headers
Mark Proctor | 3 May 2005 12:49

Re: [drools-dev] Extractors and Extraction

Carlos,

Sorry about that, did announce several times on the list before we 
removed them - we hope to add something like extractors back, in the 
near future.

In the mean time all you can really do is just repeat what is in your 
extraction as a condition.

Sorry for not being more help.

Mark
Carlos Merida Campos wrote:

>Hi,
>
>I have to update an application that uses an old Drools version. This
>version still makes use of Extraction, and Extractor classes. I've seen
>that it was fixed from beta 18. Is there any place in which I could see
>some example on how to re-program my application in order not to use
>those classes (using the last beta 21). Could someone please give me a
>tip about this?
>thank you very much
>
>carlos.
>
>
>--
>--
>********************************
(Continue reading)

Permalink | Reply | 0 comments |
headers
Mark Proctor | 3 May 2005 12:52

Re: [drools-dev] [jira] Created: (DROOLS-331) Don't use spi Exception hierarchy

I'm going to do the release candidate this week, and drools 2.1 can drop 
JDK1.3 compatability - ie we can go full jdk1.4 and nested exceptions. 
We plan to release 2.1 quite soon after 2.0,  hopefully within two/three 
months - so it might make sense concentrating your efforts on 2.1 now.

Mark
Andy Barnett wrote:

> It is true that Drools strives to remain JDK 1.3 compatible, and so 
> far this is true except for the drools-spring/annotation module which 
> requires JDK 1.5's annotation support and the drools-groovy module 
> which uses the groovy-all-1.0-beta-9.jar file (which requires JDK 1.4).
>
> As for nested exceptions, the org.drools.DroolsException class (in the 
> drools-core module) supports nested exceptions (which are compatible 
> with JDK 1.3 and up).  So if there are any Drools exception classes 
> that are not extended from org.drools.DroolsException then that seems 
> to me to be the issue to be addressed -- not avoiding the use of them 
> altogether.
>
> I'm not sure what was meant by "no clean support nested exceptions".  
> If org.drools.DroolsException's support of nested exceptions is not 
> "clean" enough, then I again recommend that that be the issue 
> addressed -- in a JDK 1.3-compatible manner, of course.
>
> Or am I failing to see some other issues here?
>
> ~Andy
>
> ---
(Continue reading)

Permalink | Reply | 0 comments |
headers
Mark Proctor | 3 May 2005 13:05

[drools-dev] Please Test CVS prior to RC1

CVS branch-2_0 is now ready for one last test before I do the Release 
Candidate for 2.0 tomorrow. Would be great if you could all give it a 
try. We now have a full ant build which is useable interoprably with 
maven; ie you can compile with ant and run examples with maven, or 
compile with maven and run examples with ant. The Ant build is purely 
building and example running at this stage. Maven still supports the 
"run-all" goal which is the best way to check that everything is running ok.

Jeff Brown has commited an example using Conway's Game Of Life with a 
great gui, it has a java and a dsl implementation.

Ant junit's run slower than on maven, it seems the xml format report is 
slower under ant than maven, if anyone can shed light on that it would 
be appreciated.

http://drools.org/CVS+Download+Instructions

cvs -d :pserver:anonymous <at> cvs.drools.codehaus.org:/home/projects/drools/scm co -r branch-2_0 drools

Thanks

Mark
Permalink | Reply | 0 comments |
headers
Peter Lin | 5 May 2005 04:12
Picon

[drools-dev] JIT algorithm

I finally wrote down the algorithm for JIT Rete alpha nodes.

http://woolfel.blogspot.com/2005/05/rete-jit-algorithm-ok-now-that-i-am.html

peter
Permalink | Reply | 0 comments |
headers
Jeff Brown | 5 May 2005 05:06
Favicon

[drools-dev] commiting changes

I am writing documentation for the new Conway's Game Of Life example 
that has been added to 2.0 
(http://drools.org/Conway%27s+Game+Of+Life+Example).  The docs are not 
complete but there is a good start there.  While working on those docs I 
noticed a couple of things in the code that are leftovers from an early 
iteration of the app that really shouldn't be there.  These things don't 
really cause a problem but they are no longer relevant and make no sense 
being there.  Earlier today I think memelet merged the 2.0 branch with 
the trunk.  If I make code changes to clean this stuff up, should that 
go in the 2.0 branch or the trunk at this point?

