100% found this document useful (1 vote)

206 views140 pages

Adv SQL and Functions

This document summarizes an introduction to advanced SQL concepts presented by Stephen Frost and Joe Conway. It covers window functions, common table expressions, lateral joins, JSONB, set-generating functions, and more. Examples are provided for basic query syntax including SELECT statements, joins, set operations, and subqueries. An overview of join types such as inner, outer, left, right, and full joins is also presented.

Uploaded by

roberto1159

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

100% found this document useful (1 vote)

206 views140 pages

Adv SQL and Functions

Uploaded by

roberto1159

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

You are on page 1/ 140

Adv.

SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
JSONB and SP-GIST
Functions - Overview
Function Basics
Functions - By Example

Advanced SQL and Functions

Stephen Frost
[email protected]
Joe Conway
[email protected]

Crunchy Data Solutions, Inc.

credativ International

September 17, 2014

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Query Syntax
Adv. SQL - Window Functions, CTEs, LATERAL
Simple Queries
JSONB and SP-GIST
Joins
Functions - Overview
Set Operations
Function Basics
Subqueries
Functions - By Example

Queries
Syntax Overview

[ WITH [ RECURSIVE ] with_query [, ...] ]

SELECT [ ALL | DISTINCT [ ON ( expression [, ...] ) ] ]
* | expression [ [ AS ] output_name ] [, ...]
[ FROM from_item [, ...] ]
[ WHERE condition ]
[ GROUP BY expression, ... ]
[ HAVING condition [, ...] ]
[ WINDOW window_name AS ( window_definition ) [, ...] ]
[ { UNION | INTERSECT | EXCEPT } [ ALL | DISTINCT ] .. ]
[ ORDER BY expression [ ASC | DESC | USING op ], ... ]
[ LIMIT num ] [ OFFSET num ]

http://www.postgresql.org/docs/9.4/interactive/sql-select.html

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Query Syntax
Adv. SQL - Window Functions, CTEs, LATERAL
Simple Queries
JSONB and SP-GIST
Joins
Functions - Overview
Set Operations
Function Basics
Subqueries
Functions - By Example

Queries
Syntax Overview- from item

[ ONLY ] table_name [ * ]
[ [ AS ] alias [ ( column_alias [, ...] ) ] ]
[ LATERAL ] ( select )
[ AS ] alias [ ( column_alias [, ...] ) ]
[ LATERAL ] function_name ( [ argument [, ...] ] )
[ AS ] alias
[ ( column_alias [, ...] | column_definition [, ...] ) ]
[ LATERAL ] function_name ( [ argument [, ...] ] )
AS ( column_definition [, ...] )
with_query_name [ [ AS ] alias [ ( col_alias [, ...] ) ] ]
from_item [ NATURAL ] join_type
from_item [ ON join_condition | USING ( column [, ...] ) ]

with_query_name [ ( column_name [, ...] ) ]

AS ( select | values | insert | update | delete )

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Query Syntax
Adv. SQL - Window Functions, CTEs, LATERAL
Simple Queries
JSONB and SP-GIST
Joins
Functions - Overview
Set Operations
Function Basics
Subqueries
Functions - By Example

Queries
Syntax Overview- VALUES, TABLE

Last, but not least, the most complicated ones of all. VALUES returns a table, after
evaluating all expressions:

VALUES ( expression [, ...] ) [, ...]

TABLE is essentially SELECT * FROM, but shorter:

TABLE table_name
[ ORDER BY expression [ ASC | DESC | USING op ], ... ]
[ LIMIT num ] [ OFFSET num ]

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Query Syntax
Adv. SQL - Window Functions, CTEs, LATERAL
Simple Queries
JSONB and SP-GIST
Joins
Functions - Overview
Set Operations
Function Basics
Subqueries
Functions - By Example

Queries
Examples

VALUES (1), (2), (3);

TABLE author;

SELECT * FROM author;

SELECT name AS designation FROM publisher;

SELECT DISTINCT authorid FROM bookauthor;

SELECT DISTINCT ON (authorid) authorid, bookname

FROM bookauthor ORDER BY authorid, bookpublishdate;

SELECT title FROM book

WHERE net_price < 10.00;

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Query Syntax
Adv. SQL - Window Functions, CTEs, LATERAL
Simple Queries
JSONB and SP-GIST
Joins
Functions - Overview
Set Operations
Function Basics
Subqueries
Functions - By Example

Queries
Examples

SELECT * FROM book

WHERE lower(title) LIKE ’%sql%’
ORDER BY lastname ASC, firstname ASC;

SELECT book.title FROM bookauthor, author, book

WHERE author.lastname = ’Eisentraut’
AND bookauthor.author_id = author.id
AND book.id = bookauthor.book_id
ORDER BY book.title DESC;

SELECT author, sum(price) AS price_sum

FROM bookpricelist
GROUP BY author HAVING sum(price) > 20
ORDER BY author;

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Query Syntax
Adv. SQL - Window Functions, CTEs, LATERAL
Simple Queries
JSONB and SP-GIST
Joins
Functions - Overview
Set Operations
Function Basics
Subqueries
Functions - By Example

Join Types

cross join
inner join
outer join
left
right
full

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Query Syntax
Adv. SQL - Window Functions, CTEs, LATERAL
Simple Queries
JSONB and SP-GIST
Joins
Functions - Overview
Set Operations
Function Basics
Subqueries
Functions - By Example

Cross Joins

Joins each row from the first table with each row from the second table

SELECT * FROM tab1 CROSS JOIN tab2;

is equivalent to

SELECT * FROM tab1, tab2;

Limited practical uses- generally used in error. particularly when comma-joins

Use of comma-joins makes this more likely
Useful for enumeration by creating a cartesian product

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Query Syntax
Adv. SQL - Window Functions, CTEs, LATERAL
Simple Queries
JSONB and SP-GIST
Joins
Functions - Overview
Set Operations
Function Basics
Subqueries
Functions - By Example

Inner Joins

Joins each row of the first table with each row from the second table for which the
condition matches

SELECT ... FROM tab1 [ INNER ] JOIN tab2 ON condition;

SELECT ... FROM tab1 [ INNER ] JOIN tab2 USING (column list);

SELECT ... FROM tab1 NATURAL [ INNER ] JOIN tab2;

Or using the “traditional” (horrible) comma-join notation:

SELECT ... FROM tab1, tab2 WHERE condition;

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Query Syntax
Adv. SQL - Window Functions, CTEs, LATERAL
Simple Queries
JSONB and SP-GIST
Joins
Functions - Overview
Set Operations
Function Basics
Subqueries
Functions - By Example

Inner Joins
Examples

SELECT * FROM book INNER JOIN publisher

ON book.publisher_id = publisher.id;

SELECT * FROM bibo INNER JOIN author

USING (book_id);

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Query Syntax
Adv. SQL - Window Functions, CTEs, LATERAL
Simple Queries
JSONB and SP-GIST
Joins
Functions - Overview
Set Operations
Function Basics
Subqueries
Functions - By Example

Outer Joins

Joins each row from the first table with each row from the second table for which the
condition matches. Furthermore, nonmatching rows are added to the result.
left join all rows from the left table
right join all rows from the right table
full join all rows from both tables
Rows without a join partner are filled up with null values.

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Query Syntax
Adv. SQL - Window Functions, CTEs, LATERAL
Simple Queries
JSONB and SP-GIST
Joins
Functions - Overview
Set Operations
Function Basics
Subqueries
Functions - By Example

Outer Joins
Syntax

SELECT ... FROM tab1 LEFT/RIGHT/FULL [ OUTER ] JOIN tab2

ON condition;

SELECT ... FROM tab1 LEFT/RIGHT/FULL [ OUTER ] JOIN tab2

USING (column list);

SELECT ... FROM

tab1 NATURAL LEFT/RIGHT/FULL [ OUTER ] JOIN tab2;

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Query Syntax
Adv. SQL - Window Functions, CTEs, LATERAL
Simple Queries
JSONB and SP-GIST
Joins
Functions - Overview
Set Operations
Function Basics
Subqueries
Functions - By Example

Outer Joins
Examples

SELECT * FROM book RIGHT JOIN bookcategory

ON book.id = bookcategory.category_id;

SELECT * FROM publisher LEFT JOIN book

ON publisher.id = book.publisher_id;

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Query Syntax
Adv. SQL - Window Functions, CTEs, LATERAL
Simple Queries
JSONB and SP-GIST
Joins
Functions - Overview
Set Operations
Function Basics
Subqueries
Functions - By Example

Set Operations
Example Data

sfrost=# table book;

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Query Syntax
Adv. SQL - Window Functions, CTEs, LATERAL
Simple Queries
JSONB and SP-GIST
Joins
Functions - Overview
Set Operations
Function Basics
Subqueries
Functions - By Example

Set Operations
UNION

SELECT title FROM book

UNION
SELECT authorname FROM book;

title
------------------
Joe Conway
Kostenlos Laufen
Angst Lauf
Wildlauf
Running Free
Running Scared
Running Wild
Stephen Frost
(8 rows)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Query Syntax
Adv. SQL - Window Functions, CTEs, LATERAL
Simple Queries
JSONB and SP-GIST
Joins
Functions - Overview
Set Operations
Function Basics
Subqueries
Functions - By Example

Set Operations
UNION ALL

SELECT title FROM book

UNION ALL
SELECT authorname FROM book;
title
------------------
Running Free
Running Wild
Running Scared
Kostenlos Laufen
Wildlauf
Angst Lauf
Stephen Frost
Stephen Frost
Stephen Frost
Joe Conway
Joe Conway
Joe Conway
(12 rows)
Stephen Frost, Joe Conway Postgres Open 2014
Adv. SQL - Basics
Query Syntax
Adv. SQL - Window Functions, CTEs, LATERAL
Simple Queries
JSONB and SP-GIST
Joins
Functions - Overview
Set Operations
Function Basics
Subqueries
Functions - By Example

Set Operations
INTERSECT

SELECT title FROM book

INTERSECT
SELECT authorname FROM book;

title
-------
(0 rows)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Query Syntax
Adv. SQL - Window Functions, CTEs, LATERAL
Simple Queries
JSONB and SP-GIST
Joins
Functions - Overview
Set Operations
Function Basics
Subqueries
Functions - By Example

Set Operations
EXCEPT

SELECT title FROM book

EXCEPT
SELECT authorname FROM book;

title
------------------
Running Free
Running Scared
Wildlauf
Running Wild
Angst Lauf
Kostenlos Laufen
(6 rows)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Query Syntax
Adv. SQL - Window Functions, CTEs, LATERAL
Simple Queries
JSONB and SP-GIST
Joins
Functions - Overview
Set Operations
Function Basics
Subqueries
Functions - By Example

