search

clawpack / classic

Classic single-grid Fortran routines
http://www.clawpack.org
BSD 3-Clause "New" or "Revised" License
11 stars 25 forks source link

Change Riemann Function Interface #55

Closed mandli closed 10 years ago

mandli commented 10 years ago

This PR updates classic to conform to the new Riemann function call interface introduced in clawpack/riemann#56.

rjleveque commented 10 years ago

Note that the travis build passed even though nosetests in classic failed, because .travis.yml in this directory hasn't been fixed to run the tests the way they are now run in amrclaw and geoclaw.

Once .travis.yml is fixed, travis will fail here, and also for new PRs to amrclaw and geoclaw until the new Riemann solver PR is merged and clawpack/clawpack is suitably updated.

We could follow the suggestion of @ahmadia and create a new branch rp_reorder or something in all these repositories and make sure they all work together before merging anything into master.

But then it's still not clear to me how we can ever merge these changes into master without travis failing on at least one of them in the process, since they are all mutually dependent.

Maybe someone can explain better how to proceed.

ketch commented 10 years ago

Are there any 3D tests in classic? I'm having trouble getting the corresponding code update in PyClaw; see https://github.com/clawpack/pyclaw/pull/330.

mandli commented 10 years ago

There are no tests in classic currently for 3D. I checked to see if the examples ran to completion and some of the output looked correct but I still have not been able to plot yet. I will look into clawpack/pyclaw#330.

rjleveque commented 10 years ago

examples/advection_2d_annulus has a local rpn2 and rpt2 that should be removed and instead point to proper solver in riemann.

Otherwise all examples run, and the tests in PR #56.

rjleveque commented 10 years ago

I just fixed the examples/advection_2d_annulus Makefile in PR #56 and it now works with the new riemann changes too.

mandli commented 10 years ago

Move to new PR #57, closing this one.