Thanks for the clarification.

jb
--

-- 
Autism Strikes 1 in 250
Find The Cause ~ Find The Cure
http://www.autism-society.org/

Jeff Brown
brown_j <at> ociweb.com
Principal Software Engineer
Object Computing Inc.
http://www.ociweb.com/
Permalink | Reply | 15 comments |
headers
Barry Kaplan | 5 May 2005 06:57

Re: [drools-dev] commiting changes

If its just cleanup and not new functionality, I would say go ahead and commit it to the branch (2.0). But be
sure to do the merge correctly.

--

-- 
barry kaplan
groups1 <at> memelet.com
Permalink | Reply | 0 comments |
headers
Mark Proctor | 5 May 2005 12:36

Re: [drools-dev] commiting changes

Jeff,

Commit to 2.0 and merge with head and they will be part of the RC2, 
which I want to do next week. To late for RC1 as I'm just working on the 
release notes at the moment.

Mark
Jeff Brown wrote:

> I am writing documentation for the new Conway's Game Of Life example 
> that has been added to 2.0 
> (http://drools.org/Conway%27s+Game+Of+Life+Example).  The docs are not 
> complete but there is a good start there.  While working on those docs 
> I noticed a couple of things in the code that are leftovers from an 
> early iteration of the app that really shouldn't be there.  These 
> things don't really cause a problem but they are no longer relevant 
> and make no sense being there.  Earlier today I think memelet merged 
> the 2.0 branch with the trunk.  If I make code changes to clean this 
> stuff up, should that go in the 2.0 branch or the trunk at this point?
>
> Thanks for the clarification.
>
>
>
> jb
Permalink | Reply | 0 comments |
headers
Barry Kaplan | 5 May 2005 16:20

Re: [drools-dev] commiting changes

Jeff, 

In the commit message, you said "copied late trivial changes from branch-2_0". Did you do a full merge,
including moving up the merge point tag? 

see http://docs.codehaus.org/display/DROOLS/CVS+Download+Instructions, section "Working on the
Release Candidate Branch".

--

-- 
barry kaplan
groups1 <at> memelet.com
Permalink | Reply | 12 comments |
headers
Mark Proctor | 5 May 2005 20:59

[drools-dev] Drools 2.0-rc1 Released

The Drools <http://drools.org> development team is proud to announce the 
release of Drools 2.0-rc1.

Please see the Release Notes <http://drools.org/Release+Notes> for more 
details on this release.

The Project Downloads <http://drools.org/Project+Downloads>  page 
contains full information on obtaining Drools.

Happy Drooling

The Drools Team
Permalink | Reply | 0 comments |
headers
Barry Kaplan | 6 May 2005 00:24

[drools-dev] working memory in multi-threaded applications?

Is anybody working with drools in a multi-threaded application? Currently the synchronization in
WorkingMemoryImpl is very incomplete and, in some cases, incorrect. (There is a jira issue for this, but I
don't know the number offhand).

FYI, I intend to:

1) Remove all synchronization from WorkingMemoryImpl such that is only correct when access by a single
thread (which is really the case right now with the incomplete syncing). 
2) I will then provide an adapter that syncs all public methods.
3) Finally, if by user request and/or poor performance measurements, I will correctly synchronize the
class efficiently (probalby still with some sort of adaptor/AOP).

If you think this will cause you grief, let me know -- including why you think so.

--

-- 
barry kaplan
groups1 <at> memelet.com
Permalink | Reply | 0 comments |

Gmane