Subqueries
Uncorrelated

Uncorrelated subquery:
Subquery calculates a constant result set for the upper query
Executed only once

SELECT title, authorname, price

FROM book
WHERE book.price >
(SELECT AVG(book.price) FROM book);

title | authorname | price

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Query Syntax
Adv. SQL - Window Functions, CTEs, LATERAL
Simple Queries
JSONB and SP-GIST
Joins
Functions - Overview
Set Operations
Function Basics
Subqueries
Functions - By Example

Subqueries
Correlated

Correlated subquery:
Subquery references variables from the upper query
Subquery has to be repeated for each row of the upper query
Could be rewritten as a join

SELECT title, authorname, price

FROM book book_outer
WHERE EXISTS
(SELECT * FROM book WHERE
book_outer.language = book.language
AND book.id <> book_outer.id
AND abs(book_outer.price - book.price) <= 20);

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Query Syntax
Adv. SQL - Window Functions, CTEs, LATERAL
Simple Queries
JSONB and SP-GIST
Joins
Functions - Overview
Set Operations
Function Basics
Subqueries
Functions - By Example

Subqueries
Correlated

Results:

title | authorname | price

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Window Functions
Adv. SQL - Window Functions, CTEs, LATERAL
Window Function - Examples
JSONB and SP-GIST
Common Table Expressions (CTEs)
Functions - Overview
Writable CTEs
Function Basics
LATERAL
Functions - By Example

Window Functions - Basics

Window functions are like ordinary aggregates, but are restricted to operate on a
portion of the tuples only.

function_name ([expression [, expression ... ]]) OVER ( window_definition )

function_name ([expression [, expression ... ]]) OVER window_name
function_name ( * ) OVER ( window_definition )
function_name ( * ) OVER window_name

Where window name is an identifier and window definition is:

[ existing_window_name ]
[ PARTITION BY expression [, ...] ]
[ ORDER BY expression [ ASC | DESC | USING operator ] [ NULLS { FIRST | LAST } ] [, ...] ]
[ frame_clause ]

Each Window function scans all tuples belonging to the “group”

the current tuple is part of.

Stephen Frost, Joe Conway Postgres Open 2014

Window Functions - Frame Clause

Range vs. Rows

RANGE UNBOUNDED PRECEDING

select all rows from the partition start up through the last peer in
the order of its ORDER BY clause (or all if omitted)
RANGE BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED
FOLLOWING
select all rows in the partition
ROWS UNBOUNDED PRECEDING
select all rows (regardless of duplicates) up through the current row
in order of its ORDER BY clause
BETWEEN UNBOUNDED PRECEDING AND CURRENT ROW
same as UNBOUNDED PRECEDING

Stephen Frost, Joe Conway Postgres Open 2014

Window Functions Basic Example

Select all books and compare its price against the average price of all books in the
same language:

SELECT
title, language, price,
AVG(price) OVER(PARTITION BY language) FROM book;

title | language | price | avg

Stephen Frost, Joe Conway Postgres Open 2014

Window Functions Basic Example

Window Function results can be operated against similar to functions-

SELECT
title, language, price,
ROUND(AVG(price) OVER(PARTITION BY language),2) FROM book;

title | language | price | round

Stephen Frost, Joe Conway Postgres Open 2014

Window Functions Range vs Row Example

With RANGE, all duplicates are considered part of the same group and the function is
run across all of them, with the same result used for all members of the group.

SELECT
title, language, price,
AVG(price) OVER(ORDER BY language RANGE UNBOUNDED PRECEDING)
FROM book;

title | language | price | avg

Stephen Frost, Joe Conway Postgres Open 2014

Window Functions Basic Example

With rows, can get a ”running” average even with an ORDER BY over duplicates:

SELECT
title, language, price,
AVG(price) OVER(ORDER BY language ROWS UNBOUNDED PRECEDING)
FROM book;

title | language | price | avg

Stephen Frost, Joe Conway Postgres Open 2014

Window Functions Window Clause

Select all books and compare its price against the average price and total price of all
books in the same language:

SELECT
title, language, price,
AVG(price) OVER mywindow,
SUM(price) OVER mywindow
FROM book
WINDOW mywindow AS (PARTITION BY language);

http://www.postgresql.org/docs/9.4/interactive/tutorial-window.html

Stephen Frost, Joe Conway Postgres Open 2014

Window Functions Window Clause

Results:

title | language | price | avg | sum

------------------+----------+--------+---------------------+--------
Running Free | English | 100.00 | 76.6666666666666667 | 230.00
Running Wild | English | 80.00 | 76.6666666666666667 | 230.00
Running Scared | English | 50.00 | 76.6666666666666667 | 230.00
Kostenlos Laufen | German | 95.00 | 71.6666666666666667 | 215.00
Wildlauf | German | 75.00 | 71.6666666666666667 | 215.00
Angst Lauf | German | 45.00 | 71.6666666666666667 | 215.00
(6 rows)

Stephen Frost, Joe Conway Postgres Open 2014

Window Functions Row Number

Multiple Window clauses can be in the same query, or even some with a named
window clause and some without one.

SELECT
row_number() OVER () as row, title, language, price,
AVG(price) OVER mywindow,
SUM(price) OVER mywindow
FROM book
WINDOW mywindow AS (PARTITION BY language);

Stephen Frost, Joe Conway Postgres Open 2014

Window Functions Row Number

Results:

row | title | language | price | avg | sum

-----+------------------+----------+--------+---------------------+--------
1 | Running Free | English | 100.00 | 76.6666666666666667 | 230.00
2 | Running Wild | English | 80.00 | 76.6666666666666667 | 230.00
3 | Running Scared | English | 50.00 | 76.6666666666666667 | 230.00
4 | Kostenlos Laufen | German | 95.00 | 71.6666666666666667 | 215.00
5 | Wildlauf | German | 75.00 | 71.6666666666666667 | 215.00
6 | Angst Lauf | German | 45.00 | 71.6666666666666667 | 215.00
(6 rows)

Stephen Frost, Joe Conway Postgres Open 2014

Window Functions Rank

SELECT
rank() OVER (ORDER BY title), title, language, price,
AVG(price) OVER mywindow,
SUM(price) OVER mywindow
FROM book
WINDOW mywindow AS (PARTITION BY language);

rank | title | language | price | avg | sum

------+------------------+----------+--------+---------------------+--------
1 | Angst Lauf | German | 45.00 | 71.6666666666666667 | 215.00
2 | Kostenlos Laufen | German | 95.00 | 71.6666666666666667 | 215.00
3 | Running Free | English | 100.00 | 76.6666666666666667 | 230.00
4 | Running Scared | English | 50.00 | 76.6666666666666667 | 230.00
5 | Running Wild | English | 80.00 | 76.6666666666666667 | 230.00
6 | Wildlauf | German | 75.00 | 71.6666666666666667 | 215.00
(6 rows)

Stephen Frost, Joe Conway Postgres Open 2014

Window Functions Rank Order By

ORDER BY window clause may re-order the rows, but an explicit overall ORDER BY
can still be used to achieve the desired result ordering.

SELECT
rank() OVER (ORDER BY title), title, language, price,
AVG(price) OVER mywindow,
SUM(price) OVER mywindow
FROM book
WINDOW mywindow AS (PARTITION BY language) ORDER BY price;

Stephen Frost, Joe Conway Postgres Open 2014

Window Functions Rank

Note that the rank value remains correct even though the final ordering is changed.
Results:

rank | title | language | price | avg | sum

------+------------------+----------+--------+---------------------+--------
1 | Angst Lauf | German | 45.00 | 71.6666666666666667 | 215.00
4 | Running Scared | English | 50.00 | 76.6666666666666667 | 230.00
6 | Wildlauf | German | 75.00 | 71.6666666666666667 | 215.00
5 | Running Wild | English | 80.00 | 76.6666666666666667 | 230.00
2 | Kostenlos Laufen | German | 95.00 | 71.6666666666666667 | 215.00
3 | Running Free | English | 100.00 | 76.6666666666666667 | 230.00
(6 rows)

Stephen Frost, Joe Conway Postgres Open 2014

Window Functions Rank Duplicates

Rank handles duplicates also. Note we are ranking over language now.

SELECT
rank() OVER (ORDER BY language), title, language, price,
AVG(price) OVER mywindow,
SUM(price) OVER mywindow
FROM book
WINDOW mywindow AS (PARTITION BY language);

rank | title | language | price | avg | sum

------+------------------+----------+--------+---------------------+--------
1 | Running Free | English | 100.00 | 76.6666666666666667 | 230.00
1 | Running Wild | English | 80.00 | 76.6666666666666667 | 230.00
1 | Running Scared | English | 50.00 | 76.6666666666666667 | 230.00
4 | Kostenlos Laufen | German | 95.00 | 71.6666666666666667 | 215.00
4 | Wildlauf | German | 75.00 | 71.6666666666666667 | 215.00
4 | Angst Lauf | German | 45.00 | 71.6666666666666667 | 215.00
(6 rows)

Stephen Frost, Joe Conway Postgres Open 2014

Introduction - What is a CTE?

SQL Standard defines table expressions declared by WITH

A table expression is noted as Common Table Expression, CTE
A CTE could contain recursive references to itself
Can be seen as a temp table or view private to a query

Stephen Frost, Joe Conway Postgres Open 2014

Syntax

[ WITH [ RECURSIVE ] with_query [, ...] ]

with with query as:

with_query_name [ ( column_name [, ...] ) ]

AS ( select )

Using a self-reference within a RECURSIVE query needs the following syntax in the
inner WITH definition:

non_recursive_term UNION [ ALL ] recursive_term

Stephen Frost, Joe Conway Postgres Open 2014

Simple non-recursive example

SELECT (SELECT ’Stephen’::text)

|| ’ ’
|| (SELECT ’Frost’::text);

WITH as syntactic sugar:

WITH with_1(prename)
AS ( SELECT ’Stephen’::text ),
with_2(fullname)
AS ( SELECT with_1.prename || ’ ’ || ’Frost’ from with_1 )
SELECT fullname FROM with_2;

Result: Stephen Frost

Stephen Frost, Joe Conway Postgres Open 2014

Another non-recursive example

Use WITH clauses to calculate the average by language, then another to pull the sum
by language, and finally join them with the original table.

