Diff a tar file with the current contents on the filesystem

tardiff.py

#!/usr/bin/env python

import difflib
import sys
import tarfile

def read(fn):
    with open(fn) as f:
        return f.readlines()

def tardiff(fn):
    tf = tarfile.open(fn)
    for ti in tf:
        if not ti.isreg():
            continue
        orig = tf.extractfile(ti)
        orig_contents = orig.readlines()
        cur_contents = read(ti.name)
        if orig_contents == cur_contents:
            continue

        #print ti.name, 'is diff'
        for line in difflib.unified_diff(orig_contents, cur_contents, fromfile='a/' + ti.name, tofile='b/' + ti.name):
            sys.stdout.write(line)

if __name__ == "__main__":
    tardiff(sys.argv[1])