Re. subsurface: FTBFS in experimental

[Lutz Vieweg]

On 08/26/2015 07:52 PM, Pierre-Yves Chibon wrote:
>> Yesterday I endeavored to build the Subsurface git master head on CentOS,
>> and it was kind of a nightmare - everything, from just-one-patchlevel-version-of-
>> cmake-ahead-of-what-CentOS-delivers, continuing with of course Qt 5 in a
>> very certain version, and with lots of compile-time options to be tweaked,
>> up the customized libgit/gmerlin/marble libraries had to be built from
>> scratch, it took me ~50 attempts of error-spilling compiles until
>> all the strange dependencies were finally resolved, resulting in a
>> ~500MB sized installation directory.
>
> Which version of CentOS do you run?

The latest-and-hopefully-greatest, 7.1.1503 at this time.

> If I get the copr running, we could easily add an epel7 branch and see what
> fails to build there, this might help a bit.

I don't know anything about copr yet, and I am kind of sure that the
way I built Subsurface for CentOS (in a freely relocateable directory
which holds everything including a script which deduces the required
environment-variable settings from its own path, setting up a symbolic
link under /tmp/ to work as an "anchor" for the --prefix=... setting)
will probably violate tons of RHEL distribution policies ;-)

But if anyone is interested in my build, I'm happy to provide a
.tar.xz to unpack and run (with only CentOS-provided dependencies
needed as prerequisites).

> (For EL6 I have little hope we can get it to run w/o spending days on it, but
> maybe I could be surprised :) )

This I once tried, but would have ended up with basically an
installation directory containing a complete distribution,
starting at glibc of a certain release, which would have
been required to install one of the dozens of other libraries...
I had a working Subsurface 4.0 for CentOS 6, and would't go
beyond that.

Regards,

Lutz Vieweg