3:23 p.m.
Hi,
I was wondering if calls to a Helper class are allowed in LHS of the rule?
Eg:
Rule myRule
When
MyObject(objField == Helper.transform("value"))
Then
// smth
End
Thanks,
Vlad
3:33 p.m.
The JBoss guys can correct me if I'm wrong, but although I think that works,
what you've just done is eliminate any chance to precompute matches and trim
down the checks that need to happen to find a rule match. Since the engine
can't know that Helper.transform("value") returns a constant value, it has
to re-run that every time and it has to reject for matches to the rule
constraint every time.
I'm sure this would really sink performance, in other words. You can easily
see the effect -- write the rule this way and time it, and then write it
with the return value of Helper.transform being inlined, and I'm sure
there's a big difference. It's the same as any other eval(...) situation in
Drools; possible, but not good for performance.
On 2/12/07, Olenin, Vladimir (MOH) <Vladimir.Olenin(a)moh.gov.on.ca> wrote:
Hi,
I was wondering if calls to a Helper class are allowed in LHS of the rule?
Eg:
Rule myRule
When
MyObject(objField == Helper.transform("value"))
Then
// smth
End
Thanks,
Vlad
_______________________________________________
rules-users mailing list
rules-users(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users
4:40 p.m.
I believe there is a caveat on functions and helper methods called from the
LHS that they need to be constant over time (or at least over the life of
the working memory) which then allows matches to be computed as per normal.
Steve
On 2/13/07, Michael Suzio <msuzio(a)gmail.com> wrote:
The JBoss guys can correct me if I'm wrong, but although I think that
works, what you've just done is eliminate any chance to precompute matches
and trim down the checks that need to happen to find a rule match. Since
the engine can't know that Helper.transform("value") returns a constant
value, it has to re-run that every time and it has to reject for matches to
the rule constraint every time.
I'm sure this would really sink performance, in other words. You can
easily see the effect -- write the rule this way and time it, and then write
it with the return value of Helper.transform being inlined, and I'm sure
there's a big difference. It's the same as any other eval(...) situation in
Drools; possible, but not good for performance.
On 2/12/07, Olenin, Vladimir (MOH) <Vladimir.Olenin(a)moh.gov.on.ca> wrote:
>
> Hi,
>
> I was wondering if calls to a Helper class are allowed in LHS of the
> rule?
> Eg:
>
> Rule myRule
> When
> MyObject(objField == Helper.transform("value"))
> Then
> // smth
> End
>
>
> Thanks,
>
> Vlad
> _______________________________________________
> rules-users mailing list
> rules-users(a)lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/rules-users
>
_______________________________________________
rules-users mailing list
rules-users(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users
--
Steven Williams
Supervising Consultant
Object Consulting
Office: 8615 4500 Mob: 0439 898 668 Fax: 8615 4501
stevenw(a)objectconsulting.com.au
www.objectconsulting.com.au
consulting | development | training | support
our experience makes the difference
5:52 p.m.
I see this in the docs:
"A Predicate constraint can use any valid Java expression as long as it
evaluated to a primitive boolean - avoid using any Drools keywords as
Declaration identifiers. Previously bound declarations can be used in the
expression. Functions used in a Predicate Constraint must return time
constant results."
and under the discussion of 'eval':
"Evals cannot be indexed and thus are not as optimal as using Field
Constraints."
and this in general discussion of constraints and accessors of your objects
used in them:
"Do please make sure that you are accessing methods that take no parameters,
and are in-fact "accessors" (as in, they don't change the state of the
object in a way that may effect the rules - remember that the rule engine
effectively caches the results of its matching inbetween invocations to make
it faster)."
So, if invoking a helper method is like an eval, I'm thinking Drools
computes the match, but then throws it away because it knows this is not a
constant expression. Whereas, if it were just a match of an accessor's
value vs. a constant String, it could cache that.
I'm wildly speculating, only because I want to understand fully what is
happening, so I'm throwing stuff out there so the Drools team will have to
correct my misapprehensions before I confuse the whole list (*g*).
Sometimes reading the Drools documentation is like a Talmudic scholar trying
to find the deeper meaning of the sacred words... as much as I want
the 3.2milestone, I want real documentation even more (and *hint hint*
-- would
probably be willing to pay for it, guys! Can we get an O'Reilly book?)
