INTRODUCTION TO SQL

CS 564- Fall 2020

ACKs: Dan Suciu, Jignesh Patel, AnHai Doan

WHAT IS THIS LECTURE ABOUT

• The Relational Model

• SQL: Basics
– creating a table
– primary keys
• SQL: Single-table queries
– SELECT-FROM-WHERE structure
– DISTINCT/ORDER BY/LIMIT
• SQL: Multi-table queries
– foreign keys
– joins
CS 564 [Fall 2020] - Paris Koutris 2
RELATIONAL MODEL

CS 564 [Fall 2020] - Paris Koutris 3

RELATIONAL MODEL

• first proposed by Codd in 1969

• has just a single concept: relation
• the world is represented as a collection of tables
• well-suited for efficient manipulations on
computers

CS 564 [Fall 2020] - Paris Koutris 4

RELATION

The data is stored in tables (or relations)

table name attribute name
PRODUCT
name category price manufacturer
iPad tablet $399.00 Apple
Surface tablet $299.00 Microsoft
… … … …

record/tuple

CS 564 [Fall 2020] - Paris Koutris 5

DOMAINS

• Each attribute has an atomic type called domain

• A domain specifies the set of values allowed
• Examples:
– integer
– string
– real
PRODUCT(name: string,
category: string,
price: real,
manufacturer: string)

CS 564 [Fall 2020] - Paris Koutris 6

SCHEMA

The schema of a relation:

– relation name + attribute names
– Product (name, price, category, manufacturer)
– In practice we add the domain for each attribute

The schema of a database:

– a collection of relation schemas

CS 564 [Fall 2020] - Paris Koutris 7

INSTANCE

The instance of a relation:

– a set of tuples or records

The instance of a database:

– a collection of relation instances

CS 564 [Fall 2020] - Paris Koutris 8

EXAMPLE

PRODUCT(name: string, schema

category: string,
price: real,
manufacturer: string)
instance

name category price manufacturer

iPad tablet $399.00 Apple
Surface tablet $299.00 Microsoft
… … … …

CS 564 [Fall 2020] - Paris Koutris 9

SCHEMA VS INSTANCE

• Analogy with programming languages:

– schema ~ type
– instance ~ value
• Important distinction
– schema: stable over long periods of time
– instance: changes constantly, as data is
inserted/updated/deleted

CS 564 [Fall 2020] - Paris Koutris 10

SQL: BASICS

CS 564 [Fall 2020] - Paris Koutris 11

WHAT IS SQL?

• The most widely used database language

• Used to query and manipulate data

• SQL stands for Structured Query Language

– many SQL standards: SQL-92, SQL:1999, SQL:2011
– vendors support different subsets
– we will discuss the common functionality

CS 564 [Fall 2020] - Paris Koutris 12

 CREATING A TABLE

table name

CREATE TABLE Author(

attributes authorid INTEGER PRIMARY KEY,
firstname CHAR(20), atomic types
lastname CHAR(30),
);

CS 564 [Fall 2020] - Paris Koutris 13

PRIMARY KEYS

A primary key is a minimal subset of attributes that

is a unique identifier of tuples in a relation

• A key is an implicit constraint on which tuples can

be in the relation
• In SQL we specify that an attribute is the primary
key with the keyword PRIMARY KEY

CS 564 [Fall 2020] - Paris Koutris 14

UNIQUE KEYS

• We can also define a unique key: a subset of attributes that

uniquely defines a row:
CREATE TABLE Author(
authorid INTEGER UNIQUE,
firstname CHAR(20)) ;
• There can be only one primary key, but many unique keys!

CS 564 [Fall 2020] - Paris Koutris 15

NULL VALUES

• tuples in SQL relations can have NULL as a value

for one or more attributes
• The meaning depends on context:
– missing value: e.g. we know that Greece has
some population, but we don’t know what it is
– inapplicable: e.g. the value of attribute spouse
for an unmarried person

CS 564 [Fall 2020] - Paris Koutris 16

NULL VALUES

When creating a table in SQL, we can assert that a

particular attribute takes no NULL values

CREATE TABLE Author(

authorid INTEGER PRIMARY KEY,
firstname CHAR(20) NOT NULL,
lastname CHAR(30)
);

CS 564 [Fall 2020] - Paris Koutris 17

POPULATING A TABLE

• To insert a single tuple:

INSERT INTO <relation>
VALUES ( <list of values>);
• We may add to the relation name a list of
attributes (if we forget the order)

INSERT INTO Author

VALUES(001, 'Dan’, 'Brown');

CS 564 [Fall 2020] - Paris Koutris 18

SQL: SINGLE-TABLE QUERIES

CS 564 [Fall 2020] - Paris Koutris 19

BASIC SQL QUERY

optional

SELECT [DISTINCT] attributes

FROM one or more tables
WHERE conditions on the tables

conditions of the form: Attr1 op Attr2

CS 564 [Fall 2020] - Paris Koutris 20

EXAMPLE

What is the population of USA?

