search

danieljprice / phantom

Phantom Smoothed Particle Hydrodynamics and Magnetohydrodynamics code
https://phantomsph.github.io
Other
103 stars 223 forks source link

Automated test that ./phantom blah.in succeeds after running phantomsetup #451

Closed danieljprice closed 1 year ago

danieljprice commented 1 year ago

Type of PR: Bug fix / new test

Description: See #450 and #449: In the buildbot we now check not only that ./phantomsetup blah succeeds, but also that ./phantom blah.in successfully runs for at least one timestep

Testing: Checked that this works with SETUP=turb, will be checked fully in the GitHub actions

UPDATE: a LOT of the setups turned out to be broken when run this way, particularly with DEBUG=yes. Slowly working through the backlog of complaints from the actions... tidying up a few other things (especially #55) as I go along

Did you run the bots? yes

Did you update relevant documentation in the docs directory? yes

danieljprice commented 1 year ago

@lsiess three of the remaining test failures are from "wind", "ismwind" and "radwind", which I think will be fixed when #446 is merged by the looks of your code (i.e. your new set_defaults routine)

...so please can you get #446 to pass so it can be merged...?

lsiess commented 1 year ago

The problem with the test setups is that you have to compile and link with utils_binary.f90 set_binary.f90 inject_wind.f90 because they are needed for test_wind. I found a dirty way of doing it in the Makefile with 

ifndef SRCINJECT
    SRCINJECT=utils_binary.f90 set_binary.f90 inject_wind.f90
endif

and the use of

INJECT_PARTICLES=yes

but I am sure you will find a more elegant way of doing it. cheers

danieljprice commented 1 year ago

got it, I think I fixed this bit at least