Closed kkappler closed 2 years ago
Alternatively, a test could be added to the synthetic mth5 processing that explicitly uses data from two separate mth5 files. For example, the remote reference processing can be done with test12rr.h5 (this is currently working), and then can be done by referencing the data in test1.h5 and test2.h5.
Since we want this functionality (the ability to process data from more than one h5, I think adding the additional test is a better pathway forward than removing the individual mth5s.
Originally this issue was about redundant synthetic mth5 files. Rather than remove the mth5 files that associate with a single station, tests were modified so that the processing runs:
This has been implemented and is tested in tests/synthetic/test_compare_aurora_vs_archived_emtf.py
The synthetic tests create several mth5s data/test12rr.h5 data/test1.h5 data/test1_nan.h5 data/test2.h5
However, the data in test12rr.h5 is actually the data from test1.h5 and test2.h5.
The data/test1.h5, data/test2.h5 files can be removed if we point the processing to data/test12rr.h5, which could itself be given a nicer name like test_data.h5