Lorenzo Bettini wrote:
> I tried it and got no problem, but as far as I understand this is not a
> proof, since
>
> #if defined HAVE_DECL_STRDUP && !HAVE_DECL_STRDUP && !defined strdup
> /* Duplicate S, returning an identical malloc'd string. */
> extern char *strdup (const char *s);
> #endif
>
> so the extern definition is not parsed, since it finds the definition of
> strdup in string.h (included in strdup.h).
Yes, the absence of a failure doesn't prove anything.
> However, I tried this test program:
>
> #include <iostream>
>
> extern char *strdup (const char *s);
>
> int main()
> {
> char *copy = strdup("foobar");
> std::cout << copy << std::endl;
>
> return 0;
> }
>
> and got no problem
This doesn't prove anything either: <iostream> may include <string.h>, which
may have an extern "C" declaration for strdup.
Can you tell us the list of .h files that you want to use from C++? It's
easier to make them C++ safe than to consider when exactly the extern "C"
will be needed or not.
Bruno
