[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: libcifpp transition



Hi Andrius,

Okay, finally had time to work on this.

I patched the libcifpp many times but I now feel confident it should work. It builds on an experimental box (with i386 to make things more difficult). When I install the packages from this new patch (v2.0.4-6) I can build all the dependencies as well.


Op 24-01-2022 om 12:46 schreef Andrius Merkys:
To complete libcifpp1 -> libcifpp2 we need to ensure we can build all
reverse dependencies with new libcifpp. The list of these dependencies
is conveniently given in [1]. I have attempted building cif-tools
1.0.0-4, but failed due to the following:

In file included from src/cif2pdb.cpp:28:
src/cif-tools.hpp:34:10: fatal error: cif++/Config.hpp: No such file or
directory
     34 | #include "cif++/Config.hpp"
        |          ^~~~~~~~~~~~~~~~~~
The Config file was no longer needed thanks to the switch to cmake. The
API should be roughly the same.

As suggested by other, I opted to put a stub file in the debian package. That was the easiest thing to do.

Which brings me to:

[1]https://release.debian.org/transitions/html/auto-libcifpp.html
When is this auto transition thing starting? I mean, if I visit this
page I get only red boxes. Don't understand what needs to be done here.

Would be nice to have some overview to see what needs to be done.
There is a nice documentation for transitions [2]. For now, we have to
make sure we have everything needed to perform the transition: we need
to know what to do to make all reverse-dependencies build with the new
libcifpp. Then we start the transition by asking a slot, and then we
upload new libcifpp to unstable (it is in experimental now). Afterwards
we have to perform all the adjustments to reverse-dependencies (here the
red boxes will turn green, hopefully). Provided all red boxes turn
green, the transition is complete.

[2]https://wiki.debian.org/Teams/ReleaseTeam/Transitions

The page for the auto-package mentions a conflict with the also updated libpdb-redo. Is this going to work?

Anyway, now someone needs to file that bug to ask for a time slot. Who is supposed to do that? Is that me? Or does anyone else volunteer?

-maarten


Reply to: