Tony Harminc wrote:
Something not really discussed (yet) is that this is a NEW dataset with directory blocks specified. Surely there is no conceivable reason to want to write to this PDS as a sequential dataset... is there?
There are a number of reasons; e.g., fast directory initialization (one EXCP per track), writing the directory prior to writing the members (backwards, but works if done correctly), etc.
All the reasons for writing to a PDS directory are for existing datasets. Would it be unreasonable for either OPEN or the application program (IDCAMS in this case) to check for this case?
What should it check? If I use BSAM or EXCP to write, then the DCB parameters aren't even relevant; what else could Open check? Ditto for IDCAMS - it's a utility that performs a requested function, and has no way of detection inapplicable usage. You might just as easily ask for the Initiator to request confirmation when it hits a disposition of DELETE for a data set, or any number of other deleterious conditions that could occur due to carelessness.
Gerhard Postpischil Bradford, VT ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
