Re: Missing free_var() at end of accum_sum_final()?

From: Peter Eisentraut <peter(dot)eisentraut(at)enterprisedb(dot)com>
To: Joel Jacobson <joel(at)compiler(dot)org>, Dean Rasheed <dean(dot)a(dot)rasheed(at)gmail(dot)com>
Cc: Michael Paquier <michael(at)paquier(dot)xyz>, pgsql-hackers(at)lists(dot)postgresql(dot)org, Andres Freund <andres(at)anarazel(dot)de>
Subject: Re: Missing free_var() at end of accum_sum_final()?
Date: 2023-03-02 08:32:44
Message-ID: dabadcf6-1bcc-c275-56ab-07bd2a93870c@enterprisedb.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On 20.02.23 23:16, Joel Jacobson wrote:
> In the new attached patch, Andres fixed buffer idea has been implemented
> throughout the entire numeric.c code base.

I think the definition of the "preinitialized constants" could be
adapted to this as well. For example, instead of

static const NumericDigit const_one_data[1] = {1};
static const NumericVar const_one =
{1, 0, NUMERIC_POS, 0, NULL, (NumericDigit *) const_one_data};

it could be something like

static const NumericVar const_one =
{0, 0, NUMERIC_POS, 0, NULL, NULL, 1, {1}};

Or perhaps with designators:

static const NumericVar const_one =
{.sign = NUMERIC_POS, .buflen = 1, .fixed_buf = {1}};

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Amit Kapila 2023-03-02 09:15:57 Re: [PATCH] Use indexes on the subscriber when REPLICA IDENTITY is full on the publisher
Previous Message Peter Eisentraut 2023-03-02 08:21:55 Re: Make some xlogreader messages more accurate