Bug#890878: RFS: company-irony

2018-03-14 Thread Alberto Luaces
Hi Nicholas,

Nicholas D Steeves writes:

[...]

Thanks again for your advice!

>> > I: company-irony source: testsuite-autopkgtest-missing
>> 
>> This is N/A, I think.
>
> It's not required at this point in time, but someday it's possible
> that self tests will be required.  Dh_elpa_test runs the tests as part
> of a package build, and autopkgtest is a framework that automates
> testing of packages in a container or virtual machine.  Because this
> is Informational level lint it's not high priority for Lintian, but if
> you ever want to write a test that gets an company-irony autocompleted
> list for something, and then compares that against the expected list,
> in the expected order.  Tests that provide assurances it won't do
> hilarious/embarrassing autocompletion like cell phones do.  A Nice to
> have, later, if you have time and find the challenge interesting ;-)
>

Ok.  I don't have currently the skills for writing those tests, but
maybe in the future I can try to learn from some other packages.

[...]

>
> Because for now the most pressing issue is that it doesn't initialise
> properly...
>
>   Company backend 'company-clang' could not be initialized:
>   Company found no clang executable
>
> This was both with no configuration (and M-x company-mode), and with
> following upstream's README in a clean sid chroot.  I opened a random
> cpp from kdeconnect to test.  I suspect a documentation of
> configuration issue because I would have expected company-irony to
> load rather than company-clang...but it's possibly a bug.
>
> Please let me know how you made company-irony work.

Oh, yes, I think this is expected.  From the documentation at

/usr/share/doc/elpa-company-irony/README.md

--8<---cut here---start->8---

## Configuration

Add `company-irony` to your company backends.

~~~el
(eval-after-load 'company
  '(add-to-list 'company-backends 'company-irony))
~~~

--8<---cut here---end--->8---


Regards,

Alberto



Bug#890878: RFS: company-irony

2018-03-08 Thread Alberto Luaces
Thanks again, Nicholas:

Nicholas D Steeves writes:

> W: company-irony source: out-of-date-standards-version 4.1.1 (current is 
> 4.1.3)

Ok.  Upgraded to 4.1.3 while checking that it complains with the Policy
upgrades: basically none of them applied, except for the Vcs-* one, that
this package is already compliant with.

> I: company-irony source: testsuite-autopkgtest-missing

This is N/A, I think.

> W: elpa-company-irony: new-package-should-close-itp-bug

Now I have a number assigned.

> I: elpa-company-irony: extended-description-is-probably-too-short

Ok.  Rephrased.

Regards,

Alberto



Bug#890878: RFS: company-irony

2018-03-01 Thread Alberto Luaces
Nicholas D Steeves writes:

[...]

>
> Hi Alberto,
>
> Welcome to the team, and thank you for packaging company-irony!  I
> consider it a valuable addition to the archive :-)  The following
> might be something you already know, but if not, here's a neat trick:
>
> Make your changes, and then while in emacs, M-x magit-status, then d u
> (diff unstaged).  Stage the changes that are part of one logical
> operation with C-, select region, then s (or just s on a hunk to
> stage the whole hunk).  Finally c c (commit staged), write your commit
> message, and finally C-c C-c.  Later you can use gbp dch -a [-N
> $upstream_version-$debian_revision, if necessary] to generate a nice 
> changelog.
>

Thanks for the tips!

>
> Hi Sean and David,
>
> I'm willing to do reviews, and want to encourage best practises and our
> team's high standards.  Please feel free to comment.
>
> debian/copyright:
>   Author's email is directly underneath Copyright in
>   company-irony.el's header.  I would either Add it to the Copyright:
>   for the 'Files: *' section, or add an Upstream-Contact field. (
>   https://www.debian.org/doc/packaging-manuals/copyright-format/1.0/#examples
>   ) Sean, what's your best practices stance on this?  I'm guessing
>   Upstream-Contact.
>

I went the route of adding the email address, but of course it can be
changed at any time.

