There are two ways to serialize Python objects: binary or JSON.

pickle module

Python standard library that can store native datatypes, lists/tuples/dictionaries/sets, and functions, classes and instance of classes.

File

  • pickle.dump(data, file) serializes a serializable Python data structure, data, to a binary writable file, file, using a Python-specific binary format, the pickle protocol. The protocol cannot cope with other programming language, and older version of Python may not support its latest version.
  • pickle.load(file) loads a pickle file opened in binary read mode. DO NOT use pickle.load() on untrusted data!

Memory

  • pickle.dumps(data) does the same thing except it returns a byte array.
  • pickle.loads(array) loads a pickle byte array.

json module

Python standard library to work with JSON file. Since JSON file must be encoded with Unicode, encoding should be specified when opening files (encoding='utf-8').

Note: JSON doesn’t support tuples and bytes. Lists are mapped as arrays, and dictionaries are mapped as objects.

json.dump(data, file): same as pickle.dump() but it produces a JSON file.

  • Optional indent argument, 0 means “put each value on its own line”, a number greater than 0 means “put each value on its own line, and use this number of spaces to indent nested data structures”.
  • Optional default argument, a function that transforms the data before serialization. The function should transform an unsupported datatype into a dictionary, preferrably with the following format:
  {'__class__': class_name,
   '__value__': value_transformation}

json.load(file) takes a stream object and creates a new Python object. To parse unsupported datatype, hook a transformation function to object_hook optional argument.