WITH avg_price(language, avg)

AS ( SELECT language, avg(price)
FROM book
GROUP BY language ),
sum_price(language, sum)
AS ( SELECT language, sum(price)
FROM book
GROUP BY language )
SELECT book.*, round(avg,2) as avg, round(sum,2) as sum
FROM book
JOIN avg_price USING (language)
JOIN sum_price USING (language);

Stephen Frost, Joe Conway Postgres Open 2014

Another non-recursive example

Results:

title | language | price | authorname | id | avg | sum

------------------+----------+--------+---------------+----+-------+--------
Running Free | English | 100.00 | Stephen Frost | 1 | 76.67 | 230.00
Running Wild | English | 80.00 | Stephen Frost | 2 | 76.67 | 230.00
Running Scared | English | 50.00 | Stephen Frost | 3 | 76.67 | 230.00
Kostenlos Laufen | German | 95.00 | Joe Conway | 4 | 71.67 | 215.00
Wildlauf | German | 75.00 | Joe Conway | 5 | 71.67 | 215.00
Angst Lauf | German | 45.00 | Joe Conway | 6 | 71.67 | 215.00
(6 rows)

Stephen Frost, Joe Conway Postgres Open 2014

Simple recursive example

List all numbers from 1 to 100:

WITH RECURSIVE foo_with(n)

AS
(
VALUES(1)
UNION
SELECT
n+1
FROM
foo_with
WHERE n < 100
) SELECT * FROM foo_with ORDER BY n;

Stephen Frost, Joe Conway Postgres Open 2014

Recursion - formal explanation

IT is the Intermediate Table, WT the Working Table and RT the Result Table.

1 Initialize
IT is initialized as an empty set
Execute the non-recursive query
Assign results to both RT and WT ;
2 Execute recursive query
Replace recursive self-reference with WT
Assign results during execution to IT
Append IT to RT
Replace WT with current IT
Truncate IT
3 Check recursion
Repeat 2) until WT is an empty set
Return RT
Stephen Frost, Joe Conway Postgres Open 2014
Adv. SQL - Basics
Window Functions
Adv. SQL - Window Functions, CTEs, LATERAL
Window Function - Examples
JSONB and SP-GIST
Common Table Expressions (CTEs)
Functions - Overview
Writable CTEs
Function Basics
LATERAL
Functions - By Example

Simple recursive example

Detailed example:
WITH RECURSIVE foo_with(n)
AS (

-- non-recursive query, assign results to WT, RT

VALUES(1)

-- recursive query with self reference to foo_with

-- self-reference substituted by WT, results
-- assigned to IT, WT and appended to RT

UNION
SELECT
n+1
FROM
foo_with
WHERE n < 100

-- empty IT and execute recursive term as long

-- as WT contains any tuple.

-- produces result set RT

SELECT * FROM foo_with ORDER BY n;

Stephen Frost, Joe Conway Postgres Open 2014
Adv. SQL - Basics
Window Functions
Adv. SQL - Window Functions, CTEs, LATERAL
Window Function - Examples
JSONB and SP-GIST
Common Table Expressions (CTEs)
Functions - Overview
Writable CTEs
Function Basics
LATERAL
Functions - By Example

Another example (1)

CREATE TABLE parts_list

(whole text, part text, count int);

INSERT INTO parts_list VALUES

(’car’, ’engine’, 1),
(’car’, ’wheel’, 4),
(’engine’, ’cylinder head’, 1),
(’cylinder head’, ’screw’, 14),
(’wheel’, ’screw’, 5),
(’car’, ’doors’, 4),
(’car’, ’steering wheel’, 1),
(’doors’, ’window’, 1);

parts list is a self-referencing table, cannot be easily retrieved with plain SQL.

Stephen Frost, Joe Conway Postgres Open 2014

Another example (2)

Return the number of screws needed to assemble a car

WITH RECURSIVE list(whole, part, num)

AS
(
SELECT whole, part, count AS num
FROM parts_list
WHERE whole = ’car’

UNION

SELECT d.whole, d.part, d.count * list.num AS num

FROM list
JOIN parts_list d ON (d.whole = list.part)
)
SELECT SUM(num) FROM list WHERE part = ’screw’;

Result: 34

Stephen Frost, Joe Conway Postgres Open 2014

Caveats

Recursive queries use iteration in reality

UNION vs. UNION ALL
Only one recursive self-reference allowed
Primary query evaluates subqueries defined by WITH only once
Name of a WITH-Query hides any “real” table
No aggregates, GROUP BY, HAVING, ORDER BY, LIMIT, OFFSET in a
recursive query allowed
No mutual recursive WITH-Queries allowed
Recursive references must not be part of an OUTER JOIN
http://www.postgresql.org/docs/9.4/interactive/queries-with.html

Stephen Frost, Joe Conway Postgres Open 2014

Simple Writable CTE

Delete from one table and insert into another

WITH archive_rows()
AS
(
DELETE
FROM parts_list
WHERE whole = ’car’
RETURNING *
)
INSERT INTO parts_list_archive
SELECT * FROM archive_rows;

Stephen Frost, Joe Conway Postgres Open 2014

Recursive Writable CTE

Insert the parts needed to assemble a car into another table

WITH RECURSIVE list(whole, part, num)

AS
(
SELECT whole, part, count AS num
FROM parts_list
WHERE whole = ’car’

UNION

SELECT d.whole, d.part, d.count * list.num AS num

FROM list
JOIN parts_list d ON (d.whole = list.part)
)
INSERT INTO parts_list_car SELECT * FROM list;

Stephen Frost, Joe Conway Postgres Open 2014

LATERAL

LATERAL is a new JOIN method (aka ’LATERAL JOIN’) which allows a subquery in
one part of the FROM clause to refernce columns from earlier items in the FROM
clause.
Refer to earlier table
Refer to earlier subquery
Refer to earlier set-returning function
Implicitly added when a SRF is referring to an earlier item in the FROM clause

Stephen Frost, Joe Conway Postgres Open 2014

LATERAL Table example

Refer to earlier table’s column in arguments to SRF

CREATE TABLE numbers AS

SELECT generate_series as max_num FROM generate_series(1,10);

SELECT *
FROM numbers, LATERAL generate_series(1,max_num);

SELECT *
FROM numbers, generate_series(1,max_num);

Stephen Frost, Joe Conway Postgres Open 2014

LATERAL Table example

SELECT *
FROM numbers, generate_series(1,max_num);

max_num | generate_series
---------+-----------------
1 | 1
2 | 1
2 | 2
3 | 1
3 | 2
3 | 3
[...]
(55 rows)

Stephen Frost, Joe Conway Postgres Open 2014

LATERAL Subquery example

Refer to earlier subquery’s column in arguments to SRF

Returns same results as previous query, but without the table.

SELECT *
FROM (SELECT generate_series as max_num
FROM generate_series(1,10)) as numbers,
LATERAL generate_series(1,max_num);

SELECT *
FROM (SELECT generate_series as max_num
FROM generate_series(1,10)) as numbers,
generate_series(1,max_num);

Stephen Frost, Joe Conway Postgres Open 2014

LATERAL Subquery example

Refer to earlier subquery’s column in arguments to SRF

SELECT *
FROM (SELECT generate_series as max_num
FROM generate_series(1,10)) as numbers,
LATERAL generate_series(1,max_num);

SELECT *
FROM (SELECT generate_series as max_num
FROM generate_series(1,10)) as numbers,
generate_series(1,max_num);

SELECT * FROM generate_series(1,10) as numbers,

generate_series(1,numbers);

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
JSONB and SP-GIST JSONB
Functions - Overview SP-GIST
Function Basics
Functions - By Example

JSONB vs. JSON

JSONB is a new data type in 9.4 which is nearly identical to the JSON data type.
There are a few specific difference which are important to note:
JSON is stored as a regular ’text’ blob, making it slow to utilize
JSONB is stored much more efficiently in a binary data format
JSONB is very slightly slower to input
JSONB normalizes input, reduces whitespace, does not preserve order or
duplicates
JSON can only be sensibly indexed through functional indexes
JSONB can be directly indexed
JSONB number output depends on PostgreSQL numeric data type
JSONB has containment and existance operators

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
JSONB and SP-GIST JSONB
Functions - Overview SP-GIST
Function Basics
Functions - By Example

JSONB Example

As mentioned, JSONB does not preserve whitespace (or lack of it), for example:

postgres=# SELECT ’{"bar":"baz","balance":7.77,"active":false}’::json;

json
---------------------------------------------
{"bar":"baz","balance":7.77,"active":false}
(1 row)

postgres=# SELECT ’{"bar":"baz","balance":7.77,"active":false}’::jsonb;

jsonb
--------------------------------------------------
{"bar": "baz", "active": false, "balance": 7.77}
(1 row)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
JSONB and SP-GIST JSONB
Functions - Overview SP-GIST
Function Basics
Functions - By Example

JSONB Example

JSONB uses the numeric data type’s output format, see these two identical inputs:

postgres=# SELECT ’{"reading": 1.230e-5}’::json, ’{"reading": 1.230e-5}’::jsonb

json | jsonb
-----------------------+-------------------------
{"reading": 1.230e-5} | {"reading": 0.00001230}
(1 row)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
JSONB and SP-GIST JSONB
Functions - Overview SP-GIST
Function Basics
Functions - By Example

JSONB Containment and Existance Examples

Array on the right side is contained within the one on the left.

postgres=# SELECT ’[1, 2, 3]’::jsonb @> ’[1, 3]’::jsonb;

?column?
----------
t
(1 row)

JSONB structure on the left contains the element on the right.

postgres=# SELECT ’["foo", "bar", "baz"]’::jsonb ? ’bar’;

?column?
----------
t
(1 row)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
JSONB and SP-GIST JSONB
Functions - Overview SP-GIST
Function Basics
Functions - By Example

SP-GIST

SP-GIST differs from other index types by decomposing the given space into disjoint
partitions.
SP-GIST index creation is generally faster than GIST
SP-GIST index size is comparable to GIST
SP-GIST query time is much faster than GIST

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
JSONB and SP-GIST JSONB
Functions - Overview SP-GIST
Function Basics
Functions - By Example

SP-GIST Example

postgres=# create table geo (point point);

CREATE TABLE
postgres=# create index pt_gist_idx on geo using gist(point);
CREATE INDEX
postgres=# create index pt_spgist_idx on geo using spgist(point);
CREATE INDEX
postgres=# insert into geo
postgres-# select (random()*180-90 || ’,’ || random()*360-180)::point
postgres-# from generate_series(1,1000000);
INSERT 0 1000000

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
JSONB and SP-GIST JSONB
Functions - Overview SP-GIST
Function Basics
Functions - By Example

SP-GIST Example

Performance depends on the amount of data and the size of the overall space of the
data which is indexed. A simple 1,000,000 point example shows improved
performance, where smaller data sets showed little difference:

postgres=# explain analyze select point from geo where point ~= ’(-29.549120804
[...]
Execution time: 0.245 ms
postgres=# create index pt_spgist_idx on geo using spgist(point);
CREATE INDEX
postgres=# explain analyze select point from geo where point ~= ’(-29.549120804
[...]
Execution time: 0.158 ms

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Introduction
JSONB and SP-GIST Uses
Functions - Overview Varieties
Function Basics Languages
Functions - By Example

What are Functions?

