Bug#813815: Open ITPs (turned into RFPs)

2019-09-18 Thread Andreas Tille
Hi Joost,

On Wed, Aug 14, 2019 at 01:07:29AM +0200, Joost van Baal-Ilić wrote:
> 
> Thanks for this reminder.  For both r-cran-nfactors and r-cran-semplot there
> already _is_ work in our g...@salsa.debian.org:r-pkg-team : at
> r-cran-nfactors.git and r-cran-semplot.git .

By chance I stumbled upon these packages as pre-dependencies for some
other package.  They also needed quite some more pre-depends, but all is
uploaded now.

> So no need for
> prepare_missing_cran_package for those.

For sure I forgot and just noticed when injecting my repository failed
since there was an existing one (same for some pre-dependencies you had
prepared before).  I kept you as second Uploader for all those packages
you created repositories.

Admittedly the existing repositories were not really a help but rather
extra work.  Dh-make-R became better over time and so I just have
overridden your work with a single commit.  I guess the detailed history
is not that important.
 
> I cannot commit to maintaining these packages appear in a stable Debian
> release: I do all my R packaging work during $dayjob; I'm mainly interested in
> keeping the packages working on Debian stable, and somewhat in sync with
> upsteam.  So that's technically what would end up in debian backports.

I would really appreciate if you would do more official backports.  I
need to do this as well (currently even more for Debian 9 than for
Debian 10).  It would be really help if you would join this so we could
share the workload.  I have some **rudimentary** script which really
needs some work (or some research whether something better exists) here:

   
https://salsa.debian.org/med-team/community/helper-scripts/blob/master/do_backport

It helps a bit to automatise things but its not really reliably neither
well testet.

> The
> reason I stick the work in git @ salsa is I feel there's a chance others might
> benefit from it too.  That's the reason I've recently not been very active in
> actually uploading it to NEW.

That's fine in general.  However, as I said, dh-make-R becomes better
and better and there is probably no point to inject something that is
not really actively used in production.
 
> There some packages of me in our git now for which I've not yet posted ITP's.
> I believe this interferes badly with some of your workflows.  Would it help 
> you
> if I actually _would_ post ITP's?  I'd be glad to.

I do not think its needed to file ITPs.  Its just a script call away.
This can be done by the maintainer who actually wants to care for the
package.

> Also, debian/copyright of most of my recent packages is still lacking.  I'll
> try to find time to work on that (note to self: have another look at cme).

That's partly what I mean.  I just added LGPL-2.1 to dh-make-R since it
was lacking before.  In case you might have generated the packaging with
an older dh-make-R and it has this license you need to copy-n-paste the
text manually.  Creating from scratch would be less work. :-P
 
> What do you think?

I hope you can join official backporting for common profit. :-)
 
> PS: thanks for your talk @ debconf, I enjoyed the video!

Nice.

Kind regards

   Andreas.
 

-- 
http://fam-tille.de



Bug#813815: Open ITPs (turned into RFPs)

2019-08-13 Thread Joost van Baal-Ilić
Hi Andreas,

On Mon, Jul 22, 2019 at 07:42:06PM +0200, Andreas Tille wrote:
> 
> you have at least two long standing ITPs (#813815 - r-cran-nfactors,
> #835082 - r-cran-semplot, may be others) which were turned into RFPs
> automatically.  Could you give some statement whether these packages
> might remain interesting for you and whether we should fire up
> prepare_missing_cran_package for these?

Thanks for this reminder.  For both r-cran-nfactors and r-cran-semplot there
already _is_ work in our g...@salsa.debian.org:r-pkg-team : at
r-cran-nfactors.git and r-cran-semplot.git .  So no need for
prepare_missing_cran_package for those.

I cannot commit to maintaining these packages appear in a stable Debian
release: I do all my R packaging work during $dayjob; I'm mainly interested in
keeping the packages working on Debian stable, and somewhat in sync with
upsteam.  So that's technically what would end up in debian backports.  The
reason I stick the work in git @ salsa is I feel there's a chance others might
benefit from it too.  That's the reason I've recently not been very active in
actually uploading it to NEW.

There some packages of me in our git now for which I've not yet posted ITP's.
I believe this interferes badly with some of your workflows.  Would it help you
if I actually _would_ post ITP's?  I'd be glad to.

Also, debian/copyright of most of my recent packages is still lacking.  I'll
try to find time to work on that (note to self: have another look at cme).

What do you think?

Bye,

Joost

PS: thanks for your talk @ debconf, I enjoyed the video!



Bug#813815: Open ITPs (turned into RFPs)

2019-07-22 Thread Andreas Tille
Hi Joost,

you have at least two long standing ITPs (#813815 - r-cran-nfactors,
#835082 - r-cran-semplot, may be others) which were turned into RFPs
automatically.  Could you give some statement whether these packages
might remain interesting for you and whether we should fire up
prepare_missing_cran_package for these?

Kind regards

   Andreas.

-- 
http://fam-tille.de