AUR (en) - gmsh

Git Clone URL:	https://aur.archlinux.org/gmsh.git (read-only, click to copy)
Submitter:	S1G1
Maintainer:	gborzi (carlosal1015, gpettinello)
Last Packager:	carlosal1015
Votes:	61
Popularity:	0.039885
First Submitted:	2006-04-04 23:31 (UTC)
Last Updated:	2024-01-21 21:19 (UTC)

Latest Comments

« First ‹ Previous 1 .. 15 16 17 18 19 20

gborzi commented on 2010-10-18 16:38 (UTC)

I forgot to add that if you have gcc-fortran without lapack+blas the package builds, but at the configuration stage it warns that lapack was not found and some algorithms may not work.

gborzi commented on 2010-10-18 13:25 (UTC)

Hi Tempel, I have reproduced your error message on my system by uninstalling both gcc-fortran and lapack+blas. Perhaps your lapack library got corrupted, or you have non-standard lapack library (e.g. a vendor library) that isn't recognized by gmsh. With atlas-lapack the package builds fine.

Tempel commented on 2010-10-18 04:24 (UTC)

Really? For me, without gcc-fortran, it doesn't complete the cmake command, ending with "-- Configuring incomplete, errors occurred!" Earlier in the same step, it says "CMake Error: your Fortran compiler: "CMAKE_Fortran_COMPILER-NOTFOUND" was not found. Please set CMAKE_Fortran_COMPILER to a valid compiler path or name." To test this, I built it successfully, then cleared out the build directory, then removed gcc-fortran (no other packages were changed). After that, it would not build again. You're right that each .f has a corresponding .c. So could this problem be from vagaries in the cmake system? Is there any reason it might behave differently in two different systems?

gborzi commented on 2010-10-17 11:28 (UTC)

The package builds fine without gcc-fortran. Please note that for each .f source file there is a corresponding .c file in the source tree.

Tempel commented on 2010-10-17 03:31 (UTC)

It looks like gcc-fortran should be in the makedepends.

gborzi commented on 2010-06-24 11:34 (UTC)

In this update the arch is 'any' because this is a architecture-independent package and added "LC_ALL=C" before "make doc" to avoid a bug in texi2dvi.

gborzi commented on 2010-04-06 18:13 (UTC)

Feel free to report any problem in the future without the fear to bother me. BTW, gmsh needed to be rebuilt after gmp upgrade.

Tempel commented on 2010-04-06 17:13 (UTC)

I just recompiled gmsh, and now it works perfectly. Sorry to have bothered you!

gborzi commented on 2010-04-04 20:25 (UTC)

I have just recompiled gmsh and tried it on a simple IGES file. It was opened without problems. Please check if you have /etc/profile.d/opencascade.sh and that it is correctly sourced.

Tempel commented on 2010-04-04 14:35 (UTC)

Import of IGES files is failing, giving the message "Error : Gmsh must be compiled with OpenCascade support to load '[file].IGS'" OpenCascade was built first, and I didn't edit the PKGBUILD for it or gmsh, so what happened to the OpenCascade support?