OpenPoker architecture

[Joel Reymont]

I plan a series of posts describing the OpenPoker architecture.

Is this of interest to anyone now that the code is public?

Thanks, Joel

--
http://wagerlabs.com

[grant michaels]

the code actually isn't as interesting for me without your description, so i'll wholeheartedly vote in favor ... especially if you can shed light on how/why you ended up w/ it being the way it is ... hearing what you were also trying, and why the alternatives failed (benchmarking etc), is perhaps more important than what you ended up w/ from a learning perspective ...

best personal regards,
 
-[ grantmichaels ]-

[Eli Liang]

I find OpenPoker interesting since it falls exactly in the domain which I think Erlang seems most suitable.  Highly scalable, multi-CPU applications for which non-Erlang technologies have only traditionally handled with a bit of difficulty.

I find Erlang most fascinating because of the scalability potential. But it seems like few applications really test these boundaries.

[Eli Liang]

I agree with Grant. It's the exposition on what you learned about this problem-space and how you discovered Erlang works in it that is so interesting.  For example, you mentioned on the Erlang-questions list how you had to reimplement at least the core of the system a few times.  Once I recall, you mentioned changing to records, or perhaps it was changing from records to tuples. I can't exactly remember. What you learned that caused you to have to trigger the extensive refactoring would be very interesting.

[Toby DiPasquale]

On Mon, Dec 8, 2008 at 9:17 AM, Joel Reymont <joe...@gmail.com> wrote:
>

> I plan a series of posts describing the OpenPoker architecture.
>
> Is this of interest to anyone now that the code is public?

+1. That would be the perfect way to open The Erlang Journal, IMHO.

--
Toby DiPasquale

[Bernhard Damberger]

Agreed.

_bernhard

[Bernhard Damberger]

Don't know if other people were playing w/ OpenPoker...but I checked
it out and followed the install.txt instructions and got stuck on step
7:

BTW, to get going I had to install eunit before step 3.
svn co http://svn.process-one.net/contribs/trunk/eunit
In ~/.erlang add the following:
code:add_pathz("/Users/bernied/Projects/eunit/ebin").
code:add_pathz("/Users/bernied/Projects/eunit/include").

_bernhard
-----
bernied@me~/Projects/openpoker/src$ erl -sname master1 -s mnesia start
Erlang (BEAM) emulator version 5.6.2 [source] [smp:2] [async-threads:
0] [kernel-poll:false]

Eshell V5.6.2 (abort with ^G)
(master1@me)1> test:all().

=INFO REPORT==== 8-Dec-2008::12:27:22 ===
application: mnesia
exited: stopped
type: temporary
hand:rank_full_house1_test...*failed*
::error:{assertEqual_failed,[{module,hand},
{line,601},
{expression,"H # hand . score"},
{expected,24},
{value,0}]}
in function hand:'-rank_full_house1_test/0-fun-3-'/2

hand:rank_full_house2_test...*failed*
::error:{assertEqual_failed,[{module,hand},
{line,608},
{expression,"H # hand . score"},
{expected,264},
{value,0}]}
in function hand:'-rank_full_house2_test/0-fun-3-'/2

hand:rank_full_house3_test...*failed*
::error:{assertEqual_failed,[{module,hand},
{line,615},
{expression,"H # hand . score"},
{expected,20},
{value,0}]}
in function hand:'-rank_full_house3_test/0-fun-3-'/2

hand:rank_full_house4_test...*failed*
::error:{assertEqual_failed,[{module,hand},
{line,622},
{expression,"H # hand . score"},
{expected,32},
{value,0}]}
in function hand:'-rank_full_house4_test/0-fun-3-'/2

hand:full_house_win1_test...*failed*
::error:{assertEqual_failed,[{module,hand},
{line,781},
{expression,"short ( H1 ) > short ( H2 )"},
{expected,true},
{value,false}]}
in function hand:'-full_house_win1_test/0-fun-2-'/3

=======================================================
Failed: 5. Aborted: 0. Skipped: 0. Succeeded: 40.
All 13 tests successful.
There were no tests to run.
There were no tests to run.
There were no tests to run.
There were no tests to run.
There were no tests to run.
There were no tests to run.
test:five_players_sb_bust_test...*timed out*

[Joel Reymont]

On Dec 8, 2008, at 8:40 PM, Bernhard Damberger wrote:
> hand:rank_full_house1_test...*failed*
> ::error:{assertEqual_failed,[{module,hand},
> {line,601},
> {expression,"H # hand . score"},
> {expected,24},
> {value,0}]}
> in function hand:'-rank_full_house1_test/0-fun-3-'/2

The implementation changed and the full house tests need to be fixed.
I thought I did but I guess I did not commit the fix.

> test:five_players_sb_bust_test...*timed out*

The sb_bust tests are a pain in the rear. There's a lot to discuss here.

Anyone afraid of The Erlang Journal converting into The OpenPoker
Architecture Journal?

--
http://wagerlabs.com

[Toby DiPasquale]

On Mon, Dec 8, 2008 at 3:47 PM, Joel Reymont <joe...@gmail.com> wrote:
> The sb_bust tests are a pain in the rear. There's a lot to discuss here.
>
> Anyone afraid of The Erlang Journal converting into The OpenPoker
> Architecture Journal?

I'm not. Its way too early for those kinds of thoughts ;-)

--
Toby DiPasquale