Open solomon-negusse opened 9 months ago
Thanks for the report. I won't have a chance to look into this for a while, but one note in case you want to look into it:
It looks like that's mutating a value in-place:
d["templates"][key] = d["templates"][key] + "?" + sas_token
So that cell probably isn't idempotent, and so it probably wouldn't be safe to run multiple times. I wonder what would happen if you avoid that (either by splitting the logic that updates the template, or by copy.deepcopy
-ing d
before running that.
Thanks for the quick response, will test that out.
Simply running twice cells 13 & 14 of the notebook that read and plot point single variable time-series for a point will reproduce this issue where the first run will have the valid values but second will be all
nan
s. I encountered this when parallelizing reading of the files with dask that results in multiple reads and the unexpected result.