DSXchange

:!: Please don't feel the need to add your problem on to other's only vaguely related posts. I've removed them so the discussion of your issue can happen here. And here are the two replies that were there that should have been here: Does Powershell need an option to return once the command is comple...

Get the Folder name using what approach? Printing from the Director? You'd have to double-check and see if there is a way to enable that without having to do one folder at a time... I have no DataStage access to play with. From the command line using dsjob it should be part of that is returned by th...

None that I am aware of though there may be some minor things. The big change in "architecture" came between the 7.x and 8.x releases.

Please don't jump on the end of older posts. I've split your post out on its own and linked it back.

Which is not as simple as it sounds.

Yes, you can use a sequence automagically in the insert statement however the fact that you want it to always start at 1 rules that approach out. Which is why I asked.

It's hard to tell from your examples so far but does the "sequence" need to start from 1 each time? Meaning this part:

RejOra
1
2

When you run it a second time then can it look like this:

RejOra
3
4
5

Or should it start with 1 again?

So are you generating this 10M record file and you'd rather generate 10 1M record files? Or do you simply have a file that needs splitting? For the latter I don't see any need for an ETL tool when the O/S command split will do that for you. For the former, I don't recall anything that would make tha...

If your source and target are both VARCHAR2(9) fields and it throws the error noted, my first question would be what are the NLS settings in the source, target and DataStage.

Just wanted to make a comment on #3. Generally when we migrate the jobs from DEV to PROD env, the Project Names are always same but in this case it will not be, will this be an issue? I've never worked for a client where the projects all had the exact same name across the environments, the environme...

DataStage doesn't care if you use single or double quotes, either works but you need to be consistent. If you need double quotes inside your strings then use single quotes on the outside. And vice-versa.

Well... looks like my earlier advice on how to start your own topic wasn't clear. Once again you jumped on the end of an older, sort of related post. And once again I've split it off to your own topic so you have control of its fate. Still would be great if you could go back and edit the above post ...

Use Field() and tell it the field is "-" delimited. For the first part, take 2 fields starting from field 1, for the second take 3 fields starting with field 3. The optional argument after the field number to extract controls how many fields to extract. It defaults to 1 if you don't specif...

Know? No. Suspect?

chulett wrote:Sounds like you have a characterset issue where those 5 characters won't fit in 5 bytes but if the issue only shows up in the Oracle Connector, I'd be contacting your official support provider and see if perhaps there is a known issue that a patch addresses.

I'd start with Field(). Does your "..." at the end mean there can be an unknown number of fields in the string? I'd also be curious if you are on 8.5 or higher so you have access to transformer looping.