Fast, efficient Blowfish cipher implementation in pure Python (3.4+).

Overview

PyPI CI

blowfish

This module implements the Blowfish cipher using only Python (3.4+).

Blowfish is a block cipher that can be used for symmetric-key encryption. It has a 8-byte block size and supports a variable-length key, from 4 to 56 bytes. It's fast, free and has been analyzed considerably. It was designed by Bruce Schneier and more details about it can be found at <https://www.schneier.com/blowfish.html>.

Dependencies

  • Python 3.4+

Features

  • Fast (well, as fast you can possibly go using only Python 3.4+)
  • Efficient; generators/iterators are used liberally to reduce memory usage
  • Cipher-Block Chaining (CBC) mode
  • Cipher-Block Chaining with Ciphertext Stealing (CBC-CTS) mode
  • Propagating Cipher-Block Chaining (PCBC) mode
  • Cipher Feedback (CFB) mode
  • Output Feedback (OFB) mode
  • Counter (CTR) mode
  • Electronic Codebook (ECB) mode
  • Electronic Codebook with Ciphertext Stealing (ECB-CTS) mode

Installation

If you just need a Blowfish cipher in your Python project, feel free to manually copy blowfish.py to your package directory (license permitting).

distutils

To install the module to your Python distribution, use the included distutils script:

$ python setup.py install

pip

Stable versions can be installed from pypi using pip:

$ pip install blowfish

pip can also install the latest development version directly from git:

$ pip install 'git+https://github.com/jashandeep-sohi/python-blowfish.git'

Development

Want to add a mode of operation? Speed up encryption?

Make your changes to a clone of the repository at https://github.com/jashandeep-sohi/python-blowfish and send me a pull request.

Tests

Tests are written using the Python unittest framework. All tests currently reside in the test.py file and can be run using the distutils script:

$ python setup.py test

Bugs

Are you having problems? Please let me know at https://github.com/jashandeep-sohi/python-blowfish/issues

Usage

Warning

Cryptography is complex, so please don't use this module in anything critical without understanding what you are doing and checking the source code to make sure it is doing what you want it to.

Note

This is just a quick overview on how to use the module. For detailed documentation please see the docstrings in the module.

First create a Cipher object with a key.

import blowfish

cipher = blowfish.Cipher(b"Key must be between 4 and 56 bytes long.")

By default this initializes a Blowfish cipher that will interpret bytes using the big-endian byte order. Should the need arrise to use the little-endian byte order, provide "little" as the second argument.

cipher_little = blowfish.Cipher(b"my key", byte_order = "little")

Block

To encrypt or decrypt a block of data (8 bytes), use the encrypt_block or decrypt_block methods of the Cipher object.

from os import urandom

block = urandom(8)

ciphertext = cipher.encrypt_block(block)
plaintext = cipher.decrypt_block(ciphertext)

assert block == plaintext

As these methods can only operate on 8 bytes of data, they're of little practical use. Instead, use one of the implemented modes of operation.

Cipher-Block Chaining Mode (CBC)

To encrypt or decrypt data in CBC mode, use encrypt_cbc or decrypt_cbc methods of the Cipher object. CBC mode can only operate on data that is a multiple of the block-size in length.

data = urandom(10 * 8) # data to encrypt
iv = urandom(8) # initialization vector

data_encrypted = b"".join(cipher.encrypt_cbc(data, iv))
data_decrypted = b"".join(cipher.decrypt_cbc(data_encrypted, iv))

assert data == data_decrypted

Cipher-Block Chaining with Ciphertext Stealing (CBC-CTS)

To encrypt or decrypt data in CBC-CTS mode, use encrypt_cbc_cts or decrypt_cbc_cts methods of the Cipher object. CBC-CTS mode can operate on data of any length greater than 8 bytes.

data = urandom(10 * 8 + 6) # data to encrypt
iv = urandom(8) # initialization vector

data_encrypted = b"".join(cipher.encrypt_cbc_cts(data, iv))
data_decrypted = b"".join(cipher.decrypt_cbc_cts(data_encrypted, iv))

assert data == data_decrypted

Propagating Cipher-Block Chaining Mode (PCBC)

To encrypt or decrypt data in PCBC mode, use encrypt_pcbc or decrypt_pcbc methods of the Cipher object. PCBC mode can only operate on data that is a multiple of the block-size in length.

