serialization - What difference between pickle and _pickle in python 3?

Question

I am new in python and want implement fast object serialization. I was trying to use json, but it was too slow, also was trying to use marshall module, but the size of the objects serialized by marshall 6-7 times more than pickle, so i decided to use pickle in my project. I read about cPickle module, read that it quite fast, but there is no such module in python 3 and docs says that module named _pickle is written in C. So in my projects i use

import _pickle as pickle

Is any difference between pickle and _pickle? How i can implement faster objects serialization/deserialization?

Answer 1

The pickle module already imports _pickle if available. It is the C-optimized version of the pickle module, and is used transparently.

From the pickle.py source code:

# Use the faster _pickle if possible
try:
    from _pickle import *
except ImportError:
    Pickler, Unpickler = _Pickler, _Unpickler

and from the pickle module documentation:

The pickle module has an transparent optimizer (_pickle) written in C. It is used whenever available. Otherwise the pure Python implementation is used.

In Python 2, _pickle was known as cPickle, but has been updated to allow the transparent use as an implementation detail.