Arch Linux User Repository

Pre-built binaries of this package and its dependencies can be found in the arch4edu repository.

Important note: Is recommended to receive the following key before to install:

$ gpg --recv-keys BF38D4D02A328DFF

Also is possible skip the verification, adding the flag for (e.g makepkg, yay) --skippgpcheck, --nopgpfetch, respectively.

Trying to build the latest version I got this error:

[...]
stellarium/src/stellarium-24.3/src/core/modules/SolarSystem.cpp'
In file included from /usr/include/oneapi/tbb/detail/_small_object_pool.h:23,
                 from /usr/include/oneapi/tbb/detail/_task.h:23,
                 from /usr/include/oneapi/tbb/parallel_for.h:23,
                 from /usr/include/tbb/parallel_for.h:17,
                 from /usr/include/c++/14.2.1/pstl/parallel_backend_tbb.h:25,
                 from /usr/include/c++/14.2.1/pstl/parallel_backend.h:20,
                 from /usr/include/c++/14.2.1/pstl/algorithm_impl.h:22,
                 from /usr/include/c++/14.2.1/pstl/glue_execution_defs.h:50,
                 from /usr/include/c++/14.2.1/execution:39,
                 from stellarium/src/stellarium-24.3/src/core/modules/SolarSystem.cpp:51:
/usr/include/oneapi/tbb/profiling.h:229:15: error: expected unqualified-id before ‘)’ token
  229 |     void emit() { }

Looks like QT's emit-macro is getting in the way of TBB defining that as a function name. Wrapping the include in src/core/modules/SolarSystem.cpp fixes it:

#undef emit
#include <execution>
#define emit

As this seems to be specific to building with TBB, not sure where that should be fixed. Is there even a way to use <execution> without linking against TBB?

sorry about the unformatted log:

:: (2/2) Parsing SRCINFO: stellarium gpg: error reading key: No public key

:: PGP keys need importing: -> 79151C2E6351E7278DA1A730BF38D4D02A328DFF, required by: stellarium :: Import? [Y/n] :: Importing keys with gpg... gpg: error writing keyring '[keyboxd]': Attempt to write a readonly SQL database gpg: error reading '[stream]': Attempt to write a readonly SQL database gpg: Total number processed: 0 -> problem importing keys

Hello. Stellarium requires calcmysky but there is a conflict with /usr/lib/libShowMySky-Qt6.so: calcmysky-0.3.2-1 stellarium-24.1-1. Can you solve it?

error: failed to commit transaction (conflicting files)
/usr/lib/libShowMySky-Qt6.so exists in both 'calcmysky' and 'stellarium'
/usr/lib/libShowMySky-Qt6.so.15 exists in both 'calcmysky' and 'stellarium'
/usr/lib/libShowMySky-Qt6.so.15.0.0 exists in both 'calcmysky' and 'stellarium'

@afontenot You are right, that line "weak-digest SHA1" is there in my gpg.conf. I have added it (in addition other lines) years ago, based on this article, but now did not think of it anymore; thank you for reminding me and pointing me to this my deviation from default behavior/configuration.

Nevertheless: Although using SHA1 for self-signatures of PGP keys is not insecure, i.e. SHA1's weakness not problematic for this particular purpose, I found many voices on the web claiming that SHA1 should generally not be used anymore for anything, or banned altogether. So it might be worth considering re-self-signing all keys anyway, found these useful instructions on how to achieve this (only in German unfortunately).

Thanks a lot for your help, and for providing Stellarium, I do love this beautiful program! <3

@CaliforniaCoach: I'm almost certain you have weak-digest SHA1 set somewhere in your GPG config file. I haven't changed my config file (in fact I don't have one), and I have no issue with running gpg --verify -v stellarium-23.4.tar.gz.asc.

Check the Arch Wiki for some possible locations of these files.

You shouldn't need to allow weak digest algorithms, because according to GPG documentation, "MD5 is the only digest algorithm considered weak by default."

Coming back to my problem from 2023-10-07: I am still having the same issue, makepkg still aborts with an error "One or more PGP signatures could not be verified" (yeah, I know, I could skip the check with --skippgpcheck, but it does not feel good and lowers security).

My outputs of gpg --verify ... from 2023-10-07 are still (almost) the same, sig check fails with standard options and passes with --allow-weak-digest-algos option. Using newer gpg version 2.4.3 this time, though.

I think it would be easy to fix it on your (or Alexander Wolf's) side, you (or Alexander) would just have to (cross-)sign the keys again with the proper algos (other than SHA-1), and an up-to-date gpg version. I also found this related article: https://www.gnupg.org/faq/subkey-cross-certify.html

Could you do so, please?

I am getting stuck here when compiling:

[65/934] Building CXX object plugins/Exoplanets/src/CMakeFiles/Exoplanets-static.dir/Exoplanets-static_autogen/mocs_compilation.cpp.o ninja: build stopped: subcommand failed. ==> ERROR: A failure occurred in build(). Aborting... error: failed to build 'stellarium-23.4-4': error: packages failed to build: stellarium-23.4-4

Alright, nlopt is a dependency.

stellarium wouldn't start because of missing libnlopt.so.0, so i guess it depends on nlopt package

Hi. I had to rebuild stellarium since I ran across the same startup-error as this guy: https://www.indilib.org/forum/ccds-dslrs/14096-asi-driver-crashed-with-symbol-lookup-error.html so I guess, my INDI-library also updated since the last time I successfully used stellarium. (I also rebuilt calcmysky, not sure if that was necessary.) works again. Greetings