Closed surgura closed 3 months ago
Sorry for the one million files in the test zarr. I could also generate that on the fly instead of saving? maybe that's beter, not sure
Yes, I think that creating the zarr on the fly is better; as it will then also check whether zarr files are created as expected. Alternative is to put .zarr files in the gitignore?
Oh, and should the CNV
is the title of this PR not be CSV
?
Yes, I think that creating the zarr on the fly is better; as it will then also check whether zarr files are created as expected. Alternative is to put .zarr files in the gitignore?
Oh, and should the
CNV
is the title of this PR not beCSV
?
The file format is CNV, a file format created by Seabird(the microcat brand) The zarr file is in the test directory and used for the test. There already is a test for the creation of the zarr so that is not needed. I can either create it before running the rest by using the ctd function, but I think it might be better to replace this huge file with a very short zarr. That way a we have a fast test without having too much data on git
By changing the chunks
parameter in the ParticleFile object creation, you can also limit the number of files (=chunks) generated. See the note at https://docs.oceanparcels.org/en/latest/examples/tutorial_parcels_structure.html#4.-Execution-and-output
By changing the
chunks
parameter in the ParticleFile object creation, you can also limit the number of files (=chunks) generated. See the note at https://docs.oceanparcels.org/en/latest/examples/tutorial_parcels_structure.html#4.-Execution-and-output
Ok makes sense I will keep that in mind. I have create a very small one now and I think this should be fine as test input?
30 for CTD only