Full fledged SQL objects

Many other database objects are implemented with them
Fundamental part of PostgreSQL’s system architecture
Created with CREATE FUNCTION
Executed through normal SQL
target-list:
SELECT myfunc(f1) FROM foo;
FROM clause:
SELECT * FROM myfunc();
WHERE clause:
SELECT * FROM foo WHERE myfunc(f1) = 42;

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Introduction
JSONB and SP-GIST Uses
Functions - Overview Varieties
Function Basics Languages
Functions - By Example

How are they Used?

Functions
Operators
Data types
Index methods
Casts
Triggers
Aggregates
Ordered-set Aggregates
Window Functions

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Introduction
JSONB and SP-GIST Uses
Functions - Overview Varieties
Function Basics Languages
Functions - By Example

What Forms Can They Take?

PostgreSQL provides four kinds of functions:
SQL
Procedural Languages
Internal
C-language
Arguments
Base, composite, or combinations
Scalar or array
Pseudo or polymorphic
VARIADIC
IN/OUT/INOUT
Return
Singleton or set (SETOF)
Base or composite type
Pseudo or polymorphic
http://www.postgresql.org/docs/9.4/static/sql-createfunction.html
Stephen Frost, Joe Conway Postgres Open 2014
Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Introduction
JSONB and SP-GIST Uses
Functions - Overview Varieties
Function Basics Languages
Functions - By Example

SQL Functions
Behavior
Executes an arbitrary list of SQL statements separated by
semicolons
Last statement may be INSERT, UPDATE, or DELETE with
RETURNING clause
Arguments
Referenced by function body using name or $n: $1 is first arg,
etc. . .
If composite type, then dot notation $1.name used to access
Only used as data values, not as identifiers
Return
If singleton, first row of last query result returned, NULL on no
result
If SETOF, all rows of last query result returned, empty set on
no result
http://www.postgresql.org/docs/9.4/static/xfunc-sql.html
Stephen Frost, Joe Conway Postgres Open 2014
Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Introduction
JSONB and SP-GIST Uses
Functions - Overview Varieties
Function Basics Languages
Functions - By Example

Procedural Languages

User-defined functions
Written in languages besides SQL and C
Task is passed to a special handler that knows the details of
the language
Dynamically loaded
Could be self-contained (e.g. PL/pgSQL)
Might be externally linked (e.g. PL/Perl)
http://www.postgresql.org/docs/9.4/static/xplang.html

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Introduction
JSONB and SP-GIST Uses
Functions - Overview Varieties
Function Basics Languages
Functions - By Example

Internal Functions

Statically linked C functions

Could use CREATE FUNCTION to create additional alias
names for an internal function
Most internal functions expect to be declared STRICT

CREATE FUNCTION square_root(double precision)

RETURNS double precision AS
’dsqrt’
LANGUAGE internal STRICT;

http://www.postgresql.org/docs/9.4/static/xfunc-internal.html

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Introduction
JSONB and SP-GIST Uses
Functions - Overview Varieties
Function Basics Languages
Functions - By Example

C Language Functions

User-defined functions written in C

Compiled into dynamically loadable objects (also called shared
libraries)
Loaded by the server on demand
contrib is good source of examples
Same as internal function coding conventions
Require PG MODULE MAGIC call
Short example later, but deserves separate tutorial
http://www.postgresql.org/docs/9.4/static/xfunc-c.html

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Introduction
JSONB and SP-GIST Uses
Functions - Overview Varieties
Function Basics Languages
Functions - By Example

Language Availability
PostgreSQL includes the following server-side procedural
languages:
http://www.postgresql.org/docs/9.4/static/xplang.html

PL/pgSQL
Perl
Python
Tcl
Other languages available:
http://pgfoundry.org/softwaremap/trove_list.php?form_cat=311

Java
V8 (Javascript)
Ruby
R
Shell
others . . .
Stephen Frost, Joe Conway Postgres Open 2014
Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Creating New Functions

CREATE [ OR REPLACE ] FUNCTION
name ( [ [ argmode ] [ argname ] argtype [ { DEFAULT | = } default_expr ]
[, ...] ] )
[ RETURNS rettype
| RETURNS TABLE ( column_name column_type [, ...] ) ]
{ LANGUAGE lang_name
| WINDOW
| IMMUTABLE | STABLE | VOLATILE | [ NOT ] LEAKPROOF
| CALLED ON NULL INPUT | RETURNS NULL ON NULL INPUT | STRICT
| [ EXTERNAL ] SECURITY INVOKER | [ EXTERNAL ] SECURITY DEFINER
| COST execution_cost
| ROWS result_rows
| SET configuration_parameter { TO value | = value | FROM CURRENT }
| AS ’definition’
| AS ’obj_file’, ’link_symbol’
} ...
[ WITH ( attribute [, ...] ) ]

http://www.postgresql.org/docs/9.4/static/sql-createfunction.html
Stephen Frost, Joe Conway Postgres Open 2014
Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Dollar Quoting
Works for all character strings
Particularly useful for function bodies
Consists of a dollar sign ($), ”tag” of zero or more characters,
another dollar sign
Start and End tag must match
Nest dollar-quoted string literals by choosing different tags at
each nesting level
CREATE OR REPLACE FUNCTION dummy () RETURNS text AS
$_$
BEGIN
RETURN $$Say ’hello’$$;
END;
$_$
LANGUAGE plpgsql;

http://www.postgresql.org/docs/9.4/static/sql-syntax-lexical.html#SQL-SYNTAX-DOLLAR-QUOTING
Stephen Frost, Joe Conway Postgres Open 2014
Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Anonymous Functions

DO [ LANGUAGE lang_name ] code

Keyword DO executes anonymous code block

Transient
Any procedural language with support, defaults to plpgsql
No parameters, returns void
Parsed and executed once
LANGUAGE clause can be before or after code block

http://www.postgresql.org/docs/9.4/static/sql-do.html

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Anonymous Functions

DO $_$
DECLARE r record;
BEGIN
FOR r IN SELECT u.rolname
FROM pg_authid u
JOIN pg_auth_members m on m.member = u.oid
JOIN pg_authid g on g.oid = m.roleid
WHERE g.rolname = ’admin’
LOOP
EXECUTE $$ ALTER ROLE $$ || r.rolname ||
$$ SET work_mem = ’512MB’ $$;
END LOOP;
END$_$;

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Anonymous Functions

SELECT u.rolname, s.setconfig as setting

FROM pg_db_role_setting s
JOIN pg_authid u on u.oid = s.setrole
JOIN pg_auth_members m on m.member = u.oid
JOIN pg_authid g on g.oid = m.roleid
WHERE g.rolname = ’admin’;
rolname | setting
----------+------------------
rockstar | {work_mem=512MB}
(1 row)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Changing Existing Functions

Once created, dependent objects may be created

Must do DROP FUNCTION ... CASCADE to recreate
Or use OR REPLACE to avoid dropping dependent objects
Very useful for large dependency tree
Can’t be used in some circumstances (must drop/recreate
instead). You cannot:
change function name or argument types
change return type
change types of any OUT parameters

CREATE OR REPLACE FUNCTION ...;

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Function Arguments - argmode

( [ [ argmode ] [ argname ] argtype [ { DEFAULT | = } default_expr ]
[, ...] ] )

argmode (optional): IN, OUT, INOUT, or VARIADIC

IN is the default if argmode is omitted
OUT and INOUT cannot be used with RETURNS TABLE
VARIADIC can only be followed by OUT
Not required (but good style): IN, then INOUT, then OUT
Func name + IN/INOUT/VARIADIC arg sig identifies function
CREATE FUNCTION testfoo (IN int, INOUT int, OUT int)
RETURNS RECORD AS $$
VALUES ($2, $1 * $2);
$$ language sql;
SELECT * FROM testfoo(14, 3);
column1 | column2
---------+---------
3 | 42
(1 row)
Stephen Frost, Joe Conway Postgres Open 2014
Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Function Arguments - argname

( [ [ argmode ] [ argname ] argtype [ { DEFAULT | = } default_expr ]
[, ...] ] )

argname (optional):
Most, but not all, languages will use in function body
Use named notation to improve readability and allow reordering
Defines the OUT column name in the result row type
CREATE FUNCTION testfoo (IN a int, INOUT mult int = 2, OUT a int)
RETURNS RECORD AS $$
VALUES (mult, a * mult);
$$ language sql;
SELECT * FROM testfoo(mult := 3, a := 14);
mult | a
------+----
3 | 42
(1 row)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Function Arguments - argtype

( [ [ argmode ] [ argname ] argtype [ { DEFAULT | = } default_expr ]
[, ...] ] )

argtype (required) (optionally schema-qualified):

base, array, composite, or domain types
can reference the type of a table column:
table_name.column_name%TYPE
Polymorphic ”pseudotypes”:
⇒ anyelement, anyarray, anynonarray, anyenum, anyrange
CREATE FUNCTION testfoo (INOUT a anyelement, INOUT mult anyelement)
RETURNS RECORD AS $$
VALUES (a * mult, mult);
$$ language sql;
SELECT * FROM testfoo(mult := 3.14, a := 2.71828);
a | mult
-----------+------
8.5353992 | 3.14
(1 row)
Stephen Frost, Joe Conway Postgres Open 2014
Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Function Arguments - default expr

( [ [ argmode ] [ argname ] argtype [ { DEFAULT | = } default_expr ]
[, ...] ] )

default expr (optional):