data = urandom(10 * 8) # data to encrypt
iv = urandom(8) # initialization vector

data_encrypted = b"".join(cipher.encrypt_pcbc(data, iv))
data_decrypted = b"".join(cipher.decrypt_pcbc(data_encrypted, iv))

assert data == data_decrypted

Cipher Feedback Mode (CFB)

To encrypt or decrypt data in CFB mode, use encrypt_cfb or decrypt_cfb methods of the Cipher object. CFB mode can operate on data of any length.

data = urandom(10 * 8 + 7) # data to encrypt
iv = urandom(8) # initialization vector

data_encrypted = b"".join(cipher.encrypt_cfb(data, iv))
data_decrypted = b"".join(cipher.decrypt_cfb(data_encrypted, iv))

assert data == data_decrypted

Output Feedback Mode (OFB)

To encrypt or decrypt data in OFB mode, use encrypt_ofb or decrypt_ofb methods of the Cipher object. OFB mode can operate on data of any length.

data = urandom(10 * 8 + 1) # data to encrypt
iv = urandom(8) # initialization vector

data_encrypted = b"".join(cipher.encrypt_ofb(data, iv))
data_decrypted = b"".join(cipher.decrypt_ofb(data_encrypted, iv))

assert data == data_decrypted

Counter Mode (CTR)

To encrypt or decrypt data in CTR mode, use encrypt_ctr or decrypt_ctr methods of the Cipher object. CTR mode can operate on data of any length. Although you can use any counter you want, a simple increment by one counter is secure and the most popular. So for convenience sake a simple increment by one counter is implemented by the blowfish.ctr_counter function. However, you should implement your own for optimization purposes.

from operator import xor

data = urandom(10 * 8 + 2) # data to encrypt

# increment by one counters
nonce = int.from_bytes(urandom(8), "big")
enc_counter = blowfish.ctr_counter(nonce, f = xor)
dec_counter = blowfish.ctr_counter(nonce, f = xor)

data_encrypted = b"".join(cipher.encrypt_ctr(data, enc_counter))
data_decrypted = b"".join(cipher.decrypt_ctr(data_encrypted, dec_counter))

assert data == data_decrypted

Electronic Codebook Mode (ECB)

Note: ECB mode does not provide strong confidentiality, regardless of the cipher, and is not recommended for use in applications.

To encrypt or decrypt data in ECB mode, use encrypt_ecb or decrypt_ecb methods of the Cipher object. ECB mode can only operate on data that is a multiple of the block-size in length.

data = urandom(10 * 8) # data to encrypt

data_encrypted = b"".join(cipher.encrypt_ecb(data))
data_decrypted = b"".join(cipher.decrypt_ecb(data_encrypted))

assert data == data_decrypted

Electronic Codebook Mode with Cipher Text Stealing (ECB-CTS)

Note: the warning pertaining to ECB mode above also applies to ECB-CTS.

To encrypt or decrypt data in ECB-CTS mode, use encrypt_ecb_cts or decrypt_ebc_cts methods of the Cipher object. ECB-CTS mode can operate on data of any length greater than 8 bytes.

data = urandom(10 * 8 + 5) # data to encrypt

data_encrypted = b"".join(cipher.encrypt_ecb_cts(data))
data_decrypted = b"".join(cipher.decrypt_ecb_cts(data_encrypted))

assert data == data_decrypted
Comments
  • Bad Pi!

    Bad Pi!

    After using 'PI' sign in source, i can't install module via pip on Kubuntu/Pythin3.5, and QPython/Python3.5. When i remove 'PI' sign, importing of module works well. `

    import blowfish Traceback (most recent call last): File "", line 1, in File "blowfish.py", line 32 SyntaxError: Non-ASCII character '\xcf' in file blowfish.py on line 32, but no encoding declared; see http://python.org/dev/peps/pep-0263/ for details

    `

    opened by Moneetor 3
  • Bytes to str and back, ctr_counter

    Bytes to str and back, ctr_counter

    Hoping you might be able to point me in the right direction.

    I'm using encrypt_ctr successfully, however the storage medium I'm using requires me to store into str. I'm not sure what encoding/decoding to use to make it go to a str, and then be able to be converted from str back to bytes for decrypt_ctr to handle.

