From: | Jonathan Vanasco <postgres(at)2xlp(dot)com> |
---|---|
To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
Cc: | pgsql general <pgsql-general(at)postgresql(dot)org> |
Subject: | Re: bug in schema export? |
Date: | 2007-04-17 23:49:26 |
Message-ID: | 7ED3BB4D-BB69-47B1-912B-71885850BEB3@2xlp.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
On Apr 17, 2007, at 2:51 PM, Tom Lane wrote:
> Jonathan Vanasco <postgres(at)2xlp(dot)com> writes:
>> i believe i found a bug
>> psql> create table test_failure ( id integer primary key );
>> psql> alter table test_failure alter id rename to id_new ; # not
>> relevant, just illustrating the reason how i got to step 2
>> psql> alter table test_failure_pkey rename to
>> test_failure_id_new_pkey;
>
>> $ pg_dump --schema-only -ttest_failure -Utest > test.sql
>
>> the pkey in the schema is added as 'test_failure_pkey' , not
>> 'test_failure_id_new_pkey';
>
> You didn't rename the constraint. If we were going to do anything
> about
> this, it'd be to forbid directly renaming indexes that're owned by
> constraints. However, since we don't currently have a RENAME
> CONSTRAINT
> command, the effect of that would be that you couldn't do it at
> all ...
>
> In the meantime you can probably do an UPDATE on pg_constraint if you
> really care.
Extremely annoying.
In the example above,
\d test_failure
will not show a separate constraint named by the old value. it only
shows the current name of the index.
Either renaming primary keys should be disabled, or that constraint
forced to be visible.
Under the current system, there is no intuitive way for someone to be
aware of an old constraint on a renamed index like this.
I discovered this while trying to re-merge 2 databases that were
split onto a new bigger cluster and having them complain about
conflicting constraints.
// Jonathan Vanasco
| - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
- - - - - - - - - - - - - - - - - - -
| SyndiClick.com
| - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
- - - - - - - - - - - - - - - - - - -
| FindMeOn.com - The cure for Multiple Web Personality Disorder
| Web Identity Management and 3D Social Networking
| - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
- - - - - - - - - - - - - - - - - - -
| RoadSound.com - Tools For Bands, Stuff For Fans
| Collaborative Online Management And Syndication Tools
| - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
- - - - - - - - - - - - - - - - - - -
From | Date | Subject | |
---|---|---|---|
Next Message | Bruce Momjian | 2007-04-17 23:56:53 | Re: Status of Postgres 8.2.4 and pg_standby |
Previous Message | CAJ CAJ | 2007-04-17 23:40:00 | Status of Postgres 8.2.4 and pg_standby |