02-08-04 01:42 PM
Yes,
I think you understood the question, but let me give you
some more details.
I used the Editor to create a document spec for a flat
file and then generated an instance in both the XML
version and the native version. I then created a map
using this specification as the source and then tested
this map within the Mapper. I found the Mapper gave
different results. Specifically, the XML file was
processed correctly and the Native Instance was not. By
processed correctly, I mean data in the source fields are
transformed to data in the destination fields. Not
processed correctly means the fields in the instance file
were ignored. I was able to see the same behavior using
the map within a channel as well.
I later found out if I add a value for the Source Tag
Identifier property of the individual records in the flat
file doc spec, the editor-generated native instance files
were processed correctly.
Jim
>-----Original Message-----
>On Mon, 22 Dec 2003 10:35:28 -0800, "Jim"
><anonymous@discussions.microsoft.com> wrote:
>
>I am not sure I have understood correctly, so I will just
write how I
>understand your problem;
>
>You have a documentspecification, and from this
documentspecification
>you have created both a native instance and an xml-
instance of the
>document in BizTalk Editor, right? Which documenttype is
it? Is it a
>flat file? An X12 file? Or somehitng else?
>
>Then, when you say that the xml-instance is correctly
handled, what do
>you mean? Do you mean that a receive function takes it
and it is sent
>through a channel, or are you only talking about BizTalk
Editor
>validating the file? Since you are talking about
envelopes and
>document definitions, I will assume the former. What is
the
>errormessage in the eventlog? That could probably be
helpful. Have you
>looked at the generated native instance and the generated
XML-instance
>and found the differences? If so, what are they? If not,
then try
>that.
>
>
>
XML
is
into
definition
>
>--
>Jan Eliasen, representing himself and not the company he
works for
>.
>
[ Post a follow-up to this message ]
|