[rules-users] 答复: Many to many matching using Drools rules

[Wolfgang Laun]

I could not propose rules for fitting the Reps to the Accs because
I don't know how it must be done, and which values one has to expect,
and how they could be related. Drools is not going to abolish the
need for precise specifications...

Unless you can't write it down in English or whatever, you can't
program it in Drools...

-W





On 14/12/2011, Zhuo Li <milanello1998 at gmail.com> wrote:
> Yep, this was what I was thinking but in that case I rely more on the
> pairing algorithm I gave - might be in java function - rather than drools'
> rules and advantage of RETE. There are many algorithms can resolve my
> problem and I can implement them in java, but that has nothing to do with
> drools or rules. I was wondering if Drools provide any natural way to
> resolve this problem within its own rule but seems no luck so far.
>
> Best
> Abe
>
> -----邮件原件-----
> 发件人: rules-users-bounces at lists.jboss.org
> [mailto:rules-users-bounces at lists.jboss.org] 代表 Wolfgang Laun
> 发送时间: 2011年12月14日 15:02
> 收件人: Rules Users List
> 主题: Re: [rules-users] Many to many matching using Drools rules
>
> You can collect objects of both classes with matching id and date.
> Each of these collected objects has the original id and date a list of
> amount figures.
>
> Pairing the collected object from repository and accounting is simple.
> Then, you'll have to implement an algorith that selects the items from
> both sides in an acceptable way. I presume that one or more items from
> one side must exactly be equal to one or more items from the other
> side, which may not always be possible even though there would be
> "coverage" of an item of one side by items on the other side (e.g.
> 7,11 vs.3,3,3,3).
>
> But assuming that you detect matching sums, you retract the original
> facts. In any case, the collected objects must be retracted, too, at
> the end of the processing.
>
> (Some details depend on the way the original facts are inserted.)
>
> HTH
> -W
>
>
>
>
>
>
> On 14/12/2011, Zhuo LI <milanello1998 at gmail.com> wrote:
>> i want to find txns from both sides whose portfolio asset and sum of amt
>> are the same.
>>
>> 在 2011-12-14 上午9:18，"Salaboy" <salaboy at gmail.com>写道：
>>
>>> Do you want to compare to set of transactions to find which of them are
> in
>>> both sets and which of them are not?
>>> Is that what are you trying to achieve?
>>> Cheers
>>>
>>> - CTO @ http://www.plugtree.com
>>> - MyJourney @ http://salaboy.wordpress.com
>>> - Co-Founder @ http://www.jbug.com.ar
>>> - Mauricio "Salaboy" Salatino -
>>>
>>> On 13/12/2011, at 10:27, Wolfgang Laun <wolfgang.laun at gmail.com> wrote:
>>>
>>> > Huh?
>>> > -W
>>> >
>>> > 2011/12/13 Zhuo Li <milanello1998 at gmail.com>:
>>> >> Folks,
>>> >>
>>> >>
>>> >>
>>> >> I am recently working on a BI project which is doing reconciliation
>>> between
>>> >> repository and accounting. I'm using Drools 5.1.1 to maintain all
> these
>>> >> matching rules. One scenario I have no clue to implement in Drools is
>>> that I
>>> >> need to find x transactions from repository, to match y transactions
>>> from
>>> >> accounting, while existing repository transactions and accounting
>>> >> transactions are bigger than x and y. Does anybody have experience to
>>> >> do
>>> >> similar thing in Drools?
>>> >>
>>> >>
>>> >>
>>> >> Best Abe
>>> >>
>>> >>
>>> >> _______________________________________________
>>> >> rules-users mailing list
>>> >> rules-users at lists.jboss.org
>>> >> https://lists.jboss.org/mailman/listinfo/rules-users
>>> >>
>>> > _______________________________________________
>>> > rules-users mailing list
>>> > rules-users at lists.jboss.org
>>> > https://lists.jboss.org/mailman/listinfo/rules-users
>>>
>>> _______________________________________________
>>> rules-users mailing list
>>> rules-users at lists.jboss.org
>>> https://lists.jboss.org/mailman/listinfo/rules-users
>>>
>>
>
> _______________________________________________
> rules-users mailing list
> rules-users at lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/rules-users
>
>
> _______________________________________________
> rules-users mailing list
> rules-users at lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/rules-users
>