usnistgov / fipy

FiPy is a Finite Volume PDE solver written in Python
http://pages.nist.gov/fipy/en/latest
Other
504 stars 148 forks source link

Add Logging #875

Closed guyer closed 2 years ago

guyer commented 2 years ago

See also a tiny new mpilogging package.

wd15 commented 2 years ago

mpilogging doesn't seem to be in pypi yet. Is it supposed to be?

guyer commented 2 years ago

mpilogging doesn't seem to be in pypi yet. Is it supposed to be?

No. Bootstrapping issue. I'd like somebody else's impression before I transfer it to usnistgov, and I want to do that before it goes to PyPI, but it seems silly to write (hopefully) very temporary instructions about how to get it.

wd15 commented 2 years ago

Would is make sense to include the version number of the dependencies at the start of the log file in debug mode(like for python setup.py test)?

guyer commented 2 years ago

Would is make sense to include the version number of the dependencies at the start of the log file in debug mode(like for python setup.py test)?

Do you not get something like

{'python': '3.10.4 | packaged by conda-forge | (main, Mar 24 2022, 17:45:10) [Clang 12.0.1 ]', 'fipy': '3.4.2.1+50.g2f7bf56b1', 'numpy': '1.22.4', 'pysparse': 'not installed', 'scipy': '1.8.1', 'matplotlib': '3.5.2', 'mpi4py': '3.1.3', 'petsc4py': '3.17.1', 'pyamgx': 'not installed', 'PyTrilinos': 'not installed', 'mayavi': '4.7.4', 'gmsh': '4.10.2', 'solver': 'petsc'}

logged at the top of a run?

wd15 commented 2 years ago

Do you not get something like

{'python': '3.10.4 | packaged by conda-forge | (main, Mar 24 2022, 17:45:10) [Clang 12.0.1 ]', 'fipy': '3.4.2.1+50.g2f7bf56b1', 'numpy': '1.22.4', 'pysparse': 'not installed', 'scipy': '1.8.1', 'matplotlib': '3.5.2', 'mpi4py': '3.1.3', 'petsc4py': '3.17.1', 'pyamgx': 'not installed', 'PyTrilinos': 'not installed', 'mayavi': '4.7.4', 'gmsh': '4.10.2', 'solver': 'petsc'}

logged at the top of a run?

Funny that you asked, but yes I do! I thought it was a good idea, but didn't think you would also have this good idea.

wd15 commented 2 years ago

FYI, there's a stray DEBUG=True here. Might be left over from the debug statements.