Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!ames!lll-winken!uunet!tektronix!tekcrl!tekfdi!videovax!bart
From: bart@videovax.tv.Tek.com (Bart Massey)
Newsgroups: comp.lang.c++
Subject: Re: More nits to pick at...
Message-ID: <5423@videovax.tv.Tek.com>
Date: 5 May 89 16:59:57 GMT
References: <190@riunite.ACA.MCC.COM> <9279@alice.UUCP> <193@riunite.ACA.MCC.COM> <9291@alice.UUCP>
Reply-To: bart@videovax.tv.tek.com (Bart Massey)
Organization: Tektronix TV Measurement Systems, Beaverton OR
Lines: 34

In article <9291@alice.UUCP> ark@alice.UUCP (Andrew Koenig) writes:
> In article <193@riunite.ACA.MCC.COM>, rfg@riunite.ACA.MCC.COM (Ron Guilmette) writes:
> > >How can you declare a function which can accept a pointer
> > >to itself as an argument (in a type-safe manner of course)?
> > >
> > >	You can't in C++ or C.
> > 
> > Well, so is anybody (other than me) concerned about this?
> 
> Sure.
> 
> But suppose C++ came up with a syntax for doing such a thing.
> How would you translate it into C if C can't express it?

Sigh.  You've already done the syntactic and semantic analysis once, in your
translator.  You don't need C's type checking, and in fact don't want it in
this case (you can always safely cast the parameter to any type you need).
So, you just make the type be (void *).  Or (int (*)()).  Or whatever.

In fact, you're probably trying to emit K&R C, for portability.  It's not
possible to declare a function which *won't* accept itself as an argument in
K&R C, since old-style function declarations don't allow argument type
checking. So...

					Bart Massey
					..tektronix!videovax.tek.com!bart
					..tektronix!reed.bitnet!bart

P.S. -- This is the kind of thing I was talking about in an earlier article
when I claimed that cfront had at least as a secondary goal the generation
of C code which corresponded closely to the original and was easily
understood by people.  As I said before, there's nothing wrong with this.
But this kind of reasoning makes it pretty apparent that this secondary goal
does exist.