Used if arg not provided
An expression coercible to arg type
All input (IN/INOUT/VARIADIC) can have default
Following args must also have defaults
CREATE FUNCTION testfoo (IN a int, INOUT mult int = 2, OUT a int)
RETURNS RECORD AS $$
VALUES (mult, a * mult);
$$ language sql;
SELECT * FROM testfoo(14);
mult | a
------+----
2 | 28
(1 row)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Function Overloading
Input argument (IN/INOUT/VARIADIC) signature used
Avoid ambiguities:
Type (e.g. REAL vs. DOUBLE PRECISION)
Function name same as IN composite field name
VARIADIC vs same type scalar
CREATE OR REPLACE FUNCTION foo (text) RETURNS text AS $$
SELECT ’Hello ’ || $1
$$ LANGUAGE sql;
CREATE OR REPLACE FUNCTION foo (int) RETURNS text AS $$
SELECT ($1 / 2)::text || ’ was here’
$$ LANGUAGE sql;

SELECT foo(’42’), foo(84);

foo | foo
----------+-------------
Hello 42 | 42 was here
(1 row)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Function Return Type

[ RETURNS rettype
| RETURNS TABLE ( column_name column_type [, ...] ) ]

rettype (required) (optionally schema-qualified):

base, array, composite, or domain types
can reference the type of a table column:
table_name.column_name%TYPE
Polymorphic ”pseudotypes”:
⇒ anyelement, anyarray, anynonarray, anyenum, anyrange
Special ”pseudotypes”:
language handler: procedural language call handler
fdw handler: foreign-data wrapper handler
record: returning an unspecified row type
trigger: trigger function
void: function returns no value
Stephen Frost, Joe Conway Postgres Open 2014
Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Function Return Type

[ RETURNS rettype
| RETURNS TABLE ( column_name column_type [, ...] ) ]

rettype (required) (optionally schema-qualified):

INOUT/OUT args: RETURNS clause may be omitted
⇒ Note: does not return a set
If RETURNS present, must agree with OUT
SETOF modifier - ”set returning” or ”table” function

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Function Return Type - OUT + No RETURNS

CREATE FUNCTION testbar1 (OUT f1 int, OUT f2 text) AS $$

VALUES (42, ’hello’), (64, ’world’);
$$ language sql;
SELECT * FROM testbar1();
f1 | f2
----+-------
42 | hello
(1 row)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Function Return Type - OUT + SETOF RECORD

CREATE FUNCTION testbar2 (OUT f1 int, OUT f2 text)

RETURNS SETOF RECORD AS $$
VALUES (42, ’hello’), (64, ’world’);
$$ language sql;

SELECT * FROM testbar2();

f1 | f2
----+-------
42 | hello
64 | world
(2 rows)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Function Return Type - Custom Type

CREATE TYPE testbar3_type AS (f1 int, f2 text);

CREATE FUNCTION testbar3 ()
RETURNS SETOF testbar3_type AS $$
VALUES (42, ’hello’), (64, ’world’);
$$ language sql;
SELECT * FROM testbar3();
f1 | f2
----+-------
42 | hello
64 | world
(2 rows)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Function Return Type - RETURNS TABLE

CREATE FUNCTION testbar4 ()

RETURNS TABLE (f1 int, f2 text) AS $$
VALUES (42, ’hello’), (64, ’world’);
$$ language sql;
SELECT * FROM testbar4();
f1 | f2
----+-------
42 | hello
64 | world
(2 rows)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Function Return Type - unspecified RECORD

CREATE FUNCTION testbar5 ()

RETURNS SETOF RECORD AS $$
VALUES (42, ’hello’), (64, ’world’);
$$ language sql;
SELECT * FROM testbar5() as t(f1 int, f2 text);
f1 | f2
----+-------
42 | hello
64 | world
(2 rows)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Function Return Type - RETURNS scalar

CREATE FUNCTION testbar6 ()

RETURNS SETOF int AS $$
VALUES (42), (64);
$$ language sql;
SELECT * FROM testbar6();
testbar6
----------
42
64
(2 rows)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Function Return Type - RETURNS scalar with alias

CREATE FUNCTION testbar7 ()

RETURNS SETOF int AS $$
VALUES (42), (64);
$$ language sql;
SELECT * FROM testbar7() AS t(f1);
f1
----
42
64
(2 rows)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Function Return Type - Targetlist

SELECT testbar2();
testbar2
------------
(42,hello)
(64,world)
(2 rows)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Function Return Type - Targetlist, expanded

SELECT (testbar2()).*;
f1 | f2
----+-------
42 | hello
64 | world
(2 rows)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

LANGUAGE
LANGUAGE lang_name

Language of function body

Native: Internal, SQL
Interpreted, core: PL/pgSQL, PL/Perl, PL/Python, PL/Tcl
Intepreted, external: PL/Java, PL/J, PL/V8, PL/Ruby, PL/R,
PL/Sh
Compiled, external: Custom C loadable libraries
Some (e.g. perl, tcl) have ”trusted” and ”untrusted” variants
CREATE FUNCTION ...
LANGUAGE sql;
LANGUAGE plpgsql;
LANGUAGE plperlu;
LANGUAGE plr;
LANGUAGE C;
LANGUAGE internal;

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

WINDOW

Window Functions
Indicates function is a window function rather than ”normal”
function
Provides ability to calculate across sets of rows related to
current row
Similar to aggregate functions, but does not cause rows to
become grouped
Able to access more than just the current row of the query
result
Window functions can be written in C, PL/R, PL/V8, others?

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

WINDOW
Serveral window functions built-in
select distinct proname from pg_proc where proiswindow order by 1;
proname
--------------
cume_dist
dense_rank
first_value
lag
last_value
lead
nth_value
ntile
percent_rank
rank
row_number
(11 rows)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Volatility

VOLATILE (default)
Each call can return a different result
Example: random() or timeofday()
Functions modifying table contents must be declared volatile
STABLE
Returns same result for same arguments within single query
Example: now()
Consider configuration settings that affect output
IMMUTABLE
Always returns the same result for the same arguments
Example: lower(’ABC’)
Unaffected by configuration settings
Not dependent on table contents

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Volatility

select distinct proname, provolatile

from pg_proc
where proname in (’lower’, ’now’, ’timeofday’) order by 1;
proname | provolatile
-----------+-------------
lower | i
now | s
timeofday | v
(3 rows)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Volatility

select lower(’ABC’), now(), timeofday() from generate_series(1,3);

lower | now | timeofday
-------+-------------------------------+-------------------------------------
abc | 2014-08-17 12:26:08.407439-07 | Sun Aug 17 12:26:08.408005 2014 PDT
abc | 2014-08-17 12:26:08.407439-07 | Sun Aug 17 12:26:08.408042 2014 PDT
abc | 2014-08-17 12:26:08.407439-07 | Sun Aug 17 12:26:08.408048 2014 PDT
(3 rows)

select lower(’ABC’), now(), timeofday() from generate_series(1,3);

lower | now | timeofday
-------+-------------------------------+-------------------------------------
abc | 2014-08-17 12:26:13.215355-07 | Sun Aug 17 12:26:13.215566 2014 PDT
abc | 2014-08-17 12:26:13.215355-07 | Sun Aug 17 12:26:13.215586 2014 PDT
abc | 2014-08-17 12:26:13.215355-07 | Sun Aug 17 12:26:13.215591 2014 PDT
(3 rows)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Behavior with Null Input Values

CALLED ON NULL INPUT (default)
Function called normally with the null input values
RETURNS NULL ON NULL INPUT
Function not called when null input values are present
Instead, null is returned automatically
CREATE FUNCTION sum1 (int, int) RETURNS int AS $$
SELECT $1 + $2
$$ LANGUAGE SQL RETURNS NULL ON NULL INPUT;
CREATE FUNCTION sum2 (int, int) RETURNS int AS $$
SELECT COALESCE($1, 0) + COALESCE($2, 0)
$$ LANGUAGE SQL CALLED ON NULL INPUT;

SELECT sum1(9, NULL) IS NULL AS "true", sum2(9, NULL);

true | sum2
------+------
t | 9
(1 row)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Security Attributes - LEAKPROOF

LEAKPROOF requirements
No side effects
Reveals no info about args other than by return value
Planner may push leakproof functions into views created with
the security barrier option
Can only be set by the superuser

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Security Attributes - LEAKPROOF

\c - postgres
DROP TABLE IF EXISTS all_books CASCADE;
CREATE TABLE all_books(id serial primary key,
luser text,
bookname text,
price int);
INSERT INTO all_books
SELECT g.f,
CASE WHEN g.f % 2 = 0 THEN ’joe’ ELSE ’tom’ END,
’book-’ || g.f::text,
40 + g.f % 20
FROM generate_series(1,8) as g(f);

DROP VIEW IF EXISTS user_books;

CREATE VIEW user_books AS
SELECT id, luser, bookname, price FROM all_books
WHERE luser = CURRENT_USER;
GRANT ALL ON user_books TO public;

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Security Attributes - LEAKPROOF

Note the ”COST 1” below . . .

CREATE OR REPLACE FUNCTION leak_info(text, text) returns int AS $$
BEGIN
IF $1 != CURRENT_USER THEN
RAISE NOTICE ’%:%’, $1, $2;
END IF;
RETURN 0;
END;
$$ COST 1 LANGUAGE plpgsql;

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Security Attributes - LEAKPROOF

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Security Attributes - LEAKPROOF

Note the ”WITH (security barrier)” below . . .

