Perhaps "high fructose corn syrup" would be more correct (ie a yuk tasting sweetener that americans like to use)
eg:
Person(age < parentsAge + 1 - legalMinimum)
where all the above are fields of Person. We could compile that out to a predicate or an eval, but there may be other issues (parser ambiguities).
On 11/12/06,
Mark Proctor <[EMAIL PROTECTED]> wrote:
I don't know if we would ever do such a thing, to get something like
that working with Rete is really messy.
Mark
Geoffrey De Smet wrote:
> In that case, mind if I open a long-term issue called
> "Connective constraints between fields"?
>
> For now I 've just implemented all 4 combinations (2^2 = 4)
> as a separate rule and it works :)
>
>
> With kind regards,
> Geoffrey De Smet
>
>
> Edson Tirelli wrote:
>>
>> Geoffrey,
>>
>> Mark already provided the samples link, but it is important to note
>> that connective constraints allow multiple _expression_ per field, not
>> between fields.
>>
>> []s
>> Edson
>>
>> Geoffrey De Smet wrote:
>>
>>> I opened the first issue (bug) with testcase patch here:
>>> http://jira.jboss.com/jira/browse/JBRULES-554
>>>
>>> As for the second issue (feature request), it looks like "connective
>>> constraints" is already done on the trunk, so I 'll not open it.
>>> Are there any rule/query examples of such "connective constraints"?
>>>
>>> With kind regards,
>>> Geoffrey De Smet
>>>
>>>
>>> Michael Neale wrote:
>>>
>>>> In 3.2 (trunk) we have added what are known as "connective
>>>> constraints" which will do what you want (yes they are different
>>>> from conditional elements in behaviour).
>>>>
>>>> But not yet, in 3.0 can you do that.
>>>>
>>>> On 11/10/06, *John Cocktolstoy* <[EMAIL PROTECTED]
>>>> <mailto:[EMAIL PROTECTED]>> wrote:
>>>>
>>>> Hi guys,
>>>> I got the same errors but thought it is how it should work - if
>>>> you
>>>> bind different names in each column there is no error. That was
>>>> the
>>>> reason why I were convinced that JBossRules don't support nested
>>>> "and"s and "or"s :-). Is this bug registered on JIRA yet? Any
>>>> chances this bug will be fixed soon?
>>>>
>>>> regards
>>>> John
>>>>
>>>>
>>>> On 11/10/06, *Geoffrey De Smet* < [EMAIL PROTECTED]
>>>> <mailto: [EMAIL PROTECTED]>> wrote:
>>>>
>>>>
>>>>
>>>> Edson Tirelli wrote, On 2006-11-10 12:37 AM:
>>>> >
>>>> > Hey Geoffrey,
>>>> >
>>>> > You guys are doing a great job testing things. Thanks.
>>>> >
>>>> > The correct syntax would be like that:
>>>> >
>>>> > query "multipleMatchesPerTeamPerDay"
>>>> > $team : Team()
>>>> > Match( $id : id, homeTeam == $team, $day : day ) ||
>>>> > Match( $id : id, awayTeam == $team, $day : day )
>>>> > exists Match( id > $id, homeTeam == $team, day == $day
>>>> ) ||
>>>> > exists Match(id > $id, awayTeam == $team, day == $day )
>>>> > end
>>>> >
>>>> > Although, seems there is a regression as it is raising
>>>> an error
>>>> > saying variables are being declared twice. Although, when
>>>> using "||"
>>>> > (or), variables MUST be bound in each logical outcome.
>>>> >
>>>> > Do you mind opening a JIRA for that?
>>>>
>>>> I will and I 'll add a testcase patch us usual :)
>>>>
>>>> Still I believe something like
>>>> Match($id : id, homeTeam == $team || awayTeam == $team,
>>>> $day
>>>> : day)
>>>> should be possible in the long term, because:
>>>> - it avoids duplicating the id and day constraints, which
>>>> makes
>>>> the rule
>>>> a lot easier to read
>>>> - drools could be smart enough to compile that in a way to
>>>> reuse
>>>> the
>>>> beta nodes of id and day maybe? - if it doesn't already
>>>> Do you mind if I open a second (long-term feature request)
>>>> JIRA
>>>> for that
>>>> too?
>>>>
>>>> >
>>>> > Thank you,
>>>> >
>>>> > Regards,
>>>> > Edson
>>>> >
>>>> > Geoffrey De Smet wrote:
>>>> >
>>>> > > I got a simple domain model:
>>>> > >
>>>> > > public class Match {
>>>> > >
>>>> > > private Long id;
>>>> > >
>>>> > > private Team homeTeam;
>>>> > > private Team awayTeam;
>>>> > >
>>>> > > private Day day;
>>>> > >
>>>> > > // getter, setters, ...
>>>> > > }
>>>> > >
>>>> > > And a rule (query) that checks if a team has 2 matches
>>>> on the
>>>> same day
>>>> > > (no matter if they are the home or the away team):
>>>> > >
>>>> > > query "multipleMatchesPerTeamPerDay"
>>>> > > $team : Team();
>>>> > > Match($id : id, homeTeam == $team || awayTeam ==
>>>> $team,
>>>> $day : day);
>>>> > > exists Match(id > $id, homeTeam == $team ||
>>>> awayTeam ==
>>>> $team, day
>>>> > > == $day);
>>>> > > end
>>>> > >
>>>> > > But apparently I can't do this?
>>>> > >
>>>> > > unknown:12:50 Unexpected token '=='
>>>> > > unknown:13:39 Unexpected token '$team'
>>>> > > unknown:13:57 Unexpected token '=='
>>>> > > unknown:18:33 Unexpected token 'day'
>>>> > >
>>>> > > Line 12 is: Match($id : id, homeTeam == $team ||
>>>> awayTeam ==
>>>> $team,
>>>> > > $day : day);
>>>> > >
>>>> > > Is there any way to avoid having to write a separate rule
>>>> (query) for
>>>> > > every combination (home-home, away-home, home-away,
>>>> away-away)?
>>>> > >
>>>> > > Thanks for any help/advice.
>>>> > >
>>>> >
>>>> >
>>>>
>>>> --
>>>> With kind regards,
>>>> Geoffrey De Smet
>>>>
>>>>
>>>>
>>>> ---------------------------------------------------------------------
>>>> To unsubscribe from this list please visit:
>>>>
>>>> http://xircles.codehaus.org/manage_email
>>>>
>>>>
>>>>
>>>
>>>
>>> ---------------------------------------------------------------------
>>> To unsubscribe from this list please visit:
>>>
>>> http://xircles.codehaus.org/manage_email
>>>
>>>
>>
>>
>
>
> ---------------------------------------------------------------------
> To unsubscribe from this list please visit:
>
> http://xircles.codehaus.org/manage_email
>
> --
> This email has been verified as Virus free
> Virus Protection and more available at http://www.plus.net
>
---------------------------------------------------------------------
To unsubscribe from this list please visit:
http://xircles.codehaus.org/manage_email
