Re: Simple Atomic Relationship Insert

[Roxanne Reid-Bennett <rox@xxxxxxxxxxx>]

On 1/15/2015 6:12 PM, Robert DiFalco wrote:

FWIW I was using the select_hometown_id FUNCTION like this:

INSERT INTO users(...) values(..., select_hometown_id('Portland, OR'));

try this:  (if you still get deadlocks, uncomment the advisory lock [thanks Daniel] and try again)
Logically I suppose it might run faster to do the select, then insert "if".  I almost always write these as insert first - because it's the more restrictive lock.

CREATE OR REPLACE FUNCTION select_hometown_id(hometown_name VARCHAR) RETURNS
INTEGER AS
$BODY$
DECLARE
    v_id integer;
BEGIN
--        perform pg_advisory_xact_lock(hashtext(hometown_name));
      BEGIN
        insert into hometowns (name)
        select hometown_name where not exists (select id from hometowns where name = hometown_name)
        returning id into v_id;
        IF (v_id IS NULL) THEN
            select id into v_id from hometowns where name = hometown_name;
        END IF;
        return v_id;
      EXCEPTION
         WHEN OTHERS THEN
           -- choose your poison, this really shouldn't get here
      END;
   
    return null;
END;
$BODY$
LANGUAGE plpgsql;

On Tue, Jan 13, 2015 at 2:21 PM, Brian Dunavant <brian@xxxxxxxxxx> wrote:

The loop to run it twice handles that yes.  I don't think that buys
you anything over a more traditional non-cte method though.  I'd run
them a few thousand times to see if there's any difference in runtimes
but my guess is the CTE version would be slightly slower here.

v_id integer;
BEGIN;
  select id into v_id from hometowns where name = hometown_name;
  BEGIN
    insert into hometowns (name)
    select hometown_name where v_id is null
    returning id into v_id;
  EXCEPTION WHEN unique_violation
  THEN
     select id into v_id from hometowns where name = hometown_name;
  END;
  insert into users (name, hometown_id)
  values ('Robert', v_id);
END;

On Tue, Jan 13, 2015 at 5:06 PM, Robert DiFalco

<robert.difalco@xxxxxxxxx> wrote:
> This seems to get rid of the INSERT race condition.
>
> CREATE OR REPLACE FUNCTION select_hometown_id(hometown_name VARCHAR) RETURNS
> INTEGER AS $
> DECLARE hometown_id INTEGER;
> BEGIN
>   LOOP
>     BEGIN
>       WITH sel AS (
>           SELECT id FROM hometowns WHERE name = hometown_name
>       ), ins AS (
>         INSERT INTO hometowns (name)
>           SELECT hometown_name
>           WHERE NOT EXISTS(SELECT 1 FROM sel)
>         RETURNING id
>       )
>       SELECT id INTO hometown_id FROM ins UNION ALL SELECT id FROM sel;
>       RETURN hometown_id;
>
>     EXCEPTION WHEN unique_violation
>       THEN
>     END;
>   END LOOP;
> END;
> $ LANGUAGE plpgsql;
>
>
> On Tue, Jan 13, 2015 at 1:53 PM, Brian Dunavant <brian@xxxxxxxxxx> wrote:
>>
>> With the single CTE I don't believe you can do a full upsert loop.  If
>> you're doing this inside of a postgres function, your changes are
>> already atomic, so I don't believe by switching you are buying
>> yourself much (if anything) by using a CTE query instead of something
>> more traditional here.
>>
>> The advantages of switching to a CTE would be if this code was all
>> being done inside of the app code with multiple queries.
>>
>> On Tue, Jan 13, 2015 at 4:39 PM, Robert DiFalco
>> <robert.difalco@xxxxxxxxx> wrote:
>> > Well, traditionally I would create a LOOP where I tried the SELECT, if
>> > there
>> > was nothing I did the INSERT, if that raised an exception I would repeat
>> > the
>> > LOOP.
>> >
>> > What's the best way to do it with the CTE? Currently I have the
>> > following
>> > which gives me Duplicate Key Exceptions when two sessions try to insert
>> > the
>> > same record at the same time.
>> >
>> > CREATE OR REPLACE FUNCTION select_hometown_id(hometown_name VARCHAR)
>> > RETURNS
>> > INTEGER AS $
>> > DECLARE hometown_id INTEGER;
>> >     BEGIN
>> >       WITH sel AS (
>> >           SELECT id FROM hometowns WHERE name = hometown_name
>> >       ), ins AS (
>> >         INSERT INTO hometowns (name)
>> >           SELECT hometown_name
>> >           WHERE NOT EXISTS(SELECT 1 FROM sel)
>> >         RETURNING id
>> >       )
>> >       SELECT id INTO hometown_id FROM ins UNION ALL SELECT id FROM sel;
>> >       RETURN hometown_id;
>> >     END;
>> > $ LANGUAGE plpgsql;
>> >
>> > And that is no bueno. Should I just put the whole thing in a LOOP?
>
>

-- 
[At other schools] I think the most common fault in general is to teach students how to pass exams instead of teaching them the science.
Donald Knuth