Closed mvuorre closed 1 year ago
I think it would be possible to refactor the code so that some of the intermediate steps are stored in either temporary or permanent tables in the db. Right now dbplyr just keeps addding subqueries. See compute
and rows_append
or _insert
.
Great idea, I'll look into it.
Fixed in #9.
In an early version there was a bug that caused too many game_saved events (many per second).
I tried to fix it in 842bc63e3cf218ad9f17f61f1dea005807fc9b18, but the data is too large to do this in R (with methods I used.)
Find out another way of removing game_saved rows where lag from previous was too short. Optimally in the database (@rpsychologist?).