Re: pgsql: Introduce hash_search_with_hash_value() function - Mailing list pgsql-committers

From Pavel Stehule
Subject Re: pgsql: Introduce hash_search_with_hash_value() function
Date
Msg-id CAFj8pRBka02SrWnu0iTCTCOfBUco_A4uLNjuZj3UH0CkVJMF6Q@mail.gmail.com
Whole thread Raw
In response to Re: pgsql: Introduce hash_search_with_hash_value() function  (Alexander Korotkov <aekorotkov@gmail.com>)
Responses Re: pgsql: Introduce hash_search_with_hash_value() function
List pgsql-committers


st 7. 8. 2024 v 22:25 odesílatel Alexander Korotkov <aekorotkov@gmail.com> napsal:
On Wed, Aug 7, 2024 at 1:34 PM Pavel Stehule <pavel.stehule@gmail.com> wrote:
> st 7. 8. 2024 v 12:22 odesílatel Alexander Korotkov <aekorotkov@gmail.com> napsal:
>> Thank you.
>> Please see 40064a8ee1 takes special efforts to match HTAB hash
>> function to syscache hash function.  By default, their hash values
>> don't match and you can't simply use syscache hash value to search
>> HTAB.  This probably should be mentioned in the header comment of
>> hash_seq_init_with_hash_value().
>
>
> yes, enhancing doc should be great + maybe assert

Please check the patch, which adds a caveat to the function header
comment.  I don't particularly like an assert here, because there
could be use-cases besides syscache callbacks, which could legally use
default hash function.

looks well

Regards

Pavel

------
Regards,
Alexander Korotkov
Supabase

pgsql-committers by date:

Previous
From: Michael Paquier
Date:
Subject: Re: pgsql: Fix more holes with SLRU code in need of int64 for segment numbe
Next
From: Peter Eisentraut
Date:
Subject: pgsql: Revert ECPG's use of pnstrdup()