Nick Hodgskin
Nick Hodgskin
Yeah, I'm struggling to understand what the test (and underlying fieldfilebuffer.py) is doing. The fact that we have failure here: ```py if chunk_mode != "specific_different": assert len(fieldset.U.grid._load_chunk) == len(fieldset.V.grid._load_chunk) ```...
> are there ever any mutable items in `kwargs` that should be copied (and not referenced) in creating `_repeatkwargs`? I don't think so, but can someone else think about it?...
Yes, thats also a possibility. Let's save this particular discussion for down the line where I review all attributes/methods and suggest actions to take (deprecate to make private, make read...
Done in #1709
Hmm, actually this can't really be removed at the moment since it patches the FieldSet. Let's still keep this open so that we can reconsider how its done in v4
I think it would be good to go ahead with this migration.
I think there's a distinction here between Field input and output. I think: - don't cast dtypes for the input field data - write trajectory output in float64 Looking at...
Thanks for the input @JamiePringle - we'll keep the flexibility of float32 trajectory outputs in v4
Awesome 🤩 > Where should I put these test cases? Below `test_invalid_lon_lat` in `tests/v4/test_xgrid.py`
cc @jimena-medinarubio - was running into difficulties getting 2D advection to work in XY space. Currently just setting to work with 2 depth slices at the surface