convert.py

#!/usr/bin/env python

from __future__ import with_statement
import os, re, argparse

# Convert one machine code format to another - 
# rename this with no py extension, make it executable, and put it in your bash PATH to make it a commandline tool

# This uses the argparse module to parse args to the commandline tool
parser = argparse.ArgumentParser()
parser.add_argument("src_file", help="The source text file of machine code to convert.")
parser.add_argument('-d', '--destination', help="Specify the destination file. Uses <origname>_conv.txt otherwise.")

# Grab the arguments to use (could add more for different conversion formats or whatever)
args = parser.parse_args()
destination = args.destination
src_file = args.src_file

if destination:
	destination_name = destination
else:
	destination_name = os.path.splitext(src_file)[0]+'_conv.txt'


# Open the original file for reading, and the new file for writing
with open(src_file, 'r+') as src_f:
	with open(destination_name, 'w+') as dst_f:
		
		# Loop through each line of the file		
		for line in src_f:
			# Compile a regular expression that matches Y## or X##
			query = re.compile('([X,Y][0-9]{2})')
			
			# Use re.sub to take the match and add a '.' character to the end
			line_converted = re.sub(query, r'\1.', line)
			# You could also use match group 0 instead of \1 like this but it's more verbose
			# lambda creates a function in-line
			# line_converted = re.sub(query, lambda match: match.group(0)+'.', line)

			# Write our converted line to the new file
			dst_f.write(line_converted)
	

# Or you could use bash with sed if you put this in your ~/.bash_profile
# function convert { (sed -E 's/([X,Y][0-9]{2})/\1./g' $1 > ${1}_conv;) }
# -E uses extended modern regular expressions so you don't have to escape the group definitions
# \1 refers to the first group (same as $1 in sublime/perl/java)


# Example to open each line of a text file and do stuff to it.
from __future__ import with_statement
   bg_db_path = '/Volumes/af/show/wash/common/nuke/RF_SEQ_BG_DB'
    bg_db = {}
    with open(bg_db_path) as f:
        for line in f:
            line = line.strip()
            # Skip commented and blank lines
            if line.startswith('#') or not line:
                continue
            line_info = line.split('\t')
            bg_db[line_info[0]] = line_info[1:3]