Re: [Debian-science-sagemath] Sagemath Build-Depends backported - unfortunately build error (Was: Help needed to backport ipywidgets ... (Was: Backporting sagemath))
On 04/18/2017 02:25 PM, Andreas Tille wrote:
> Hi Tobias,
>
> On Tue, Apr 18, 2017 at 01:45:01PM +0100, Tobias Hansen wrote:
>>>
>>> Despite I'm not sure whether I fully understand this table I'll grab my
>>> hat as Debian Med maintainer and backport biopython.
>>
>> The left column shows the versions of the packages that
>> sage-the-distribution 7.4 uses. The right column shows either the
>> version in jessie or (if the version in jessie is smaller than the one
>> of sage and there is a package in jessie-backports) the version in
>> jessie-backports.
>>
>> The colors just mean: red or orange = potentially bad.
>
> So far for the legend but I was thinking how bad are so many potentially
> bad entries. I also stumbled about "missing" gcc, and what missing
> "readline" might mean.
>
> Kind regards
>
> Andreas.
>
gcc and readline are just listed as missing because the names of the
Debian source packages changed between jessie and stretch.
I added two more checkboxes so now you can hide all the blue and green
entries.
General purpose packages such as general python packages normally have a
better chance to maintain compatibility than the math packages. I would
start by fixing brial and backporting the list I gave you yesterday and
then look at the tests and see which of the other red entries need to be
backported.
You should compare this to
https://people.debian.org/~thansen/debian-sage-7.4-status.html
That means at least the red/orange entries for configparser, cvxopt and
ecl are fine. You can probably get away with many more, the question is
just which ones.
Best,
Tobias
Reply to: