assm-test/src/main.c

#include <bits/types/FILE.h>
#include <stdlib.h>
#include <stdio.h>
#include <string.h>
#include <sysexits.h>
#include "../includes/token.h"
#include "../includes/futil.h"
#include "../includes/scanner.h"
#include "../includes/parser.h"

const char* MagicStartName = "__start";
TokenList* LIST;
SymbolTable* Symbols = NULL;

unsigned char mem[128] = {0};
void print(void);
void assemble(void);

int main(int argc, char* args[]) {
    if (argc == 1) {
        printf("Usage: assm <file1.asm>\n");
        return EX_USAGE;
    }

    atexit(print);

    char* source_code;
    size_t bytes_read;

    if (!ReadAllString(args[1], &source_code, &bytes_read)) return EX_IOERR;

    LIST = GenerateTokenList(source_code);
    
    ParseTokens(LIST, &Symbols);

    free(source_code);

    assemble();

    printf("\nBytes:\n");
    for(unsigned long i = 0; i < sizeof(mem); i++) { 
        if (i != 0 && i % 8 == 0) printf("\n");
        printf("%02X ", mem[i] & 0xFF);
    }
    printf("\n");
}

void print(void) {
    char mnemonic[12];
    printf("printing tokens...\n");


    for(int i = 0; i < LIST->size; i++) {
        Token* t = (Token*) LIST->content[i];

        if (t->EndOfFile) {
            printf("EOF\n");
            break;
        }

        if (t->Class == PunctuationClass){
            if(t->Value.Punctuation == NewLine) {
                printf("<%d>\n", t->LineNumber);
                continue;
            }

            printf("[P]%c", t->Value.Punctuation);
            continue;
        }

        if (t->Class == LabelClass) {
            printf("[L]%s*", t->Lemexe);
            continue;
        }

        if (t->Class == IdentifierClass) {
            printf("[I]%s ", t->Lemexe);
            continue;
        }
        
        if (t->Class == RegisterClass) {
            printf("[R]%d", t->Value.Register);
            continue;
        }

        if (t->Class == NumberClass) {
            printf("[N]%d ", t->Value.Number);
            continue;
        }

        if (t->Class == MnemonicClass) {
            GetMnemonicText(t->Value.Mnemonic, mnemonic);
            printf("[M]%s ", mnemonic);
            continue;
        }

        if (t->Class == DirectiveClass) {
            printf("[D]%d ", t->Value.Directive);

        }

        if (t->Class == CharacterClass) {
            printf("[C]'%s'", t->Lemexe);
        }
    } 
}

int CurrentIndex = 0;
int PC2 = 0;

void WriteByte(unsigned char byte) {
    mem[PC2] = byte;
    PC2++;
}

void WriteWord(unsigned short word) {
    mem[PC2] = (word >> 8) & 0xFF;
    mem[PC2 + 1] = word & 0xFF;
    PC2 += 2;
}

unsigned short GetValueFromToken(Token* token) {
    if (token->Class & IdentifierClass) {
        Symbol* symbol;

        if (TryGetSymbol(token->Lemexe, Symbols, &symbol)) {
            return symbol->Address;
        } 
        else {
            fprintf(stderr, "Failed to find symbol %s while assembling.\n", token->Lemexe);

            exit(1);
        }
    }
    else if (token->Class == NumberClass) return token->Value.Number;

    fprintf(stderr, "Failed to find token '%d' while assmbling.\n", token->Class);

    exit(1);
}

int IsAtEnd(void){
    return CurrentIndex >= LIST->size;
}

void assemble() {
    while(!IsAtEnd()) {
        Token* current = LIST->content[CurrentIndex];
        switch(current->Value.Mnemonic) {
            case COPYA:
            case COPYAB:
                CurrentIndex++;

                WriteByte(current->Value.Mnemonic | ((Token*) LIST->content[CurrentIndex])->Value.Register);
                CurrentIndex++;
                WriteWord(GetValueFromToken(LIST->content[CurrentIndex]));
                break;
            case COPYRA:
            case COPYRAB:
            case COPYRARA:
            case COPYRARAB:
            case COPY:
                CurrentIndex++;

                WriteByte(current->Value.Mnemonic | ((Token*) LIST->content[CurrentIndex])->Value.Register);
                CurrentIndex++;
                WriteByte(((Token*) LIST->content[CurrentIndex])->Value.Register);
                break;
            case COPYI: //TODO: should explicitly get number from value I feel
                CurrentIndex++;

                WriteByte(current->Value.Mnemonic | ((Token*) LIST->content[CurrentIndex])->Value.Register);
                CurrentIndex++;
                WriteWord(GetValueFromToken(LIST->content[CurrentIndex]));
                break;
            case COPYB:
                CurrentIndex++;

                WriteByte(current->Value.Mnemonic | ((Token*) LIST->content[CurrentIndex])->Value.Register);
                CurrentIndex++;
                WriteByte((unsigned char)GetValueFromToken(LIST->content[CurrentIndex]));
                break;
            case CMP:
                CurrentIndex++;

                WriteByte(current->Value.Mnemonic | ((Token*) LIST->content[CurrentIndex])->Value.Register);
                CurrentIndex++;
                WriteByte(((Token*) LIST->content[CurrentIndex])->Value.Register);
                break;
            case CMPI:
                CurrentIndex++;

                WriteByte(current->Value.Mnemonic | ((Token*) LIST->content[CurrentIndex])->Value.Register);
                CurrentIndex++;
                WriteWord(GetValueFromToken(LIST->content[CurrentIndex]));
                break;
            case ADD:
            case SUB:
            case AND:
            case OR:
            case XOR:
                CurrentIndex++;

                WriteByte(current->Value.Mnemonic | ((Token*) LIST->content[CurrentIndex])->Value.Register);
                CurrentIndex++;
                WriteByte(((Token*) LIST->content[CurrentIndex])->Value.Register);
                break;
            case SHL:
            case SHR:
                CurrentIndex++;

                WriteByte(current->Value.Mnemonic | ((Token*) LIST->content[CurrentIndex])->Value.Register);
                CurrentIndex++;
                WriteWord(GetValueFromToken(LIST->content[CurrentIndex]));
                break;
            case INC:
            case DEC:
            case PUSH:
            case POP:
            case JMPI:
                CurrentIndex++;

                WriteByte(current->Value.Mnemonic | ((Token*) LIST->content[CurrentIndex])->Value.Register);
                break;
            case JMP:
            case JZ:
            case JG:
            case JL:
                WriteByte(current->Value.Mnemonic);
                CurrentIndex++;
                WriteWord(GetValueFromToken(LIST->content[CurrentIndex]));
                break;
            case NOP:
            case RET:
                WriteByte(current->Value.Mnemonic);
                break;
            case CALL:
                WriteByte(current->Value.Mnemonic);
                CurrentIndex++;
                WriteWord(GetValueFromToken(LIST->content[CurrentIndex]));
            default:
                break;
        }

        CurrentIndex++;
    }
}