Closed njvrzm closed 6 months ago
As for benchmarking, I'd love that. I don't have any sense of how much this might affect performance, and it depends entirely on the characteristics of the data: like, (to take it to the extreme case) if every single EntryId
happened to be the same, this change would do virtually nothing. Also since the aggregation is happening to data coming in from the network, IO latency might make this insignificant. I think my next step is probably to get the plugin running locally and do some profiling.
We can revive this if it comes up but for now I don't think this is worth pursuing.
Aims to speed up result aggregation in the client. Previously this scanned the list of entries to look for a match for each new result; now it will maintain a map of EntryId to index.