Daniel Lemire
Daniel Lemire
@novoj That would be a bug. If you provide a buffer, it should fill it up as much as possible. If you can identify such a case, please report it...
> IIRC that iterator yields at container boundaries, but returns how many elements were filled. Ah. Interesting.
@okrische If it is just a matter of adding support for an offset, I think it can be done with an extra function without too much fuss.
@okrische If you do in-place ORs, it shouldn't be too bad performance-wise.
@okrische I understand. I think. Don't we just want to add `BitSetUtil.bitmapOf(offset, long[])` which is nearly a trivial change?
@maciej Sure. Here is a demo with actual code... https://github.com/lemire/JavaFastPFOR/tree/master/examples/Axelbrooke Here are a few references... - Daniel Lemire, Nathan Kurz, Leonid Boytsov, SIMD Compression and the Intersection of Sorted Integers,...
@TuomasKiviaho It is not clear what you mean by "other way around". What is the final location?
When I ask what you mean by "the other way around", that has to do with the fact that if we add a new format to Roaring, it goes without...
@ponkin > Does method 'RoaringBitmap.serialize(out)' save bitmap in compressed format? It does. The format uses approximatively as much memory as the in-memory bitmaps (a bit less), in fact, it is...
@ponkin You can have a look at the memory usage in the following paper... https://arxiv.org/abs/1603.06549