DSXchange

If all else fails, look it up yourself!


Custom SQL?

Well... our current job design does have the ORABULK stage building only the control file and the data file is spit out by a Sequential File stage. Almost 500 jobs pull data from Oracle 'CDC' views and then in a single subsequent step (korn scripted) they are loaded in parallel using sqlldr. I suppo...

You don't do assignments in a Contraint, they should be an expression that equates to True or False. Your If-Then-Else logic is meant to be used in a Derivation to set the values appropriately. To turn that into a Constraint to control when data is allowed to flow out of the link, do something like ...

Not really... not if the record should go to Table B only on an Update action. If that's the case... You really should move away from the "Insert or Update" actions as they don't give you a way to differentiate between the two. Try this - Preload the existing keys into a Hash and do a lookup against...

When promoting a job, is there a way to grab all dependent components? I see there is a way to select those when checking-in a job. Not that I am aware of... but then I don't believe you should really need it. As you mentioned, when importing you have the opportunity to bring in 'dependant objects'...

Something in the way of answers for you... 1) Sure! Since this is a client-side tool, it doesn't care where all of your projects are... and they typically reside on different physical servers. 2) You don't.... and I'm not sure why you would want to. The only place where "multiple job versions" exist...

Does anyone know if the problem with the ORABULK stage generating bad control files for VARCHAR2 fields still exists in 7.1? It is an issue in 7.0.1 and just majorly bit me in the behind. All VARCHAR2 fields are generated as CHAR in the control file, so if the data exceeds 256 characters the load fa...

In your version, you should have an OCI and an OCI8 stage to load into Oracle 7 and 8, respectively. They are optional plugins, so if you don't seem them then they probably weren't loaded when your server was installed. You can go back and rerun the install script, which will allow you to add new pl...

Well... seeing as how it is a Tranform, the end result is the same as if it was 'inlined', so no - no overhead savings. Just easier to understand.

Arg, you're right - "MCN". No need to create one, as I said it already exists - check under "Built-In\String" for the DIGITS Transform.

Mine? Hmm... I don't really know. Be great if someone with the ability to check could do so and let us know!

I'm assuming this will work as it does for the OCI stages... Don't "Load" the metadata from inside the stage. Instead, make sure Table Definitions are showing in your Repository Browser and then drag the metadata to the link on your ODBC Stage. Doing it this way will leave the Derivation column alone.

This is a common problem and there are quite a number of posts on this particular subject here. If you search for something like "releasing locks" you should turn them up.

Use the Transform called DIGITS, it uses 'Oconv' and 'MTC' (I believe) to do exactly that.

You'd need to contact Ken directly. I don't know if your posting in the 'Using an array as a job handler' thread is enough or if you should send him a note directly.