RE: CHKOBJ in RPG IV programme

[Vernon Hamberg <vhamberg@xxxxxxxxx>]

Hi Rod

I've not seen ChkIFSObj, so I assumed there was no object type in its
parameters. Further, I assumed that you put the full path in to see if  an
object exists. Hence, it _could_ be used to check existence of a QSYS.LIB
object.

My guess is that, unlike QSYS.LIB objects, IFS objects must be unique by
filename. This is interesting, as the type is not part of the name, as it
is in the library system (I know - only in the name as stated in a path)

I think we're mostly on the same plane here - maybe about a 1/2 inch
offset?   ;-)

In C/C++ it's not as easy to use CL commands. I DO, in fact, use the
access() function to test existence of QSYS objects, because it gives me a
return code, and I don't need to write a CL program to do this with a
passed argument (no return codes possible from CL). I actually wrap it in a
nice little #define called exists(), so I get a statement like if
(exists("/qsys.lib/vern.lib/object.usrspc") )

Regards

Vern

At 05:44 PM 8/29/02 +1000, you wrote:

>Vern, Quite right, the objtype for a QSys object is implicit in the Path
>name.
>I wasn't suggesting people use ChkIfsObj (meaning stat()) instead of ChkObj
>for QSys objects. ChkObj is clearly the obvious choice, merely that
>ChkIfsObj wasn't necessarily doing what people thought it did (i.e. checking
>for a stream file existence). If you were using ChkIfsObj in a CLP to check
>a user-entered path parameter (e.g. StmF) you could get (theoretically) a
>"found" condition on an object that wasn't, say, a streamfile
>I have modified ChkIfsObj to include an ObjType parm (only supported  *STMF,
>*DSTMF, *DIR, *DDIR, *SYMLNK, *FLR, *DOC)