On Apr 10, 12:49 am, Richard Heathfield <r...@[EMAIL PROTECTED]
> wrote:
> ymunt...@[EMAIL PROTECTED]
said:
>
> > On Apr 10, 12:24 am, Richard Heathfield <r...@[EMAIL PROTECTED]
> wrote:
> >> Juha Nieminen said:
>
> >> > Richard Heathfield wrote:
> >> >> Please
> >> >> also explain where I can find a definition for "stack-allocated
> >> >> object" in the C Standard.
>
> >> > You certainly know what I meant.
>
> >> Presumably you intended to say "objects with automatic storage
class".
>
> > There is no such thing as "objects with automatic storage class".
>
> There are five storage-class-specifier keywords: typedef, extern,
static,
> auto, and register. The typedef specifier is an anomaly, being called a
> storage-class-specifier only for reasons of syntactic convenience. The
> other four denote the storage class of an object by decorating its
> declaration. If none is present, the defaults are that a file scope
object
> has static storage duration and external linkage, and that a block scope
> object has automatic storage duration and no linkage.
Blah blah blah. Some information about what storage
class specifiers do. Still, storage duration is not
storage class (as you rightly note here, typedef is
a funny one, because storage class is a syntax thing).
You're the pedant here, remember?
> Although the Standard does not use the term "automatic storage class"
> normatively, it does use it in a footnote:
>
> * An object of automatic storage class that does not have
> volatile-qualified type has been changed between a setjmp invocation
> and a longjmp call and then has its value accessed ($4.6.2.1).
>
> My apologies for saying "with" rather than "of", but I think this is
> sufficient to demonstrate that the concept of automatic storage class
> exists, is meaningful, and is associated with objects.
Bull****. The standard is using term "storage duration"
for this concept, and it's using "storage class" only
when talking about storage class specifiers, like
"register storage class" or "storage class extern".
It's funny it says "automatic storage class" in the
annex, that's clearly a mistake/typo/nobody-cared;
but it's even more funny how you try to use that to
back up your fancy theory about a "concept of automatic
storage class", a concept which doesn't exist. If you
were talking human language, I would accept "automatic
storage class" or "automatic object", just like I
accept "allocated on stack". But you aren't, you
are talking standardese here and inventing new
"concepts". So, bull****.
"objects with automatic storage class" do not exist,
just like "global objects" do not. Or do they?
> If you're declaring
> open season on misuse of prepositions, that's fine by me, but don't you
> think it could get a little tedious?
And this is typical RH's trick: twist it. I don't
care about "with" or "of" here, as you perfectly
well understand. But you composed two paragraphs
of text about that in order to hide the fact that
you're making up your own "standard" concepts while
not letting other use "allocated on stack" which is
perfectly well understood by everybody including you.
Yevgen
|
