Is full data range necessary for tests?

[prakaa]

Three tests in test_processing_info_maps.py draw data from the start of the data model (see code below) to 2018. In some cases this is 9 years worth of data and quite large. This slows testing down. Is this range necesary for these tests?

Based on tests run for #15, testing time incuding downloads is ~20 minutes on a high end computer.

start_test_window = defaults.nem_data_model_start_time

where

nem_data_model_start_time = '2009/07/01 00:00:00

[mdavis-xyz]

I've noticed that the .zip files aren't actually cached. Only the CSV files are.

One way of speeding up tests is to not re-download the zips, but to just re-extract them.