How to get a rule to fail on blank responsibilities?

Michael Pokraka workflow at quirky.me.uk
Fri Jan 28 09:08:57 EST 2005


Hi Richard, 
I feel like I'm deliberately trying to be difficult here, but I promise I'm not
:-)

The rule is used in an application -  a QM notification uses rules to determine
the various agents when it is created. So workflow doesn't feature at this
point. 

As to filling in values... I suggested it, but they don't want to have to do
that for umpteen responsibilities and remember not to leave blanks for ever
more. I also kinda find it difficult to argue the requirement of 'if it isn't
explicitly specified by a responsibility, leave it blank'.
Oh, and there are a heck of a lot of elements to fill in ranges for - we've
already created our own 'slimmed down' copy of the SAP standard rule because it
has well over 200 container elements! (rules 03100024/6/7).

For now I've done the terrible deed and checked for blanks in a FM-based rule
which calls RH_GET_ACTORS for the responsibility rule. But still I remain
curious if there's a neater way for something seemingly obvious.

Thanks for your feedback!
Cheers
Mike

--- Richard Marut <rvmarut at earthlink.net> wrote:

> Mike,
> 
> What I meant to say was, check the data in a preceding condition step. If
> blank, send something (mail or work item) to the workflow initiator or an
> administrator.
> 
> By the way, have you tried not using an *? How about using a range for Plant
> such as 0001 - 9999 and DocType of 01 - ZZ which would eliminate a blank as
> being a valid piece of data? I didn't try this when I ran into the same
> issue because the users wanted the work item to go to everyone when the data
> was blank.
> 
> Richard...
> 
> -----Original Message-----
> From: sap-wug-bounces at mit.edu [mailto:sap-wug-bounces at mit.edu] On Behalf Of
> Michael Pokraka
> Sent: Friday, January 28, 2005 4:51 AM
> To: SAP Workflow Users' Group
> Subject: RE: How to get a rule to fail on blank responsibilities?
> 
> Hi Richard, 
> I suppose I should have added that I have little control over the data. To
> make
> life interesting the rule also gets used outside of WF, it's used to
> populate a
> field in a document.
> 
> The biggest issue is that if the abuser doesn't fill in all the info, it
> just
> picks the first agent. Workflow is supposed to go and get someone to fill in
> an
> agent if it's blank, but so far the rule will always find an agent. 
> 
> I'm just desperately trying to avoid coding a FM-based rule which checks the
> data and then calls the real rule... it's just way too big a hammer for
> cracking a peanut. 
> 
> Thanks for the input,
> Cheers
> Mike
> 
> --- Richard Marut <rvmarut at earthlink.net> wrote:
> 
> > Mike,
> > 
> > How about checking the data in a previous step before the activity step
> with
> > the responsibility role/rule? You can then take a different path or action
> > when the data is blank.
> > 
> > Richard...
> > 
> > -----Original Message-----
> > From: sap-wug-bounces at mit.edu [mailto:sap-wug-bounces at mit.edu] On Behalf
> Of
> > Michael Pokraka
> > Sent: Thursday, January 27, 2005 12:52 PM
> > To: SAP Workflow User Group
> > Subject: How to get a rule to fail on blank responsibilities?
> > 
> > G'day all, 
> > Passing a blank value to a responsibility will always result in a positive
> > test. Can't get them to fail on blank data. Unless of course I'm having a
> > daft
> > day (quite possible, it's been a long one!)? 
> > 
> > Now, the tricky bit is that I have many values to test against, but lets
> > keep
> > it simple: 
> > 
> > Resp1  Plant = 1000, DocType = *,  Agent 1
> > Resp2  Plant = *,    DocType = ZZ, Agent 2
> > 
> > If I test this without entering any data, it will return all agents, even
> > with
> > 'Terminate if no resolution' switched on. It somehow feels like a stupid
> > question... but how to get around that? 
> > 
> > In short I want only those with MATCHING data to return agents. I.e. in
> the
> > example above, it I test it with Plant 9999, it will return Agent 2 (since
> > the
> > doctype is blank). For my purposes, this should also fail (but isn't as
> > important as both blanks above). 
> > 
> > Any obvious answer I might have missed? 
> > Cheers
> > Mike
> > 
> > _______________________________________________
> > SAP-WUG mailing list
> > SAP-WUG at mit.edu
> > http://mailman.mit.edu/mailman/listinfo/sap-wug
> > 
> > 
> > _______________________________________________
> > SAP-WUG mailing list
> > SAP-WUG at mit.edu
> > http://mailman.mit.edu/mailman/listinfo/sap-wug
> > 
> > 
> 
> _______________________________________________
> SAP-WUG mailing list
> SAP-WUG at mit.edu
> http://mailman.mit.edu/mailman/listinfo/sap-wug
> 
> 
> _______________________________________________
> SAP-WUG mailing list
> SAP-WUG at mit.edu
> http://mailman.mit.edu/mailman/listinfo/sap-wug
> 
> 



More information about the SAP-WUG mailing list