\c - postgres
DROP VIEW user_books;
CREATE VIEW user_books WITH (security_barrier) AS
SELECT id, luser, bookname, price FROM all_books
WHERE luser = CURRENT_USER;
GRANT ALL ON user_books TO public;

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Security Attributes - LEAKPROOF

\c - joe
EXPLAIN ANALYZE SELECT * FROM user_books
WHERE leak_info(luser, bookname) = 0;
QUERY PLAN
------------------------------------------------------------------------
Subquery Scan on user_books (cost=0.00..1.16 rows=1 width=72) (actual ...
Filter: (leak_info(user_books.luser, user_books.bookname) = 0)
-> Seq Scan on all_books (cost=0.00..1.14 rows=1 width=72) (actual ...
Filter: (luser = ("current_user"())::text)
Rows Removed by Filter: 4
Planning time: 0.648 ms
Execution time: 1.903 ms
(7 rows)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Security Attributes - LEAKPROOF

\c - postgres
ALTER FUNCTION leak_info(text, text) LEAKPROOF;

\c - joe
EXPLAIN ANALYZE SELECT * FROM user_books
WHERE leak_info(luser, bookname) = 0;
NOTICE: tom:book-1
NOTICE: tom:book-3
NOTICE: tom:book-5
NOTICE: tom:book-7
QUERY PLAN
------------------------------------------------------------------
Seq Scan on all_books (cost=0.00..1.18 rows=1 width=72) (actual ...
Filter: ((leak_info(luser, bookname) = 0) AND
(luser = ("current_user"())::text))
Rows Removed by Filter: 4
Planning time: 0.646 ms
Execution time: 2.145 ms
(5 rows)
Stephen Frost, Joe Conway Postgres Open 2014
Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Security Attributes - LEAKPROOF

Lesson
Be sure function really is leak proof before making
LEAKPROOF
Why use LEAKPROOF at all?
Performance (predicate push down)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Security Attributes - SECURITY INVOKER/DEFINER

SECURITY INVOKER (default)
Function executed with the rights of the current user
SECURITY DEFINER
Executed with rights of creator, like ”setuid”
\c - postgres
CREATE TABLE foo (f1 int);
INSERT INTO foo VALUES(42);
REVOKE ALL ON foo FROM public;
CREATE FUNCTION see_foo() RETURNS TABLE (luser name, f1 int) AS $$
SELECT CURRENT_USER, * FROM foo
$$ LANGUAGE SQL SECURITY DEFINER;
\c - guest
SELECT * FROM foo;
ERROR: permission denied for relation foo
SELECT CURRENT_USER AS me, luser AS definer, f1 FROM see_foo();
me | definer | f1
-------+----------+----
guest | postgres | 42
(1 row) Stephen Frost, Joe Conway Postgres Open 2014
Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Optimizer Hints

COST execution_cost
ROWS result_rows

execution cost
Estimated execution cost for the function
Positive floating point number
Units are cpu operator cost
Cost is per returned row
Default: 1 unit for C-language/internal, 100 units for all others
result rows
Estimated number rows returned
Positive floating point number
Only allowed when declared to return set
Default: 1000
Stephen Frost, Joe Conway Postgres Open 2014
Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Optimizer Hints

CREATE FUNCTION testbar8 ()

RETURNS SETOF int AS $$
VALUES (42), (64);
$$ LANGUAGE sql COST 0.1 ROWS 2;

SELECT procost, prorows FROM pg_proc WHERE proname = ’testbar8’;

procost | prorows
---------+---------
0.1 | 2
(1 row)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Function Local Configs

SET configuration_parameter
{ TO value | = value | FROM CURRENT }

SET clause
Specified config set to value for duration of function
SET FROM CURRENT uses session’s current value
CREATE FUNCTION testbar9 ()
RETURNS SETOF int AS $$
VALUES (42), (64);
$$ LANGUAGE sql SET work_mem = ’512MB’;

SELECT proconfig FROM pg_proc WHERE proname = ’testbar9’;

proconfig
------------------
{work_mem=512MB}
(1 row)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Function Body

AS definition
| AS obj_file, link_symbol

definition
String literal
Parse by language parser
Can be internal function name
Can be path to object file if C language function name matches
Dollar quote, or escape single quotes and backslashes

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Function Body
AS definition
| AS obj_file, link_symbol

obj file, link symbol

Used when C language function name does not match SQL
function name
obj file is path to object file
⇒ $libdir: replaced by package lib dir name, determined at
build time
link symbol is name of function in C source code
When more than one FUNCTION call refers to same object
file, file only loaded once
# pg_config --pkglibdir
/usr/local/pgsql-REL9_4_STABLE/lib

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL Creation
JSONB and SP-GIST Arguments
Functions - Overview Return Types
Function Basics Attributes
Functions - By Example

Function Body

CREATE FUNCTION foobar ()

RETURNS int AS $$
SELECT 42;
$$ LANGUAGE sql;

CREATE OR REPLACE FUNCTION plr_version ()

RETURNS text
AS ’$libdir/plr’,’plr_version’
LANGUAGE C;

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
SQL Functions
JSONB and SP-GIST
PL/pgSQL Functions
Functions - Overview
Miscellaneous
Function Basics
Functions - By Example

Simple

CREATE FUNCTION sum (text, text)

RETURNS text AS $$
SELECT $1 || ’ ’ || $2
$$ LANGUAGE SQL;

SELECT sum(’hello’, ’world’);

sum
-------------
hello world
(1 row)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
SQL Functions
JSONB and SP-GIST
PL/pgSQL Functions
Functions - Overview
Miscellaneous
Function Basics
Functions - By Example

Custom Operator

CREATE OPERATOR + (
procedure = sum,
leftarg = text,
rightarg = text
);

SELECT ’hello’ + ’world’;

?column?
-------------
hello world
(1 row)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
SQL Functions
JSONB and SP-GIST
PL/pgSQL Functions
Functions - Overview
Miscellaneous
Function Basics
Functions - By Example

Custom Aggregate

CREATE OR REPLACE FUNCTION concat_ws_comma(text, ANYELEMENT)

RETURNS text AS $$
SELECT concat_ws(’,’, $1, $2)
$$ LANGUAGE sql;

CREATE AGGREGATE str_agg (ANYELEMENT) (

sfunc = concat_ws_comma,
stype = text);

SELECT str_agg(f1) FROM foo;

str_agg
---------
41,42
(1 row)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
SQL Functions
JSONB and SP-GIST
PL/pgSQL Functions
Functions - Overview
Miscellaneous
Function Basics
Functions - By Example

SETOF with OUT Arguments

CREATE OR REPLACE FUNCTION sql_with_rows(OUT a int, OUT b text)

RETURNS SETOF RECORD AS $$
values (1,’a’),(2,’b’)
$$ LANGUAGE SQL;

select * from sql_with_rows();

a | b
---+---
1 | a
2 | b
(2 rows)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
SQL Functions
JSONB and SP-GIST
PL/pgSQL Functions
Functions - Overview
Miscellaneous
Function Basics
Functions - By Example

INSERT RETURNING

CREATE TABLE foo (f0 serial, f1 int, f2 text);

CREATE OR REPLACE FUNCTION

sql_insert_returning(INOUT f1 int, INOUT f2 text, OUT id int) AS $$
INSERT INTO foo(f1, f2) VALUES ($1,$2) RETURNING f1, f2, f0
$$ LANGUAGE SQL;

SELECT * FROM sql_insert_returning(1,’a’);

f1 | f2 | id
----+----+----
1 | a | 1
(1 row)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
SQL Functions
JSONB and SP-GIST
PL/pgSQL Functions
Functions - Overview
Miscellaneous
Function Basics
Functions - By Example

Composite Argument

CREATE TABLE emp (name text,

salary numeric,
age integer,
cubicle point);

CREATE FUNCTION double_salary(emp) RETURNS numeric AS $$

SELECT $1.salary * 2 AS salary;
$$ LANGUAGE SQL;

SELECT name, double_salary(emp.*) AS dream

FROM emp WHERE emp.cubicle ~= point ’(2,1)’;

SELECT name,
double_salary(ROW(name, salary*1.1, age, cubicle)) AS dream
FROM emp;

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
SQL Functions
JSONB and SP-GIST
PL/pgSQL Functions
Functions - Overview
Miscellaneous
Function Basics
Functions - By Example

Polymorphic

CREATE FUNCTION myappend(anyarray, anyelement) RETURNS anyarray AS

$$
SELECT $1 || $2;
$$ LANGUAGE SQL;

SELECT myappend(ARRAY[42,6], 21), myappend(ARRAY[’abc’,’def’], ’xyz’);

myappend | myappend
-----------+---------------
{42,6,21} | {abc,def,xyz}
(1 row)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
SQL Functions
JSONB and SP-GIST
PL/pgSQL Functions
Functions - Overview
Miscellaneous
Function Basics
Functions - By Example

Target List versus FROM Clause

CREATE FUNCTION new_emp() RETURNS emp AS $$

SELECT ROW(’None’, 1000.0, 25, ’(2,2)’)::emp;
$$ LANGUAGE SQL;
SELECT new_emp();
new_emp
--------------------------
(None,1000.0,25,"(2,2)")

SELECT * FROM new_emp();

name | salary | age | cubicle
------+--------+-----+---------
None | 1000.0 | 25 | (2,2)

SELECT (new_emp()).name;
name
------
None

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
SQL Functions
JSONB and SP-GIST
PL/pgSQL Functions
Functions - Overview
Miscellaneous
Function Basics
Functions - By Example

VARIADIC

CREATE FUNCTION mleast(VARIADIC numeric[]) RETURNS numeric AS $$

SELECT min($1[i]) FROM generate_subscripts($1, 1) g(i);
$$ LANGUAGE SQL;

SELECT mleast(10, -1, 5, 4.4);

mleast
--------
-1
(1 row)

SELECT mleast(42, 6, 42.42);

mleast
--------
6
(1 row)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
SQL Functions
JSONB and SP-GIST
PL/pgSQL Functions
Functions - Overview
Miscellaneous
Function Basics
Functions - By Example

DEFAULT Arguments

CREATE FUNCTION foo(a int, b int DEFAULT 2, c int DEFAULT 3)

RETURNS int LANGUAGE SQL AS $$SELECT $1 + $2 + $3$$;

SELECT foo(10, 20, 30);

foo
-----
60
(1 row)

SELECT foo(10, 20);

foo
-----
33
(1 row)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
SQL Functions
JSONB and SP-GIST
PL/pgSQL Functions
Functions - Overview
Miscellaneous
Function Basics
Functions - By Example

PL/pgSQL

PL/pgSQL is SQL plus procedural elements

variables
if/then/else
loops
cursors
error checking
Loading the language handler into a database:
CREATE EXTENSION plpgsql;
ERROR: extension "plpgsql" already exists

http://www.postgresql.org/docs/9.4/static/plpgsql.html

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
SQL Functions
JSONB and SP-GIST
PL/pgSQL Functions
Functions - Overview
Miscellaneous
Function Basics
Functions - By Example

Simple

CREATE OR REPLACE FUNCTION sum (text, text)

RETURNS text AS $$
BEGIN
RETURN $1 || ’ ’ || $2;
END;
$$ LANGUAGE plpgsql;

SELECT sum(’hello’, ’world’);

sum
-------------
hello world
(1 row)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
SQL Functions
JSONB and SP-GIST
PL/pgSQL Functions
Functions - Overview
Miscellaneous
Function Basics
Functions - By Example

Parameter ALIAS

CREATE OR REPLACE FUNCTION sum (int, int)

RETURNS int AS $$
DECLARE
i ALIAS FOR $1;
j ALIAS FOR $2;
sum int;
BEGIN
sum := i + j;
RETURN sum;
END;
$$ LANGUAGE plpgsql;

SELECT sum(41, 1);

sum
-----
42
(1 row)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
SQL Functions
JSONB and SP-GIST
PL/pgSQL Functions
Functions - Overview
Miscellaneous
Function Basics
Functions - By Example

Named Parameters

CREATE OR REPLACE FUNCTION sum (i int, j int)

RETURNS int AS $$
DECLARE
sum int;
BEGIN
sum := i + j;
RETURN sum;
END;
$$ LANGUAGE plpgsql;

SELECT sum(41, 1);

sum
-----
42
(1 row)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
SQL Functions
JSONB and SP-GIST
PL/pgSQL Functions
Functions - Overview
Miscellaneous
Function Basics
Functions - By Example

Control Structures: IF ...

