DSXchange

I am using container as lookup . if no rows are returned from container . the job is running for infinite time . There must be something else going on. The simple fact that a lookup in a shared container returns no rows shouldn't cause something like this. We'd need a better idea of exactly what's ...

I wish I was creating an XML export, it would be very useful. No, you are missing the point. You can only create a .dsx export from the command line. However, in your client software on the machine you are doing the export from, the 'Export as XML' option is checked. This causes the stupid error yo...

Waaaay back at the beginning of the thread, I mentioned this. You can do the exact same thing in DataStage because all you are doing is asking the source database (in this case Oracle) to do the conversion for you. In your source OCI stage, in the derivation of the QUARTER field, you could put: TO_C...

Luckily, you don't need to get your date out in a different format, you just need to match your Iconv date mask with what is coming in. We'll try substringing off the time, since we don't need it and then changing the mask around a bit: OCONV(ICONV(InLink.InDate[1,10],"D-YMD[4,2,2]"...

Yes the OS is HP/UX. Is this problem due to OS? Yes, this is why I asked. Several others (including myself) have had the same problem with Korn shell scripts core dumping on HP/UX - only from DataStage. This post may help shed some light on the issue and documents what we have done that corrected t...

What option in the source OCI stage are you using to generate the sql? Column Generated? Full? Custom? If you let DataStage build the sql and you tell it the field is a Timestamp then you are not getting it in MM/DD/YYYY format. It supplies it in "YYYY-MM-DD HH24:MI:SS" format via a TO_CHAR() functi...

I've had some weird problems, too. Seems like problems posting, "cannot find host", yada yada, then it finally goes through. Come back and I've posted the same thing three times, then have to scrape out two.

No worries.

No, afraid there's no such thing as a 'transaction size' or a commit for sequential files from DataStage that you could be seeing the result of. What you may be seeing is a product of your disk subsystem, especially with 'enterprise storage' like EMC and the like. Depending on the settings and the a...

Dang, there's an echo in here.

It means the conversion failed. Which should mean the incoming data doesn't fit the mask provided. If you are certain it is coming in MM/DD/YYYY format, try this slightly different derivation:

I tried putting the expression OutLink.Quarter = OCONV(ICONV(InLink.InDate,"D4/MDY"),"DQ") with my equivalent link names in the Derivation of the transformer, but still it is red in color. Hope you don't mean that literally. All you should have in the derivation is what you have on the right side o...

I almost posted the link for you, Chuck.

Operating system? HP/UX by chance?

In Informatica, I just had to give the SQL statement TO_CHAR(date_column, 'Q') and it extracted the quarter information from the source. That's Oracle - not Informatica or DataStage - doing the work for you. If you want to do it that way, go ahead! Use the same sql statement in the Derivation of th...

Sure, do this all the time to generate data. Two things you need to know: 1) Your transformer needs a Stage Variable defined. You never need to use it, simply define it so the job will compile. 2) Make sure you put a Constraint in the transformer or the job will run forever. Simplest way is to const...