PROJECTION: keeps only the
specified attributes
SELECT Population
FROM Country
WHERE Code = 'USA';

SELECTION: filters the tuples of the relation

CS 564 [Fall 2020] - Paris Koutris 21

SEMANTICS

1. Think of a tuple variable ranging over each tuple

of the relation mentioned in FROM
2. Check if the current tuple satisfies the WHERE
clause
3. If so, compute the attributes or expressions of the
SELECT clause using this tuple

CS 564 [Fall 2020] - Paris Koutris 22

* IN SELECT CLAUSES
When there is one relation in the FROM clause, * in
the SELECT clause stands for “all attributes of this
relation”

SELECT *
FROM City
WHERE Population >= '1000000'
AND CountryCode = 'USA';

CS 564 [Fall 2020] - Paris Koutris 23

RENAMING ATTRIBUTES

If we want the output schema to have different

attribute names, we can use AS <new name> to
rename an attribute

SELECT Name AS LargeUSACity

FROM City
WHERE Population >= '1000000'
AND CountryCode = 'USA';

CS 564 [Fall 2020] - Paris Koutris 24

ARITHMETIC EXPRESSIONS

We can use any arithmetic expression (that makes

sense) in the SELECT clause

SELECT Name,
(Population/ 1000000) AS PopulationInMillion
FROM City
WHERE Population >= '1000000’ ;

CS 564 [Fall 2020] - Paris Koutris 25

WHAT CAN WE USE IN WHERE CLAUSES?

• attribute names of the relations that appear in the

FROM clause
• comparison operators: =, <>, <, >, <=, >=
• arithmetic operations (+, -, /, *)
• AND, OR, NOT to combine conditions
• operations on strings (e.g. concatenation)
• pattern matching: s LIKE p
• special functions for comparing dates and times

CS 564 [Fall 2020] - Paris Koutris 26

PATTERN MATCHING

s LIKE p: pattern matching on strings

– % = any sequence of characters
– _ = any single character

SELECT Name, GovernmentForm

FROM Country
WHERE GovernmentForm LIKE '%Monarchy%';

CS 564 [Fall 2020] - Paris Koutris 27

USING DISTINCT

• The default semantics of SQL is bag semantics

(duplicate tuples are allowed in the output)
• The use of DISTINCT in the SELECT clause
removes all duplicate tuples in the result, and
returns a set
SELECT DISTINCT GovernmentForm
FROM Country;

CS 564 [Fall 2020] - Paris Koutris 28

ORDER BY

The use of ORDER BY orders the tuples by the

attribute we specify in decreasing (DESC) or
increasing (ASC) order

SELECT Name, Population

FROM City
WHERE Population >= ‘1000000’
ORDER BY Population DESC;

CS 564 [Fall 2020] - Paris Koutris 29

LIMIT

• The use of LIMIT <number> limits the output to

be only the specified number of tuples
• It can be used with ORDER BY to get the
maximum or minimum value of an attribute!

SELECT Name, Population

FROM City
ORDER BY Population DESC
LIMIT 2;

CS 564 [Fall 2020] - Paris Koutris 30

SQL: MULTI-TABLE QUERIES

CS 564 [Fall 2020] - Paris Koutris 31

FOREIGN KEYS

Suppose that we want to create a table Book, and

make sure that the author of the book exists in the
table Author

CREATE TABLE Book(

bookid INTEGER PRIMARY KEY,
title TEXT,
authorid INTEGER,
FOREIGN KEY (authorid) REFERENCES
Author(authorid));
CS 564 [Fall 2020] - Paris Koutris 32
FOREIGN KEYS

• Use the keyword REFERENCES, as:

FOREIGN KEY ( <list of attributes> )

REFERENCES <relation> ( <attributes> )

• Referenced attributes must be declared PRIMARY

KEY or UNIQUE

CS 564 [Fall 2020] - Paris Koutris 33

ENFORCING FK CONSTRAINTS

If there is a foreign-key constraint from attributes of

relation R to the primary key of relation S, two
violations are possible:
1. An insert or update to R introduces values not
found in S
2. A deletion or update to S causes some tuples of R
to dangle

There are 3 ways to enforce foreign key constraints!

