Archives of the TeradataForum

Message Posted: Sun, 08 May 2005 @ 09:46:09 GMT

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

We do not use the -m option. We tried, but felt it was too cumbersome to use.

I opened an incident about 2 years ago asking why it would not allow me to use the previous macros (I don't know if anything has changed since then). In our case we use the .NAME option to distinguish one tpump vs. another. Also, we did not want to adjust the tpump script between the first run and subsequent runs.

The response to my incident is below.

If I remember back, I think this option is only useful when the tpump stops where it will not take time to drop the macros. However, If I remember right it will want to create them every time (unless you change the script so call the macros yourself).

The problem with leaving the old macros is that it fills up the dictionary with all these previous macro definitions.

From NCR:

"I have discussed with Tpump developer about an option to use an existing macro if it already exists and not to create a new one. He said that this option is not acceptable because it increases the risk of error.

He suggests:

- first run, use '-m' option to keep the macro.

- after first run, comment out the DML and add an 'EXEC macroname' statement.

..kiki..

Kiki Sanchez
Data Warehouse Group
Technology Division
Continental Airlines

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