Re: Bug#785275: ITP: python-ipaddress -- Backport of the ipaddress module from Python 3.3

[Scott Kitterman <debian@kitterman.com>]

On May 22, 2015 10:06:34 AM EDT, Tristan Seligmann <mithrandi@mithrandi.net> wrote:
>On 22 May 2015 at 15:43, Donald Stufft <donald@stufft.io> wrote:
>> I'm not sure if I was looking at the wrong ipaddr or something but
>the
>> documentation I found was pretty crappy and looking at it I didn't
>see how to
>> actually use ipaddr the same way I was using ipaddress in pip. In
>addition to
>> that, the API doesn't really look the same as the ipaddress to me so
>it's going
>> to require some sort of compatability shim maintained in pip in order
>to be
>> able to run one module on Python 2 and another module on Python 3.
>
>If you're talking about the documentation linked to from PyPI (at
><http://pythonhosted.org//ipaddr/>), the problem is just that the
>documentation is for an ancient version of ipaddr. I was also thrown
>off by this originally, but if you look at an actually-recent version
>of ipaddr, the API seems to be the same (or at least very similar, I
>didn't look into it in detail so there may be differences in
>unicode/bytes handling, although I hope not).
>
>On the other hand, if upstream software is going to use ipaddress, it
>really doesn't seem like Debian's place to patch everything to use
>ipaddr.

There are far more users of ipaddr than ipaddress.  There's exactly two API differences.  I'm willing to look at pip and propose a patch to work with either. 

>On 22 May 2015 at 03:50, Scott Kitterman <debian@kitterman.com> wrote:
>> I'm also concerned that because of type processing differences
>(strings versus
>> bytes and UTF-8 by default versus not) there is potential for subtle
>> incompatibilities in the backported ipaddress.  I had complaints on
>an
>> upstream project that uses ipaddress with python3 when it was
>available
>> failing with ipaddress in python2.
>
>I don't quite follow this. If ipaddr isn't compatible with ipaddress,
>then the claim that you can simply import ipaddr instead of ipaddress
>is obviously not true for software that is currently using ipaddress.
>On the other hand, if they are compatible, then having ipaddress
>installed shouldn't break anything regardless of whether it imports
>ipaddress or ipaddr.
>
>Or are you saying that the ipaddress backport is not compatible with
>python3 stdlib's ipaddress? (This would be a very unfortunate state of
>affairs, but impossible to imagine)

Keep in mind that ipaddr came first and became ipaddress in python3.3.  My experience was that for my use case there were problems with the backport that don't occur with ipaddr. I don't recall the details as I considered ipaddress on python2 an unsupported configuration and didn't investigate in great detail. 

Scott K