DSXchange

You can abort a job many ways, use a constraint on a link with Abort after N rows set to 1 where the constraint checks the reference value against the expected values and acts accordingly. You could also use a custom DS Function to check the values and uses the API DSLogFatal to log a fatal message ...

The view must be forcing the entire result set to resolve before streaming data, indications of this are use of GROUP BY in thedata. A SELECT * from the view must be more work than your current configuration can handle. If you can't increase the TEMP space any more, than consider extracting subsets ...

Please describe all stages used in this job, as your value may be getting lost in a prior stage and the error message relates to that stage.

Please paste the EXACT and ENTIRE message, as it often names the stage and link where the error is occurring.

The autogenerated SQL puts to_date functions on your DATE or TIMESTAMP data type columns. If your data matches the format token in the SQL, everything works. I suspect your source data is ready to load, just verify. You won't have to do any DS functions to change the data.

Preload is only beneficial for performance if the same rows are going to be referenced multiple times. If the reference only occurs once for each row in the reference, it's not beneficial. The more times a row is referenced, the more time you have "saved" by not doing the mechnical reference, but us...

You have data type issues. Add Transformer stages and adjust/cleanup the data accordingly.

Who needs a Premium Poster when you've got DSguru2B???

Maybe it's time to be renamed to DSguruIAM???

You're going to have licensing issues because Ascential supplied drivers are licensed only for internal use by DataStage. Unless you supply your own drivers, your ODBC method won't work. What's the issue with command line calls to sqlplus and osql/isql to run scripts that do this work? You could eas...

To change a text string to another, use CHANGE or EREPLACE. Example: text_string=CHANGE(text_string, "|", ")(") To add the leading and trailing parenthesis, just append: text_string="(":text_string:")" To add the "pre" text, just replac...

You'll need to beg for an end-of-record terminator, because of the confusion around the CR/LF not being a true end-of-record terminator as it's embedded within your custom delimiters. The Sequential stage will fail to recognize a row because you won't be able to declare a single character column del...

If you look at that phantom message, and post it here, we could tell you what it is saying.

Chances are that your If statements are referencing variables that don't exist within the function.

account_type,book_code, counterparty_code, etc are not declared, are you passing them as arguments?

Using DS BASIC, the OPEN and OPENPATH statements are used. The hashed file must be a local file on the server, not on a remote database. READ, READU, WRITE, and WRITEU are used to read and write rows, whereas READV and WRITEV can write individual attributes. Read your DS BASIC manual for extended in...

I was wondering why the script doesn't handle other object like shared containers, table definitions, stuff like that? Because that's how I had him write it. Table definitions are completely recreatable and DS is not the system of record. MetaStage, ERWin, etc is the place for that. Feel free to en...

Your design under DS7+ would connect the two streams in the design using an IPC stage, which is nothing more than a pipe. The creation of the named pipe happens via a before-job or before- transformer routine? If it's before-job then that happens before either stream beings. If it's before-transform...

Okay, Ray's solution won't work because we misunderstood what you are doing. You are actually "connecting" rows, but DataStage reads 1 row at a time. You're going to need to use stage variables to handle this. You need to hold a row in a stage variable if it contains the condition you desire, the "N...