Re: Set Returning Functions and joins

Awesome, thanks that works and is quite clear. The plan looks a bit funny on this. Any high-level synopsis on the performance of this?

From: David Johnston [mailto:polobo(at)yahoo(dot)com]
Sent: Wednesday, August 15, 2012 4:57 PM
To: David Greco
Cc: pgsql-general(at)postgresql(dot)org
Subject: Re: [GENERAL] Set Returning Functions and joins

Inline comments:

On Aug 15, 2012, at 16:24, David Greco <David_Greco(at)harte-hanks(dot)com<mailto:David_Greco(at)harte-hanks(dot)com>> wrote:
Not sure how to write the CTE form of this. This query gives an error that dave is missing a from clause entry.
WITH o as (
SELECT getRecord(1, dave.field1)
)
SELECT
id, o.*
FROM
dave
;

With o (id, result) as (select id, getRecord(...) from dave)
Select id, (o.result).*
From dave
Left? Join o Using (id)

Regarding the INNER JOIN, actually it appears the opposite is true:
SELECT
id, generate_series(1, dave.field1)
FROM
Dave
Works great if generate_series returns rows, but does not return any rows if generate_series does not return rows (i.e. if dave has a row with field1=null). So in fact, I need to figure out how to rewrite a LEFT JOIN version of this query

Agreed. See above example that can handle both.

From: David Johnston [mailto:polobo(at)yahoo(dot)com]<mailto:[mailto:polobo(at)yahoo(dot)com]>
Sent: Wednesday, August 15, 2012 4:16 PM
To: David Greco
Cc: pgsql-general(at)postgresql(dot)org<mailto:pgsql-general(at)postgresql(dot)org>
Subject: Re: [GENERAL] Set Returning Functions and joins

On Aug 15, 2012, at 15:55, David Greco <David_Greco(at)harte-hanks(dot)com<mailto:David_Greco(at)harte-hanks(dot)com>> wrote:
I’m porting some code from an Oracle application and we have many uses of set returning function. In particular, we are using them in joins of the form:

CREATE TABLE dave ( id integer, field1 integer );
INSERT INTO dave VALUES (1, 10);

SELECT
id, g.*
FROM
dave
INNER JOIN generate_series( 1, dave.field1 ) ON (1=1)

In reality, the examples are not trivial like this, and the set returning function returns sets of records, not single values.
Now, in the case of a LEFT JOIN and a function returning a setoff a simple value, I can rewrite it simply as:
SELECT
id, generate_series(1, dave.field1)
FROM
dave

In the case of a LEFT JOIN and a function returning a setoff a record, I can rewrite it as:
SELECT
id, ( getRecord(1, dave.field1) ).*
FROM
dave

I then figured I can rewrite INNER JOINs as:
SELECT
id, ( getRecord(1, dave.field1) ).*
FROM
dave
WHERE
Exists ( SELECT 1 FROM getRecord(1, dave.field1) )

Though I suppose this is running getRecord once for every row in dave, then another time for every row being returned.

Now in some non-trivial examples involving multiple joins on set returning functions, this gets pretty complicated.

Is there any alternative? Or I can suggest that a query the original form should be allowed?

SELECT
id, g.*
FROM
dave
INNER JOIN generate_series( 1, dave.field1 ) ON (1=1)

I suggest putting your function calls within a CTE (common table expression: SQL command WITH) structure.

I do not get why you think you need an "EXISTS" in the INNER JOIN situation. At worse you should make it a sub-query and add a IS NOT NULL condition on one of the function result columns. There is no way to avoid evaluating once per record in dave but you should never have to evaluate more frequently than that. CTE and sub-selects are your friends.

The original form is currently being developed but will not be available until at least 9.3

David J.