8:06 a.m.
Hello list,
I'm using Drools.5.0.1 and I am trying to comprehend how to used the
"from" keyword over a named query.
I've googled this:
http://blog.athico.com/2007/06/chained-from-accumulate-collect.html
where it states:
p : Person( )
Restaurant( food == p.favouriteFood )
from hs.getNamedQuery( "list restaurants by postcode" )
.setProperties( [ "postcode" : p.address.zipcode ] )
.list()
If I understand the above correctly, the "hs.getNamedQuery" returns
"Restaurant" objects using values from "Person" object.
There are a couple of issues that I would like to resolve.
Where and how is the "hs" object in "hs.getNamedQuery" statement is
declared?
If the "Person" object itself is the result of a NamedQuery, could the
above statement be rewritten as:
p : Person( ) from ps.getNamedQuery("return list of persons")
Restaurant( food == p.favouriteFood )
from hs.getNamedQuery( "list restaurants by postcode" )
.setProperties( [ "postcode" : p.address.zipcode ] )
.list()
Thank you for your time.
-Stathis
PS: Please take note that this is a newbie (in Drools) speaking.
2:17 p.m.
Don't be too keen on using "from" in connection with queries. The simple
reason is that queries return nothing but what rules can do just as well.
Thus, getting all pairs of Person's and Restaurant's with matching
"food"
and "zipcode", the straightforward LHS would be
p:Person( $favFood : favouriteFood, $zip : address.zipcode )
r: Restaurant( food == $favFood, postcode == $zip )
"from" is required if the objects aren't facts but, for instance, elements
in a collection contained in a fact. So, if you have Restaurants only
as a List<Restaurant> in a (say) City object, you would write
p:Person( $favFood : favouriteFood, $zip : address.zipcode )
c: City( zipcodeSet contains $zip ) # a person's home town
r: Restaurant( food == $favFood, postcode == $zip ) from c.restaurantList
assuming that a (biggish) City typically has many zipcodes.
-W
On 28 July 2010 15:06, <rouvas(a)di.uoa.gr> wrote:
Hello list,
I'm using Drools.5.0.1 and I am trying to comprehend how to used the
"from" keyword over a named query.
I've googled this:
http://blog.athico.com/2007/06/chained-from-accumulate-collect.html
where it states:
p : Person( )
Restaurant( food == p.favouriteFood )
from hs.getNamedQuery( "list restaurants by postcode" )
.setProperties( [ "postcode" : p.address.zipcode ] )
.list()
If I understand the above correctly, the "hs.getNamedQuery" returns
"Restaurant" objects using values from "Person" object.
There are a couple of issues that I would like to resolve.
Where and how is the "hs" object in "hs.getNamedQuery" statement is
declared?
If the "Person" object itself is the result of a NamedQuery, could the
above statement be rewritten as:
p : Person( ) from ps.getNamedQuery("return list of persons")
Restaurant( food == p.favouriteFood )
from hs.getNamedQuery( "list restaurants by postcode" )
.setProperties( [ "postcode" : p.address.zipcode ] )
.list()
Thank you for your time.
-Stathis
PS: Please take note that this is a newbie (in Drools) speaking.
_______________________________________________
rules-users mailing list
rules-users(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users
2:45 p.m.
Wolfang,
Wolfgang Laun wrote:
Don't be too keen on using "from" in connection with
queries. The simple
reason is that queries return nothing but what rules can do just as well.
Thus, getting all pairs of Person's and Restaurant's with matching
"food"
and "zipcode", the straightforward LHS would be
p:Person( $favFood : favouriteFood, $zip : address.zipcode )
r: Restaurant( food == $favFood, postcode == $zip )
That was(is) my initial approach, and it is clearly the most
straightforward one.
Then I learned that the (theoretical) number of facts for "Person" objects
the client has, is about 65 billion with the potential of doubling every
year.
Of course, they are in a database and I was looking into using the "from"
keyword to do lookups in the database to retrieve the facts I need.
By the way, I have looked into "eval" and I am not sure how to use it in
respect to inline vs conditional element.
Regardless, my primary issue is how to reason over the "facts" as records
in a database.
Any ideas will be greatly appreciated.
-Stathis
"from" is required if the objects aren't facts but, for instance, elements
in a collection contained in a fact. So, if you have Restaurants only
as a List<Restaurant> in a (say) City object, you would write
p:Person( $favFood : favouriteFood, $zip : address.zipcode )
c: City( zipcodeSet contains $zip ) # a person's home town
r: Restaurant( food == $favFood, postcode == $zip ) from
c.restaurantList
assuming that a (biggish) City typically has many zipcodes.
-W
On 28 July 2010 15:06, <rouvas(a)di.uoa.gr> wrote:
> Hello list,
>
> I'm using Drools.5.0.1 and I am trying to comprehend how to used the
> "from" keyword over a named query.
>
> I've googled this:
> http://blog.athico.com/2007/06/chained-from-accumulate-collect.html
>
> where it states:
>
> p : Person( )
> Restaurant( food == p.favouriteFood )
> from hs.getNamedQuery( "list restaurants by postcode" )
> .setProperties( [ "postcode" : p.address.zipcode ] )
> .list()
>
> If I understand the above correctly, the "hs.getNamedQuery" returns
> "Restaurant" objects using values from "Person" object.
>
> There are a couple of issues that I would like to resolve.
>
> Where and how is the "hs" object in "hs.getNamedQuery" statement
is
> declared?
>
> If the "Person" object itself is the result of a NamedQuery, could the
> above statement be rewritten as:
>
> p : Person( ) from ps.getNamedQuery("return list of persons")
> Restaurant( food == p.favouriteFood )
> from hs.getNamedQuery( "list restaurants by postcode" )
> .setProperties( [ "postcode" : p.address.zipcode ] )
> .list()
>
> Thank you for your time.
> -Stathis
>
> PS: Please take note that this is a newbie (in Drools) speaking.
>
>
> _______________________________________________
> rules-users mailing list
> rules-users(a)lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/rules-users
>
5235
days inactive
5235
days old
2 comments
2 participants
participants (2)
-
rouvas＠di.uoa.gr -
Wolfgang Laun