DSXchange

Hello,

I am seeing the following error messaged on rejects link :
I cannot see a direct way to change the null_field length for individual fields on reject links (you can set it at the file level but not for individual fields). One way around this is to change these fields to be varchar but the extra level of complexity is that the initial information is coming from a nullable date field in an Oracle table so that I have to import it as a timestamp (which then seems to insist on exporting as a char(19)).

Any ideas?

Cheers,

Adrian

Is this for a sequential file export?
-BP

adscot wrote:Hello,

I am seeing the following error messaged on rejects link :

Code: Select all

Insert_Rejects: When checking operator: When validating export schema: At field "DT_ADDED": "null_field" length (1) must match field's fixed width (19)

If it is a database field - Make it Varchar (or something equivalent to this).

If it is a flat file, use the function Space(19).

-T.J.

bigpoppa wrote:Is this for a sequential file export?
-BP

Yes it is a sequential file export but it is a reject file...

Teej wrote:
If it is a database field - Make it Varchar (or something equivalent to this).

If it is a flat file, use the function Space(19).

-T.J.

The problem is that you can't set field level attributes on a reject file.

If this is a sequential file just change the metadata type to varchar instead of timestamp. Actually you shouldn't bet getting this error since null dates are valid.

adscot wrote:The problem is that you can't set field level attributes on a reject file.

I was referring to doing this BEFORE the flat file output in a Transform Stage. This way, your records will NOT be rejected for that particular reason.

-T.J.

Teej wrote:

adscot wrote:The problem is that you can't set field level attributes on a reject file.

I was referring to doing this BEFORE the flat file output in a Transform Stage. This way, your records will NOT be rejected for that particular reason.

-T.J.

Ah ha I understand - sorry me being a bit slow. I can put a transform on the reject link prior to the output to resolve the output issues

Cheer,

Adrian

adscot wrote:Ah ha I understand - sorry me being a bit slow. I can put a transform on the reject link prior to the output to resolve the output issues

Let us know of the results in applying my suggestion.

Thanks!

-T.J.