Re: Handling null-capable fields with program-described file--reading externally described using program defined layout

Peter Dow (ML) wrote:

You said "The current behaviour may seem like a bug for an application that wants to behave like DSPPFM; but it is a feature for an application that was never intended to work if a record had a null-valued field." My feeling is that the application that wants an error when a null value is encountered can just use ALWNULL(*NO); the application that wants to behave like DSPPFM (mine!) should be able to specify ALWNULL(*INPUTONLY).

Was Chuck Pence's explanation of what's going on correct? Is it just that the open for a program-described file is not requesting ALWNULL(*INPUTONLY) for that file?

Peter, I personally agree that ALWNULL(*INPUTONLY) _should_ have also applied to program described files; I don't know why it was restricted to externally-described files. But since it didn't apply to pgm-desc files when *YES was first introduced in V2R1 or when *INPUTONLY (= *YES) was introduced in V3R7(?), it can't start applying to them now. Some new syntax would have to be added that would make alwnull apply to program described files.

Yes, Chuck's explanation was correct.