When I run make all in Sina/python/ on a fresh clone I get failures related to missing files. See the following output:
ValueError: No example data store found for noaa/data.sqlite in example dirs ('/collab/usr/gapps/wf/examples/', '/home/joe/projects/sina/Sina/examples') ValueError: No example data store found for noaa/data.sqlite in example dirs ('/collab/usr/gapps/wf/examples/', '/home/joe/projects/sina/Sina/examples')
I get similar output for the fukushima data set.
This can be fixed by:
navigate to Sina/python/
run make clean
run make
run source venv/bin/activate
navigate to Sina/examples/noaa/ and Sina/examples/fukushima/ and run their respective build_db.sh files.
navigate back to Sina/python/
run deactivate
run make clean
run make all
Though the above fixes it, I think the make file should check for those database files. If they are not found anywhere, then build them so the tests pass. I do see that building the fukushima database takes a while. A longer build time sounds better to me than failing builds, no? Not to mention the database files only need to be built the first time.
Alternatively, we could host the sqlite files somewhere.
At the very least there should be a README update stating that to have a successful make all the above database files need to be built first when operating on a Non LC network.
Environment:
When I run
make allinSina/python/on a fresh clone I get failures related to missing files. See the following output:ValueError: No example data store found for noaa/data.sqlite in example dirs ('/collab/usr/gapps/wf/examples/', '/home/joe/projects/sina/Sina/examples') ValueError: No example data store found for noaa/data.sqlite in example dirs ('/collab/usr/gapps/wf/examples/', '/home/joe/projects/sina/Sina/examples')I get similar output for the fukushima data set.This can be fixed by:
Sina/python/make cleanmakesource venv/bin/activateSina/examples/noaa/andSina/examples/fukushima/and run their respectivebuild_db.shfiles.Sina/python/deactivatemake cleanmake allThough the above fixes it, I think the make file should check for those database files. If they are not found anywhere, then build them so the tests pass. I do see that building the fukushima database takes a while. A longer build time sounds better to me than failing builds, no? Not to mention the database files only need to be built the first time.
Alternatively, we could host the sqlite files somewhere.
At the very least there should be a README update stating that to have a successful
make allthe above database files need to be built first when operating on a Non LC network.