Python 3 support #43
Python 3 support #43
Conversation
| @@ -334,16 +347,21 @@ def __call__(self): | |||
| self.changed = False | |||
| self._action = None | |||
| deleted = [self[key] for key in self._deleted_children] | |||
| for node in self.storage.values() + deleted: | |||
| for node in list(self.storage.values()) + deleted: | |||
There was a problem hiding this comment.
here I would prefer itertools.chain
| @@ -1,13 +1,17 @@ | |||
| # -*- coding: utf-8 -*- | |||
| from __future__ import print_function | |||
There was a problem hiding this comment.
I dont think this is necessary, `print('foo') works in python 2 OOTB
There was a problem hiding this comment.
I think you're right. Probably inserted by modernize and missed during cleanup.
|
|
||
| >>> str(filter) | ||
| '(|(cn=sepp)(sn=meier\xc3\xa4))' | ||
| >>> str(filter) in ['(|(sn=meierä)(cn=sepp))', '(|(cn=sepp)(sn=meier\xc3\xa4))'] |
There was a problem hiding this comment.
I can not follow here? why check both 'ä' variants?
There was a problem hiding this comment.
The representation differs in Python 2 vs. 3. We don't care what exactly it is, so we allow both.
There was a problem hiding this comment.
Hi Manuel,
Thank you very much for your efforts.
It was my original plan to work on this topic on my own at alpine city sprint. Unfortunately I was not able to attend.
My plan was to convert doctests to unittests before porting to python3. This is the reason why I did not merge your PR into master.
Unittest conversion is done now with #44, and python3 support is in there as well.
I've taken some implementation details from your PR, thanks again. The biggest difference in my python 3 support implementation is the use of bytes_mode=False in python-ldap, which simplifies things a lot, because it now provides unicode support for for most of the data OOTB, which was handled by node.ext.ldap before.
I'll close this PR
Refs #33