Re: pgsql: Fix headerscheck failure in replication/worker_internal.h

Stephen Frost <sfrost@snowman.net>

From: Stephen Frost <sfrost@snowman.net>
To: Alvaro Herrera <alvherre@alvh.no-ip.org>
Cc: pgsql-hackers@lists.postgresql.org
Date: 2021-11-16T17:14:31Z
Lists: pgsql-hackers
Greetings,

* Alvaro Herrera (alvherre@alvh.no-ip.org) wrote:
> On 2021-Nov-16, Alvaro Herrera wrote:
> 
> > Fix headerscheck failure in replication/worker_internal.h
> 
> The other failure is in src/include/libpq/be-gssapi-common.h:
> 
> In file included from /tmp/headerscheck.a6f0um/test.c:2:
> ./src/include/libpq/be-gssapi-common.h:20:10: fatal error: gssapi/gssapi.h: No existe el fichero o el directorio
>    20 | #include <gssapi/gssapi.h>
>       |          ^~~~~~~~~~~~~~~~~
> compilation terminated.
> 
> One possible solution for this one is to add an exclusion in
> headerscheck (and cpluspluscheck?); the other possible solution seems to
> be to wrap the whole contents of the file in "#ifdef ENABLE_GSS".  I
> think the latter is roughly the approach used for OpenSSL inclusions.

Short answer is yes, inclusion of be-gssapi-common.h is typically
wrapped in a #ifdef, see src/backend/libpq/auth.c

Thanks,

Stephen

Commits

  1. Harden be-gssapi-common.h for headerscheck

  2. Fix headerscheck failure in replication/worker_internal.h