Re: pgsql 'prefix' error - Mailing list pgsql-general

From Harald Fuchs
Subject Re: pgsql 'prefix' error
Date
Msg-id puk4xfg773.fsf@srv.protecting.net
Whole thread Raw
In response to How is the right query for this condition ?  (Bino Oetomo <bino@indoakses-online.com>)
Responses Re: pgsql 'prefix' error  (Bino Oetomo <bino@indoakses-online.com>)
List pgsql-general
In article <4B0BBC8E.6010803@indoakses-online.com>,
Bino Oetomo <bino@indoakses-online.com> writes:

> I downloaded pgfoundry's prefix, postgresql-8.3-prefix_1.0.0-1_i386.deb
> I install it using dpkg , and run the prefix.sql

> Create database .. named 'prefbino', and

> CREATE TABLE myrecords (
>  record prefix_range NOT NULL,
>  PRIMARY KEY (record)
> );


> Looks good, next
> I try to create some records, But I got this error :
> -----------------------START------------------
> prefbino=# COPY myrecords (record) FROM stdin;
> Enter data to be copied followed by a newline.
> End with a backslash and a period on a line by itself.
>>> 1
>>> 12
>>> 123
>>> 1234
>>> \.
> ERROR: duplicate key value violates unique constraint "myrecords_pkey"
> CONTEXT: COPY myrecords, line 2: "12"

> -----------------------STOP------------------

> Kindly please give me further enlightment

At least in prefix 1.0.0 unique indexes seem to be broken.  Just drop
the primary key and add a separate index:

CREATE INDEX myrecords_record_ix ON myrecords USING gist (record);

pgsql-general by date:

Previous
From: Tom Lane
Date:
Subject: Re: [HACKERS] Updating column on row update
Next
From: "Kevin Grittner"
Date:
Subject: Re: [HACKERS] Updating column on row update