rinclude
argument in collect_results
finds matches in the absolute path
#386
Labels
data
Related to data management
running-listing
Functionality for running and listing simulation runs
I have a DrWatson project named
GradientSensing
which resides in a directory namedPhycosphereRadius
(where other projects also live).Now I have run some analysis and let's say in my datadir I have some files, one named
Phycosphere.jld2
, the otherOtherstuff.jld2
(there will be multiple files with same suffix and different parameters etc).If I do
collect_results(datadir(), rinclude=[r"Phycosphere"])
I would expect that onlyPhycosphere.jld2
is loaded, instead alsoOtherstuff.jld2
is loaded, because therinclude
matches the "Phycosphere" keyword somewhere along the absolute path (even if it's outside of the project:~/PhycosphereRadius/GradientSensing/data/Otherstuff.jld2
).Is this a bug or intended behavior? For the moment I can just rename my file, but in general this behavior means that the results of my code are not independent of the absolute path of the project, which I guess is non-optimal. But maybe I'm just missing something obvious: is there a way to have
collect_results
ignore anything that happens outside of the project directory?The text was updated successfully, but these errors were encountered: