[Debian-med-packaging] Bug#936798: Please port KisSplice to Python3 (Was: Bug#936798: kissplice: Python2 removal in sid/bullseye)

[Nilesh Patra]

Hi,

On Fri, 29 May 2020, 01:08 Andreas Tille, <andreas at an3as.eu> wrote:

> Hi Moritz,
>
> On Thu, May 28, 2020 at 08:48:41PM +0200, Moritz Mühlenhoff wrote:
> >
> > JFTR, this is now fixed upstream in 2.5.0 and later:
> > http://kissplice.prabi.fr/download/:
> >
> > --------------------------------------
> > Kissplice Version 2.5.0 (2020-04-06)
> >
> > New features:
> >    Now compatible with python3
> > --------------------------------------
>
> Thanks for the hint.  Unfortunately there are remaining issues
> in build time test and autopkgtest. even of version 2.5.1 .:-(
>

I've Fixed the autopkgtest

For the failing build tests: this is what I observe - this try matching
strings containing a few values with kissplice result.

For instance,  this is one of the failures:

Expected:
0a: Single SNPs, Inexact Repeats or sequencing substitution errors, 69 found

Actual:
0a: Single SNPs, Inexact Repeats or sequencing substitution errors, 22 found

There's definitely a difference between the outputs of previous kissplice
version and current kissplice version, I tried this manually as well.

Just _maybe_, the upstream data has been changed, and the correct data
hasn't been injected; or the implementation has changed in version,
2.5.1and this should output 22, but the tests haven't been changed
accordingly ; or maybe this is a bug with kissplice itself.

In either case, I'm not really about this. Maybe we should ask upstream
about it.

Thanks and regards
Nilesh
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://alioth-lists.debian.net/pipermail/debian-med-packaging/attachments/20200529/905df97b/attachment.html>