DSXchange

Doh! You so right... too much in a hurry.

Sure... more like this is how I do it:

That's why I suggested the "alter session set rollback segment" route. Should work fine in the before-sql area.

I'll have a play with your other suggestions to see what I come up with but the data volumes will be large for the initial population of the Warehouse (50GB) but quite small on a nightly basis so hashed files may be too slow. Cool. Certainly couldn't hurt to 'play' and at the very least expand your...

No, you should be able to issue specific 'alter session' commands in the before-sql area to use specific rollback segments if that's really what you need to do.

Have any other Sybase users come up against these restrictions re stored procedures/flat files as tables and if so how have you got around them ? Please don't take this the wrong way, but these are not so much 'restrictions' as a lack of understanding on your part. The only way you can treat this f...

No. Stage variables are local variables and don't exist outside of the stage they are defined in.

Are you sure that you've applied the SP2 patch from Ascential? When you say something like this, it doesn't sound like it: I'm just had all of the available SP2 patches installed but unfortunately this hasn't made any difference. Or: I finally got a PC built with Service Pack 1 and all its patches i...

Job Parameters are (in almost all cases) enclosed in pound signs or 'hashes' ( # ) when referenced. The colon is used for concatenation. So, if you simply wanted to refer to a job parameter called DATA, you would do it like this: #DATA# If you wanted to include it as part of a string in a command, t...

Do you mean automatically, perhaps as part of a job sequence? Or is this a general question?

In general, you can Validate a job rather than Run it to see if it can connect to all of the various source and target stages in the job.

Use replace option and use replacement value of something like "XXXXYYYYYZZZ" or something very noticeable and unlikely to appear in the missing column. Then, put a constraint on the following transformer to place the rows that have that column value in your reject file. And here I was thinking, wh...

No, the funny thing is... "after job" is still during the run-time of the job, so job parameters can be used by them. The job hasn't finished running until any after job routines finish as well.

You can also get a product from Ascential that does log scraping from the Oracle transaction logs which can make delta processing a lot easier. Vincent, is Ascential still selling this? I was under the impression that they may not be maintaining it and (thus) may no longer be offering it for sale, ...

But does any of this help the OP treat a flat file like a table?

I don't think so and assume my advice regarding how to proceed still stands - in addition to yours, of course.

The Dynamic Relational Stage, new in 7.5, something which allows you to use a single stage to connect to "any" relational source.