The following two functions are used to compress arbitrary Python objects and send them safely via socket or email, using only printable chars. In my specific message protocol, '=' signs are also not allowed; I could have replaced them or truncate them - I chose to truncate them.

I think they work quite well speed-wise and have no unpredicted errors. I post it here for your judgement.

## Framework
import cPickle as pickle
import zlib as zl
import base64 as b64

def compress(o):
    ## First pickle
    p = pickle.dumps(o, pickle.HIGHEST_PROTOCOL)

    ## Then zip
    z = zl.compress(p, zl.Z_BEST_COMPRESSION)

    ## Then encode
    e = b64.b64encode(z)

    ## Then truncate
    t = e.rstrip("=")

    ## Then return
    return t

def decompress(s):
    ## First pad
    e = s + "=" * (-len(s) % 4)

    ## Then decode
    z = b64.b64decode(e)

    ## Then unzip
    p = zl.decompress(z)

    ## Then unpickle
    o = pickle.loads(p)

    ## Then return
    return o

Avoid single letter variable names. Spell out the words they stand for to make your code easier to read.

Your comments are useless. You are just restating the code in less detail. The only comment you should have is explaining about the '=' like you did in your post.

Don't double-space your code.

Pickle isn't safe. You can construct a pickle that does arbitrary things like delete files or what-not. So you should pretty much never be sending them over a socket.

| improve this answer | |
  • \$\begingroup\$ pep8 says "Separate top-level function and class definitions with two blank lines."; that's why I am double-spacing my code... \$\endgroup\$ – Bach Feb 13 '14 at 18:06
  • 1
    \$\begingroup\$ I'm referring to all the newlines in your function. Not the top level space. \$\endgroup\$ – Winston Ewert Feb 14 '14 at 0:49

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.