DSXchange

dslearner - I am trying to help, but some information is still missing: 1. Have you tried using the old Oracle Enterprise Stage. "Yes" or "No". If "No", then try the old Oracle Enterprise Stage which you know used to work at the previous release. Does it still work corr...

You need to change your file to either have only one record type per row OR to remove the line breaks (with sed or similar) and change the CFF definition as well.

Do you have the same problem when using the Oracle Enterprise Stage? You should check that stage in order to see if your problem really is with the Oracle connector or a general issue at 8.5

Hmm... I'm still not quite sure of what your goal is here. 1. SALE 1 Units Sold 4500 2. 3. Record 4 PUR 1 has 449.0203000 So it SALE 1 written now with the modified value of 4050.9797? Or is the second PUR record now processed and so on? It would seem, at first, that this isn't really possible withi...

This would only work if all your text files are of the same format; unless you want to store the schema of all the files and use that parameterized. One simple method would be to use a transform stage variable to keep a running total, then when you read the trailer record (which you would need to so...

What methods are you using in the Oracle connector stage? Inserts? Updates? Was DataStage the only thing changed and the database remained the same? Do you have problems using the Oracle enterprise stage at 8.5 as well or is the problem restricted to the Oracle Connector stage?

Have you tried adding quotes to the column values in the DFF stage and does that make a difference?

I confess to not understanding what you are trying to do, despite the good description. Could you explain in a bit more detail the steps that you need to process for "SALE" "1"?

In that case the CFF Stage is your answer, it will automatically correctly read each record type. The documentation does a barely acceptable job describing how to do this but it should get you started.

I assumed those were line terminators, not column delimiters.
Change your sequential stage to declare "~#" as your Field Delimiter String.

yes. Declare it as an ODBC source and analyze away.

As stated several times before - avoid floating point. Your data in the sequential source is in decimal format and should be read, and stored, as such. Sequential files can read decimal data, so if you are not getting values then there is a problem in the column definitions that you should correct. ...

In your sequential file stage, specify Format -> Record Level -> Record Delimiter String as "~#" (without the quotes).

There are several approaches you can take. What you need to determine r answer is how you differentiate between your 3 record types. Is the first field of each record type a single digit, "1", "2" or "3"? If you have such a field then this problem is easily solved using...

Although this is a pretty esoteric question, perhaps someone here might know the answer. When creating the GUI interface to a custom BuildOp there are quite a few options available to the developer, it isn't too difficult to make the parameter declarations look like that of the CDC Stage or Sort Sta...