RE: CF-Spec - another call for opinions

["Leland, David" <dleland@xxxxxxxxxx>]

Title: RE: CF-Spec - another call for opinions

I'd vote for number 2.  If I'm dumb enough to create a user BIF with a name which conflicts with a future IBM supplied BIF, then I SHOULD get a compile error.  While you're at it, if you allow EVAL and CALLP to be omitted, why not allow the "CF" to be omitted also?  That would be AWESOME!!!!

Dave

1) Make EVAL and CALLP always required:

      CF eval KeepLooping = *ON
      CF dow KeepLooping
      CF    read MasterFile
      CF    if %eof
      CF       callp HandleEndOfFile()
      CF    endif
      CF enddo

2) Allow EVAL and CALLP to be omitted:

      CF KeepLooping = *ON
      CF dow KeepLooping
      CF    read MasterFile
      CF    if %eof
      CF       HandleEndOfFile()
      CF    endif
      CF enddo

3) Option 2 with a commitment from us that no new opcodes will
   conflict with any possible variable or procedure name.  For
   example, opcode ON-ERROR could never be confused with a var or
   proc name.  This may mean some goofy looking opcodes in the
   future.  On the other hand, since most enhancements these days
   seem to be in BIF's, this may not be too big a deal.

4) Optional EVAL and CALLP, but use some special character to
   distinguish opcode names from var or proc names:

      CF KeepLooping = *ON
      CF /dow KeepLooping
      CF    /read MasterFile
      CF    /if %eof
      CF       HandleEndOfFile()
      CF    /endif
      CF /enddo

   If you don't like "/", what about some other character?

5) Is there some other alternative we've missed?