On Fri, Mar 13, 2015 at 9:22 AM, Ion Gaztañaga
El 13/03/2015 a las 8:51, Giacomo Drago escribió:
My library is not ready for a formal review process, and I'd like to ask whether there is some potential interest first, thus saving everybody a lot of time.
I have been working on a flat_map (http://www.boost.org/doc/libs/1_57_0/doc/html/boost/ container/flat_map.html) variant that keeps some of the distinctive features of the original one (contiguous storage/cache friendliness, zero memory overhead/shrink_to_fit support, vector-based representation, etc...) but makes insertion and erasure of the elements much faster, at the price of a slower lookup and in-order traversal.
Just a question. Are your insertions one by one or by range?
Might also be interesting to know if your perf numbers are based on a primitive key or a UDT, and whether the relative performance changes depending on key/value sizes. --DD