XML File read follow up

[wjfitzgerald]

Good morning,

I am working on transforming an xml file to a flat file, and have it working for a basic file. i am now trying to move on to the more complex file structure that is actually going to be used and, of course i am getting an error!

can any one tell me what the following means, please?
[i]xmlreadsepa..XML_Input_0: The name '{0}' is not a valid NCName.
expression = '/:SCTIcfBlkCredTrf' Remaining tokens are: ('/')[/i]

as an aside could anyone explain the concept of repitition element required to me please? i have read the XML pdf and the duke notes, but fear i am missing the point somehow.

As always, thanks in advance.

Regards[color=black][/color]

[chulett]

First question, in my mind, would be to confirm how your XPath Expressions were generated. Automatically from an xsd?

[wjfitzgerald]

Hi,

I had a sample file and used the import an xml file definiton option. Is this what you mean by automatically from aN XSD?

(I am a cobol mainframe programmer and so am out of the comfirt zone slightly and don't have all the lingo, i'm afraid.)

thanks again

[chulett]

No. While reading the sample file directly can be "ok" you really need to get your hands on an ".xsd" which is the "schema definition" file. Any chance of requsting that from whomever is creating the file?

[wjfitzgerald]

I'll go chase that down and see if that solves my problems.

Thanks again.

[eostic]

The symptom above with the {0} in the message looks like one of the 8.0 anomalies that there is a patch for. You will want to contact support for the 8.0 patches to XML.

One of the issues was namespaces.....if that's your current issue, you can get around it temporarily to test your Job's logic by manually editing out the namespaces in the document (all the xmlns:xyz="http://x.y.z" and the prefixes where xyz: is then used below in the document).

As for repetition element, I generally refer to it as "an element at the lowest level node in the document that I am interested in." It's the node that you expect will "repeat" for "n" rows. An xml document for employees in a company might go deep thru a hierarchy of division, department, team, and then have employee as the lowest interesting node. EmployeeName might then be the repeating element (give it key=yes --- it has nothing do to with "keys").

As suggested above, an xsd is a good idea, but you can import from the document --- just be careful about elements that are not in the document that you imported from, and also give careful consideration to you datatypes.

Ernie

[wjfitzgerald]

thanks ernie, i will chase up the patch this morning.

these things always make sense once you understand them!

regards

john fitz