Re: Hash Table Collisions (n.runs-SA-2011.004)

[Patrick Rapin <toupie300@...>]

> Well, this is not an official opinion, but one possible approach would
> be to break strings into two variants: short strings and long strings,
> where long strings would not be interned any more.

Thank you Roberto for the distance you take from the mainstream ideas.
All propositions made in this thread to solve the (potential)
collision problem look like patches to me, increasing both code size
and execution time.
Your idea is much smarter : this would still increase code size, but
slightly decrease execution time for long strings and open the door
for a new feature.
External strings might be useful in particular on embedded systems !
I also think that long strings used as keys or compared for equality
are not so common.
Would it be a possibility for the community to measure that usage in
real applications, for example using a patch ?