<arpa/nameser.h>: Remove RR type classification macros [BZ #20592]

  The macros are no longer up-to-date, and the classification is not
useful.  In this particular case, removal without prior deprecation
seems the right approach.

2016-09-09  Florian Weimer  <fweimer@redhat.com>

	[BZ #20592]
	Remove RR type classification macros.
	* resolv/arpa/nameser.h (ns_t_qt_p, ns_t_mrr_p, ns_t_rr_p)
	(ns_t_udp_p, ns_t_xfr_p): Remove.
  

On 9.9.2016 13:43, Florian Weimer wrote:
> The macros are no longer up-to-date, and the classification is not
> useful.  In this particular case, removal without prior deprecation
> seems the right approach.
> 
> 2016-09-09  Florian Weimer  <fweimer@redhat.com>
> 
> 	[BZ #20592]
> 	Remove RR type classification macros.
> 	* resolv/arpa/nameser.h (ns_t_qt_p, ns_t_mrr_p, ns_t_rr_p)
> 	(ns_t_udp_p, ns_t_xfr_p): Remove.
> 
> diff --git a/NEWS b/NEWS
> index 5566125..a98ca3c 100644
> --- a/NEWS
> +++ b/NEWS
> @@ -46,6 +46,12 @@ Version 2.25
>    only implemented minimal support for the previous version of DNSSEC, which
>    is incompatible with the currently deployed version.
>  
> +* The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
> +  ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
> +  <arpa/nameser.h> header file because the distinction between RR types and
> +  meta-RR types is not officially standardized, subject to revision, and
> +  thus not suitable for encoding in a macro.
> +
>  Security related changes:
>  
>    On ARM EABI (32-bit), generating a backtrace for execution contexts which
> diff --git a/resolv/arpa/nameser.h b/resolv/arpa/nameser.h
> index f11b9f0..a866ce8 100644
> --- a/resolv/arpa/nameser.h
> +++ b/resolv/arpa/nameser.h
> @@ -283,17 +283,6 @@ typedef enum __ns_type {
>  	ns_t_max = 65536
>  } ns_type;
>  
> -/* Exclusively a QTYPE? (not also an RTYPE) */
> -#define	ns_t_qt_p(t) (ns_t_xfr_p(t) || (t) == ns_t_any || \
> -		      (t) == ns_t_mailb || (t) == ns_t_maila)
> -/* Some kind of meta-RR? (not a QTYPE, but also not an RTYPE) */
> -#define	ns_t_mrr_p(t) ((t) == ns_t_tsig || (t) == ns_t_opt)
> -/* Exclusively an RTYPE? (not also a QTYPE or a meta-RR) */
> -#define ns_t_rr_p(t) (!ns_t_qt_p(t) && !ns_t_mrr_p(t))
> -#define ns_t_udp_p(t) ((t) != ns_t_axfr && (t) != ns_t_zxfr)
> -#define ns_t_xfr_p(t) ((t) == ns_t_axfr || (t) == ns_t_ixfr || \
> -		       (t) == ns_t_zxfr)
> -
>  /*%
>   * Values for class field
>   */

It makes sense to me as well (but I did not thoroughly review the code).
  

On 09/20/2016 09:06 AM, Petr Spacek wrote:

> It makes sense to me as well (but I did not thoroughly review the code).

Thanks, committed as well.

Florian