I sometimes wonder if the error/warning/notification logging code I wrote actually logs errors and warnings for the cases, and only for the cases, that it should. Would help a lot to have a test that feeds the whole pipeline some mock input from the Forecast and Github API queries, does dump_the_log, and checks that everything comes out as expected. Obviously could do much more detailed testing too, but that test I'm actively missing right now.
I sometimes wonder if the error/warning/notification logging code I wrote actually logs errors and warnings for the cases, and only for the cases, that it should. Would help a lot to have a test that feeds the whole pipeline some mock input from the Forecast and Github API queries, does
dump_the_log, and checks that everything comes out as expected. Obviously could do much more detailed testing too, but that test I'm actively missing right now.