[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: using WSDL2PRG
Larry,
tns:NotifyPhoneBasic uses messages tns:NotifyPhoneBasicSoapIn and
tns:NotifyPhoneBasicSoapOut.
The parameter of tns:NotifyPhoneBasicSoapIn is tns:NotifyPhoneBasic which
is a complex element made of several string elements. Each of these sub
elements are defined with maxOccurs set to 1. Hence these elements are no
arrays and no DIM_A* constant is generated.
The parameter of tns:NotifyPhoneBasicSoapOut is
tns:NotifyPhoneBasicResponse. That element contains sub element
NotifyPhoneBasicResult of type tns:NotifyReturn which contains sub element
VariableArray. VariableArray is of type tns:ArrayOfVariable. That is the
point where we hit the array! tns:ArrayOfVariable contains element Variable
of type tns:Variable and is defined with maxOccurs "unbounded"!
tns:Variable is just a key/value pair and does specify "unbounded".
So the only array used in NotifyPhoneBasic is Variable, used by
ArrayOfVariable, used by VariableArray, etc.
That is why there is only one DIM_A* namely DIM_A1.
When I generate and compile the program with the default settings, the size
of tns_notifyPhoneBasicResponse_t is 34040 byte. After having changed
DIM_A1 from 128 to 32 the new size of tns_notifyPhoneBasicResponse_t
dropped to 9080. It is up to you to decide which string size and number of
elements to use.
The _t variables are used as type definitions. All real variables are
declared using 'like' or 'likeds' and a _t type definition as the reference
field. You find all _t type definitions in the generated stub module. E.g:
*
D tns_NotifyPhoneBasicResponse_t...
D DS based(pDummy)
D qualified
D NotifyPhoneBasicResult...
D likeds(tns_NotifyReturnRnmd_t)
*
D tns_NotifyReturnRnmd_t...
D DS based(pDummy)
D qualified
D CallAnswered N
D CallComplete N
D Country 128A varying
D Demo N
D DigitsPressed 128A varying
D Duration 10I 0
D EndTime Z
D MachineDetection...
D 128A varying
D MinuteRate 31S15
D QueueID 20I 0
D ResponseCode 10I 0
D ResponseText 128A varying
D StartTime Z
D TextToSay 128A varying
D TryCount 10I 0
D VariableArray likeds(tns_ArrayOfVariable_t)
Please notice that 'NotifyPhoneBasicResult' if defined as
'tns_NotifyReturnRnmd_t'.
In your programs you have to include the generated stub module as a copy
book and set a compiler condition prior to that. Let us assume that the
name of the stub module is PHONE. The your program have to include PHONE as
shown below to get the prototype and type definitions:
/define prototype_phone
/copy PHONE
Hope that helps.
Thomas.
ftpapi-bounces@xxxxxxxxxxxxxxxxxxxxxx schrieb am 16.03.2009 15:03:06:
>
> Thomas - I appreciate the help so far, and apologize for my ignorance,
but I
> am having a hard time understanding what you suggest. (All of this is
very
> new to me.) I only see one DIM_A in the generated code - it is DIM_A1
for
> 128 bytes - not one for each array. Also, you suggest that I try to
change
> the _t type definition for that one variable, but I don't know where to
find
> it or what to change.
>
> If this is of any help to you, the URL of the WSDL that I am using is
> http://ws.cdyne.com/NotifyWS/PhoneNotify.asmx?wsdl
>
> the operation that I am using is
> tns:NotifyPhoneBasic()
>
> and the variable that I want to be really big is
> TextToSay
>
>
>
> Larry Kleinman
> Kleinman Associates, Inc.
> 212-949-6469
> 203-255-4100
>
>
>
> thomas.raddatz@gf
> d.de
> Sent by:
To
> ftpapi-bounces@li ftpapi@xxxxxxxxxxxxxxxxxxxxxx
> sts.scottklement.
cc
> com
>
Subject
> Re: using WSDL2PRG
> 03/16/2009 07:06
> AM
>
>
> Please respond to
> HTTPAPI and
> FTPAPI Projects
> <ftpapi@xxxxxxxxx
> ttklement.com>
>
>
>
>
>
>
>
>
> The DIM parameter specifies the default array dimension of array elements
> that have "maxOccurs" set to "unbounded". Sample:
>
> <xs:element name="person" maxOccurs="unbounded">
>
> WSDL2RPG generates a DIM_A* constant for each array it encounters. You
can
> change that constant to any value you like except for 0. The problem is
to
> find the right balance between the length of strings and the number of
> array elements. Usually programmers want both values sets to "unlimited"
> which unfortunately is not possible in RPG. With V6 we are close to
> "unlimited" because IBM increased the length of strings to 16MB. On the
> other hand it may not be a good idea to declare each and every string as
> 16MB, isn't it?
>
> Thomas.
>
>
> >
> > Thomas - what does the DIM parm do? Can i make that smaller to avoid
> > hitting the 64K limit?
> >
> > Larry Kleinman
> > Kleinman Associates, Inc.
> > 212-949-6469
> > 203-255-4100
> >
> >
> >
>
> > Thomas Raddatz
>
> > <thomas.raddatz@t
>
> > ools400.de>
> To
> > Sent by: HTTPAPI and FTPAPI Projects
>
> > ftpapi-bounces@li <ftpapi@xxxxxxxxxxxxxxxxxxxxxx>
>
> > sts.scottklement.
