data-preservation-programs / fil-onboarding-report Goto Github PK

The "last day"/"last week" metrics currently take the last entry in the dataframe. This means in the scenario where a client has not been onboarding data for a few days instead of displaying 0 we display the amount of data they onboarded on the last day when they onboarded data.

In #13 we decided to pre calculate the id to name mapping using calculate_client_id_to_name_map but then we also kept the get_client_name

This results in clunky code in main.py, where the user of StatsClient now needs to call calculate_client_id_to_name_map before they can use get_client_name but that's not really clear from the function signatures.

Instead we should go back to the original implementation of making the function get the id to name mapping lazily, but then cache it when we get it.

It would be good to find out if /api/getVerifiedClients has a filter parameter as well. Something like /api/getVerifiedClients?client_id=f0.... .

1. If go to example: https://filecoin-tracker.com/?client_id=2130992 but then enter a new client id in the search box on the lower left such as f02088809 the url still stays the same as https://filecoin-tracker.com/?client_id=2130992

2. Ideally when updating the search box the url would also update the long url. This may create a problem that I can not think of right now. It does make it faster to share URLs on specific client ids after a search.

We should have a figma of the current design so we can leave feedback, and have candidate next versions