DSXchange

What do you mean by "standalone"? Hi Chandru, Well, it is difficult to make out exactly what you want. 1> Before/After routine: Define your routine type as before/after routine and the routine will become a before/after routine. 2> Standalone call from sequence: This will work by default ...

Magesh_bala wrote:Its not the right one.
before posting, Please check and post.

Magesh_bala wrote:
Please revert back with an clear unix commend to be used.

What, is that an order? and by the way, this is not a unix forum.

Try : sort -u

Or you could, after job, start a scripted sqlplus session and run that file from inside sqlplus. Very true. And, in the lines of what Craig is suggesting, have worked on project where we were doing very similar stuff and we created a complete oracle package for the replication of tables. The whole ...

Yes

Well, it is complaining about the fact that there is a URL and not an SQL statement as it expects. It is not complaining about the SQL inside the file since it has not accessed it at all. This is not how it works, that you just put the URL to an SQL file and it will execute it. What you can do is, p...

If you are already able to see the column defined, it should be out there, probably you have to scroll or something.

Cool and welcome to dsxchange !!

Another strategy: Since you are concerned about field names, I will assume that you are about to insert the data into database. In that case, you can create a job that reads the old to new column mapping from the table and create a insert/update statement which you can pass as a parameter to the job...

Well, from a testing perspective I would not be very concerned about the feature/functionality changes since the bottomline will be to test "output" data and validate all business rules. The conversion will typically involve regression testing - runing through all test cases again, and as ...

Well, you can check out the lookup stage and use dummy columns in both primary and secondary lookup inputs to achieve a cross join and set lookup condition in the lookup stage.

But, what do you mean by "generate a undetermined number of output rows per input"

Before you can test in control center, you have to find exactly for which values datastage is throwing that error. Because the value you tested with may not return multiple values ... and datastage works in the normal fashion, taking each value from the primary link and looking up the recordset in t...

what is hidden? the input links or the place where you want to make "fail then continue"?

Sounds like an outer join.

You should double click on the "Condition" box in the reference metadata box within the lookup stage.