Split an Amiga ROM file into even/odd words

amiromsplit.py

#!/usr/bin/env python

import os
import sys
import re

REGEXP = "AMIGA"
BUFSIZE = 4096
ODD_SUFFIX = "_odd_hi_uXa"
EVEN_SUFFIX = "_even_low_uXb"

if len (sys.argv) != 2:
	print "Usage: %s <romfile>" % sys.argv[0]
	sys.exit (1)

fn_in = sys.argv[1]
split = os.path.splitext (fn_in)
fn_odd = split[0] + ODD_SUFFIX + split[1]
fn_even = split[0] + EVEN_SUFFIX + split[1]

print ("Writing to %s and %s..." % (fn_odd, fn_even))

warningIssued = False
with open (fn_in) as fin, open (fn_odd, "w") as fodd, open (fn_even, "w") as feven:
	while True:
		buf = fin.read (BUFSIZE)
		if not buf:
			break
		if not warningIssued and re.search (REGEXP, buf, re.IGNORECASE):
			print ("WARNING: ROM does not look byteswapped, be careful!")
			warningIssued = True
		pairs = zip (buf[::2], buf[1::2])
		odd = pairs[0::2]
		even = pairs[1::2]
		for pair in odd:
			fodd.write (bytearray (pair))
		for pair in even:
			feven.write (bytearray (pair))

Like 13 - missing parens. Like 26 - problem with python because it defaults to utf-8 codec. I'm not a python guy though, so trying to figure out how to not make the split be on text but on binary instead

This was written for Python 2, try running it through the 2to3 tool.

Change w to wb and add rb in the other call for binary mode.

Thanks! I'll try in a bit. I used a tool written in C on OSX for now. It also does CRC checksums of a lot of the ROMs provided, so it'll ensure that it's not a wrong one. Pretty cool stuff.

Yeah, this was a quiiiick'n'dirty hack I wrote maaaany years ago (5+), I'm sure there's better stuff around but I could not find anything at the time.