Re: Unusual behaviour with intarray - Mailing list pgsql-general

From Teodor Sigaev
Subject Re: Unusual behaviour with intarray
Date
Msg-id 3EA7EBBB.4050501@sigaev.ru
Whole thread Raw
In response to Re: Unusual behaviour with intarray  (Teodor Sigaev <teodor@sigaev.ru>)
List pgsql-general
Please apply patch for 7.3 CVS brach. Patch fixes the problem, for 7.4 it will
be later (with other changes).


Teodor Sigaev wrote:
> Ok, this is a bug.
> I'll post a patch as soon as possible.
>
> Adam Witney wrote:
>
>> Hi,
>>
>> I am using intarray in contrib, and have seen some strange behaviour with
>> one of the operations. Anyone know whats going on here?
>>
>> cabbage=# create table test (id int, writers int[]);
>> CREATE TABLE
>> cabbage=# insert into test values(1, '{}');
>> INSERT 1545023 1
>> cabbage=# select * from test;
>>  id | writers ----+---------
>>   1 | {}
>> (1 row)
>>
>> cabbage=# update test set writers = (select writers - '{1}'::int[]);
>> UPDATE 1
>> cabbage=# select * from test;
>>  id | writers ----+---------
>>   1 | {}
>> (1 row)
>>
>> cabbage=# update test set writers = (select writers - '{1}'::int[]);
>> UPDATE 1
>> cabbage=# select * from test;
>>  id |  writers  ----+-----------
>>   1 | {3044207}
>> (1 row)
>>
>> Thanks for any help
>>
>> Adam
>>
>>
>

--
Teodor Sigaev                                  E-mail: teodor@sigaev.ru

Attachment

pgsql-general by date:

Previous
From: Devrim GUNDUZ
Date:
Subject: Re: Maximum number of connections
Next
From: Thomas Kellerer
Date:
Subject: Re: Maximum number of connections