DSXchange

Hi all,
i have job which reads the sequential file and writes data to the oracle Database.metadata for the sequential stage holds nearly 50 columns but i may get only 30 columns from another file,which i need to load it from the same job.so my question is like is their any option in parallel extender like 'Incomplete column' as in server edition sequential stage.I dont think parallel extender have this kind of option,if no then is there any alternate simple way to handle this kind of files.

If it is just direct load, then you can use RCP to propagate the columns. Don't define any meta data in the job. This might work, not very sure though. How are you loading the data into Oracle? what options are you using?

Maveric,

Its not really a direct load,i am doing pivoting after extracting the data from the file.so i cant go ahead without mentionig the metadata.I know we can handle it very easily in server jobs but not exactly in parallel extender.Is anything i can do with format,i dont have any delimiter as End of record.I have only single record in every file.

Now i decided to read whole data in a single field and carrying out the further transformations.But now encountered with another warning message as :
Missing record delimiter "\n", saw EOF instead

My file has no Record End Delimiter,and contains ony a single record.
so what should i do to suppress this warning.I dont want to carry this warning.any suggestions with PX.

Thanks

Your record lacks a record delimiter. DataStage is informing you of that fact. You could try setting or unsetting the Final Delimiter property. Otherwise use a local message handler to demote the message to informational.