cmake: compat 11 vs 13 (-indep)

[Mathieu Malaterre]

[cc me please]

Symptoms:

I've recently discovered the following change in behavior for
dh_missing + cmake + -indep. Basically using compat 13 it is now an
error when build-indep rule install stuff other than just the -indep
files (1).

This means that the following d/rules is going to fails when using compat=13:

https://salsa.debian.org/multimedia-team/openvdb/-/blob/master/debian/rules

As you can see, during debuild -A I only build one target (not 'all'):

https://salsa.debian.org/multimedia-team/openvdb/-/blob/master/debian/rules#L94

In compat 13 I was able to take advantage of `make -C`:

https://salsa.debian.org/debian-phototools-team/imath/-/blob/master/debian/rules#L35-37

This 'trick' will fail if someone switch from --buildsystem=cmake into
--buildsystem=cmake+ninja.

---

Root issue. As far as I understand cmake install() command (2)
documentation, there is a single 'install' for the entire project. In
other words, it is not possible using regular cmake logic to do:

$ cmake . && make  doc && make install_doc_related_stuff_only

Could someone with better understanding of dh_missing + cmake please
review the above and suggest a potential better solution than my
correct "hack" of using `-C docs`. I could also remove stuff installed
in debian/tmp just before `dh_missing` is run, but this is also a
hack.

Thanks much in advance,

(1) 
https://buildd.debian.org/status/fetch.php?pkg=imath&arch=amd64&ver=3.1.3-7&stamp=1642087594&raw=0
(2) https://cmake.org/cmake/help/latest/command/install.html

Bug#1003612: RFS: libview/0.6.6-3 [ITA] -- VMware's Incredibly Exciting Widgets

[Bastian Germann]

On 14.01.22 14:18, Ileana Dumitrescu wrote:

I did not originally notice #956789. The upstream code has not been updated 
since 2010, so it looks like nobody is maintaining it. I ran this through 
apt-rdepends, and no other package depends on libview to build. It should be 
removed since it appears that nobody is using it.


So you do not use it? Why would you then adopt it?

Should this RFS then be closed and the ITA set back to orphaned?

Bug#1003090: RFS: ffcvt/1.7.5-1

[Tong Sun]

Trying again.

On Thu, Jan 6, 2022 at 1:16 PM Tong Sun
 wrote:
>
> On Mon, Jan 3, 2022 at 3:45 PM Tong Sun wrote:
> >
> > On Mon, Jan 3, 2022 at 3:20 PM Tong Sun
> >  wrote:
> > >
> > > Package: sponsorship-requests
> > > Severity: normal
> > >
> > > Dear mentors,
> > >
> > > I updated my ffcvt to a newer version, and am now looking for a sponsor.
> > >
> > > Here is from the d/changelog
> > >
> > > ffcvt (1.7.5-1) unstable; urgency=medium
> > >
> > >   * New upstream version 1.7.5
> > > - add --Speed for speeding up playback (v1.7.5)
> > > - add a copy target type that can speed up operations (v1.7.4)
> > > - add option -S,Seg -- split video into multiple segments (v1.7.3)
> > > - add option -sel, subtitle encoding language (v1.7.2)
> > > - add option -C which allows cutting multiple segments (v1.7.1)
> > > - add wx type for weixin (v1.7.0)
> > > - x264-mp3 type set ext to .mp4 (v1.6.2)
> > >   * fix "source: out-of-date-standards-version" problem
> > >   * fix "source: update-debian-copyright" problem
> > >   * fix "source: prefer-uscan-symlink filenamemangle" problem
> > >
> > > Note
> > >
> > > 1) My GPG key expired a few days ago, and I've already submit the
> > > update to keyring.debian.org, yet I understand it might take a while
> > > to really get updated, so please
> > >
> > > go directly to
> > > https://salsa.debian.org/go-team/packages/ffcvt/-/commits/master
> > >
> > > 2) Please disregard the .gitlab-ci.yml build failure as I'm not
> > > supposed to touch it.
> >
> > Ops, please hold off reviewing it, as I've found out where the problem
> > is -- ffcvt depends on the latest easygen to build yet I haven't
> > upgraded easygen in Debian yet. and my expired GPG key is preventing
> > me from fixing it promptly.
> >
> > Will update when the situation is fixed.
> >
> > Sorry & thanks
> >
> > > The build is fine, check out here --
> > > https://github.com/suntong/ffcvt/runs/4687405723?check_suite_focus=true
>
> Hi,
>
> The situation should have been fixed with the new upload of easygen.
>
> However, the CI build is still failing in salsa. This is something
> that I don't understand as it builds OK on github.
>
> Sorry I've run out of ideas why it is happening like this, and am now
> thinking to remove the build dependency of easygen, to fix this and to
> make things easier...

I still don't have a clue why it builds OK on github but fails in
salsa CI build, and I still hope that somebody can help.

I'll wait for one or two weeks more, and if still nobody can help,
I'll remove the build dependency of easygen as planned, as I know for
sure it can fix the issue.

> Somebody help please.

thanks