[MITgcm-devel] Moving remaining pieces to github

[Jean-Michel Campin]

Hi Dimitris,

Thanks for pointing to this. However, I prefer to have the regression-test bits directly
under MITgcm GitHub. And regarding the old web-site, since there is already a "manual_archive"
under MITgcm GitHub, I thought the old web-site would fit better there too.

Cheers,
Jean-Michel

On Fri, Nov 29, 2024 at 04:29:23PM -0800, Dimitris Menemenlis wrote:
> Just a comment (not a suggestion or request): for directories that used to exist under the old MITgcm_contrib CVS repo, Chris and Oliver created:
> https://github.com/MITgcm-contrib
> 
> D.
> 
> > On Nov 28, 2024, at 8:26???AM, Oliver Jahn <jahn at mit.edu> wrote:
> > 
> > ???Hi Jean-Michel,
> > 
> > sounds good.  I've made a couple of repos along these lines:
> > 
> > https://github.com/jahn/mitgcm.org
> > https://github.com/jahn/regression_tests
> > 
> > The mitgcm.org one has some empty commit messages early on which I didn't touch.  I fixed one higher up where the conversion tool got confused (2002-09-20 Importing web-site building process.)
> > 
> > Let me know what you think (Jean-Michel and others).
> > 
> > Cheers,
> > Oliver
> > 
> >> On 26/11/2024 12.24, Jean-Michel Campin wrote:
> >> Hi Oliver,
> >> I hope you are doing well.
> >> There are still few pieces that are under CVS that would be good to move to GitHub:
> >> 1) the old "mitgcm.org" with the original web-site src files.
> >> 2) some scripts to (a) run daily regression tests (currently in: MITgcm_contrib/test_scripts)
> >>                    (b) to collect the regression-test output, organize and check these.
> >>    this later part is mostly in "mitgcm.org/scripts/", and I could take care of bringing
> >>    in the 1 or 2 missing files once mitgcm.org/scripts/ is moved to git.
> >> The original web-site src files are still used to make the test web-page, this might
> >> change one day, but this should not prevent us to move this to GitHub.
> >> We could decide to convert mitgcm.org/ CVS to git (as you did for MITgcm and verification_other
> >> plus few other repos) and put this under MITgcm, at the same level as verification_other
> >>  ( git at github.com:MITgcm/verification_other.git ). Regarding name, may be "web_legacy" ?
> >> For the other bits, they could go to an other GitHub repos, also at the same level,
> >> called "regression_test", with one directory "run_tests" (<- from MITgcm_contrib/test_scripts/)
> >> and an other directory "collect_tests" (<- from mitgcm.org/scripts).
> >> I don't know if it's possible to make a "converted" git repos from 2 CVS dirs, and
> >> if it is, how much work this requires.
> >> so what is your opinion about this ? and do you (or others) have better suggestions ?
> >> And just for information, the location of CVS repos are, for
> >>  1)  at: zany.mit.edu:/data/ORWELL/export-9/gcmpack/mitgcm.org/
> >>  2a) at: zany.mit.edu:/data/ORWELL/export-9/gcmpack/MITgcm_contrib/test_scripts/
> >>  2b) at: zany.mit.edu:/data/ORWELL/export-9/gcmpack/mitgcm.org/scripts/
> >> Cheers,
> >> Jean-Michel
> > _______________________________________________
> > MITgcm-devel mailing list
> > MITgcm-devel at mitgcm.org
> > http://mailman.mitgcm.org/mailman/listinfo/mitgcm-devel