[pgpool-hackers: 189] Re: compile error in pgpool-recovey

[Tatsuo Ishii]

I think pgpool-walrecrunning() is not used any more.
--
Tatsuo Ishii
SRA OSS, Inc. Japan
English: http://www.sraoss.co.jp/index_en.php
Japanese: http://www.sraoss.co.jp

> Thanks Ishii. However there is another compile issue with
> pgpool-walrecrunning that I failed to notice it earlier. Enum WalRcvState
> has been changed in postgres 9.3 as well and it has removed WALRCV_RUNNING
> state. Instead added WALRCV_STREAMING state.
> 
> Please see the attached patch to fix that. This will make
> it compile-able but I am not sure if its an appropriate change. Because
> according to WalRcvRunning() function in postgres "if (state !=
> WALRCV_STOPPED)" means its running but since previously only
> WALRCV_RUNNING was being checked, I have tested the WALRCV_STREAMING state
> only.
> 
> Regards,
> --Asif
> 
> On Wed, Apr 3, 2013 at 12:53 PM, Tatsuo Ishii <ishii at postgresql.org> wrote:
> 
>> I have made slight change to your patch because of compiler
>> warnings(htup_details.h needed). Also I think we shoudl apply the
>> patch to V3.0 or higher as well because people might want to use the
>> pgpool version and PostgreSQL 9.3 together.
>>
>> I'm waiting for git.postgresql.org allows to push again...
>> --
>> Tatsuo Ishii
>> SRA OSS, Inc. Japan
>> English: http://www.sraoss.co.jp/index_en.php
>> Japanese: http://www.sraoss.co.jp
>>
>> > Looks good. Thanks.
>> > --
>> > Tatsuo Ishii
>> > SRA OSS, Inc. Japan
>> > English: http://www.sraoss.co.jp/index_en.php
>> > Japanese: http://www.sraoss.co.jp
>> >
>> >> Sure. Here is the updated patch.
>> >>
>> >> Regards,
>> >> --Asif
>> >>
>> >> On Tue, Apr 2, 2013 at 5:33 PM, Tatsuo Ishii <ishii at postgresql.org>
>> wrote:
>> >>
>> >>> In this you can put ifdef to adopt different version of PostgreSQL
>> >>> source codes. Can you please try in that way? Following is the similar
>> >>> technique used in sql/pgpool-regclass/pgpool-regclass.c.
>> >>>
>> >>> #if !defined(PG_VERSION_NUM) || (PG_VERSION_NUM < 90200)
>> >>>         result = RangeVarGetRelid(rel, true);
>> >>> #else
>> >>>         /* RangeVarGetRelid() of PostgreSQL 9.2 or later, has third
>> >>>          * argument "missing_ok" which suppresses ERROR exception, but
>> >>>          * returns invlaid_oid. See include/catalog/namespace.h */
>> >>>         result = RangeVarGetRelid(rel, true, true);
>> >>> #endif
>> >>> --
>> >>> Tatsuo Ishii
>> >>> SRA OSS, Inc. Japan
>> >>> English: http://www.sraoss.co.jp/index_en.php
>> >>> Japanese: http://www.sraoss.co.jp
>> >>>
>> >>>
>> >>> > Yes its not yet released. I am compiling against latest postgres
>> source
>> >>> > i.e. 9.3devel master branch.
>> >>> > Since pgpool-II master branch is also yet under development I
>> thought it
>> >>> > should be able to compile with it.
>> >>> >
>> >>> > Regards,
>> >>> > --Asif
>> >>> > On Apr 2, 2013 4:50 PM, "Tatsuo Ishii" <ishii at postgresql.org> wrote:
>> >>> >
>> >>> >> > Here is the patch to fix a compilation error I faced while
>> compiling
>> >>> the
>> >>> >> > pgpool2/sql/pgpool-recovery. I faced the following error:
>> >>> >> >
>> >>> >> > pgpool-recovery.c:201:2: error: too few arguments to function
>> >>> >> > ‘LookupExplicitNamespace’
>> >>> >> >
>> >>> >> > I looked into it and found out that the signature
>> >>> >> > of LookupExplicitNamespace function has been changed in the
>> postgres
>> >>> >> code.
>> >>> >>
>> >>> >> What version of PostgreSQL are you compiling pgpool-recovery.c
>> >>> >> against? I suspect you are compiling it against not-yet-released
>> >>> >> version of PostgreSQL.
>> >>> >> --
>> >>> >> Tatsuo Ishii
>> >>> >> SRA OSS, Inc. Japan
>> >>> >> English: http://www.sraoss.co.jp/index_en.php
>> >>> >> Japanese: http://www.sraoss.co.jp
>> >>> >>
>> >>>
>> > _______________________________________________
>> > pgpool-hackers mailing list
>> > pgpool-hackers at pgpool.net
>> > http://www.pgpool.net/mailman/listinfo/pgpool-hackers
>>