Without being familiar with the Egret internals, it looks like this was instead expected to be a pandas.Series (at least from what I understand from https://github.com/grid-parity-exchange/Egret/blob/main/egret/parsers/rts_gmlc/parser.py#L772): if this is the case, it's possible that some change introduced in pandas 1.5 somehow causes numpy arrays to be returned/created instead, which are not compatible with the timestamp-based indexing used by Egret.
pandas was recently (yesterday?) updated to version 1.5, which seems to have introduced changes that are not compatible with the Egret internals.
We detected this when running the IDAES test suite, which runs (a version of) the Prescient 5bus example, resulting in the following stack trace:
After running
pytest --pdb
, it looks like the contents ofself.timeseries_df["Series"]
are a numpy array:Without being familiar with the Egret internals, it looks like this was instead expected to be a
pandas.Series
(at least from what I understand from https://github.com/grid-parity-exchange/Egret/blob/main/egret/parsers/rts_gmlc/parser.py#L772): if this is the case, it's possible that some change introduced in pandas 1.5 somehow causes numpy arrays to be returned/created instead, which are not compatible with the timestamp-based indexing used by Egret.