News for hackers. Stuff that matters.

A comment on Finding unique items: hash vs. sort. Parent comment

Re:Finding unique items: hash vs. sort by noctune on Wednesday January 08, 2020 @02:16AM (#21989087)
One reason why unordered map is slow is that it has to be a chained hash map due to the spec not allowing iterator invalidation when resizing.
[ Parent | on HN ]
Re:Finding unique items: hash vs. sort by bertr4nd on Wednesday January 08, 2020 @03:28PM (#21995438)
I believe unordered_map does invalidate integrators on resize, but does not invalidate references to keys/data, which means the data can’t be stored inline with the hash table structure.
[ Parent | on HN ]
Re:Finding unique items: hash vs. sort by aidenn0 on Wednesday January 08, 2020 @03:58AM (#21989681)
that's certainly annoying...
[ Parent | on HN ]