james writes:
> Using the -Xprof switch seems to indicate java.lang.Character.hashCode
> is called a lot and i guess that's to do with the nested maps that
> represent the trie.
>
> Using jvisualvm doesn't help much either as it's not giving me a call
> graph.
> Just self counts for each function
.
> I'm sure there is lots of room for improvement to make it more
> idiomatic and perform better so I would be grateful if anyone would
> care to cast their eye over it.
>
> http://wiki.github.com/phraemer/Boggle-Solver
>
> thanks,
>
> James
--
You received thi
jure version. That would be interesting
> for comparison ...
>
> On Nov 7, 12:11 am, james wrote:
>
>
>
> > Hey!
>
> > As a learning exercise I wrote a simple boggle solver in clojure.
> > I'm sure there is lots of room for improvement to make it more
>
.github.com/phraemer/Boggle-Solver
I tested this against a friends very neat (but not optimised) c++
implimentation and it's only slightly faster as far as I can tell.
I suspect that some clever people on here could easily get better
performance out of my version and yet keep it idiomatic. (
top.
Haven't yet implemented a Clojure version. That would be interesting
for comparison ...
On Nov 7, 12:11 am, james wrote:
> Hey!
>
> As a learning exercise I wrote a simple boggle solver in clojure.
> I'm sure there is lots of room for improvement to make it more
> idiom
Very nice, I had actually been implementing a boggle solver for my
optimal boggle grid generator (5x5 with no limits on number of letters
on dice). I hadn't discovered the wonders of assoc-in though which
makes things look much nicer when creating nested maps. I really like
the layout of
Hey!
As a learning exercise I wrote a simple boggle solver in clojure.
I'm sure there is lots of room for improvement to make it more
idiomatic and perform better so I would be grateful if anyone would
care to cast their eye over it.
http://wiki.github.com/phraemer/Boggle-Solver
thanks,
