«

In article <u0fqs9->,
Russell Shaw <rjshawN_o@s_pam.netspace.net.au> wrote:
>Hi,
>In gcc-4.7 C99, i get an error (in a function scope):
>
> struct {
> int a;
> } sa;
>
> struct {
> int a;
> } sb;
>
> sb = sa;
>
>error: incompatible types when assigning to type 'struct <anonymous>' from type
>'struct <anonymous>'

You're asking rather a lot for the compiler to recognize that two different
structures just happen to have been defined the same way.

Try this:

struct foo {
int a;
};

struct foo sa;
struct foo sb;

sb = sa;

--
-Ed Falk,
http://thespamdiaries.blogspot.com/

Edward A. Falk <> wrote:
> In article <u0fqs9->,
> Russell Shaw <rjshawN_o@s_pam.netspace.net.au> wrote:

>>In gcc-4.7 C99, i get an error (in a function scope):

>> struct {
>> int a;
>> } sa;

>> struct {
>> int a;
>> } sb;

>> sb = sa;

>>error: incompatible types when assigning to type 'struct <anonymous>' from type
>>'struct <anonymous>'

> You're asking rather a lot for the compiler to recognize that two different
> structures just happen to have been defined the same way.

Well, it doesn't have to 'just' recognize it, but only test when it
finds an actual operation.

I believe that PL/I can do it. I am not sure what happens if you give
PL/I a structure expression of incompatible structures. My guess is
that it converts as appropriate.

DCL 1 SA, 2 A FIXED BINARY(31,0);
DCL 1 SB, 2 A FIXED BINARY(31,0);

I am pretty sure that PL/I will let you do either

SA=SB;

or even

SA=SA+SB;

and, for extra challange:

DCL 1 SC, 2 A FIXED BINARY(31,0);
DCL 1 SD, 2 A FLOAT BINARY(53);

SD=SD+SQRT(SC);

-- glen

On 01/19/2013 06:41 PM, Edward A. Falk wrote:
> In article <u0fqs9->,
> Russell Shaw <rjshawN_o@s_pam.netspace.net.au> wrote:
>> Hi,
>> In gcc-4.7 C99, i get an error (in a function scope):
>>
>> struct {
>> int a;
>> } sa;
>>
>> struct {
>> int a;
>> } sb;
>>
>> sb = sa;
>>
>> error: incompatible types when assigning to type 'struct <anonymous>' from type
>> 'struct <anonymous>'
>
> You're asking rather a lot for the compiler to recognize that two different
> structures just happen to have been defined the same way.

Perhaps - yet that is precisely what the standard would have required if
they had been declared in two different translation units; it's only
because they are in the same translation unit that they're not
compatible (6.2.7p1).
--
James Kuyper

On 01/19/2013 06:41 PM, Edward A. Falk wrote:
> In article <u0fqs9->,
> Russell Shaw <rjshawN_o@s_pam.netspace.net.au> wrote:
>> Hi,
>> In gcc-4.7 C99, i get an error (in a function scope):
>>
>> struct {
>> int a;
>> } sa;
>>
>> struct {
>> int a;
>> } sb;
>>
>> sb = sa;
>>
>> error: incompatible types when assigning to type 'struct <anonymous>' from type
>> 'struct <anonymous>'
>
> You're asking rather a lot for the compiler to recognize that two different
> structures just happen to have been defined the same way.

Perhaps - yet that is precisely what the standard would have required if
they had been declared in two different translation units; it's only
because they are in the same translation unit that they're not
compatible (6.2.7p1).
--
James Kuyper

James Kuyper <> writes:

> On 01/19/2013 06:41 PM, Edward A. Falk wrote:
>> In article <u0fqs9->,
>> Russell Shaw <rjshawN_o@s_pam.netspace.net.au> wrote:
>>> Hi,
>>> In gcc-4.7 C99, i get an error (in a function scope):
>>>
>>> struct {
>>> int a;
>>> } sa;
>>>
>>> struct {
>>> int a;
>>> } sb;
>>>
>>> sb = sa;
>>>
>>> error: incompatible types when assigning to type 'struct
>>> <anonymous>' from type 'struct <anonymous>'
>>
>> You're asking rather a lot for the compiler to recognize that
>> two different structures just happen to have been defined the
>> same way.
>
> Perhaps - yet that is precisely what the standard would have
> required if they had been declared in two different translation
> units; [snip]

I expect M. Kuyper means something different from what he is
saying. C compilers don't even try to determine whether structs
declared in different translation units are defined the same way,
and certainly the Standard doesn't require them to do so.