Re: [Gnumed-bugs] <bug>: Folinine is surely not there in database. Y

[Karsten Hilbert]

On Mon, May 27, 2013 at 01:47:00PM -0400, Dr.Vaibhav Banait wrote:

> client version: 1.3.3
>
> user comment  : Folinine is surely not there in database. You have already 
> seen this log before.

If what you want to say is that you see GNUmed reporting
something that is truly not in the database then you will
have to report a bug to the PostgreSQL project because the
database clearly says it very much IS there:

> 2013-05-27 23:15:50  ERROR     gm.db 
> (d:\workplace\gnumed-client.1.3.3\build\pyi.win32\gnumed\outpyz1.pyz\gnumed.pycommon.gmpg2::run_rw_queries()
>  #1208): RW query failed: [INSERT INTO ref.branded_drug (description, 
> preparation) VALUES (E'Folinine', E'capsule') RETURNING pk]
> 2013-05-27 23:15:50  ERROR     gm.db 
> (d:\workplace\gnumed-client.1.3.3\build\pyi.win32\gnumed\outpyz1.pyz\gnumed.pycommon.gmpg2::run_rw_queries()
>  #1211): PG error code: 23505
> 2013-05-27 23:15:50  ERROR     gm.db 
> (d:\workplace\gnumed-client.1.3.3\build\pyi.win32\gnumed\outpyz1.pyz\gnumed.pycommon.gmpg2::run_rw_queries()
>  #1213): PG error message: ERROR:  duplicate key value violates unique 
> constraint "substance_brand_pkey"

What you seem to be saying is that you assert that
PostgreSQL is reporting a unique key violation which does
not exist in reality.

This would be a bug in PostgreSQL (which I am hard pressed
to believe).

Given the fact that we haven't yet cleared up the suspicion
that your database is corrupted (due to not having provided
all the data I asked for) I would seem to think there's
hardly any way to tell for sure.

The first step would be to reindex your database because
index corruption might make PostgreSQL *think* there's
duplication when there is not. I don't know whether
reindexing will do anything to your otherwise-possibly-corrupted
database state so I would not recommend it just so.

Regards,
Karsten
-- 
GPG key ID E4071346 @ gpg-keyserver.de
E167 67FD A291 2BEA 73BD  4537 78B9 A9F9 E407 1346