From: | vodevsh(at)gmail(dot)com |
---|---|
To: | pgsql-docs(at)postgresql(dot)org |
Subject: | suggestion to make an example easier reproducible |
Date: | 2017-05-04 11:55:33 |
Message-ID: | 20170504115533.24360.68520@wrigleys.postgresql.org |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-docs |
The following documentation comment has been logged on the website:
Page: https://www.postgresql.org/docs/9.6/static/plpgsql-implementation.html
Description:
Documentation is good. But if you change `INSERT INTO logtable VALUES
(logtxt, 'now');` to `RAISE INFO '%','now'::timestamptz;` in example, people
will probably try it out to see how caching works (no need to create table).
Also could explain why RETURNED value is reparsed and recalculated, while
RAISEd - not:
CREATE FUNCTION logfunc1() RETURNS timestamptz AS $$
BEGIN
RAISE INFO '%','now'::timestamptz;
RETURN 'now';
END;
$$ LANGUAGE plpgsql;
CREATE FUNCTION logfunc2() RETURNS timestamptz AS $$
DECLARE
curtime timestamp;
BEGIN
curtime := 'now';
RAISE INFO '%',curtime::timestamptz;
RETURN curtime;
END;
$$ LANGUAGE plpgsql;
t=# select logfunc1(),logfunc2();
INFO: 2017-05-04 11:49:34.828784+00
INFO: 2017-05-04 11:49:34.828784+00
logfunc1 | logfunc2
-------------------------------+-------------------------------
2017-05-04 11:49:34.828784+00 | 2017-05-04 11:49:34.828784+00
(1 row)
t=# select logfunc1(),logfunc2();
INFO: 2017-05-04 11:49:34.828784+00
INFO: 2017-05-04 11:49:39.661306+00
logfunc1 | logfunc2
-------------------------------+-------------------------------
2017-05-04 11:49:39.661306+00 | 2017-05-04 11:49:39.661306+00
(1 row)
Thank you
From | Date | Subject | |
---|---|---|---|
Next Message | Magnus Hagander | 2017-05-05 10:43:30 | Re: Wording in pg_upgrade IMPLEMENTATION |
Previous Message | Peter Eisentraut | 2017-05-02 16:35:06 | release date formatting |