Archives of the TeradataForum

Message Posted: Mon, 26 Jul 2010 @ 15:53:13 GMT

	<Prev	Next>		<<First	<Prev	Next>	Last>>

back on this trail again ...

In terms of an OUTMOD, we've been told that a 'C' Program Outmod (from FastExport) can only RECEIVE data (and then write it out to a file by the C Program), but then we see this in the manual :

Output Data Length

Output data length specifies the length of the output data record. On entry to the OUTMOD routine, this parameter has a zero value and the output data record parameter points to an empty buffer. If the OUTMOD routine modifies an input data record, and the new length is:

- Less than or equal to the input length, the response row can be modified in place. In this case, the OUTMOD routine should set the input data length buffer to the new value before returning to FastExport.

- Greater than the input length, the output response row must be generated in the output data buffer. In this case, the OUTMOD routine should set the output data length buffer to the new value and reset the input data length buffer to zero before returning to FastExport.

Does this infer that we can get the C OutMod to modify the data - stripping off the two-byte length field, generated by FastExport - prior to passing it back to FastExport, or are we misreading this and the C Program must write the data out to a file for us ?

Not that we mind too much, but it would be a lot easier for us if the C Program can modify the supplied row data and then just 'hand it back' to FastExport, for it to do the writing out to a file (particularly since that's the way we've written the FastExports)

Regards

David Clough
Database Developer
Database Design Group

	<Prev	Next>		<<First	<Prev	Next>	Last>>