Re: ON CONFLICT DO SELECT (take 3)
| От | Dean Rasheed |
|---|---|
| Тема | Re: ON CONFLICT DO SELECT (take 3) |
| Дата | |
| Msg-id | CAEZATCWsybvZP3ce8rGcVNx-QHuDOJZDz8y=p1SzqHwjRXyV4Q@mail.gmail.com обсуждение исходный текст |
| Ответ на | Re: ON CONFLICT DO SELECT (take 3) (Viktor Holmberg <v@viktorh.net>) |
| Список | pgsql-hackers |
On Sun, 23 Nov 2025 at 20:34, Viktor Holmberg <v@viktorh.net> wrote:
>
> I’ve update the docs in all the cases you mentioned. I’ve also grepped through the docs for “ON CONFLICT” and “DO
UPDATE”and fixed upp all mentions where it made sense
>
--- a/doc/src/sgml/ref/create_table.sgml
+++ b/doc/src/sgml/ref/create_table.sgml
@@ -1380,7 +1380,7 @@ WITH ( MODULUS <replaceable
class="parameter">numeric_literal</replaceable>, REM
clause. <literal>NOT NULL</literal> and
<literal>CHECK</literal> constraints are not
deferrable. Note that deferrable constraints cannot be used as
conflict arbitrators in an <command>INSERT</command> statement that
- includes an <literal>ON CONFLICT DO UPDATE</literal> clause.
+ includes an <literal>ON CONFLICT DO UPDATE / SELECT</literal> clause.
</para>
</listitem>
</varlistentry>
Actually, a deferrable constraint cannot be used with ON CONFLICT DO
NOTHING either, which makes this a pre-existing documentation bug,
that should be fixed and back-patched separately as follows:
- includes an <literal>ON CONFLICT DO UPDATE</literal> clause.
+ includes an <literal>ON CONFLICT</literal> clause.
In addition, I think we should change the word "arbitrators" to
"arbiters". It means pretty-much the same thing, but the latter is the
term used everywhere else.
I'll take care of that separately, so the above diff won't be needed
in this patch.
Regards,
Dean
В списке pgsql-hackers по дате отправления: