Anyone familiar with partioned datasets on a mainframe?
I've been asked to migrate some partitioned datasets from the mainframe to flat files in Windows.
Trying to gather information about accessing them through WebFOCUS.

We have FOCUS installed on our mainframe.
The mainframe is setup as a Remote Server in WebFOCUS.

I know nothing about moving around in the mainframe.

In the past a mainframe developer had to create some datasets in mainframe FOCUS so that I could create data adapters in WebFOCUS to write my reports from.

What is the step by step process to access mainframe partitioned datasets through a Windows WebFOCUS environment?

What needs to be setup on the mainframe side?
What needs to be setup on the Windows WebFOCUS side?

This message has been edited. Last edited by: Kerry,

A separate output file for each member?

Using the -REMOTE syntax, you can DYNAM ALLOC each member separately to a generic master (you might be able to find a command to print all of the members into a file and then use them as input to a DM loop). The master (on the mainframe) would have a single field equal to the record length of the PDS members. Often this is 80 but it could be another record lenth.

Then issue a TABLE and do an ON TABLE PCHOLD to get it down to the Windows platform. You would need to APP HOLD or APP FI where you want the file to go on the Windows side.

You can find the syntax for the DYNAM command in a mainframe FOCUS manual, the chapter about the MVS environment.

I hope this is enough to get you started. I haven't done any of this in years but apparently I still have a few cells left that remember a little bit.

What type of data is stored in the members of the datasets ?

As Ginny mentions PDF members are typically 80 byte record length, but may vary.

Ginny, it seems that FOCUS on MF is like riding a bike, it tends to get lost in the memory, just needing the right key to unlock it.

Though I can't remember the technique I used to use to list all the members of a PDS.

Rather than worrying about the member contents for the mainframe piece, I'd still use a generic master, unless of course there were packed fields, then create custom masters when the files where down on the WF box.

Thanks for the help. I'll let you know how my testing goes.