At 12:44 PM 5/26/98 -0700, you wrote:
>Since you can pass the FD name to a routine, I passed it to... PUTJCW.
>FINDJCW was then used to retrieve the value.  To verify this was indeed
>the "real" file number, I passed it to PRINTFILEINFO, which gave a
>tombstone for the expected file.
>
>The call to PUTJCW was pretty vanilla:
>
>    FD  MYFILE.
>        ...
>        CALL INTRINSIC "PUTJCW"  USING  JCW-NAME
>                                        MYFILE
>                                        JCW-STATUS.
>
>and FINDJCW expects a 16-bit logical:
>
>        05  MYFILE-FILENUM      PIC 9(4) COMP.
>        ...
>        CALL INTRINSIC "FINDJCW" USING  JCW-NAME
>                                        MYFILE-FILENUM
>                                        JCW-STATUS.

Of course, this has a side-effect (ie setting the JCW!) JCL may raise the
occasional eye-brow. I also like the ASCII/BINARY combination, but it uses
the same idea, that intrinsics are smart enough to pass the FD name as a
value parameter.!

Isn't this more fun than working?
TOny