>
> debian/gbp.conf:
>   gbp:info: Tarballs 'company-irony_1.1.0.orig.tar.xz' not found
>   gbp:warning: Pristine-tar branch "pristine-tar" not found
>   gbp:info: Creating
>   /home/sten/devel/build-area/company-irony_1.1.0.orig.tar.xz
>   gbp:error: v1.1.0 is not a valid treeish
>
>   Alberto, if you're using pristine-tar you need to push the branch;
>   alternatively, if you got upstream source from git and are not using
>   pristine-tar you need to push the upstream tag to our repo and also
>   modify gbp.conf to indicate you're not using pristine-tar.  Also,
>   for future reference, if you choose the git-only workflow you'll
>   need to push each new upstream version tag as you update the
>   package.
>

Ok, I pushed a "upstream" branch and all of its tags, so gbp should not
choke this time. 

> debian/watch:
>   Missing, please add one.  Between the one for irony-mode (watch
>   version 3, Guillaume is also the upstream for this one) and
>   fountain-mode (version 4) you should be able to figure out how to
>   produce a working v4 one ;-)  The only reason I mention
>   fountain-mode is because it's the one I've checked most recently.
>

I think I managed to get one working, from reading other packages and
the uscan manpage.

Regards,

Alberto



Bug#890878: RFS: company-irony

2018-02-26 Thread Alberto Luaces
Sean Whitton writes:

> Hello,
>
> On Wed, Feb 21 2018, Alberto Luaces wrote:
>
>> Thanks, Sean.  It is now located at
>>
>> https://salsa.debian.org/aluaces-guest/company-irony
>>
>> I guess someone else has to clone it under the team project folders,
>> so I created that personal repository first.
>
> You should be able to do it yourself... are you saying that you were
> unable to create a repo under emacsen-team?  I just bumped your
> permission level.
>

Thanks for that.  Yes, I think salsa's default permissions for non-DDs
are much more stricter than Alioth were.  Nevertheless, I have been able
to create and populate the repository under the Team's group.

>
> I don't want to upload the package with the wrong Vcs-* headers, so
> let's get this fixed first.

I have refreshed those fields.  I have not still refreshed the changelog
date in order to wait for more potential changes.



Bug#890878: RFS: company-irony

2018-02-21 Thread Alberto Luaces
Sean Whitton writes:

> Hello,
>
> On Tue, Feb 20 2018, Alberto Luaces wrote:
>
>> I would need someone to sponsor "company-irony", which is now packaged
>> and uploaded to Alioth.
>
> It should be on salsa.  alioth is shutting down in a matter of weeks.
>
> If you request access to https://salsa.debian.org/emacsen-team/ I'll
> grant it.

Thanks, Sean.  It is now located at

https://salsa.debian.org/aluaces-guest/company-irony

I guess someone else has to clone it under the team project folders, so
I created that personal repository first.



Bug#890878: RFS: company-irony

2018-02-20 Thread Alberto Luaces
Geert Stappers writes:

> On Tue, Feb 20, 2018 at 10:15:37AM +0100, Alberto Luaces wrote:
>> I would need someone to sponsor "company-irony",
>
> What is "company-irony"?
> Please tell more about it  ( think "try to sell it" )
>

Sorry about that: "company" is a completion framework for emacs.
"irony" is a completion system for C++.  "company-irony" makes irony
analysis of the source code available to the company framework, so it
makes emacs behave as a sort of IDE for C++. 

>
>> which is now packaged and uploaded to Alioth.
>
> Where?
> Please provide URL  ( think "help those you want to help you" )
>

It is in the debian-emacsen git repo:

https://anonscm.debian.org/git/pkg-emacsen/pkg/company-irony.git/

Sorry again for the confusion: the guidelines I followed implied that I
had to fill a RFS, CCing to debian-emacsen.  I understand that for other
people it can be a bit cryptic.

Alberto



Bug#890878: RFS: company-irony

2018-02-20 Thread Alberto Luaces
Package: sponsorship-requests
Severity: normal

Hello,

I would need someone to sponsor "company-irony", which is now packaged and 
uploaded to Alioth.

Thank you!

P.S.: I have followed the guidelines at the dh-make-elpa manpage.



Bug#848687: RFS: yasnippet-snippets_0~git20161123-1

2017-01-02 Thread Alberto Luaces
Sean Whitton writes:


[... Elided completed tasks ...]