CREATE OR REPLACE FUNCTION even (i int)

RETURNS boolean AS $$
DECLARE
tmp int;
BEGIN
tmp := i % 2;
IF tmp = 0 THEN RETURN true;
ELSE RETURN false;
END IF;
END;
$$ LANGUAGE plpgsql;

SELECT even(3), even(42);

even | even
------+------
f | t
(1 row)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
SQL Functions
JSONB and SP-GIST
PL/pgSQL Functions
Functions - Overview
Miscellaneous
Function Basics
Functions - By Example

Control Structures: FOR ... LOOP

CREATE OR REPLACE FUNCTION factorial (i numeric)

RETURNS numeric AS $$
DECLARE
tmp numeric; result numeric;
BEGIN
result := 1;
FOR tmp IN 1 .. i LOOP
result := result * tmp;
END LOOP;
RETURN result;
END;
$$ LANGUAGE plpgsql;
SELECT factorial(42::numeric);
factorial
------------------------------------------------------
1405006117752879898543142606244511569936384000000000
(1 row)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
SQL Functions
JSONB and SP-GIST
PL/pgSQL Functions
Functions - Overview
Miscellaneous
Function Basics
Functions - By Example

Control Structures: WHILE ... LOOP

CREATE OR REPLACE FUNCTION factorial (i numeric)
RETURNS numeric AS $$
DECLARE tmp numeric; result numeric;
BEGIN
result := 1; tmp := 1;
WHILE tmp <= i LOOP
result := result * tmp;
tmp := tmp + 1;
END LOOP;
RETURN result;
END;
$$ LANGUAGE plpgsql;

SELECT factorial(42::numeric);
factorial
------------------------------------------------------
1405006117752879898543142606244511569936384000000000
(1 row)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
SQL Functions
JSONB and SP-GIST
PL/pgSQL Functions
Functions - Overview
Miscellaneous
Function Basics
Functions - By Example

Recursive
CREATE OR REPLACE FUNCTION factorial (i numeric)
RETURNS numeric AS $$
BEGIN
IF i = 0 THEN
RETURN 1;
ELSIF i = 1 THEN
RETURN 1;
ELSE
RETURN i * factorial(i - 1);
END IF;
END;
$$ LANGUAGE plpgsql;

SELECT factorial(42::numeric);
factorial
------------------------------------------------------
1405006117752879898543142606244511569936384000000000
(1 row)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
SQL Functions
JSONB and SP-GIST
PL/pgSQL Functions
Functions - Overview
Miscellaneous
Function Basics
Functions - By Example

Record types

CREATE OR REPLACE FUNCTION format ()

RETURNS text AS $$
DECLARE
tmp RECORD;
BEGIN
SELECT INTO tmp 1 + 1 AS a, 2 + 2 AS b;
RETURN ’a = ’ || tmp.a || ’; b = ’ || tmp.b;
END;
$$ LANGUAGE plpgsql;

select format();
format
--------------
a = 2; b = 4
(1 row)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
SQL Functions
JSONB and SP-GIST
PL/pgSQL Functions
Functions - Overview
Miscellaneous
Function Basics
Functions - By Example

PERFORM
CREATE OR REPLACE FUNCTION func_w_side_fx() RETURNS void AS
$$ INSERT INTO foo VALUES (41),(42) $$ LANGUAGE sql;

CREATE OR REPLACE FUNCTION dummy ()

RETURNS text AS $$
BEGIN
PERFORM func_w_side_fx();
RETURN ’OK’;
END;
$$ LANGUAGE plpgsql;

SELECT dummy();
SELECT * FROM foo;
f1
----
41
42
(2 rows)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
SQL Functions
JSONB and SP-GIST
PL/pgSQL Functions
Functions - Overview
Miscellaneous
Function Basics
Functions - By Example

Dynamic SQL

CREATE OR REPLACE FUNCTION get_foo(i int)

RETURNS foo AS $$
DECLARE
rec RECORD;
BEGIN
EXECUTE ’SELECT * FROM foo WHERE f1 = ’ || i INTO rec;
RETURN rec;
END;
$$ LANGUAGE plpgsql;

SELECT * FROM get_foo(42);

f1
----
42
(1 row)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
SQL Functions
JSONB and SP-GIST
PL/pgSQL Functions
Functions - Overview
Miscellaneous
Function Basics
Functions - By Example

Cursors
CREATE OR REPLACE FUNCTION totalbalance()
RETURNS numeric AS $$
DECLARE
tmp RECORD; result numeric;
BEGIN
result := 0.00;
FOR tmp IN SELECT * FROM foo LOOP
result := result + tmp.f1;
END LOOP;
RETURN result;
END;
$$ LANGUAGE plpgsql;

SELECT totalbalance();
totalbalance
--------------
83.00
(1 row)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
SQL Functions
JSONB and SP-GIST
PL/pgSQL Functions
Functions - Overview
Miscellaneous
Function Basics
Functions - By Example

Error Handling

CREATE OR REPLACE FUNCTION safe_add(a integer, b integer)

RETURNS integer AS $$
BEGIN
RETURN a + b;
EXCEPTION
WHEN numeric_value_out_of_range THEN
-- do some important stuff
RETURN -1;
WHEN OTHERS THEN
-- do some other important stuff
RETURN -1;
END;
$$ LANGUAGE plpgsql;

http://www.postgresql.org/docs/9.4/static/errcodes-appendix.html

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
SQL Functions
JSONB and SP-GIST
PL/pgSQL Functions
Functions - Overview
Miscellaneous
Function Basics
Functions - By Example

Nested Exception Blocks

CREATE FUNCTION merge_db(key integer, data text)
RETURNS void AS $$
BEGIN
LOOP
UPDATE db SET b = data WHERE a = key;
IF found THEN RETURN;
END IF;
BEGIN
INSERT INTO db (a, b) VALUES (key, data);
RETURN;
EXCEPTION WHEN unique_violation THEN
-- do nothing
END;
END LOOP;
EXCEPTION WHEN OTHERS THEN
-- do something else
END;
$$ LANGUAGE plpgsql;

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
SQL Functions
JSONB and SP-GIST
PL/pgSQL Functions
Functions - Overview
Miscellaneous
Function Basics
Functions - By Example

Window Function
CREATE TABLE mydata (
pk int primary key,
mydate date NOT NULL,
gender text NOT NULL CHECK(gender IN (’M’,’F’)),
mygroup text NOT NULL,
id int NOT NULL
);

INSERT INTO mydata VALUES

(1, ’2012-03-25’,’F’,’A’,1),(2, ’2005-05-23’,’F’,’B’,2),
(3, ’2005-09-08’,’F’,’B’,2),(4, ’2005-12-07’,’F’,’B’,2),
(5, ’2006-02-26’,’F’,’C’,2),(6, ’2006-05-13’,’F’,’C’,2),
(7, ’2006-09-01’,’F’,’C’,2),(8, ’2006-12-12’,’F’,’D’,2),
(9, ’2006-02-19’,’F’,’D’,2),(10,’2006-05-03’,’F’,’D’,2),
(11,’2006-04-23’,’F’,’D’,2),(12,’2007-12-08’,’F’,’D’,2),
(13,’2011-03-19’,’F’,’D’,2),(14,’2007-12-20’,’M’,’A’,3),
(15,’2008-06-15’,’M’,’A’,3),(16,’2008-12-16’,’M’,’A’,3),
(17,’2009-06-07’,’M’,’B’,3),(18,’2009-10-09’,’M’,’B’,3),
(19,’2010-01-28’,’M’,’B’,3),(20,’2007-06-05’,’M’,’A’,4);
Stephen Frost, Joe Conway Postgres Open 2014
Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
SQL Functions
JSONB and SP-GIST
PL/pgSQL Functions
Functions - Overview
Miscellaneous
Function Basics
Functions - By Example

Window Function
SELECT id, gender, obs_days, sum(chgd) as num_changes FROM
(SELECT id, gender,
CASE WHEN row_number() OVER w > 1
AND mygroup <> lag(mygroup) OVER w THEN 1
ELSE 0 END AS chgd,
last_value(mydate) OVER w - first_value(mydate) OVER w AS obs_days
FROM mydata
WINDOW w AS
(PARTITION BY id, gender ORDER BY id, gender, mydate
ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING)
) AS ss GROUP BY id, gender, obs_days ORDER BY id, gender;
id | gender | obs_days | num_changes
----+--------+----------+-------------
1 | F | 0 | 0
2 | F | 2126 | 5
3 | M | 770 | 1
4 | M | 0 | 0
(4 rows)

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
SQL Functions
JSONB and SP-GIST
PL/pgSQL Functions
Functions - Overview
Miscellaneous
Function Basics
Functions - By Example

Lateral

SELECT d.datname, u.rolname, c.config

Stephen Frost, Joe Conway Postgres Open 2014

Adv. SQL - Basics
Adv. SQL - Window Functions, CTEs, LATERAL
SQL Functions
JSONB and SP-GIST
PL/pgSQL Functions
Functions - Overview
Miscellaneous
Function Basics
Functions - By Example

Thank You

Questions?

