Persistent store and web administration
by Schwenker, Stephen
Hello,
We have been using jboss rules for 6 months now and find it to be a good
tool overall. On the other hand we have noticed a few down sides.
Currently our rule engine runs on a jboss server and is deployed in a
sar. Every time we want to change the rules, we have to re-package and
re-deploy the app.
We want to advance our solution by providing a web application to
administer the rules. The first step of this will be to create a form
of persistent store for the rules. My first thought is to put it into a
database because we can then manage it easily with a web admin. Has
anyone done some work on this or can give me some ideas on how I should
start working on this? I'm willing to contribute some of my code to the
project if anyone is interested.
Thank you,
Steve.
16 years, 5 months
Persistent store and web administration
by Schwenker, Stephen
Hello,
We have been using jboss rules for 6 months now and find it to be a good
tool overall. On the other hand we have noticed a few down sides.
Currently our rule engine runs on a jboss server and is deployed in a
sar. Every time we want to change the rules, we have to re-package and
re-deploy the app.
We want to advance our solution by providing a web application to
administer the rules. The first step of this will be to create a form
of persistent store for the rules. My first thought is to put it into a
database because we can then manage it easily with a web admin. Has
anyone done some work on this or can give me some ideas on how I should
start working on this? I'm willing to contribute some of my code to the
project if anyone is interested.
Thank you,
Steve.
16 years, 5 months
Persistent store and web administration
by Schwenker, Stephen
Hello,
We have been using jboss rules for 6 months now and find it to be a good
tool overall. On the other hand we have noticed a few down sides.
Currently our rule engine runs on a jboss server and is deployed in a
sar. Every time we want to change the rules, we have to re-package and
re-deploy the app.
We want to advance our solution by providing a web application to
administer the rules. The first step of this will be to create a form
of persistent store for the rules. My first thought is to put it into a
database because we can then manage it easily with a web admin. Has
anyone done some work on this or can give me some ideas on how I should
start working on this? I'm willing to contribute some of my code to the
project if anyone is interested.
Thank you,
Steve.
16 years, 8 months
externalize solver configuration - which technology?
by Geoffrey De Smet
For the solver I 'd like to externalize the configuration into an xml(?)
file.
something like
<localsearch decisionMode="OPTIMAL">
<finish steps="500"/>
<ruleBase location="/org/foo/bar/drools.drl"/>
</localsearch>
This is a bit inspired by how ehcache is configured, clean and easy,
without any class implementation details: the way users like it.
I 've been thinking on doing this with XStream, but that's not versatile
enough to make user friendly configuration files. Although Xstream works
out great to serialize solutions to and from xml files.
Next I was thinking about is JDom or Xom, but that's yet another
dependency. What would you guys recommend?
--
With kind regards,
Geoffrey De Smet
16 years, 8 months
Re: [drools-dev] blogging
by Michael Neale
I think blogspot offers per tag RSS, its just not public (you kind of have
to guess some URLs) - so yes, that can be done !
On 1/29/07, Paul Browne <paulb(a)firstpartners.net> wrote:
>
> Michael,
>
> One blog , tag as appropriate (e.g. Dev, user) would be a good way to
> good.
>
> Even better (depending on your blogging platorm) would be to offer an
> RSS feed for each.
>
> Paul
>
> On 1/29/07, Michael Neale <michael.neale(a)gmail.com> wrote:
> Hi All. I like to update the blog with progress of various things (and
> want to, more) - but this is not of general interest to most people
> (other then to know that "stuff that is cool is happening").
> >
> > I know that people like to read our blogs for "end user" type useful
> tips with rules - which is fine- but do you think its still a good idea to
> do lots of "dev status" type blogs? I think its useful, as long as we tag it
> accordingly (otherwise we can create a "sub blog" or something).
>
> >
> > Thoughts? I just want to update it more regularly, helps keep me
> accountable to the community ;) Reduces the chance of slacking ;)
> >
> > Michael.
> >
>
> ---------------------------------------------------------------------
> To unsubscribe from this list please visit:
>
> http://xircles.codehaus.org/manage_email
>
>
16 years, 8 months
blogging
by Michael Neale
Hi All. I like to update the blog with progress of various things (and want
to, more) - but this is not of general interest to most people (other then
to know that "stuff that is cool is happening").
