Re: Compiling table_log for PG 11 / attisdropped

From: Adrian Klaver <adrian(dot)klaver(at)aklaver(dot)com>
To: Ken Tanzer <ken(dot)tanzer(at)gmail(dot)com>, PG-General Mailing List <pgsql-general(at)postgresql(dot)org>
Subject: Re: Compiling table_log for PG 11 / attisdropped
Date: 2019-07-12 20:42:26
Message-ID: 3a2d53d8-8aca-5978-7e87-7ad9f961c93a@aklaver.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

On 7/12/19 1:19 PM, Ken Tanzer wrote:
> Hi.  I'm trying to update some databases from 9.6 to 11, and they use
> table_log for tracking changes to tables.
>
> I started with the most recent version I could find, a fork labeled
> version 0.5.  [1]

What OS.version and compiler/version?

Some digging found the below, which is similar:

https://trac.osgeo.org/postgis/ticket/3814

https://trac.osgeo.org/postgis/changeset/15653
>
> This version compiles on both 9.6, and 10, but on 11 it fails with these
> errors:
>
> table_log.c: In function ‘count_columns’:
> table_log.c:80:27: error: invalid type argument of ‘->’ (have
> ‘FormData_pg_attribute’)
>    if (!tupleDesc->attrs[i]->attisdropped)

If I am following the issue correctly then the above becomes:

if (!tupleDesc->attrs[i].attisdropped)

>                            ^
> table_log.c: In function ‘__table_log’:
> table_log.c:339:56: error: invalid type argument of ‘->’ (have
> ‘FormData_pg_attribute’)
>     if (trigdata->tg_relation->rd_att->attrs[col_nr - 1]->attisdropped)
>                                                         ^
> table_log.c:373:56: error: invalid type argument of ‘->’ (have
> ‘FormData_pg_attribute’)
>     if (trigdata->tg_relation->rd_att->attrs[col_nr - 1]->attisdropped)

and above becomes:

*[col_nr - 1].attisdropped

>
> Can anyone shed any light on these errors, and/or how they might be fixed?
>
> (I did try filing an issue with this project [2], but since there is no
> activity there I'm turning to this mailing list.)
> (Also, I know table_log is quite old and there are better alternatives.
> If I were starting fresh I would surely use them.  But it works for my
> simple needs, and I have a lot of legacy data accumulated, so I would
> really like to avoid switching to a new or different logging system.)
>
> Thanks in advance,
> Ken
>
> [1] Table log: https://github.com/psoo/table_log
> [2] Issue filed: https://github.com/psoo/table_log/issues/2
> [3] table_log.c: https://github.com/psoo/table_log/blob/master/table_log.c
>
>
>
> --
> AGENCY Software
> A Free Software data system
> By and for non-profits
> /http://agency-software.org//
> /https://demo.agency-software.org/client/
> ken(dot)tanzer(at)agency-software(dot)org <mailto:ken(dot)tanzer(at)agency-software(dot)org>
> (253) 245-3801
>
> Subscribe to the mailing list
> <mailto:agency-general-request(at)lists(dot)sourceforge(dot)net?body=subscribe> to
> learn more about AGENCY or
> follow the discussion.

--
Adrian Klaver
adrian(dot)klaver(at)aklaver(dot)com

In response to

Responses

Browse pgsql-general by date

  From Date Subject
Next Message Benedict Holland 2019-07-12 21:08:44 Re: Machine learning top of db logs
Previous Message Ken Tanzer 2019-07-12 20:19:43 Compiling table_log for PG 11 / attisdropped