I am currently experiencing some difficulties adapting our old universe/fetcher setup to quantopian 2.0.
I have a csv that I fetch with two columns: date, symbol.
This represents a list of symbols to trade on a particular date. Signals are inferred from other market sources directly in quantopian.
My problem is that the data.fetcher_assets contains way too much information as old securities also seem to be in the list; i.e. for a particular date where we only have 74 symbols listed we get back more than 3000 symbols which roughly corresponds to the complete list.
How do I most easily make this work? It worked perfectly before the upgrade.