Hi,
The way I see it, this is the way your template should look like. I do not
see the need to use eval() at all.
rule "DC_0"
when
$service: Service(name == "@{service}", value == "@{value}")
$riskType: RiskType()
then
$riskType.setName($service.getType("@{value}"));
$riskType.setAtt(0,"@{value}");
end
I am assuming that RiskType is already inserted in WM, but you could just as
well create a new object in the then-part of the rule.
In stead of using the java-function to determine the name, you might
actually just as well create an other rule that sets the name according to
the value, thus putting the entire decision logic in rules.
Hope this helps.
Regards,
Frank
ChrisMu wrote:
Hi,
I'm trying to get a template (.drt) to generate 2 rules (.drl) from 2
datasets but by doing a little bit more work than direct substitution.
Data (service,value):
DC, AAA
DC, XYZ
Idea for the template - if value is of type1 then create Type object with
'Type1' as name. If value is of type2, use 'Type2' as name:
rule "DC_(a){row.rowNumber}"
when
Service(name == "DC")
then
Type.setName(eval(Service.getType(@{value})))
Type.setAtt(0,@{value})
end template
eval(Service.getType(@{arg0})) should take 'AAA' or 'XYZ' etc as arg and
return a string Type1/Type2 etc.
This would then generate 2 rules that look like:
rule "DC_0"
when
Service(name == "DC", value == "AAA")
then
RiskType.setName("Type1")
RiskType.setAtt(0,"AAA")
rule "DC_1"
when
Service(name == "DC", value == "XYZ")
then
RiskType.setName("Type2")
RiskType.setAtt(0,"XYZ")
...is this possible at all? I'm trying to avoid the eval step being done
in the actual rule (.drl) .
Thanks
Chris
--
View this message in context:
http://drools.46999.n3.nabble.com/Smart-Templates-tp3006479p3006589.html
Sent from the Drools: User forum mailing list archive at
Nabble.com.