DSXchange

ddevdutt:

Perfect - that's the ticket!
I appreciate your help!

I would like to replicate the following:

select x.a, y.b
from x, y
where x.m = y.m
or x.n = y.n

Is there a way to incorporate an "OR" condition within a JOIN stage (or lookup)?

Thanks for the help - you sent me the answer I was looking for and it works PERFECT!

I have a case where I do not know at runtime if an input file will exist. Is there a way to dynamically test from datastage if a file exists and continue on if it does not exist? I am running a sequence, and if the file does not exist, I do not want the job accessing it to abort, which would cause t...

If running in parallel, at the Oracle stage, click the INPUT tab, then choose partition. Select HASH and select the same columns that comprise the keys for the table you are loading. This will ensure that all the duplicate keys go down the same "path".

girija: Thanks! Will this work even though nothing will go down the reject link? (I think this is what csri is getting at).

Smith: We thought of disabling the FK as you mentioned, but the powers that be will not allow this.

That is exactly the problem - the text I am capturing does have a LF that I need to preserve. We are getting the source file from another entity with the hex 1e as the record delimiter. We are using the pipe as a field terminator - could we also use a pipe for the record delimiter as well?

I have a job where I split off from a transformer to load two Oracle tables, one parent, and one child. Even though I ordered the links appropriatly in the transformer stage (parent load first, child second), the child table does not get loaded. I assume it is because the parent Oracle stage has not...

By "record delimiter", I mean

(a) the character that marks the end of the record (the record delimiter)?

In Job #1, one of my output datasets receives zero rows. In Job #2 I try to read in that dataset, and I get errors. Should an empty dataset be an error state?? (Even View Data gives an error rathter than just saying "No Rows".) Is there a way to check the dataset in Job #2 to avoid getting...

How do I use hex 1e as the record delimiter reading a sequential file? I assume I use the final delimiter string, but how do I tell it to use a hex character?

Thank you all for your help - you guys rock!

I did not realize that the sparse lookup was an option in the Oracle stage - I was looking in the lookup stage. Works great now!

Thanks!

Sorry to be so vague. Here is what I need in more detail. For every row from my input dataset, I need to add a column that has the nextval from my oracle sequence. So, for example, I need to join row 1 of my dataset with sequence.nextval, row2 with sequence.nextval, row3 with sequence.nextval and so...

Sorry to be so vague. Here is what I need in more detail. For every row from my input dataset, I need to add a column that has the nextval from my oracle sequence. So, for example, I need to join row 1 of my dataset with sequence.nextval, row2 with sequence.nextval, row3 with sequence.nextval and so...