Stephen Frost, Joe Conway Postgres Open 2014

Geographical Data Science and Spatial Data Analysis An Introduction in R (Spatial Analytics and GIS) 1st Edition
100% (1)
Geographical Data Science and Spatial Data Analysis An Introduction in R (Spatial Analytics and GIS) 1st Edition
384 pages
Chartek Quality Control Manual (Rev9) 2023-07-07
No ratings yet
Chartek Quality Control Manual (Rev9) 2023-07-07
36 pages
Teradata Notes
No ratings yet
Teradata Notes
1 page
Y3 Ug Handbook 2019 2020 Final 8october2019 45543
No ratings yet
Y3 Ug Handbook 2019 2020 Final 8october2019 45543
37 pages
Operation Handbook - Hollow Fiber Cartridges For Membrane Separations PDF
No ratings yet
Operation Handbook - Hollow Fiber Cartridges For Membrane Separations PDF
28 pages
Leetcode SQL QnA 1693149052
No ratings yet
Leetcode SQL QnA 1693149052
60 pages
Stored Procedures in MS SQL Server
No ratings yet
Stored Procedures in MS SQL Server
5 pages
Introduction To Kubernetes - KubeCon Slide Deck
No ratings yet
Introduction To Kubernetes - KubeCon Slide Deck
60 pages
Are You Addicted To Mobile Phones - Google Forms
No ratings yet
Are You Addicted To Mobile Phones - Google Forms
6 pages
Data Analysis With SQL: Postgresql Cheat Sheet
No ratings yet
Data Analysis With SQL: Postgresql Cheat Sheet
4 pages
Oracle SQL and PL/SQL
From Everand
Oracle SQL and PL/SQL
Niraj Gupta
4.5/5 (8)
Beginning Microsoft SQL Server 2012 Programming
From Everand
Beginning Microsoft SQL Server 2012 Programming
Paul Atkinson
1/5 (1)
Advanced SQL
No ratings yet
Advanced SQL
31 pages
DataWarehouseInterview Part1
No ratings yet
DataWarehouseInterview Part1
4 pages
SQL
No ratings yet
SQL
141 pages
SQL Server and ASP Net Questions & Answers
No ratings yet
SQL Server and ASP Net Questions & Answers
12 pages
41 Essential SQL Interview Questions and Answers - Toptal®
100% (1)
41 Essential SQL Interview Questions and Answers - Toptal®
26 pages
SQL Server Database Mirroring Concept
No ratings yet
SQL Server Database Mirroring Concept
75 pages
SQL and PLSQL
100% (1)
SQL and PLSQL
129 pages
SQL Material
100% (3)
SQL Material
252 pages
SQL DW
No ratings yet
SQL DW
596 pages
Analaytical Function-Pravin
No ratings yet
Analaytical Function-Pravin
24 pages
Introduction To SQL Stored Procedures
No ratings yet
Introduction To SQL Stored Procedures
3 pages
SQL Server String To Date
No ratings yet
SQL Server String To Date
15 pages
ADD Add Constraint Alter: Keyword Description
No ratings yet
ADD Add Constraint Alter: Keyword Description
5 pages
Client Server Model: Many Databases Applications Are Built in
No ratings yet
Client Server Model: Many Databases Applications Are Built in
52 pages
Query Performance Tuning
No ratings yet
Query Performance Tuning
35 pages
SQL Questions
No ratings yet
SQL Questions
4 pages
Oracle Analytical Functions 1
No ratings yet
Oracle Analytical Functions 1
18 pages
Etl Architecture
No ratings yet
Etl Architecture
76 pages
SQL Difference
No ratings yet
SQL Difference
61 pages
SSIS Creating A Custom Transformation Component
0% (1)
SSIS Creating A Custom Transformation Component
38 pages
SQL 80 Plus Queries
No ratings yet
SQL 80 Plus Queries
15 pages
SQL Statement Tunning
No ratings yet
SQL Statement Tunning
19 pages
Querying Microsoft SQL Server
No ratings yet
Querying Microsoft SQL Server
3 pages
SSRS 2012 Material
No ratings yet
SSRS 2012 Material
58 pages
Optimizing SQL Queries in Oracle
0% (1)
Optimizing SQL Queries in Oracle
69 pages
Complex SQL Writing and Performance Tuning
0% (1)
Complex SQL Writing and Performance Tuning
4 pages
Querying With T-SQL - 01
No ratings yet
Querying With T-SQL - 01
24 pages
0oracle SQL Scribe)
No ratings yet
0oracle SQL Scribe)
128 pages
SQL Questions
No ratings yet
SQL Questions
13 pages
PLSQL For Beginners - Training Guide
100% (1)
PLSQL For Beginners - Training Guide
176 pages
SSIS Cheatsheet
100% (1)
SSIS Cheatsheet
1 page
SCD Type1
100% (1)
SCD Type1
16 pages
Informatica Performance Tuning
No ratings yet
Informatica Performance Tuning
35 pages
Day65 - Day70 Power BI Interview
No ratings yet
Day65 - Day70 Power BI Interview
31 pages
Talend Data Integration: Subramanyam K
No ratings yet
Talend Data Integration: Subramanyam K
64 pages
Create Materialized View
No ratings yet
Create Materialized View
2 pages
SQL Performance Improvement
No ratings yet
SQL Performance Improvement
94 pages
TSQL Interview Questions and Answers
No ratings yet
TSQL Interview Questions and Answers
8 pages
SQL Refresher Complete Notes PDF
No ratings yet
SQL Refresher Complete Notes PDF
352 pages
Advanced SQL Functions
No ratings yet
Advanced SQL Functions
12 pages
Data Modeling 101:: Bringing Data Professionals and Application Developers Together
No ratings yet
Data Modeling 101:: Bringing Data Professionals and Application Developers Together
46 pages
WEEK 4 - What Is Common Table Expressions
No ratings yet
WEEK 4 - What Is Common Table Expressions
3 pages
Oracle Technology Best Practices
No ratings yet
Oracle Technology Best Practices
52 pages
HR Schema Script
No ratings yet
HR Schema Script
54 pages
AppDynamics Third Edition
From Everand
AppDynamics Third Edition
Gerardus Blokdyk
No ratings yet
Sql Plsql Oracle
From Everand
Sql Plsql Oracle
Andrew Igla
No ratings yet
Microsoft SQL Server 2008 R2 Administration Cookbook
From Everand
Microsoft SQL Server 2008 R2 Administration Cookbook
Satya Shyam K Jayanty
5/5 (1)
Instant Pentaho Data Integration Kitchen
From Everand
Instant Pentaho Data Integration Kitchen
Sergio Ramazzina
No ratings yet
Data Engineering with Scala and Spark: Build streaming and batch pipelines that process massive amounts of data using Scala
From Everand
Data Engineering with Scala and Spark: Build streaming and batch pipelines that process massive amounts of data using Scala
Eric Tome
No ratings yet
My Part-Time Study Notes on Mssql Server
From Everand
My Part-Time Study Notes on Mssql Server
Morris Sebenzile Mntoninzi
No ratings yet
Big Data and Analytics: The key concepts and practical applications of big data analytics (English Edition)
From Everand
Big Data and Analytics: The key concepts and practical applications of big data analytics (English Edition)
Dr. Jugnesh Kumar
No ratings yet
Oracle Quick Guides: Part 2 - Oracle Database Design
From Everand
Oracle Quick Guides: Part 2 - Oracle Database Design
Malcolm Coxall
No ratings yet
TIBCO Spotfire Complete Self-Assessment Guide
From Everand
TIBCO Spotfire Complete Self-Assessment Guide
Gerardus Blokdyk
No ratings yet
PostgreSQL 9 High Availability Cookbook
From Everand
PostgreSQL 9 High Availability Cookbook
Shaun M. Thomas
5/5 (2)
An Introduction To Kubernetes (Feb 2019)
No ratings yet
An Introduction To Kubernetes (Feb 2019)
55 pages
Intro To PostgreSQL
100% (3)
Intro To PostgreSQL
54 pages
Docker Management Design Patterns Swarm Mode On Amazon Web Services
No ratings yet
Docker Management Design Patterns Swarm Mode On Amazon Web Services
329 pages
2017 07 07 Rails On Docker Getting Started Docker Ruby Rails
No ratings yet
2017 07 07 Rails On Docker Getting Started Docker Ruby Rails
23 pages
IEEE - Ethically Aligned Design PP 17-35
No ratings yet
IEEE - Ethically Aligned Design PP 17-35
294 pages
Kanishk Jha CV
No ratings yet
Kanishk Jha CV
2 pages
Introduction To DC-DC Converter
No ratings yet
Introduction To DC-DC Converter
5 pages
Fundamental and Ratio Analysis of IFL and HDFC Life Insurance
100% (1)
Fundamental and Ratio Analysis of IFL and HDFC Life Insurance
8 pages
ROSA Aflatoxin (Quantitative) Test - Prepared For Cargill
100% (1)
ROSA Aflatoxin (Quantitative) Test - Prepared For Cargill
4 pages
Unit 4 - Entrepreneurial Risk and Stress in Tourism
No ratings yet
Unit 4 - Entrepreneurial Risk and Stress in Tourism
15 pages
BP11 Container Safety Convention 2011 PDF
100% (1)
BP11 Container Safety Convention 2011 PDF
44 pages
4.2.2.10 Packet Tracer - Configuring Extended ACLs Scenario 1
No ratings yet
4.2.2.10 Packet Tracer - Configuring Extended ACLs Scenario 1
9 pages
Chapter 4 Data Mining
No ratings yet
Chapter 4 Data Mining
5 pages
Fitanezajowetavinewul
No ratings yet
Fitanezajowetavinewul
2 pages
A. D. Gothong Manufacturing Corporation Employees Union-ALU vs. Confesor
No ratings yet
A. D. Gothong Manufacturing Corporation Employees Union-ALU vs. Confesor
11 pages
Various Types of Deped Correspondence
0% (1)
Various Types of Deped Correspondence
10 pages
Challan Income Tax
No ratings yet
Challan Income Tax
1 page
Control de Flujo F261
No ratings yet
Control de Flujo F261
12 pages
Reflection Paper 1 - Career and Vocation Plan
No ratings yet
Reflection Paper 1 - Career and Vocation Plan
3 pages
Mli102 - Management of Library and Information Centres
100% (1)
Mli102 - Management of Library and Information Centres
49 pages
Ad360 Quick Start Guide
No ratings yet
Ad360 Quick Start Guide
25 pages
Manual de Usuario Videolaringoscopio
No ratings yet
Manual de Usuario Videolaringoscopio
43 pages
RCRA Watchdog - Overview of DuPont Leland NC Hazwaste Articles
No ratings yet
RCRA Watchdog - Overview of DuPont Leland NC Hazwaste Articles
2 pages
05 v.I.P. System BASE I Processing Specifications 0847
No ratings yet
05 v.I.P. System BASE I Processing Specifications 0847
224 pages
What's New at 2019 R1 Mechanical Enhancements
No ratings yet
What's New at 2019 R1 Mechanical Enhancements
21 pages
Ipsos Pride Report 2024
No ratings yet
Ipsos Pride Report 2024
58 pages
Manwah
No ratings yet
Manwah
7 pages
Beepedia Monthly Current Affairs (Beepedia) December 2023
No ratings yet
Beepedia Monthly Current Affairs (Beepedia) December 2023
99 pages
Botswana Development Control Code Sept 2013
No ratings yet
Botswana Development Control Code Sept 2013
341 pages