> "Loic Dachary" <[EMAIL PROTECTED]> writes:
>
>   If by "thread support" you mean that the poker-eval library is not
> thread safe, I think that's a correct assumption.

That's what i thought. :=)

>   Absolutely. Let me know your gna.org account and you've won a free
> ticket
> to svn commits ;-)

At the moment, this is not needed. I work in another environment, and i'm
far than ready to commit anything to pokersource directly...so i won't
bother until the moment i'm satisfied with my work.

I'll put links on this list once the project shows some progress though.

>   The only rule here is that each line commited must be associated with a
> unit test that covers it. Very simple rule that prevents lots of trouble.

I hear you :=)

>   If you think about modifying the poker-eval code structure, there is
> an addition requirement. You must write and run benchmarks covering
> all impacted code and show that you've not introduced something that
> slows down the library. A number of poker-eval users rely on its
> performances.

Well aware of this.

I think real thread-safe implementation most probably requires quite some
changes in poker-eval itself...again, i'd like to do this and test this in
my own environement.

I will be basing my work on poker-eval version 134 (as provided in
ubuntu), but i can't find such a tag in the svn sources. does it exist?

I also assume that poker-eval module does not go through alot of changes
now...it's extensive, stable...so even if i take long, there will not be
many troubles merging my work in (if i get to the point)...it's gonna be
C++ and will also rely on Qt environment (as i see it now).

Don't expect anything soon...but maybe some day there will be a
thread-safe poker-eval library.

-- 
sumac

_______________________________________________
Pokersource-users mailing list
[email protected]
https://mail.gna.org/listinfo/pokersource-users

Reply via email to