XTRACT ZE FILES!

usage.md

xzf, XTRACT ZE FILES!, is a utility for file extraction, as you may have guessed. Inspired by xkcd and general boredom. Works on both python v2.7 and v3.3.

Usage:

$ python xzf.py file [extra_tar_flags]

Or, programatically:

from xzf import xtract_ze_files
xzf.extract_ze_files(file, extra_tar_flags)

The extra flags are passed directly to tar, so you can do something like:

$ python xzf.py some_file.tar.gz -v

And the output will be verbose. The command executed in this case will be:

tar --gzip -v -xf some_file.tar.gz

xzf.py

#!/usr/bin/python
#XTRACT ZE FILES!!!

import subprocess

#header => tar flag map
#you can find these with something along the lines of:
# hexdump -n 16 -C file
#(execute in console, of course)
headers = {
    #nothing
    b'' : '',

    #gzip
    # http://www.onicos.com/staff/iz/formats/gzip.html
    b'\x1f\x8b' : '--gzip',

    #bzip2
    # http://bzip.org/1.0.5/bzip2-manual-1.0.5.html
    #   (look for BZ_DATA_ERROR_MAGIC)
    b'BZ' : '--bzip2',
    #we assume bzip2, even though bzip1 files begin with BZ as well (the
    # distinction is in the 3rd byte, h in bzip2, 0 in bzip1)

    #TODO: check the following

    #xz
    # http://tukaani.org/xz/xz-file-format.txt
    #  (look for Header Magic Bytes)
    b'\xfd7zXZ\x00' : '--xz',
}

def xtract_ze_files (fn, extra_flags):
    tar_flag = get_tar_flag(fn)

    command = [
        'tar',

        tar_flag,
        extra_flags,
        '-xf',

        fn ]

    return subprocess.call(command)

#grabs the header, which is really the leading bytes of a file, depending on
# the longest header we have in stock.
#once you have that, go over our fine list of possible headers, find possible
# matches. I say matches in plural because files may begin with the same bytes,
# but diverge later on.
#now we basically find the best match - that's the longest header in the result,
#  naturally indicating the best match, most header chars matched file beginning
def get_tar_flag (fn):
    magic = get_header(fn)
    #python3's filter returns an iterator instead of a list. we can't handle
    # that nicely to work for both versions, so here we just extract the
    # filter functionality
    possible_heads = [k for k in headers.keys() if magic.startswith(k)]
    #default
    possible_heads.append('');

    return headers[max(possible_heads, key=len)]

def get_header (fn):
    longest = max(len(h) for h in headers)

    with open(fn, 'rb') as file:
        return file.read(longest)

if __name__ == '__main__':
    import os, sys

    status = xtract_ze_files(
        os.path.abspath(sys.argv[1]),
        ' '.join(sys.argv[2:]))

    sys.exit(status)