decoding Microsoft Safelink URL in Python

user_7679885 0
Viewed 3231
import re
import urllib
import HTMLParser

urlRegex = re.compile(r'(.+)&data=')
match=urlRegex.search('https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Foffice.memoriesflower.com%2FPermission%2F%2525%2524%255E%2526%2526*%2523%2523%255E%2524%2525%255E%2526%255E*%2526%2523%255E%2525%2525%2526%2540%255E*%2523%2526%255E%2525%2523%2526%2540%2525*%255E%2540%255E%2523%2525%255E%2540%2526%2525*%255E%2540%2Foffice.php&data=01%7C01%7Cdavid.levin%40mheducation.com%7C0ac9a3770fe64fbb21fb08d50764c401%7Cf919b1efc0c347358fca0928ec39d8d5%7C0&sdata=PEoDOerQnha%2FACafNx8JAep8O9MdllcKCsHET2Ye%2B4%3D&reserved=0')

x = match.group()
urlRegex_1 = re.compile(r'url=(.+)&data=')
match_1 = urlRegex_1.search(x)
print match1.group(1)
htmlencodedurl = urllib.unquote(urllib.unquote(match1.group(1)))
actual_url = HTMLParser.HTMLParser().unescape(htmlencodedurl)

So the 'actual_url' displays this: 'https://office.memoriesflower.com/Permission/%$^&&##^$%^&^&#^%%&@^*#&^%'

I need it to display this:

https://office.memoriesflower.com/Permission/office.php
urldecode python
2 Answers

The following is cleaner as it uses the urlparse to extract the query string and then uses path operations to remove the unwanted component:

import posixpath as path
from urlparse import urlparse, parse_qs, urlunparse
url = 'https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Foffice.memoriesflower.com%2FPermission%2F%2525%2524%255E%2526%2526*%2523%2523%255E%2524%2525%255E%2526%255E*%2526%2523%255E%2525%2525%2526%2540%255E*%2523%2526%255E%2525%2523%2526%2540%2525*%255E%2540%255E%2523%2525%255E%2540%2526%2525*%255E%2540%2Foffice.php&data=01%7C01%7Cdavid.levin%40mheducation.com%7C0ac9a3770fe64fbb21fb08d50764c401%7Cf919b1efc0c347358fca0928ec39d8d5%7C0&sdata=PEoDOerQnha%2FACafNx8JAep8O9MdllcKCsHET2Ye%2B4%3D&reserved=0'
target = parse_qs(urlparse(url).query)['url'][0]
p = urlparse(target)
q = p._replace(path=path.join(path.dirname(path.dirname(p.path)), path.basename(p.path)))
print urlunparse(q)

prints https://office.memoriesflower.com/Permission/office.php

user_3887870

I found this having a similar problem. Here's the code I used to resolve the issue. It's not particularly elegant, but you may be able to tweak it for your needs.

self.urls = (re.findall("safelinks\.protection\.outlook\.com/\?url=.*?sdata=", self.body.lower(), re.M))
if len(self.urls) > 0:
    for i, v in enumerate(self.urls):
        self.urls[i] = v[38:-11]

This works by getting the value in an ugly format and then stripping off the excess pieces of each item as a string. I believe the proper way to do this is with grouping, but this worked well enough for my needs.

user_74424830
Related