clang has parsed initializers for quite some time.
Semantic analysis for structure initialization is still not done
(there is a FIXME in the code).
clang should avoid producing bogus diagnostics. I'd prefer to have no
diagnostics until we get around to doing the proper checking.
I'll try and look at this next week,
snaroff
On Oct 26, 2007, at 7:15 PM, Keith Bauer wrote:
> ... or were initializers not even being parsed before?
>
> CookieJar:~/Desktop keith$ cat > structinit.c
> struct s { int i; };
>
> struct s a[1] =
> {
> { 3 }
> };
> CookieJar:~/Desktop keith$ clang -fsyntax-only structinit.c
> structinit.c:5:7: error: incompatible types assigning 'int' to
> 'struct s'
> { 3 }
> ^
> structinit.c:5:7: warning: excess elements in array initializer
> { 3 }
> ^
> 2 diagnostics generated.
> CookieJar:~/Desktop keith$ gcc -fsyntax-only -Wall -Wextra
> structinit.c
> CookieJar:~/Desktop keith$
>
> -Keith
> _______________________________________________
> cfe-dev mailing list
> [email protected]
> http://lists.cs.uiuc.edu/mailman/listinfo/cfe-dev
_______________________________________________
cfe-dev mailing list
[email protected]
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-dev
