Wow, I'm really glad this topic has just come up right when I'm researching the list, trying to figure out why a large IdentitySet seems to have gotten incredibly slow in the trunk image, relative to 3.9. I haven't had a chance to do any measurements yet, but I have an IdentitySet with just a quarter-million objects, it *seems* to be much slower to add than in 3.9. Did Faster Sets get integrated into the trunk image? I can't seem to find discussion or benchmark results on the list. Any info is greatly appreciated. - Chris On Wed, Mar 17, 2010 at 12:58 PM, Stéphane Ducasse <stephane.ducasse@inria.fr> wrote:
Thanks Ralph
Levente, nicolas, henrik can you help us to take a good decision. I'm not precise enough on that topic and its implications. What I would love it to have Faster Set Levente is it dependent on hash changes that martin is working on?
Stef
On Mar 17, 2010, at 6:39 PM, Ralph Boland wrote:
When I released my FasterSets package it was too late for consideration for Pharo 1.0 so it was stated that it would be reconsidered for  Pharo  1.1.
Of course, since my release, Levente has rewritten my version of FasterSets from scratch and also made many other changes to Sets.
So I am wondering what the status of FasterSets is with respect to Pharo 1.1:
 a) FasterSets is not wanted in Pharo.  b)  You will use my version of FasterSets in Pharo 1.1.  c)  You will use Levente's version of FasterSets and perhaps his many other     changes to Sets/Dictionaries in Pharo  1.1.
Regards,
Ralph Boland
_______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
_______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project