anonymize.py

"""
Python 3 script to anonymize a JSON file, keys and data. Conserve structure. Conserve the
shape of URLs found in strings, anonymizing host, path, params, query and
fragment.

Run without argument to test:

    $ python3 anonymize_json.py
    Usage: anonymize_json.py path/to/json
    running tests...
    TestResults(failed=0, attempted=18)

Run with a JSON file as argument to display the anonymized JSON to stdout:

    $ python3 anonymize_json.py b.json
    {'jccjjxfn': [{'mvszkvun': [{'pdajmknzgid': [{'fljcffiqfviuwjowkp':
    [{'flrwyvxlc': [{'zhsdkaaauramvg': ...

"""

import json
import random
import re
import string
import sys
import urllib.parse

def anonymize_str(s):
    """
    >>> anonymize_str("q")
    'p'
    >>> anonymize_str("qweqwe")
    'yopumz'
    """
    letters = string.ascii_lowercase
    return ''.join(random.choice(letters) for i in range(len(s)))

def anonymze_url(url, anonymizer=anonymize_str):
    """
    >>> def f(s): return "abc"
    >>> f("zzz")
    'abc'
    >>> anonymze_url("zzz", anonymizer=f)
    'abc'
    >>> anonymze_url("https://www.google.com", anonymizer=f)
    'https://abc'
    """
    try:
        p = urllib.parse.urlparse(url)
    except ValueError:
        ano = anonymizer(url)
    else:
        if p.netloc:   p = p._replace(netloc   = anonymizer(p.netloc))   
        if p.path:     p = p._replace(path     = anonymizer(p.path))     
        if p.params:   p = p._replace(params   = anonymizer(p.params))   
        if p.query:    p = p._replace(query    = anonymizer(p.query))    
        if p.fragment: p = p._replace(fragment = anonymizer(p.fragment)) 

        ano = urllib.parse.urlunparse(p)
    return ano

def anonymze_str_with_urls(s, anonymizer=anonymize_str):
    """
    >>> def f(s): return "abc"
    >>> f("zzz")
    'abc'
    >>> anonymze_str_with_urls("qweqwe", anonymizer=f)
    'abc'
    >>> anonymze_str_with_urls("http://www.google.com/?q=123#zzz", anonymizer=f)
    ' https://abc?abc#abc'
    >>> anonymze_str_with_urls("zzzz http://www.google.com/?q=123#rrr", anonymizer=f)
    'abc https://abc?abc#abc'
    >>> anonymze_str_with_urls("zzzz http://www.google.com/?q=123#abc qqqqqw https://www.google.fr/?z=eee&aaaa z;z;z;z", anonymizer=f)
    'abc https://abc?abc#abc https://abc?abc'
    """
    urls = re.compile("[a-z]+://").split(s)
    return " https://".join([anonymze_url(url, anonymizer=anonymizer) for url in urls])

def anonymize(obj):
    """
    >>> anonymize({})
    {}
    >>> anonymize("")
    ''
    >>> anonymize(123)
    17
    >>> anonymize([])
    []
    >>> anonymize(True)
    True
    >>> anonymize({'z': {42: [3, 'z']}})
    {'d': {32: [97, 'c']}}
    """

    if type(obj) == dict:
        return { anonymize(k): anonymize(v) for (k, v) in obj.items() }
    elif type(obj) == str:
        if "://" in obj:
            return anonymze_str_with_urls(obj)
        else:
            return anonymize_str(obj)
    elif type(obj) == int:
            return random.randint(0, 100)
    elif type(obj) == list:
        return [anonymize(item) for item in obj]
    elif type(obj) == bool:
        return bool(random.randint(0, 2))
    else:
        raise Exception("Cannot anonymize type %s" % type(obj))


random.seed(1)

if len(sys.argv) != 2:
    print("Usage: %s path/to/json" % sys.argv[0])
    print("running tests...")
    import doctest
    print(doctest.testmod())
    sys.exit(1)

with open(sys.argv[1], "r") as f:
    res = json.load(f)

print(anonymize(res))