Re: CI plan

[Andreas Tille <andreas@an3as.eu>]

Hi Tanya,

On Tue, May 31, 2016 at 08:04:21AM +0300, merlettaia wrote:
> Hi Andreas!
> 
> Sorry for not showing activity, I've experienced some strange technical
> troubles with my VM :( It turned out that switching to unstable in
> sources.list is not enough to work well with cme and testing packages from
> unstable, now I fixed it (switched to virtual box with unstable Debian on
> it).

While it seems you have sorted out your problem it might be sufficient
to install cme from unstable on top of a stable / testing system (I'm
using both (stable at work and testing at home).
 
> I've added test for checking result files and another call with non-default
> parameter set from readme (I decided not to move it to separate file
> because these two calls look almost the same, although setting up some
> additional parameters can change output file appearance. One of probable
> things worth testing is pymol script produced by concavity, it can reveal
> that file's compartibility with specific version of pymol, but it seems to
> be redundant, and on my display-less VM I was not able to call it without
> errors with xvfb, at least for now).

I do not find any question in this paragraph.  Please be more verbose in
case you might expect some sensible answer from me. ;-)
 
> 2016-05-24 18:21 GMT+03:00 Andreas Tille <andreas@an3as.eu>:
> 
> >
> > BTW, the currently *real* issue in concavity is bug #822382[1].
> >
> 
> One possible solution - to modify upstream by removing ply.* files  (and
> .ply format support) and their usage entirely: it seems that it can be used
> only as one of many possibilities for mesh serialization/deserialization.
> Concavity doesn't rely on that format for data storing or anything, typical
> input and output would be given in other formats (I've already checked
> locally, it can be quite easily removed). If this solution is ok, I can fix
> it.

Since it seems upstream will not respond to our mail this would probably
a good solution.  Packaging wise we should add the non-free files in
debian/copyright Files-Excluded field and recreate the upstream tarball.
The work-around code should be added as quilt patches - see my other mail
to Canberk if you are not comfortable with quilt.
 
> > Upstream has not answered my mail since 20 days.  I've set Laszlo in CC
> > who might have some contact or may be you can find out *working*
> > contacts (people had good experiences by Twitter or other social media).
> >
> 
> Have you tried to mail tony.capra at vanderbilt.edu?

No.  Here you can see whom I have contacted:

   https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=822382#10

> It's the best contact
> I could find. Also I found some person on github who seems to work with
> Dr.Capra: https://github.com/mikesivley - there is his email on this page,
> if this won't help, I can write him via issues.

Please let me know if you want to forward my message in the bug tracking
system yourself by adding your comments above or if I should forward it.
Your technical input and the contact information is very valuable.  If
you think Tony Capra is the best contact we should fix the
Upstream-Contact field in debian/copyright accordingly.
 
> > We need to sort out the origin of these two files - otherwise the
> > package needs to be moved to non-free.
> >
> Now I'm working on conservation-code (this time I'll do it faster >.< ).

:-)

Kind regards

       Andreas.

-- 
http://fam-tille.de