Jerven Bolleman

Actually trying to implement this I realized this is not sufficient. For collections that are backed by disk e.g. mapdb ones, we need to release these resources as soon as...

Pushed a nice idea, regarding the slow group by for the lmdb store. Quickly pushed to github before laptop battery dies.

Yes, the sparql engine does push in null values in a binding in a number of places. Which is very confusing and fragile. So I would like to fix it...

@kenwenzel would you mind having a look at this? See if you like the ideas.

Looks like github action for this pull request is hanging. I think I will close it in a bit and reopen it.

I am now at the point of actually wiring this up into the evaluation strategy. I think I don't have a good API placement or at least I am not...

lmdb query benchmark at 31c0933dfa989b2263c6a71ee107cf99979922f7 ``` Benchmark Mode Cnt Score Error Units QueryBenchmark.complexQuery avgt 5 71.214 ± 19.824 ms/op QueryBenchmark.distinctPredicatesQuery avgt 5 230.005 ± 43.627 ms/op QueryBenchmark.groupByQuery avgt 5 62.456...

@kenwenzel I think it shows there is a big potential gain. On the other hand this code is not nearly good enough to be merged. On my home machine I...

at 847dbe415a8b4328f1df77e51389dd7b5e98bb84 ``` QueryBenchmark.complexQuery avgt 5 17.374 ± 0.219 ms/op QueryBenchmark.distinctPredicatesQuery avgt 5 167.387 ± 3.132 ms/op QueryBenchmark.groupByQuery avgt 5 3.639 ± 0.135 ms/op QueryBenchmark.pathExpressionQuery1 avgt 5 59.854 ± 1.249...

Ok @kenwenzel this is a real improvement. Here the config on both the develop and this branch is to have a HashSet as the GroupBy iterator key datastructure. I noticed...