>
> Please accept my apologies for not raising these suggestions in a
> previous e-mail, and thank you for your patience with this sponsorship
> process -- I'm confident I'll be able to upload after your next update.
>
> (don't forget dch -r)

Quite the contrary, I am grateful for your guidance so far.

This time I have not set the git tag just in case some more
modifications are needed.  Now I see that it is better to leave it unset
since the uploader is the one that has the last word about it.  Please
feel free to set it if you think the package is ready for uploading.

Thanks!



Bug#848687: RFS: yasnippet-snippets_0~git20161123-1

2016-12-29 Thread Alberto Luaces
Hello Sean,

Sean Whitton writes:

> Hello Alberto,
>
> On Wed, Dec 28, 2016 at 11:43:29PM +0100, Alberto Luaces wrote:
>> I have now pushed a new release.
>
> Did someone else upload 0~git20161123-1?
>

No...

>
> If not, you should merge the changelog entries for -1 and -2.  It's just
> confusing to have changlog entries that never made it into the Debian
> archive.
>

Well, it seemed cleaner than modifying an already published history, but
I understand no solution is immune to drawbacks.  I have now pushed the
new, fixed branch.  Please note that you will have to re-synchronise, or
clone the repository again from scratch.

>
> After merging them, be sure to run `dch -r` to update the timestamp.
>

Done.  The timestamp is refreshed.

>
> Also, I think you need to put a tag '0_git20161123' on the upstream
> branch.  That's how gbp generates a tarball.

Done as well: upstream/0_git20161123, as gbp expects.

Thanks!



Bug#848687: RFS: yasnippet-snippets_0~git20161123-1

2016-12-28 Thread Alberto Luaces
Sean Whitton writes:

> Hello Alberto,
>
> On Wed, Dec 28, 2016 at 09:31:53PM +0100, Alberto Luaces wrote:
>> Thanks, Sean.  I have addressed the gbp.conf and changelog issues and I
>> think the realease is ready to go.
>
> Thanks again for your work.
>
> gbp.conf says upstream-branch=upstream but your upstream sources are on
> a branch called 'master' ...

Thanks for noticing!  Strangely enough, I missed that because gbp
buildpackage was successful nevertheless.

I have now pushed a new release.



Bug#848687: RFS: yasnippet-snippets_0~git20161123-1

2016-12-28 Thread Alberto Luaces
Thanks, Sean.  I have addressed the gbp.conf and changelog issues and I
think the realease is ready to go.

Please tell me if you think that something is still missing.



Bug#848687: RFS: yasnippet-snippets_0~git20161123-1

2016-12-19 Thread Alberto Luaces
Thanks, Sean.

After cloning the mentioned repository, one can generate both the source
and the binary packages from it issuing

gbp buildpackage --git-debian-branch=debian --git-upstream-branch=master



Bug#848687: RFS: yasnippet-snippets_0~git20161123-1

2016-12-19 Thread Alberto Luaces
Package: sponsorship-requests
Severity: normal

Hello,

I have packaged a new version of yasnippet-snippets:

https://anonscm.debian.org/git/pkg-emacsen/pkg/yasnippet-snippets.git/

I am looking for someone in the Debian Emacsen team to sponsor this
upload.

Thanks!



Bug#822634: RFS: magit-svn/2.1.1-1 [ITP]

2016-04-25 Thread Alberto Luaces
Package: sponsorship-requests
Severity: normal

  Dear mentors,

  I am looking for a sponsor for my package "magit-svn"

 * Package name: magit-svn
   Version : 2.1.1-1
   Upstream Author : The Magit Project Contributors
 * URL : https://magit.vc
 * License : GPL

  It builds those binary packages:

elpa-magit-svn - git-svn extension for magit

  To access further information about this package, please visit the following 
URL:

  http://mentors.debian.net/package/magit-svn


  Alternatively, one can download the package with dget using this command:

dget -x 
http://mentors.debian.net/debian/pool/main/m/magit-svn/magit-svn_2.1.1-1.dsc

  More information about hello can be obtained from http://www.example.com.

  Regards,
   Alberto Luaces