Describe what change or improvement would you like
Because we're reading this data unofficially, it might be kind to cache the extracted lookup key in the class. Right now the code fetches the key repeatedly (sometimes multiple times per operation), which probably causes extra load.
When this package is used on a server, it's probably as a long-lived class. Since it doesn't change often, you could store the key on the class instance until you get a 404 (which could either be real, or a signal that the url has been updated).
I did a similar approach here: xavdid/howlongtobeat@5375d74#diff-1af9a3390b92891b019f2dbae92c77ca7d05282003849d5642715613e809ab54R89, if that's useful!
Describe what change or improvement would you like
Because we're reading this data unofficially, it might be kind to cache the extracted lookup key in the class. Right now the code fetches the key repeatedly (sometimes multiple times per operation), which probably causes extra load.
When this package is used on a server, it's probably as a long-lived class. Since it doesn't change often, you could store the key on the class instance until you get a 404 (which could either be real, or a signal that the url has been updated).
I did a similar approach here: xavdid/howlongtobeat@5375d74#diff-1af9a3390b92891b019f2dbae92c77ca7d05282003849d5642715613e809ab54R89, if that's useful!