[Debian-science-sagemath] Library transition freeze coming up on November 5: ipython

[Julien Puydt]

Hi,

On 08/10/2016 16:26, Tobias Hansen wrote:
> On 10/08/2016 02:29 PM, Jerome BENOIT wrote:
>> On 08/10/16 12:58, Tobias Hansen wrote:
>>> I think it will be best to upload the newest version of fplll and
>>> ipython to unstable soon and backport sage commits to make them work if
>>> necessary.
>>
>> I have just built without issue ipython 5 in a unstable environment:
>> do I miss something ?
>
> I mean if we use ipython 5.1 and sage 7.3 (which normally has ipython
> 4.2.1), we might need to apply commits from a later sage version to make
> these versions of sage and ipython work together properly (and make
> doctests pass). That's what I mean with backport in the entire todo list.
>
>> Quick question: concerning ipython: how can we get the list of missing
> missing/blocking packages ?
>
> I hope Julien can elaborate more on that. But in the end we have to go
> through the list of reverse dependencies and reverse build dependencies
> that I sent yesterday and check if they all work with ipython 5 and if
> they need additional packages for that.

I can list a few points off the top of my head :

- the experimental ipython package in experimental is spitting warnings 
at startup, so I'm not sure it is as good as we want ;

- the previous ipython sources provided a notebook -- this is now a 
separate package ;

- the previous ipython sources provided a qt frontend -- this is now a 
separate package ;

- in experimental, we have the following source packages (in the Debian 
Python Modules Team's repository) : nbconvert, nbformat, ipykernel, 
ipython, jupyter-core, jupyter-client, ipykernel, python-qtconsole...

- there are ipywidgets and jupyter-sphinx-theme git repositories, which 
are relevant but empty -- both are relevant to jupyter-in-Debian ;

- for the jupyter notebook, there is another thread on this mailing-list 
already ;

Snark on #debian-science