Re: MERGE ... RETURNING - Mailing list pgsql-hackers

From jian he
Subject Re: MERGE ... RETURNING
Date
Msg-id CACJufxF5Gv+j2MRn+Tf1gboDGeY322GAm6_KA+0h9NbKu3zntA@mail.gmail.com
Whole thread Raw
In response to Re: MERGE ... RETURNING  (Dean Rasheed <dean.a.rasheed@gmail.com>)
Responses Re: MERGE ... RETURNING
List pgsql-hackers

Hi
mainly document issues. Other than that, it looks good!

MERGE not supported in COPY
MERGE not supported in WITH query
These entries in src/backend/po.* need to be deleted if this patch is committed?
------------------------------------------------------
  <indexterm zone="dml-returning">
   <primary>RETURNING</primary>
  </indexterm>

  <indexterm zone="dml-returning">
   <primary>INSERT</primary>
   <secondary>RETURNING</secondary>
  </indexterm>

  <indexterm zone="dml-returning">
   <primary>UPDATE</primary>
   <secondary>RETURNING</secondary>
  </indexterm>

  <indexterm zone="dml-returning">
   <primary>DELETE</primary>
   <secondary>RETURNING</secondary>
  </indexterm>

  <indexterm zone="dml-returning">
   <primary>MERGE</primary>
   <secondary>RETURNING</secondary>
  </indexterm>

in doc/src/sgml/dml.sgml, what is the point of these?
It is not rendered in the html file, deleting it still generates all the html file.
------------------------------------------------------
The following part is about doc/src/sgml/plpgsql.sgml.

    <para>
     The <replaceable>query</replaceable> used in this type of <literal>FOR</literal>
     statement can be any SQL command that returns rows to the caller:
     <command>SELECT</command> is the most common case,
     but you can also use <command>INSERT</command>, <command>UPDATE</command>, or
     <command>DELETE</command> with a <literal>RETURNING</literal> clause.  Some utility
     commands such as <command>EXPLAIN</command> will work too.
    </para>
here we need to add <command>MERGE</command>?


   <para>
    Row-level triggers fired <literal>BEFORE</literal> can return null to signal the
    trigger manager to skip the rest of the operation for this row
    (i.e., subsequent triggers are not fired, and the
    <command>INSERT</command>/<command>UPDATE</command>/<command>DELETE</command> does not occur
    for this row).  If a nonnull
here we need to add <command>MERGE</command>?


   <para>
    Variable substitution currently works only in <command>SELECT</command>,
    <command>INSERT</command>, <command>UPDATE</command>,
    <command>DELETE</command>, and commands containing one of
    these (such as <command>EXPLAIN</command> and <command>CREATE TABLE
    ... AS SELECT</command>),
    because the main SQL engine allows query parameters only in these
    commands.  To use a non-constant name or value in other statement
    types (generically called utility statements), you must construct
    the utility statement as a string and <command>EXECUTE</command> it.
   </para>
here we need to add <command>MERGE</command>?
demo:
CREATE OR REPlACE FUNCTION stamp_user2(id int, comment text) RETURNS void AS $$
    <<fn>>
    DECLARE
        curtime timestamp := now();
    BEGIN
        MERGE INTO users
        USING (SELECT 1)
        ON true
        WHEN MATCHED and (users.id = stamp_user2.id) THEN
          update SET last_modified = fn.curtime, comment = stamp_user2.comment;
        raise notice 'test';
    END;
$$ LANGUAGE plpgsql;


    <literal>INSTEAD OF</literal> triggers (which are always row-level triggers,
    and may only be used on views) can return null to signal that they did
    not perform any updates, and that the rest of the operation for this
    row should be skipped (i.e., subsequent triggers are not fired, and the
    row is not counted in the rows-affected status for the surrounding
    <command>INSERT</command>/<command>UPDATE</command>/<command>DELETE</command>).
I am not sure we need to add <command>MERGE</command >. Maybe not.

pgsql-hackers by date:

Previous
From: Amit Kapila
Date:
Subject: Re: Have pg_basebackup write "dbname" in "primary_conninfo"?
Next
From: Amit Kapila
Date:
Subject: Re: Typos in reorderbuffer.c.