drf-tracking icon indicating copy to clipboard operation
drf-tracking copied to clipboard

Latest Release doesn't have latest code

Open veris-neerajdhiman opened this issue 7 years ago • 6 comments

  • I have installed latest Release 1.3.1 but it don't have latest code as you have in your master branch.

e.g.,

  1. Earlier you have _clean_data as separate function but now you have moved it to class function and modified it too. Release 1.3.1 still have old code for this function (I have checked source code of Release too)
  2. The actual error I was getting is TypeError: expected string or bytes-like object in line No 137 if SENSITIVE_DATA.search(key):
  3. So when I checked installed Lib code and Master branch code, It was different.
  4. So please update Release code and also check if New code can fix my original error which is mentioned in 2nd point .
  5. Here data is a list of dict()
[   {
       "actor_type": "tets",
       "meeting_with": "Self",
       "first_name": "qwert",
       "last_name": "asdfg",
       "logged_at": "1448113944"
   }
   ]

veris-neerajdhiman avatar Sep 08 '17 08:09 veris-neerajdhiman

Hello @veris-neerajdhiman, if you want to check if your error is fixed, you can also install the code from master directly with pip.

Let us know if the issue is still there

triat avatar Sep 12 '17 11:09 triat

@triat yes issue was still there so for now I have forked this repo and patched a hot fix (for loop) but you must verify the same at your end and apply permanent fix.

veris-neerajdhiman avatar Sep 12 '17 12:09 veris-neerajdhiman

@veris-neerajdhiman can I propose that you do a PR if you already did the fix ? and write tests maybe ? Much appreciated

triat avatar Sep 12 '17 12:09 triat

@triat PR #73 created for this issue, Please check

veris-neerajdhiman avatar Sep 13 '17 07:09 veris-neerajdhiman

@veris-neerajdhiman @triat I think the critical issue here that we are attempting force a dict copy on a post body that might not always translate to a dict. In that scenario we could consider using the polymorphic copy.deepcopy and make _clean_data be able to handle a list of dictionaries or just ignore cleaning a list structure.

Thoughts?

avelis avatar Sep 22 '17 20:09 avelis

@avelis

  • I have used copy.deepcopy() you can check in my commits.
  • Secondly I think we should not ignore list structure as it may or may not include sensitive data.

veris-neerajdhiman avatar Sep 25 '17 07:09 veris-neerajdhiman