From mboxrd@z Thu Jan  1 00:00:00 1970
MIME-Version: 1.0
In-Reply-To: <CAOw7k5jfUXv4hWn0VRoe1c4XCtVu+2VrPmtgZ3kNKPNtLbTPzg@mail.gmail.com>
References: <CAFUtaNGtWHwKHBfup5fgn1nSuaxOxDx_cw2o6ChrXuZJqzkqPw@mail.gmail.com>
	<CAOw7k5jfUXv4hWn0VRoe1c4XCtVu+2VrPmtgZ3kNKPNtLbTPzg@mail.gmail.com>
Date: Fri, 15 Mar 2013 13:09:13 -0400
Message-ID: <CAFUtaNGvtD+2X9aoHW8NWRyfh2u4jPfmy50ssvshq983FJbGww@mail.gmail.com>
From: "Joel C. Salomon" <joelcsalomon@gmail.com>
To: Fans of the OS Plan 9 from Bell Labs <9fans@9fans.net>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Subject: Re: [9fans] ISO C and typestr
Topicbox-Message-UUID: 2a9cba3e-ead8-11e9-9d60-3106f5b1d025

On Fri, Mar 15, 2013 at 1:01 PM, Charles Forsyth
<charles.forsyth@gmail.com> wrote:
> On 15 March 2013 16:52, Joel C. Salomon <joelcsalomon@gmail.com> wrote:
>> BTW, the C11 standard includes a restricted form of 9c's anonymous
>> sub-structs (with no pointer conversion).
>
> isn't the pointer conversion most of the point of them?

Given this code:

typedef struct Foo Foo;
struct Foo {
    int flag;
    Lock;
} *foo;

C11 allows you to write `lock(foo->Lock)`, or to directly access named
members of `struct Lock` as if they were members of `struct Foo`. An
early draft of the feature did allow `lock(foo)`, but that was
rejected (I don't know why).

=E2=80=94Joel