placement of volatile in typedef of struct

[Joe Pizzi]

Is there any difference between the following two definitions? If so,
can you explain them to me?

typedef volatile struct
{
int x;
} VS_Def;

typedef struct
{
int x;
} volatile SV_Def;

Thanks.

[Jack Klein]

On 6 Feb 2004 16:56:41 -0800, jo************@ lmco.com (Joe Pizzi)
wrote in comp.lang.c:

Is there any difference between the following two definitions? If so,
can you explain them to me?

typedef volatile struct
{
int x;
} VS_Def;

typedef struct
{
int x;
} volatile SV_Def;

Thanks.

They are both extremely bad programming practice. CVR qualifiers and
in most cases pointers should NEVER be hidden behind typedefs. They
rapidly become a maintenance nightmare.

Beyond that, the first one is legal C and the second is a syntax
error.

--
Jack Klein
Home: http://JK-Technology.Com
FAQs for
comp.lang.c http://www.eskimo.com/~scs/C-faq/top.html
comp.lang.c++ http://www.parashift.com/c++-faq-lite/
alt.comp.lang.l earn.c-c++
http://www.contrib.andrew.cmu.edu/~a...FAQ-acllc.html

[Leor Zolman]

On Sat, 07 Feb 2004 04:02:00 GMT, Jack Klein <ja*******@spam cop.net>
wrote:

On 6 Feb 2004 16:56:41 -0800, jo************@ lmco.com (Joe Pizzi)
wrote in comp.lang.c:

Is there any difference between the following two definitions? If so,
can you explain them to me?

typedef volatile struct
{
int x;
} VS_Def;

typedef struct
{
int x;
} volatile SV_Def;

Thanks.

They are both extremely bad programming practice. CVR qualifiers and
in most cases pointers should NEVER be hidden behind typedefs. They
rapidly become a maintenance nightmare.

Beyond that, the first one is legal C and the second is a syntax
error.

Not sure I see how it is a syntax error; 6.7/1 doesn't seem to impose
any ordering requirements on all the different flavors of declaration
specifiers, so does it matter whether CV qualifiers come before or
after struct-or-union-specifiers?

Comeau doesn't even raise a warning on it.

That aside, I agree wholeheartedly with the advice against putting CV
qualifiers (what's the "R"?) anywhere except in plain view near the
declarator they appy to.
-leor

Leor Zolman
BD Software
le**@bdsoft.com
www.bdsoft.com -- On-Site Training in C/C++, Java, Perl & Unix
C++ users: Download BD Software's free STL Error Message
Decryptor at www.bdsoft.com/tools/stlfilt.html

[Leor Zolman]

On Sat, 07 Feb 2004 04:21:07 GMT, Leor Zolman <le**@bdsoft.co m> wrote:

CV qualifiers (what's the "R"?)
Found it, restrict. Haven't used that one much ;-)
-leor
Leor Zolman
BD Software
le**@bdsoft.co m
www.bdsoft.com -- On-Site Training in C/C++, Java, Perl & Unix
C++ users: Download BD Software's free STL Error Message
Decryptor at www.bdsoft.com/tools/stlfilt.html

Leor Zolman
BD Software
le**@bdsoft.com
www.bdsoft.com -- On-Site Training in C/C++, Java, Perl & Unix
C++ users: Download BD Software's free STL Error Message
Decryptor at www.bdsoft.com/tools/stlfilt.html

[Jack Klein]

On Sat, 07 Feb 2004 04:21:07 GMT, Leor Zolman <le**@bdsoft.co m> wrote
in comp.lang.c:

On Sat, 07 Feb 2004 04:02:00 GMT, Jack Klein <ja*******@spam cop.net>
wrote:

On 6 Feb 2004 16:56:41 -0800, jo************@ lmco.com (Joe Pizzi)
wrote in comp.lang.c:

Is there any difference between the following two definitions? If so,
can you explain them to me?

typedef volatile struct
{
int x;
} VS_Def;

typedef struct
{
int x;
} volatile SV_Def;

Thanks.
They are both extremely bad programming practice. CVR qualifiers and
in most cases pointers should NEVER be hidden behind typedefs. They
rapidly become a maintenance nightmare.

Beyond that, the first one is legal C and the second is a syntax
error.

Not sure I see how it is a syntax error; 6.7/1 doesn't seem to impose
any ordering requirements on all the different flavors of declaration
specifiers, so does it matter whether CV qualifiers come before or
after struct-or-union-specifiers?

Comeau doesn't even raise a warning on it.

OK, you caught me. Looking at the grammar, I think it is probably
quite legal. Just this once I threw it into a compiler and watched
what happened, and noticed that it kicked out an error message. That
and the prohibition against more than one storage class specifier
appearing in a declaration.
That aside, I agree wholeheartedly with the advice against putting CV
qualifiers (what's the "R"?) anywhere except in plain view near the
declarator they appy to.
-leor

I see from your own follow-up that you found it applied to the new C99
keyword restrict.

--
Jack Klein
Home: http://JK-Technology.Com
FAQs for
comp.lang.c http://www.eskimo.com/~scs/C-faq/top.html
comp.lang.c++ http://www.parashift.com/c++-faq-lite/
alt.comp.lang.l earn.c-c++
http://www.contrib.andrew.cmu.edu/~a...FAQ-acllc.html

[Leor Zolman]

On Sat, 07 Feb 2004 05:32:42 GMT, Jack Klein <ja*******@spam cop.net>
wrote:

Not sure I see how it is a syntax error; 6.7/1 doesn't seem to impose
any ordering requirements on all the different flavors of declaration
specifiers, so does it matter whether CV qualifiers come before or
after struct-or-union-specifiers?

Comeau doesn't even raise a warning on it.

OK, you caught me. Looking at the grammar, I think it is probably
quite legal.

I watched the original post sit there for hours, and wanted to reply
that the two statements were equivalent, but I'd gotten just a bit
gun-shy after several recent gaffes. Finally, seeing no one had
responded, I looked it up in the standard to my own satisfaction,
wrote it up, and was about to send it off when your post showed up. So
then I went back to the drawing board and studied it some more....

That I saw this straight off I probably have Dan Saks to thank for.
He wrote a 5-day Advanced C course last year in direct response to a
call I put out for one; in that course, he has students write their
own C declaration parsers, building them up bit-by-bit until they
handle stuff that gets pretty darned hairy. This is somewhat congruent
to the series of articles he wrote for CUJ in an attempt to de-mystify
C++ declaration syntax.

Let's just say that I never thought the day would come that I'd
actually be able to remember the difference between
declaration-specifiers and declarators, but after teaching that
course, even a year later, I still do remember. Amazing... ;-)
-leor