On Thu, 07 Dec 2006 11:40:24 +0900, Shigio YAMAGUCHI wrote...
> > When a project is large, most of the execution time is spent for the 
> > pooling. 
> > The switching after pooling seems too late.
> 
> Using idset, we can count the number of lines of the result without spooling.
> How about this?
> 
>       count the number of result using idset
>       if (number of lines of the result > RESULT_THRESHOLD)
>               use GPATH
>       else
>               (load path names and) use internal sort

Hmm. The internal sorting is not slow even on large projects. 
I can not understand why you do not focus on the speedup of pooling.

----
Hideki IWAMOTO  [EMAIL PROTECTED]


_______________________________________________
Bug-global mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/bug-global

Reply via email to