Indeed. On Python 2, has_key needs to stay.

Should I add a warning for its eventual removal?

Python 2.7 users can continue to use the in operator. It is supported.

Should I add a warning for its eventual removal?

Yes, a DeprecationWarning sounds good. But I wouldn't rush the removal itself. Let's say we'll remove it when Python 2 support is dropped?

We could only add the function under Python 2 and not define it under Python 3.

We could only add the function under Python 2 and not define it under Python 3.

We are past the stage of defining an API for python-ldap 3.0.

Sure, has_key looks ugly and backwards and "the wrong way to do it", and a good linter will complain you about it... but it's there, it's tested, it's not a source of subtle bugs, and it doesn't really cost anything to keep it.
Let's make it clear it's not the way to do things, but avoid breaking people's code for purity's sake.

Codecov Report

Merging #137 into master will increase coverage by <.01%.
The diff coverage is 66.66%.

@@            Coverage Diff             @@
##           master     #137      +/-   ##
==========================================
+ Coverage   69.35%   69.35%   +<.01%     
==========================================
  Files          49       49              
  Lines        4708     4712       +4     
  Branches      800      800              
==========================================
+ Hits         3265     3268       +3     
- Misses       1088     1089       +1     
  Partials      355      355

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 98c535b...f156d07. Read the comment docs.

Updated. Restored function. Now issues a warning.

Hi @jdufresne,
Embarrassingly, when I was rebasing your other pull request to get rid of conflicts, I managed to overwrite your branch here.
Could you please push your changes again?
I'm sorry for this. I'll be more careful in the future.

No worries. Done in #151 Thanks

That's not the one. Could you push your local branch drop-has-key to GitHub?

Oops, sorry about that. Lets try that again, #153

Thank you! The change looks good; I'll do a full review next week.