> cc
> > com
>
> >
> Subject
> > Re: using WSDL2PRG
>
> > 03/14/2009 06:30
>
> > AM
>
> >
>
> >
>
> > Please respond to
>
> > HTTPAPI and
>
> > FTPAPI Projects
>
> > <ftpapi@xxxxxxxxx
>
> > ttklement.com>
>
> >
>
> >
>
> >
> >
> >
> >
> > Scott,
> >
> > You are absolutely right. There is nothing I could add except for that
> you
> > can change the default
> > length for strings at the WSDL2RPG command prompt. Press F9 to see the
> > additional parameters STRLEN
> > and DIM. If you do that that new default length is assigned to all
> strings
> > and you may hit the RPG
> > barrier of 64k. Due to the limitations of RPG the result data structure
> > generated by WSDL2RPG can not
> > exceed 64k.
> >
> > The other option you have is to change the *_t type definition only for
> that
> > specific variable. If
> > that also does not work you have to have to get into the generated code
> and
> > replace the variable by a
> > pointer and manually allocate storage and copy the value to that
pointer.
> > Larry, let me know if you
> > hit that problem and I will try to figure out how to do that.
> >
> > Thomas.
> >
> >
> > Scott Klement schrieb:
> > > My understanding (and it may be wrong) is that WSDL documents don't
> > > usually tell you the size of the string. They simply say "data type
is
> > > String".
> > >
> > > In many languages, that's a non-issue. Java, PHP, C#, etc all let
you
> > > declare a string, and don't ask you for a size. they dynamically
> update
> > > the size as needed. Thus, the standards being the way they are.
> > >
> > > But, when Thomas created WSDL2RPG, he had to pick a number, since RPG
> > > requires it. He picked 128, figuring it would be large enough for
most
> > > of the strings used in web services, without being too big (and thus
> > > risking hitting RPG's memory limits). He figured you could easily go
> in
> > > and change that 128 to something larger if you needed to.
> > >
> > > At least, that's my understanding. I probably shouldn't speak for
him,
> > > but I think he told me that at one point :)
> > >
> > >
> > >
> > > Larry Kleinman wrote:
> > >> Hi all - I am just getting started with WSDL2RPG and my first few
> > >> simple attempts seem to work OK. I have now run into a problem -
> one
> > >> of the fields that the web service expects can be very long (a
few
> > >> thousand characters), but the "stub" created by WSDL2RPG is
defined
> as
> > >> 128A varying. Can I just change this to a bigger number in the 2
> > >> places it is defined, or is there more to it? (Where did the 128
> come
> > >> from in the first place?)
> > >> Larry Kleinman
> > >> Kleinman Associates, Inc.
> > >> 212-949-6469
> > >> 203-255-4100
> > >>
> > >>
> > >>
> > >>
> ------------------------------------------------------------------------
> > >>
> > >>
> -----------------------------------------------------------------------
> > >> This is the FTPAPI mailing list. To unsubscribe, please go to:
> > >> http://www.scottklement.com/mailman/listinfo/ftpapi
> > >>
> -----------------------------------------------------------------------
> > >
> > >
-----------------------------------------------------------------------
> > > This is the FTPAPI mailing list. To unsubscribe, please go to:
> > > http://www.scottklement.com/mailman/listinfo/ftpapi
> > >
-----------------------------------------------------------------------
> > >
> > (See attached file: smime.p7s)
> > -----------------------------------------------------------------------
> > This is the FTPAPI mailing list. To unsubscribe, please go to:
> > http://www.scottklement.com/mailman/listinfo/ftpapi
> > -----------------------------------------------------------------------
> > [Bild entfernt] [Bild entfernt] [Anhang "smime.p7s" gelöscht von
> > Thomas Raddatz/GfD/DE] [Bild entfernt]
> > -----------------------------------------------------------------------
> > This is the FTPAPI mailing list. To unsubscribe, please go to:
> > http://www.scottklement.com/mailman/listinfo/ftpapi
> > -----------------------------------------------------------------------
>
>
> --
> IMPORTANT NOTICE:
> This email is confidential, may be legally privileged, and is for the
> intended recipient only. Access, disclosure, copying, distribution, or
> reliance on any of it by anyone else is prohibited and may be a criminal
> offence. Please delete if obtained in error and email confirmation to the
> sender.
> -----------------------------------------------------------------------
> This is the FTPAPI mailing list. To unsubscribe, please go to:
> http://www.scottklement.com/mailman/listinfo/ftpapi
> -----------------------------------------------------------------------
> [Bild entfernt] [Bild entfernt] [Bild entfernt]
> -----------------------------------------------------------------------
> This is the FTPAPI mailing list. To unsubscribe, please go to:
> http://www.scottklement.com/mailman/listinfo/ftpapi
> -----------------------------------------------------------------------
--
IMPORTANT NOTICE:
This email is confidential, may be legally privileged, and is for the
intended recipient only. Access, disclosure, copying, distribution, or
reliance on any of it by anyone else is prohibited and may be a criminal
offence. Please delete if obtained in error and email confirmation to the sender.
-----------------------------------------------------------------------
This is the FTPAPI mailing list. To unsubscribe, please go to:
http://www.scottklement.com/mailman/listinfo/ftpapi
-----------------------------------------------------------------------