CS 564 [Fall 2020] - Paris Koutris 34
ACTION 1: REJECT

• The insertion/deletion/update query is rejected

and not executed in the DBMS
• This is the default action if a foreign key constraint
is declared

CS 564 [Fall 2020] - Paris Koutris 35

ACTION 2: CASCADE UPDATE

When a tuple referenced is updated, the update

propagates to the tuples that reference it

CREATE TABLE Book(

bookid INTEGER PRIMARY KEY,
title TEXT,
authorid INTEGER,
FOREIGN KEY (authorid) REFERENCES
Author(authorid)
ON UPDATE CASCADE);

CS 564 [Fall 2020] - Paris Koutris 36

ACTION 2: CASCADE DELETE

When a tuple referenced is deleted, the deletion

propagates to the tuples that reference it

CREATE TABLE Book(

bookid INTEGER PRIMARY KEY,
title TEXT,
authorid INTEGER,
FOREIGN KEY (authorid) REFERENCES
Author(authorid)
ON DELETE CASCADE);

CS 564 [Fall 2020] - Paris Koutris 37

ACTION 3: SET NULL

• When a delete/update occurs, the values that

reference the deleted tuple are set to NULL

CREATE TABLE Book(

bookid INTEGER PRIMARY KEY,
title TEXT,
authorid INTEGER,
FOREIGN KEY (authorid) REFERENCES
Author(authorid)
ON UPDATE SET NULL);
CS 564 [Fall 2020] - Paris Koutris 38
WHAT SHOULD WE CHOOSE?

• When we declare a foreign key, we may choose

policies SET NULL or CASCADE independently for
deletions and updates
ON [UPDATE, DELETE] [SET NULL, CASCADE]

• Otherwise, the default policy (reject) is used

CS 564 [Fall 2020] - Paris Koutris 39

MULTIPLE RELATIONS

• We often want to combine data from more than

one relation
• We can address several relations in one query by
listing them all in the FROM clause
• If two attributes from different relations have the
same name, we can distinguish them by writing
<relation>.<attribute>

CS 564 [Fall 2020] - Paris Koutris 40

EXAMPLE

What is the name of countries that speak Greek?

SELECT Name
FROM Country, CountryLanguage
WHERE Code = CountryCode
AND Language = 'Greek';

This is BAD style!!

CS 564 [Fall 2020] - Paris Koutris 41

EXAMPLE: GOOD STYLE

SELECT Country.Name
FROM Country, CountryLanguage
WHERE Country.Code=CountryLanguage.CountryCode
AND CountryLanguage.Language = 'Greek';

SELECT C.Name
FROM Country C, CountryLanguage L
WHERE C.Code = L.CountryCode
AND L.Language = 'Greek';

CS 564 [Fall 2020] - Paris Koutris 42

VARIABLES

Variables are necessary when we want to use two

copies of the same relation in the FROM clause

SELECT C.Name
FROM Country C, CountryLanguage L1,
CountryLanguage L2
WHERE C.Code = L1.CountryCode
AND C.Code = L2.CountryCode
AND L1.Language = 'Greek’
AND L2.Language = 'English';

CS 564 [Fall 2020] - Paris Koutris 43

SEMANTICS: SELECT-FROM-WHERE

1. Start with the cross product of all the relations in

the FROM clause
2. Apply the conditions from the WHERE clause
3. Project onto the list of attributes and expressions
in the SELECT clause
4. If DISTINCT is specified, eliminate duplicate
rows

CS 564 [Fall 2020] - Paris Koutris 44

SEMANTICS OF SQL: EXAMPLE

SELECT R.D
FROM R, S
WHERE R.A = S.B AND S.C = ‘e’ ;
A D
1 a A D B C A D B C
1 a 1 d select 2 b 2 e
2 b
cross product 1 a 2 e 2 c 2 e
2 c
2 b 1 d
project
B C 2 b 2 e
1 d 2 c 1 d D
2 e 2 c 2 e b
c
CS 564 [Fall 2020] - Paris Koutris 45
SEMANTICS OF SQL: NESTED LOOP

SELECT a1, a2, …, ak

FROM R1, R2, …, Rn
WHERE Conditions

answer := {}
for t1 in R1 do
for t2 in R2 do
…..
for tn in Rn do
if Conditions
then answer := answer ∪ {(a1,…,ak)}
return answer

CS 564 [Fall 2020] - Paris Koutris 46

SEMANTICS OF SQL

• The query processor will almost never evaluate

the query this way
• SQL is a declarative language
• The DBMS figures out the most efficient way to
compute it
• We will discuss this later in the course when we
talk about query optimization

CS 564 [Fall 2020] - Paris Koutris 47