--- Michael
On 2/12/07, Steven Williams <stevenw(a)objectconsulting.com.au> wrote:
I believe there is a caveat on functions and helper methods called from
the LHS that they need to be constant over time (or at least over the life
of the working memory) which then allows matches to be computed as per
normal.
Steve
On 2/13/07, Michael Suzio <msuzio(a)gmail.com> wrote:
>
> The JBoss guys can correct me if I'm wrong, but although I think that
> works, what you've just done is eliminate any chance to precompute matches
> and trim down the checks that need to happen to find a rule match. Since
> the engine can't know that Helper.transform("value") returns a
constant
> value, it has to re-run that every time and it has to reject for matches to
> the rule constraint every time.
>
> I'm sure this would really sink performance, in other words. You can
> easily see the effect -- write the rule this way and time it, and then write
> it with the return value of Helper.transform being inlined, and I'm sure
> there's a big difference. It's the same as any other eval(...) situation
in
> Drools; possible, but not good for performance.
>
> On 2/12/07, Olenin, Vladimir (MOH) <Vladimir.Olenin(a)moh.gov.on.ca>
> wrote:
> >
> > Hi,
> >
> > I was wondering if calls to a Helper class are allowed in LHS of the
> > rule?
> > Eg:
> >
> > Rule myRule
> > When
> > MyObject(objField == Helper.transform("value"))
> > Then
> > // smth
> > End
> >
> >
> > Thanks,
> >
> > Vlad
> > _______________________________________________
> > rules-users mailing list
> > rules-users(a)lists.jboss.org
> > https://lists.jboss.org/mailman/listinfo/rules-users
> >
>
>
> _______________________________________________
> rules-users mailing list
> rules-users(a)lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/rules-users
>
>
--
Steven Williams
Supervising Consultant
Object Consulting
Office: 8615 4500 Mob: 0439 898 668 Fax: 8615 4501
stevenw(a)objectconsulting.com.au
www.objectconsulting.com.au
consulting | development | training | support
our experience makes the difference
_______________________________________________
rules-users mailing list
rules-users(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users
6:16 p.m.
We were hoping the community would work with us and improve the
documentation over the last year - to date we have had zip :( We will
have another documentation drive once we get to the candidate release
stage for the current development cycle. We simply don't have the cycles
to work on a book, with any luck after we spend 3 to 4 weeks on
documentation during hte candidate release stage it should get better,
and maybe one day the manual can form the basis of a book. Some ideas
for the community to work include the examples in drools-examples, they
can even use what I did for drools 2.x as a starting basis, or to
provide a really good tutorial on the basics of pattern matching, you
can look the public domain clips manual for pointers here. It's not
money we need, its time, YOUR time :)
If you want to use a helper class it must be in a return value,
predicate or eval - none of which are currently indexed. return value
and predicate must also be time constant, eval does not have to be.
Mark
Michael Suzio wrote:
I see this in the docs:
"A Predicate constraint can use any valid Java expression as long as
it evaluated to a primitive boolean - avoid using any Drools keywords
as Declaration identifiers. Previously bound declarations can be used
in the expression. Functions used in a Predicate Constraint must
return time constant results."
and under the discussion of 'eval':
"Evals cannot be indexed and thus are not as optimal as using Field
Constraints."
and this in general discussion of constraints and accessors of your
objects used in them:
"Do please make sure that you are accessing methods that take no
parameters, and are in-fact "accessors" (as in, they don't change the
state of the object in a way that may effect the rules - remember that
the rule engine effectively caches the results of its matching
inbetween invocations to make it faster)."
So, if invoking a helper method is like an eval, I'm thinking Drools
computes the match, but then throws it away because it knows this is
not a constant expression. Whereas, if it were just a match of an
accessor's value vs. a constant String, it could cache that.
I'm wildly speculating, only because I want to understand fully what
is happening, so I'm throwing stuff out there so the Drools team will
have to correct my misapprehensions before I confuse the whole list
(*g*). Sometimes reading the Drools documentation is like a Talmudic
scholar trying to find the deeper meaning of the sacred words... as
much as I want the 3.2 milestone, I want real documentation even more
(and *hint hint* -- would probably be willing to pay for it, guys!
