🚀 go-pugleaf

In article <34995@uflorida.cis.ufl.EDU> bmr@thedog.cis.ufl.edu (Benedict Rafanello) writes:
>In article <23U676368B@tron.gun.de> g_dotzel@tron.gun.de (Guenter Dotzel) writes:
> 
> >MODULE mod_copy;
> >FROM ProgramArgs IMPORT IsArgPresent, ArgChan;
> >FROM IOResult IMPORT ReadResult, ReadResults;
> >FROM TextIO IMPORT WriteString, ReadToken, SkipLine, ReadString;
> >
> >TYPE fnam = ARRAY[0..254] OF CHAR;
> >VAR from, to: fnam;
> >
> >BEGIN
> >  IF (ReadResult(ArgChan()) <> endOfInput) THEN ReadToken(ArgChan(), from);
> >    IF (ReadResult(ArgChan()) <> endOfInput) THEN ReadToken(ArgChan(), to);
> >    ELSE (* ... *)
> >    END;
> >  ELSE (* ... *)
> >  END;
> >END mod_copy.
>
>I certainly don't like the look of this!  I would prefer to get the command line
>as a single string which I can parse myself.  This is the approach taken by
>Stony Brook.  Their library has a procedure, I believe its called GetCommandline,
>which returns the command line for the program as a string.  While parsing the
>command line yourself may be more work, I think it is more flexible than the
>approach above, unless I am missing something.

I agree with you completely.  I have been frustrated by C for precisely
the same reason.  But even argc and argv would be far more straightforward
than the above code.  I don't recall, is *that* the new ISO standard?  (I
hope not!)
-- 
IMHO, as always                                            _   /|
Miguel                                                     \'o.O'
mgallo@ares.calpoly.edu                                    =(___)=
"C is its own virus."                                         U