|
||||
#include <fstream>
|
||||
#include <filesystem>
|
||||
#include <regex>
|
||||
|
||||
namespace fs = std::filesystem;
|
||||
|
||||
// Funkcja pomocnicza do wczytania pliku
|
||||
std::string loadFileContent(const std::string& path) {
|
||||
std::ifstream in(path);
|
||||
if (!in) {
|
||||
std::cerr << "[PREPROCESSOR] Error: Could not open included file: " << path << "\n";
|
||||
return "";
|
||||
}
|
||||
return std::string((std::istreambuf_iterator<char>(in)), std::istreambuf_iterator<char>());
|
||||
}
|
||||
|
||||
std::string preprocessSource(const std::string& src, const std::string& projectDir, const std::string& compilerDir) {
|
||||
std::istringstream iss(src);
|
||||
std::string line;
|
||||
std::stringstream output;
|
||||
|
||||
while (std::getline(iss, line)) {
|
||||
// Szukamy: #include "..." lub #include <...>
|
||||
// U¿ywamy prostego find, ¿eby by³o szybko
|
||||
std::string trimLine = line;
|
||||
// Usuwamy bia³e znaki z pocz¹tku
|
||||
size_t first = trimLine.find_first_not_of(" \t");
|
||||
if (first != std::string::npos) trimLine = trimLine.substr(first);
|
||||
|
||||
if (trimLine.rfind("#include", 0) == 0) {
|
||||
// Mamy include!
|
||||
size_t openQuote = trimLine.find('"');
|
||||
size_t closeQuote = trimLine.rfind('"');
|
||||
size_t openAngle = trimLine.find('<');
|
||||
size_t closeAngle = trimLine.rfind('>');
|
||||
|
||||
std::string includePath;
|
||||
bool isStdLib = false;
|
||||
|
||||
// Wersja: #include "plik.pcc"
|
||||
if (openQuote != std::string::npos && closeQuote > openQuote) {
|
||||
includePath = trimLine.substr(openQuote + 1, closeQuote - openQuote - 1);
|
||||
}
|
||||
// Wersja: #include <plik.pcc>
|
||||
else if (openAngle != std::string::npos && closeAngle > openAngle) {
|
||||
includePath = trimLine.substr(openAngle + 1, closeAngle - openAngle - 1);
|
||||
isStdLib = true;
|
||||
}
|
||||
|
||||
if (!includePath.empty()) {
|
||||
std::string fullPath;
|
||||
if (isStdLib) {
|
||||
fullPath = compilerDir + "/std/" + includePath;
|
||||
std::cout << "[PREPROCESSOR] Including STD lib: " << fullPath << "\n";
|
||||
}
|
||||
else {
|
||||
// Plik lokalny: Szukamy w folderze projektu
|
||||
if (projectDir.empty()) fullPath = includePath;
|
||||
else fullPath = projectDir + "/" + includePath;
|
||||
std::cout << "[PREPROCESSOR] Including local file: " << fullPath << "\n";
|
||||
}
|
||||
std::string content = loadFileContent(fullPath);
|
||||
std::string processedContent = preprocessSource(content, projectDir, compilerDir);
|
||||
|
||||
output << "\n// --- BEGIN INCLUDE: " << includePath << " ---\n";
|
||||
output << processedContent;
|
||||
output << "\n// --- END INCLUDE ---\n";
|
||||
}
|
||||
}
|
||||
else {
|
||||
output << line << "\n";
|
||||
}
|
||||
}
|
||||
|
||||
return output.str();
|
||||
}
|
||||
8
PCCcompiler/preprocessor.h
Normal file
8
PCCcompiler/preprocessor.h
Normal file
@@ -0,0 +1,8 @@
|
||||
#ifndef PREPROCESSOR_H
|
||||
#define PREPROCESSOR_H
|
||||
|
||||
#include <string>
|
||||
std::string preprocessSource(const std::string& src, const std::string& projectDir, const std::string& compilerDir);
|
||||
|
||||
|
||||
#endif
|
||||
@@ -1,14 +1,18 @@
|
||||
# PCC Compiler (My C++ Compiler)
|
||||
|
||||

|
||||

|
||||

|
||||

|
||||
|
||||
## Docs
|
||||
https://nodrop.xyz/docs/docs.html
|
||||
|
||||
**PCC Compiler** is a custom programming language compiler built from scratch in C++. It translates PCC code into x64 Assembly (NASM), which is then linked into a standalone Windows executable.
|
||||
|
||||
## 🚀 Features
|
||||
- **Custom Syntax**: C-like syntax easy for beginners.
|
||||
- **Variables**: Support for `int` and `bool`.
|
||||
- **Variables**: Support for `int`, `bool` and `string`.
|
||||
- **include files**: you can include base files from compiler `#include <main.pcc>` or your own files `#include "myfile.pcc"`.
|
||||
- **Control Flow**: `if` statements support.
|
||||
- **Functions**: Define and call `void` functions.
|
||||
- **Native Compilation**: Compiles directly to x64 machine code.
|
||||
|
||||
Reference in New Issue
Block a user