DSXchange

Do you have the sequential stage setting "First Line Is Column Names" set to true in your job?

For both the viewing and deletion I would try using the "orchadmin" tool from the UNIX command line. I have seen the error you mentioned before, but can't recall what the cause was - but I do recall getting everything done via orchadmin. I know you can delete the dataset with embedded spaces in the ...

Usually the job is visible in the graphical tools and you can delete it from there. If it cannot be seen or deleted, then the TCL command "DS.CHECKER" will probably be able to do so. Failing all that, I would go in and manually delete the hashed files associated with that job. I'd also check the pro...

Declare the fields to be long enough to contain the blanks, then trim and modify them to the correct length in a later stage (modify or transform)

The funnel won't do that for you. But in a subsequent transform stage you can store the value from the previous row using stage variables and combine the rows there. If you explain why you need to do this (perhaps with an example) we might be able to suggest a better solution or approach.

nynali - it would be better to post in the appropriate forum, but I'm sure a moderator will find this thread and relocate it.

You can set the the default value by right-mouse clicking on the colum and then "edit row" to get to the details.

I have trouble understanding this - normally you would join (for instance in a lookup stage) on the common EMPNAME or EMPNUMBER. But you might use a funnel stage to combine the 2 streams and then combine the 2 rows in a subsequent transform stage.

Make a copy of the job, replace the hashed file stage with a sequential file stage and look at the output, if it still makes no sense please post an excerpt from the sequential file.

prams - usually tables don't have headers and footers, that is usually found in the realm of sequential files. Since you normally don't need a product such as DataStage to make a simple copy of a table there must be more involved and it would help if you could explain.

Are you certain that you clicked the "key" tick box for the two columns in the hashed file stage? Which are the 2 keys that are currently being shown in the hashed file?

It depends where you wish to compute this numeric difference. If you load the timestamp into DataStage and convert it into internal format you can subtract the two values to get an exact difference in days.

Convert() will not round, this is being done elsewhere. What is the defined datatype of the column or stage variable or derivation where you are doing thsi?

What is the string contents and what is the decimal column definition. Have you checked the other environment to see if you have a message handler set for that that job?

Good job!

Even though DS will do an implicit conversion from string to number in a transform stage, it is much better to explicitly convert using a function such as StringToDecimal(), which will take care of the sign.