NetNews Offline 2/NetNews Offline Volume 2.iso/news/comp/lang/c-part1/5815

Internet Message Format | 1996-08-05 | 3.1 KB

Path: keats.ugrad.cs.ubc.ca!not-for-mail
From: [e-mail] (Kazimir Kylheku)
Newsgroups: comp.lang.c
Subject: Re: Why does this work?
Date: 20 Feb 1996 11:36:39 -0800
Organization: Computer Science, University of B.C., Vancouver, B.C., Canada
Message-ID: <[e-mail]>
References: <4g8f7o$[e-mail]>
NNTP-Posting-Host: keats.ugrad.cs.ubc.ca

In article <4g8f7o$[e-mail]>,
Jim Rosinski <[e-mail]> wrote:
>
>Could someone please explain why the following code works?  In particular, I
>am perplexed as to why "sub1" and "sub2" declared as elements of "cmndtable"
>are valid "pointers to function returning void" (i.e. see the definition of
>"cmndstruct").  When executed, the net result is indeed to invoke two
>functions, "sub1", and "sub2" (tested on Solaris, AIX, and UNICOS).

It works because the initializers are just assigned to whatever available
fields there are in the structure. I guess what you are asking is why you
weren't forced to write the initializer as { { sub1 }, { sub 2 }, { NULL } } to
indicate that the elements are part of the structure, right?

Your usage might not be standard, even though it's silently accepted by three
compilers you know. K&R2 says, on p. 219, that

    "The initializer for a structure is is either an expression of
    the same type, or a brace-enclosed list of initializers for its
    members in order.  Unnamed bit-field members are ignored, and
    are not initialized.  If there are fewer initializers in the
    list than members of the structure, the trailing members are
    initialized with 0.  There may not be more initializers than
    members."

So far, this would indicate that your code is not compliant, unless the
function pointer's type is compatible with a single-membered structure
containing a function pointer (not true).

However, a few paragraphs later, the K&R talks about more rules regarding
aggregate initializers:

    "An _aggregate_ is a structure or array. If an aggregate contains
    members of aggregate type, the initialization rules apply recursively.
    Braces may be elided in the initialization as follows: if the
    initializer for an aggregate's member that is itself an aggregate
    begins with a left brace, the the succeeding comma-separated list of
    initializers initializes members of the subaggregate; it is erroneous
  *>    for there to be more initializers than members. If, however, the
  *>    initializer for a subaggregate does not begin with a left brace, then
  *>    only enough elements from the list are taken to account for the the
  *>    members of the subaggregate; any remaining members are left to
  *>    initialize the next member of the aggregate of which the subaggregate
  *>    is part.

The last part refers precisely to your construct. You did not begin the
structure initializers with a left brace, hence the  members of the initializer
list are simply handed to the next available fields in the next member of the
outer aggregate (i.e. next structure in the array).

Of course, K&R2 is _not_ a subsititue for the ANSI/ISO standard, but it's the
best I can do.  Dan Pop, the keeper of the standard, might have some additional
pearls of wisdom to add.

[ code omitted ]
--