Simple assembler to generate bytecode from asm text instructions for a custom-built microprocessor

tiasm.py

#!/usr/bin/env python3
#
# Valentin Dornauer <valentin@unimplemented.org>
# Sun Dec 11 22:11:27 CET 2016
#
# Simple assembler to generate bytecode from asm
# text instructions for a custom-built microprocessor.
#
# Usage: tiasm.py <input_file>

# Example programm (multiplication 5*3):
#
# # erster faktor: 5
# LDA #5
# STA 14
# STA 13
# # zweiter faktor: 3
# LDA #3
# STA 15
# SUB #1
# STA 15
# BRZ #6
# LDA 13
# ADD 14
# STA 13
# LDA 15
# JMP 5
# NOP
# NOP
# NOP

import sys
import re

def generate_mword(lower, upper):
    return lower | upper << 4

def ascii2mword(input):
    match = re.match(r"^([A-Za-z]+)\s*([#~]?)([0-9]*)$", input)

    if not match:
        print("# ERROR: don't know what to do with '%s'", input)
        return 0

    opcode = match.group(1).upper()
    is_immediate = len(match.group(2)) > 0
    try:
        immediate_value = int(match.group(3))
    except ValueError:
        immediate_value = None

    if opcode == "NOP":
        return 0
    elif opcode == "LDA":
        return generate_mword(0b0001 if is_immediate else 0b0010,
            immediate_value)
    elif opcode == "STA":
        return generate_mword(0b0011, immediate_value)
    elif opcode == "ADD":
        return generate_mword(0b0100 if is_immediate else 0b0101,
            immediate_value)
    elif opcode == "SUB":
        return generate_mword(0b0110 if is_immediate else 0b0111,
            immediate_value)
    elif opcode == "JMP":
        return generate_mword(0b1000, immediate_value)
    elif opcode == "BRZ":
        return generate_mword(0b1001, immediate_value)
    elif opcode == "BRC":
        return generate_mword(0b1010, immediate_value)
    elif opcode == "BRN":
        return generate_mword(0b1011, immediate_value)
    else:
        print("# ERROR: Unknown instruction %s" % opcode)
        return 0

if __name__ == "__main__":
    if len(sys.argv) < 2:
        print("Usage: tiasm.py <input_file>")
    input_file = sys.argv[1]
    with open(input_file) as fh:
        for line in fh.readlines():
            if line.startswith("#"):
                continue
            mword = ascii2mword(line.rstrip())
            print("# %s" % line.rstrip())
            print("{0:08b}".format(mword))

Output for example program:

# LDA #5
01010001
# STA 14
11100011
# STA 13
11010011
# LDA #3
00110001
# STA 15
11110011
# SUB #1
00010110
# STA 15
11110011
# BRZ #6
01101001
# LDA 13
11010010
# ADD 14
11100101
# STA 13
11010011
# LDA 15
11110010
# JMP 5
01011000
# NOP
00000000
# NOP
00000000
# NOP
00000000