[petsc-dev] Matt, this is an annoying BuildSystem feature
Barry Smith
bsmith at mcs.anl.gov
Tue Mar 1 14:32:28 CST 2016
> On Mar 1, 2016, at 2:24 PM, Matthew Knepley <knepley at gmail.com> wrote:
>
> On Tue, Mar 1, 2016 at 2:13 PM, Barry Smith <bsmith at mcs.anl.gov> wrote:
>
> Matt,
>
> It prints that it is looking for the directory, then that it found a copy, then that it is cloning the repository.
>
> In reality it actually is looking for the directory, then it is cloning the repository, then it found a copy.
>
> Here is what happens:
>
> 1) The retriever does a clone() and logs into a buffer
>
> 2) We call getDir(), which logs the Looking and Found lines,
>
> I can have it log Clone, Looking, and Found. Would that be better?
I'd just like the order things are presented in the log to match the order they were performed (otherwise it is very confusing).
Barry
>
> Matt
>
> Is there any way to fix the logging so its listed in the order that it actually happens?
>
>
> Looking for SUPERLU_DIST at git.superlu_dist, hg.superlu_dist or a directory starting with SuperLU_DIST
> Found a copy of SUPERLU_DIST in git.superlu_dist
> Executing: git clone https://github.com/xiaoyeli/superlu_dist /export/home/petsc/petsc.clone/arch-opensolaris-pkgs-opt/externalpackages/git.superlu_dist
> stdout: Cloning into /export/home/petsc/petsc.clone/arch-opensolaris-pkgs-opt/externalpackages/git.superlu_dist...
> Executing: ['git', 'fetch']
> Executing: ['git', 'rev-parse', 'origin/master']
> stdout: 35c3b21630d93b3f8392a68e607467c247b5e053
> Executing: ['git', 'stash']
> stdout: No local changes to save
> Executing: ['git', 'checkout', '-f', '35c3b21630d93b3f8392a68e607467c247b5e053']
> Pushing language C
> Popping language C
> Pushing language Cxx
> Popping language Cxx
> Pushing language FC
> Popping language FC
> Have to rebuild SUPERLU_DIST, /export/home/petsc/petsc.clone/arch-opensolaris-pkgs-opt/externalpackages/git.superlu_dist/superlu_dist.petscconf != /export/home/petsc/petsc.clone/arch-opensolaris-pkgs-opt/lib/petsc/conf/pkg.conf.superlu_dist
>
>
>
>
> --
> What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead.
> -- Norbert Wiener
More information about the petsc-dev
mailing list