[Debian-med-packaging] Bug#962085: bcalm, kissplice: both ship /usr/bin/bcalm

[David Parsons]

Hi,

I'm still talking with upstream to find out which solution would be 
best/simplest.

BTW, I just realized that something went wrong when migrating to the 
salsa gitlab, I'm no longer referenced a member of the project and my 
commits now appear as authored by Drew Parsons (@dparsons). My account 
is @parsons

I could still go for fork + MR but I'm not sure it's the preferred way 
for you guys (?)

Thanks,
David


On 08/06/2020 10:40, Andreas Tille wrote:
> Hi Nilesh,
>
> On Wed, Jun 03, 2020 at 03:05:34PM +0530, Nilesh Patra wrote:
>> When I was just drafting the mail regarding this, David replied to this
>> bug. I suppose they are already on this.
>> Thanks David!
> Yes, thanks David. ;-)
>   
>>> Or is it
>>> just a code copy included into kissplice?  We also should remove
>>> the code copy of gatb-core and rather link against the Debian
>>> packaged lib.
>> Yep, probably we should remove gatb-core and symlink it/tweak the build
>> system if need be.
> I guess tweaking the build system is the way to go.  A word of warning:
> gatb-core authors are not keen on a stable ABI and consistent
> versioning.  So if you volunteer to work on this I'd recommend to start
> right now ignoring the bcalm issue.  Sometimes it turns out a newer Git
> clone than we have packaged is needed (which in the worst case might
> break other packages using gatb-core).  So if you see a chance to work
> on it independently its a good idea to do so.
>
>> I simply fixed the autopkgtest and called it a day, assuming everything
>> else is in place.
>>
>> Looks like my assumption was wrong, and I apologize for not checking rest
>> of the stuff, I really should have.
> There is no need to write apology statements like this here.  We all
> are human and I personally do not see any time to spent time into writing
> such things. ;-)
>
> Kind regards
>
>        Andreas.
>