Can we get an O'Reilly book?)
--- Michael
On 2/12/07, *Steven Williams* <stevenw(a)objectconsulting.com.au
<mailto:stevenw@objectconsulting.com.au>> wrote:
I believe there is a caveat on functions and helper methods called
from the LHS that they need to be constant over time (or at least
over the life of the working memory) which then allows matches to
be computed as per normal.
Steve
On 2/13/07, *Michael Suzio* < msuzio(a)gmail.com
<mailto:msuzio@gmail.com>> wrote:
The JBoss guys can correct me if I'm wrong, but although I
think that works, what you've just done is eliminate any
chance to precompute matches and trim down the checks that
need to happen to find a rule match. Since the engine can't
know that Helper.transform("value") returns a constant value,
it has to re-run that every time and it has to reject for
matches to the rule constraint every time.
I'm sure this would really sink performance, in other words.
You can easily see the effect -- write the rule this way and
time it, and then write it with the return value of
Helper.transform being inlined, and I'm sure there's a big
difference. It's the same as any other eval(...) situation in
Drools; possible, but not good for performance.
On 2/12/07, *Olenin, Vladimir (MOH)*
<Vladimir.Olenin(a)moh.gov.on.ca
<mailto:Vladimir.Olenin@moh.gov.on.ca>> wrote:
Hi,
I was wondering if calls to a Helper class are allowed in
LHS of the rule?
Eg:
Rule myRule
When
MyObject(objField == Helper.transform("value"))
Then
// smth
End
Thanks,
Vlad
_______________________________________________
rules-users mailing list
rules-users(a)lists.jboss.org
<mailto:rules-users@lists.jboss.org>
https://lists.jboss.org/mailman/listinfo/rules-users
_______________________________________________
rules-users mailing list
rules-users(a)lists.jboss.org <mailto:rules-users@lists.jboss.org>
https://lists.jboss.org/mailman/listinfo/rules-users
--
Steven Williams
Supervising Consultant
Object Consulting
Office: 8615 4500 Mob: 0439 898 668 Fax: 8615 4501
stevenw(a)objectconsulting.com.au
<mailto:stevenw@objectconsulting.com.au>
www.objectconsulting.com.au <http://www.objectconsulting.com.au>
consulting | development | training | support
our experience makes the difference
_______________________________________________
rules-users mailing list
rules-users(a)lists.jboss.org <mailto:rules-users@lists.jboss.org>
https://lists.jboss.org/mailman/listinfo/rules-users
------------------------------------------------------------------------
_______________________________________________
rules-users mailing list
rules-users(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users
8:20 p.m.
If you want to use a helper class it must be in a return value,
predicate
or eval - none of which are currently indexed. return value and predicate
must also be time constant, eval does not have to be.
OK, I fully understand what it means to use something in an eval block --
the scoping/use of that is clear, and that section of the Drools docs seems
pretty specific. What does it mean to use a helper class in a predicate or
a return value? Any sort of example of the two probably closes the
knowledge gap here and lets me understand the specifics of the use cases.
What I do get (and thank you for being very clear on this, it helps) is that
none of these cases is performant. It's very clear that if a factual case
can be indexed, it can be optimized. If it cannot, then there is a cost
every time the condition must be evaluated. Of course, now I'm wondering...
when does *that* happen? Is the "cost" a product of # of facts x number of
computed constraints? And then pre-computed/indexed constraints are some
(small) fixed cost on top of that?
Oh... and when does this "cost" occur? Every time I call fireAllRules()?
Or does indexing persist across calls to that? This might be important,
because my app right now calls fireAllRules() after pretty much every
insertion of a new fact (the ratio of global facts applicable to all cases
vs facts for a specific case is about 1 : 1000 -- I probably call
fireAllRules at least 1000 times per instantiation of a rules engine -- is
that bad?).
OK, sorry. Lots of question packed into one email. But if you educate me, I
promise to only use that knowledge for good and to help others on this list!
--- Michael
6309
days inactive
6310
days old
5 comments
4 participants
participants (4)
-
Mark Proctor -
Michael Suzio -
Olenin, Vladimir (MOH) -
Steven Williams