    At the moment, decryption is sort of like:

    encryptedBytes = b"".join(encryptedBytesList)
    with open(metadata["filename"], "wb") as openFile:
                decoded_data = b"".join(self.cipher.decrypt_ctr(data, self.encrypt_counter))
                openFile.write(decoded_data)
    

    And encryption, sort of like:

    with open(os.path.expanduser("~/reddiSync/") + filename, "rb") as openFile:
                data = openFile.read()
                encoded_data = b"".join(self.cipher.encrypt_ctr(data, self.encrypt_counter))
            encoded_list = [encoded_data[i:i+1000] for i in range(0, len(data), 1000)]
            return [str(x) for x in encoded_list]
    

    Its ensuring the original conversion to str that's the issue.

    UTF-8 isn't compatible with some of the bytes produced, so is there an encoding I could use?

    opened by shakna-israel 2
  • DES_set_key_checked in blowfish

    DES_set_key_checked in blowfish

    I encountered a problem about DES_cfb64_encrypt, i use "cipher.encrypt_cfb(data, iv)" but incorrect result. i found that no similar function in cipher.

    opened by jmpews 2
  • Docs - electronic code book

    Docs - electronic code book

    One of the first items under "Usage" on the README is how to use ECB mode. I would like to suggest that this section should be accompanied by a warning that ECB mode is inherently insecure and does not provide strong secrecy. It is in fact very vulnerable to cryptanlysis and secrecy can be compromised even without side-channels.

    I would also recommend that this section not be placed first in the documentation.

    A set of images on wikipedia illustrates dramatically the way ECB mode leaks data: https://en.wikipedia.org/wiki/Block_cipher_mode_of_operation#ECB

    opened by leifurhauks 1
  • Add a 'test' command to setup.py to run unittests

    Add a 'test' command to setup.py to run unittests

    Now tests can be run using the distutils script:

    $ python setup.py test -vv
    running test
    running build_py
    test_decrypt_block (test.Cipher) ... ok
    test_encrypt_block (test.Cipher) ... ok
    test_decrypt_block (test.CipherLittleEndian) ... ok
    test_encrypt_block (test.CipherLittleEndian) ... ok
    test_cbc_cts_mode (test.ModesOfOperation) ... ok
    test_cbc_mode (test.ModesOfOperation) ... ok
    test_cfb_mode (test.ModesOfOperation) ... ok
    test_ctr_mode (test.ModesOfOperation) ... ok
    test_ecb_cts_mode (test.ModesOfOperation) ... ok
    test_ecb_mode (test.ModesOfOperation) ... ok
    test_ofb_mode (test.ModesOfOperation) ... ok
    test_pcbc_mode (test.ModesOfOperation) ... ok
    
    ----------------------------------------------------------------------
    Ran 12 tests in 1.816s
    
    OK
    
    enhancement 
    opened by jashandeep-sohi 0
  • Fix little-endian byte-order input encryption and decryption.

    Fix little-endian byte-order input encryption and decryption.

    This fixes bug https://github.com/jashandeep-sohi/python-blowfish/issues/9.

    u4_1_struct is used to split 32-bit integers into 4 bytes. The higher 8 bits should be the first byte, the next 8 bits should be the next byte, and so on. So, it should always be in big-endian byte order and not be dependent on the byte_order option.

    opened by jashandeep-sohi 0
  • Add custom exceptions for incorrect input data

    Add custom exceptions for incorrect input data

    Currently, the generic struct.error is raised when the input data is not the correct length. Raise a custom exception so it's a bit more clear as to why the call failed.

    enhancement 
    opened by jashandeep-sohi 0
  • Need example for encrypt/decrypt for a large file (say 1mb)

    Need example for encrypt/decrypt for a large file (say 1mb)

    Hello there this is just awesome. However, I am in a hurry and looing for a sample code which encrypts and decrypts a big file what is the cost of encryption? is it O(n) ? thanks Sriram

    opened by sriramb12 0
learn python in 100 days, a simple step could be follow from beginner to master of every aspect of python programming and project also include side project which you can use as demo project for your personal portfolio

learn python in 100 days, a simple step could be follow from beginner to master of every aspect of python programming and project also include side project which you can use as demo project for your

BDFD 6 Nov 05, 2022
The OpenAPI Specification Repository

The OpenAPI Specification The OpenAPI Specification is a community-driven open specification within the OpenAPI Initiative, a Linux Foundation Collabo

OpenAPI Initiative 25.5k Dec 29, 2022
Deduplicating archiver with compression and authenticated encryption.

