From: | Richard Huxton <dev(at)archonet(dot)com> |
---|---|
To: | Michal Hlavac <hlavki(at)medium13(dot)sk> |
Cc: | pgsql-general(at)postgresql(dot)org |
Subject: | Re: ltree bug handling nulls (was Stored procedure failure) |
Date: | 2004-08-26 10:46:18 |
Message-ID: | 412DBF7A.50006@archonet.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Michal Hlavac wrote:
> Richard Huxton wrote:
>
>>
>> Looking at the 8.0beta source, the functions (...ltree_isparent) are
>> all marked strict so they should just return null on a null parameter.
>>
>> What happens if you set my_path to some non-null but un-matched value
>> before the problem line?
>>
>
> When my_path have non-null value, everything is OK...
>
> I don't understand why one null value would crash the server???
Because it attempts to follow a pointer that isn't there.
If you installed from source, can you check contrib/ltree/ltree.sql.in
and check the "isstrict" attribute is set:
CREATE FUNCTION ltree_isparent(ltree,ltree)
RETURNS bool
AS 'MODULE_PATHNAME'
LANGUAGE 'C' WITH (isstrict,iscachable);
If it is, then you'll have to wait for a developer to take an interest,
I'm stumped. Shouldn't take long, but if you wanted to be thorough you
could report a bug via the bugs mailing list or the website.
--
Richard Huxton
Archonet Ltd
From | Date | Subject | |
---|---|---|---|
Next Message | Robert Deme | 2004-08-26 10:49:18 | unsubscribe |
Previous Message | Michal Hlavac | 2004-08-26 10:23:25 | Re: Stored procedure failure |