What does SAP run on? Java?

Is there no way to add some kind of middleware like WebORB?

--- In [email protected], Nick Middleweek <n...@...> wrote:
>
> Ah nice, thanks Valdor...
> 
> We're talking to SAP using HTTP Services so no AMF present.
> 
> 
> 
> 
> 
> 
> 2010/1/7 valdhor <valdhorli...@...>
> 
> >
> >
> > I don't use XML at all any more - I use typed objects via AMF. This cuts
> > down the data traveling across the network significantly.
> >
> > I have multiple nested objects (I don't know whether nested is the right
> > word - I have objects that contain other objects) being returned but I have
> > not found it necessary to nest more than three levels so have not come
> > across your problem.
> >
> > Tracy is the XML/e4x expert. When he has time, I'm sure he'll chime in.
> >
> >
> > --- In [email protected] <flexcoders%40yahoogroups.com>, Nick
> > Middleweek <nick@> wrote:
> > >
> > > Hey Valdor... Thanks for the tip there.
> > >
> > > Have you ever had problems receiving 4th level nested data from an HTTP
> > > Service? We're trying to receive it as e4x or xml but it's coming back as
> > > untyped Objects. I'm guessing this is a bug in the Flex framework...
> > >
> > >
> > > Cheers,
> > > Nick
> > >
> > >
> > >
> > >
> > > 2010/1/6 valdhor <valdhorlists@>
> >
> > >
> > > >
> > > >
> > > > Nick
> > > >
> > > > Just for future reference...
> > > >
> > > >
> > > > var myContact : ContactPerson = new ContactPerson();
> > > > myContact = ContactPerson(myPerson);
> > > >
> > > > is a good example of how memory leaks occur. You create a new variable
> > of a
> > > > specific type and then allocate some memory for it with the new
> > operator.
> > > > The variable is just a pointer to a specific piece of memory. You then
> > > > immediately change the pointer to point at another piece of memory. The
> > > > original piece of memory that you allocated is now dangling with no way
> > to
> > > > access it and will need to wait on the garbage collector to reclaim it.
> > > >
> > > > The proper way to do what you are trying is:
> > > >
> > > > var myContact:ContactPerson = ContactPerson(myPerson);
> > > >
> > > >
> > > > --- In [email protected] 
> > > > <flexcoders%40yahoogroups.com><flexcoders%
> > 40yahoogroups.com>, Nick
> >
> > > > Middleweek <nick@> wrote:
> > > > >
> > > > > Hello...
> > > > >
> > > > > We've just run into a problem... Has anyone else come across this
> > before?
> > > > >
> > > > > We're making HTTP Service calls and we're getting back nested data.
> > We
> > > > have
> > > > > set the resultFormat="e4x" which we then parse into known Object
> > types,
> > > > such
> > > > > a IContactData, IInvoiceDetails...
> > > > >
> > > > > The Problem: With some service calls, the data returned has 4 or more
> > > > levels
> > > > > of nested data. In these cases, Flex isn't giving us XML. It is just
> > > > > returning an untyped Object with the nested data.
> > > > >
> > > > > If the returned data has 3 levels or less of nested data then we get
> > XML.
> > > > >
> > > > >
> > > > > We then thought, ok... The untyped Object returned by Flex does have
> > all
> > > > the
> > > > > properties required to Cast it to our typed Object, e.g.
> > IContactData...
> > > > >
> > > > > But we are getting a "Coercion failed" message by the Compiler.
> > Here's a
> > > > > basic example of the problem...
> > > > >
> > > > >
> > > > > var myPerson : Object = new Object();
> > > > > myPerson.age = "25";
> > > > > myPerson.sex = "dunno"
> > > > > myPerson.name = "Nick";
> > > > >
> > > > >
> > > > > var myContact : ContactPerson = new ContactPerson();
> > > > > myContact = ContactPerson(myPerson);
> > > > > // Where ContactPerson is a typed Object with age, sex and name
> > String
> > > > > properties.
> > > > >
> > > > >
> > > > >
> > > > > So has anyone managed to solve the 4 levels of nested data problem
> > from
> > > > an
> > > > > HTTP Service call?
> > > > >
> > > > > and :)
> > > > >
> > > > > Why can't we cast an untyped Object into a typed Object? :)
> > > > >
> > > > >
> > > > > Cheers guys...
> > > > >
> > > > > Nick
> > > > >
> > > >
> > >
> >
> >  
> >
>


Reply via email to