cryptpickle 1.0

Encrypted python object serialization

CryptPickle

Encrypted python object serialization

CryptPickle allows you to easily encrypt python objects into a file and decrypt, regardless of their content. It may be any python object, including for example a Pandas DataFrame.

Migration from v0.x

Breaking change: CryptPickle 1.0 changes the encryption format from AES-ZIP (pyzipper) to AES-256-GCM with PBKDF2 (cryptography). Files created with versions prior to 1.0 are not readable with this version.

To migrate existing data: decrypt with CryptPickle 0.9, then re-encrypt with CryptPickle 1.0.

Install

pip install cryptpickle

Examples

Usage example 1 (Encrypt and Decrypt a dict with some sensible data):

import cryptpickle

# Create a dictionary with some data
# It could be any other python object. ie: a Pandas Dataframe 
ej1 = { 'name1': 'John Doe',
        'name2': 'Lisa Doe'}

# Print the data
print(ej1)

# Serialice the data in an encrypted file with a password (file.crypt)
cryptpickle.obj_to_encrypted(ej1,password="SecretPassword",path='./file.crypt')

# Load the serialiced data in other python object. Password is needed to unencrypt the data
ej2 =cryptpickle.obj_from_encrypted(password="SecretPassword",path='./file.crypt')

# Print the data
print(ej2)

Usage example 2 (Encrypt and Decrypt a Pandas DataFrame with some sensible data):

import cryptpickle
import pandas as pd

# Create a dictionary with some data
df1 = pd.DataFrame({'A': [1, 2, 3],
                   'B': ['one', 'one', 'four']})

# Print the data
print(df1)

# Serialice the data in an encrypted file (path) with a password (password)
cryptpickle.obj_to_encrypted(df1,password="SecretPassword",path='./pd_data.crypt')

# Load the serialiced data in other Pandas DataFrame.
df2 =cryptpickle.obj_from_encrypted(password="SecretPassword",path='./pd_data.crypt')

# Print the data
print(df2)

Install and try it (Linux)

1. Clone this repository and enter in the directory:

git clone https://github.com/privtools/CryptPickle.git
cd CryptPickle

2. Create a vitual environment:

python3 -m venv .venv

3. Activate the virtual environment:

source .venv/bin/activate

4. Install the package:

pip install -r requirements.txt

5. Run sample1:

python sample1.py

5. Run sample2:

python sample2.py