I know that people like to read our blogs for "end user" type useful tips
with rules - which is fine- but do you think its still a good idea to do
lots of "dev status" type blogs? I think its useful, as long as we tag it
accordingly (otherwise we can create a "sub blog" or something).
Thoughts? I just want to update it more regularly, helps keep me accountable
to the community ;) Reduces the chance of slacking ;)
Michael.
(PS sorry for the "double post" as I accidentally sent it to the codehaus
one).
16 years, 8 months
enhancing the editor
by Peter Lin
I've been working on the editor today and had some thoughts.
1. it would be nice to trigger the auto-completion when the user hits
"space" - I didn't see an easy way to do it in the editor class
2. it would be nice to further filter the auto-completion so it only shows
the phrases for that particular object. right now I show all matching
phrases. In order to be able to filter by object, the DSL format would need
to be extended to include more metadata. currently each entry looks like
this.
[when]- is empty=userId == userid, recommendation == null
In the example I created, userId is an attribute of many objects and is used
to to join between them. if we extend the format so that
object.attributemappings have extra metadata like below, the DSL
utility can then group the
mappings by object type.
[when][com.my.Reponse]- is empty=userId == userid, recommendation == null
The response object looks like this
public class Response {
private String userId;
private Recommendation recommendation;
}
the customer object
public class Customer {
private String first;
private String middle;
private String last;
private String userId;
private String emailAddress;
}
obviously, if the format is extended, it could cause backward compatability
issues.
peter
16 years, 8 months
Re: [rules-users] Re: [drools-dev] 3.2 M1
by Michael Neale
lol ! other then 3.0.x branch ?? ;)
Edson may know a branch to use, but in any case, Mark is beavering away on
MVEL integration which will be awesome (I think he wants MVEL for an M1
release).
On 1/12/07, Dirk Bergstrom <dirk(a)juniper.net> wrote:
>
> Michael Neale was heard to exclaim, On 01/02/07 05:28:
> > Guys, I am ok to do a M1 release of 3.2 whenever needed
>
> Any news on this? I've been running (in production now) on code I pulled
> from
> trunk a month or so ago, and it throws NPEs now and again. I'd really
> like to
> get something a bit more stable. Today's trunk "revision 8842" doesn't
> build,
> because the mvel code is Java 1.5.
>
> I'm kinda stuck here, and I'm hoping that someone can throw me a bone. If
> M1
> isn't coming soon, was there a particular revision number that was fairly
> stable
> that I can use?
>
> --
> Dirk Bergstrom dirk(a)juniper.net
> _____________________________________________
> Juniper Networks Inc., Computer Geek
> Tel: 408.745.3182 Fax: 408.745.8905
> _______________________________________________
> rules-users mailing list
> rules-users(a)lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/rules-users
>
16 years, 8 months
Declaration issues solved
by Edson Tirelli
Mark,
I solved the unbalanced trees problems we were talking about. I will
summarize the solutions bellow.
Unbalanced logical branches show up when we have a rule like this:
A and (B or (C and D) ) and E
In the above case, after transformation we will end up with 2 unbalanced
logical branches:
A and B and C
A and C and D and E
So, in the first branch will work with a tuple [A, B, E] and the second
will work with a tuple [A, C, D, E].
We end up with 2 problems:
a) first problem is that E column may have different attributes in each
logical branch (and in fact it has), so during the logic transformation
I'm now cloning columns when they are required in more than one branch.
b) second problem is that if someone binds the E element and uses it in
the consequence, we need to know that it has different positions in each
logical branch. The solution for this was to make each terminal node to
hold a reference to the subrule (logical branch) it belongs to. This
way, when activating the consequence, the knowledge helper will resolve
the variable using the subrule attached to the terminal node that caused
the activation.
I added test cases and it is all working fine.
[]s
Edson
--
Edson Tirelli
Software Engineer - JBoss Rules Core Developer
Office: +55 11 3124-6000
Mobile: +55 11 9218-4151
JBoss, a division of Red Hat @ www.jboss.com
16 years, 8 months
