Re: INSERT ... ON CONFLICT doesn't work
| От | David G. Johnston |
|---|---|
| Тема | Re: INSERT ... ON CONFLICT doesn't work |
| Дата | |
| Msg-id | CAKFQuwaZ=Jfw0bJDiJTzqXSAR+PvVZC7hpJcXCe8ke8mqzEG1A@mail.gmail.com обсуждение исходный текст |
| Ответ на | Re: INSERT ... ON CONFLICT doesn't work ("David G. Johnston" <david.g.johnston@gmail.com>) |
| Список | pgsql-general |
On Wed, Dec 1, 2021 at 12:59 PM David G. Johnston <david.g.johnston@gmail.com> wrote:
On Wed, Dec 1, 2021 at 12:52 PM Tom Lane <tgl@sss.pgh.pa.us> wrote:It's also possible to qualify the name in the ON CONFLICT clause,
although I think you have to parenthesize it to do so:
... ON CONFLICT (("Lockers"."Uuid"))This really needs to be confirmed and, if so, better documented on the INSERT page, since adding a schema to the index_column_name does not intuitively turn it into an index_expression which is what the syntax diagram says requires an extra pair of parentheses.
That of course should be adding a table qualifier to the column, not a schema qualifier. I get why INSERT doesn't really care, the source table is never in question in SQL-land, but as we see here there is a valid need for it in plpgsql.
David J.
В списке pgsql-general по дате отправления: