Re: typedef ... refactored

From	Keith Thompson <Keith.S.Thompson+u@gmail.com>
Newsgroups	comp.lang.c
Subject	Re: typedef ... refactored
Date	2020-08-14 12:32 -0700
Organization	None to speak of
Message-ID	<87o8ndxbts.fsf@nosuchdomain.example.com> (permalink)
References	(6 earlier) <SAvZG.2442895$fgO.1865466@fx12.am4> <rh61ij$h7d$1@dont-email.me> <rh683a$oi0$1@dont-email.me> <87wo21xhjx.fsf@nosuchdomain.example.com> <rh6nj6$q5t$1@dont-email.me>

Show all headers | View raw

David Brown <david.brown@hesbynett.no> writes:
> On 14/08/2020 19:29, Keith Thompson wrote:
>> David Brown <david.brown@hesbynett.no> writes:
>> [...]
>>> The distinction between "struct X" and "typedef X" namespaces is
>>> unavoidable in C, and useful as it lets you use the same name for a
>>> struct and its typedef.  (The alternative is to do as C++ did - make "X"
>>> automatically a typedef for "struct X".  That is, IMHO, better - but it
>>> was a change that could be made for C++ when C++ was new, not for C when
>>> C was already old.)
>> 
>> Strictly speaking, C++ doesn't make automatically create typedefs.  It
>> has the same typedef feature that C has, and it's used only when you use
>> the "typedef" keyword.  The difference is that struct tags (also enum,
>> union, and class tags) are directly visible.
>> 
>> You could achieve (mostly?) the same effect by using automatically
>> created typedefs, but that's not what C++ does.
>
> Yes - calling it "automatic typedefs" is merely a convenience to refer
> to the effect, rather than being technically accurate.

And I think that just saying that the name is directly visible is both
technically accurate and even more convenient.

> (Can you think of any way in which there would be a difference in C++ if
> it did do "automatic typedefs" here?  I know it's OT, but I'd be curious
> if you can think of any cases.)

I can't think of any such cases (which doesn't mean there aren't any).

In C it's legal to define struct foo and enum foo in the same scope; in
C++ it's illegal.  But if C++ actually did "automatic typedefs", the
effect would be the same.

The implicit typedef would have to be created when the tag name first
appears, because you can refer to "struct foo" as "foo" even inside the
definition.

    struct foo {
        foo *next; // legal in C++, illegal in C
    };
    typedef struct bar {
        bar *next; // Still illegal in C
    } bar;

Apparently this is illegal in C++:

    typedef struct foo {
        int n;
    } foo;
    struct foo obj;

because the name "foo" hides the struct tag even when it appears after
the "struct" keyword, but that's not an example of what you're asking
about.

-- 
Keith Thompson (The_Other_Keith) Keith.S.Thompson+u@gmail.com
Working, but not speaking, for Philips Healthcare
void Void(void) { Void(); } /* The recursive call of the void */

Back to comp.lang.c | Previous | Next — Previous in thread | Next in thread | Find similar

Thread

typedef ... refactored G G <gdotone@gmail.com> - 2020-08-09 01:07 -0700
  Re: typedef ... refactored G G <gdotone@gmail.com> - 2020-08-09 01:46 -0700
    Re: typedef ... refactored Richard Damon <Richard@Damon-Family.org> - 2020-08-09 08:09 -0400
      Re: typedef ... refactored Manfred <noname@add.invalid> - 2020-08-09 18:08 +0200
  Re: typedef ... refactored Jorgen Grahn <grahn+nntp@snipabacken.se> - 2020-08-09 09:29 +0000
  Re: typedef ... refactored Johann Klammer <klammerj@NOSPAM.a1.net> - 2020-08-09 13:29 +0200
  Re: typedef ... refactored Bonita Montero <Bonita.Montero@gmail.com> - 2020-08-09 13:53 +0200
    Re: typedef ... refactored Poprocks <please@replytogroup.com> - 2020-08-10 20:37 +0000
      Re: typedef ... refactored Keith Thompson <Keith.S.Thompson+u@gmail.com> - 2020-08-10 13:49 -0700
        Re: typedef ... refactored Poprocks <please@replytogroup.com> - 2020-08-10 23:09 +0000
          Re: typedef ... refactored Jorgen Grahn <grahn+nntp@snipabacken.se> - 2020-08-11 07:14 +0000
            Why not indeed??? (Was: typedef ... refactored) gazelle@shell.xmission.com (Kenny McCormack) - 2020-08-11 07:48 +0000
            Re: typedef ... refactored Ben Bacarisse <ben.usenet@bsb.me.uk> - 2020-08-11 10:56 +0100
        Re: typedef ... refactored Tim Rentsch <tr.17687@z991.linuxsc.com> - 2020-08-28 04:08 -0700
      Re: typedef ... refactored Bonita Montero <Bonita.Montero@gmail.com> - 2020-08-11 08:10 +0200
        Re: typedef ... refactored Anton Shepelev <anton.txt@gmail.com> - 2020-08-13 00:52 +0300
          Re: typedef ... refactored Bonita Montero <Bonita.Montero@gmail.com> - 2020-08-14 14:13 +0200
            Re: typedef ... refactored Bart <bc@freeuk.com> - 2020-08-14 13:37 +0100
              Re: typedef ... refactored Bonita Montero <Bonita.Montero@gmail.com> - 2020-08-14 14:54 +0200
                Re: typedef ... refactored David Brown <david.brown@hesbynett.no> - 2020-08-14 16:46 +0200
                Re: typedef ... refactored Bart <bc@freeuk.com> - 2020-08-14 16:59 +0100
                Re: typedef ... refactored David Brown <david.brown@hesbynett.no> - 2020-08-14 21:05 +0200
                Re: typedef ... refactored Bart <bc@freeuk.com> - 2020-08-14 20:51 +0100
                Re: typedef ... refactored antispam@math.uni.wroc.pl - 2020-08-16 13:36 +0000
                Re: typedef ... refactored Bart <bc@freeuk.com> - 2020-08-16 15:58 +0100
                Re: typedef ... refactored antispam@math.uni.wroc.pl - 2020-08-16 18:10 +0000
                Re: typedef ... refactored Keith Thompson <Keith.S.Thompson+u@gmail.com> - 2020-08-14 13:23 -0700
                Re: typedef ... refactored David Brown <david.brown@hesbynett.no> - 2020-08-16 19:48 +0200
                [OT] words.  Was: typedef ... refactored Ben Bacarisse <ben.usenet@bsb.me.uk> - 2020-08-16 20:08 +0100
                Re: [OT] words. Was: typedef ... refactored David Brown <david.brown@hesbynett.no> - 2020-08-17 10:50 +0200
                Re: [OT] words. Was: typedef ... refactored Ben Bacarisse <ben.usenet@bsb.me.uk> - 2020-08-17 12:00 +0100
                Re: typedef ... refactored Anton Shepelev <anton.txt@g{oogle}mail.com> - 2020-08-17 17:51 +0300
                Re: typedef ... refactored Ben Bacarisse <ben.usenet@bsb.me.uk> - 2020-08-17 16:06 +0100
                Re: typedef ... refactored Keith Thompson <Keith.S.Thompson+u@gmail.com> - 2020-08-17 10:48 -0700
                Re: typedef ... refactored scott@slp53.sl.home (Scott Lurndal) - 2020-08-17 19:36 +0000
                Re: typedef ... refactored Siri Cruise <chine.bleu@yahoo.com> - 2020-08-17 14:06 -0700
                Re: typedef ... refactored Tim Rentsch <tr.17687@z991.linuxsc.com> - 2020-08-28 03:53 -0700
                Re: typedef ... refactored Kaz Kylheku <793-849-0957@kylheku.com> - 2020-08-28 15:30 +0000
                Re: typedef ... refactored Ben Bacarisse <ben.usenet@bsb.me.uk> - 2020-08-29 02:24 +0100
                Re: typedef ... refactored Bonita Montero <Bonita.Montero@gmail.com> - 2020-08-14 18:37 +0200
                Re: typedef ... refactored Keith Thompson <Keith.S.Thompson+u@gmail.com> - 2020-08-14 10:29 -0700
                Re: typedef ... refactored David Brown <david.brown@hesbynett.no> - 2020-08-14 21:10 +0200
                Re: typedef ... refactored Keith Thompson <Keith.S.Thompson+u@gmail.com> - 2020-08-14 12:32 -0700
  Re: typedef ... refactored G G <gdotone@gmail.com> - 2020-08-09 05:44 -0700
    Re: typedef ... refactored Barry Schwarz <schwarzb@delq.com> - 2020-08-09 08:12 -0700
      Re: typedef ... refactored G G <gdotone@gmail.com> - 2020-08-09 08:32 -0700
  Re: typedef ... refactored John Bode <jfbode1029@gmail.com> - 2020-08-12 11:45 -0700
    Re: typedef ... refactored Bart <bc@freeuk.com> - 2020-08-12 19:59 +0100
    Re: typedef ... refactored Siri Cruise <chine.bleu@yahoo.com> - 2020-08-12 14:16 -0700
    Re: typedef ... refactored Tim Rentsch <tr.17687@z991.linuxsc.com> - 2020-08-28 05:05 -0700
  Re: typedef ... refactored Anton Shepelev <anton.txt@gmail.com> - 2020-08-13 00:41 +0300

csiph-web