More screencasts: installation, advanced usage What is BorgBackup? BorgBackup (short: Borg) is a deduplicating backup program. Optionally, it supports

BorgBackup 9k Jan 09, 2023
Netbox Dns is a netbox plugin for managing zone, nameserver and record inventory.

Netbox DNS Netbox Dns is a netbox plugin for managing zone, nameserver and record inventory. Features Manage zones (domains) you have. Manage nameserv

Aurora Research Lab 155 Jan 06, 2023
xeuledoc - Fetch information about a public Google document.

xeuledoc - Fetch information about a public Google document.

Malfrats Industries 651 Dec 27, 2022
A fast time mocking alternative to freezegun that wraps libfaketime.

python-libfaketime: fast date/time mocking python-libfaketime is a wrapper of libfaketime for python. Some brief details: Linux and OS X, Pythons 3.5

Simon Weber 68 Jun 10, 2022
Official Matplotlib cheat sheets

Official Matplotlib cheat sheets

Matplotlib Developers 6.7k Jan 09, 2023
An ongoing curated list of OS X best applications, libraries, frameworks and tools to help developers set up their macOS Laptop.

macOS Development Setup Welcome to MacOS Local Development & Setup. An ongoing curated list of OS X best applications, libraries, frameworks and tools

Paul Veillard 3 Apr 03, 2022
This program has been coded to allow the user to rename all the files in the entered folder.

Bulk_File_Renamer This program has been coded to allow the user to rename all the files in the entered folder. The only required package is "termcolor

1 Jan 06, 2022
Leetcode Practice

LeetCode Practice Description This is my LeetCode Practice. Visit LeetCode Website for detailed question description. The code in this repository has

Leo Hsieh 75 Dec 27, 2022
An MkDocs plugin that simplifies configuring page titles and their order

MkDocs Awesome Pages Plugin An MkDocs plugin that simplifies configuring page titles and their order The awesome-pages plugin allows you to customize

Lukas Geiter 282 Dec 28, 2022
A repository of links with advice related to grad school applications, research, phd etc

A repository of links with advice related to grad school applications, research, phd etc

Shaily Bhatt 946 Dec 30, 2022
NoVmpy - NoVmpy with python

git clone -b dev-1 https://github.com/wallds/VTIL-Python.git cd VTIL-Python py s

263 Dec 23, 2022
A curated list of awesome tools for Sphinx Python Documentation Generator

Awesome Sphinx (Python Documentation Generator) A curated list of awesome extra libraries, software and resources for Sphinx (Python Documentation Gen

Hyunjun Kim 831 Dec 27, 2022
Spin-off Notice: the modules and functions used by our research notebooks have been refactored into another repository

Fecon235 - Notebooks for financial economics. Keywords: Jupyter notebook pandas Federal Reserve FRED Ferbus GDP CPI PCE inflation unemployment wage income debt Case-Shiller housing asset portfolio eq

Adriano 825 Dec 27, 2022
Grokking the Object Oriented Design Interview

Grokking the Object Oriented Design Interview

Tusamma Sal Sabil 2.6k Jan 08, 2023
Python Eacc is a minimalist but flexible Lexer/Parser tool in Python.

Python Eacc is a parsing tool it implements a flexible lexer and a straightforward approach to analyze documents.

Iury de oliveira gomes figueiredo 60 Nov 16, 2022
DataRisk Detection Learning Resources

DataRisk Detection Learning Resources Data security: Based on the "data-centric security system" position, it generally refers to the entire security

Liao Wenzhe 59 Dec 05, 2022
Quick tutorial on orchest.io that shows how to build multiple deep learning models on your data with a single line of code using python

Deep AutoViML Pipeline for orchest.io Quickstart Build Deep Learning models with a single line of code: deep_autoviml Deep AutoViML helps you build te

Ram Seshadri 6 Oct 02, 2022
A Python Package To Generate Strong Passwords For You in Your Projects.

shPassGenerator Version 1.0.6 Ready To Use Developed by Shervin Badanara (shervinbdndev) on Github Language and technologies used in This Project Work

Shervin 11 Dec 19, 2022