Constraints and Triggers

Instructor: Vu Tuyet Trinh

Learning objectives

•Upon completion of this lesson, students will be able to:

1. Well known about different constraints and define them correctly
2. Understand triggers: What is a trigger? how it works ? When using?
3. Define simple triggers

1
 Outline
• Constraints
• Triggers

1. Introduction

Teachers

Student
DBMS Server

Clazz
Enrollment
Studying Subject
…

Integrity of the
Students
information?

Applications

2
 1. Introduction: Database Schema
student(student_id, first_name,last_name, dob, gender,
address, note, email, clazz_id)
clazz(clazz_id, name, lecturer_id, monitor_id, number_students)
subject(subject_id, name, credit, percentage_final_exam)
enrollment(student_id, subject_id, semester, midterm_score, final_score)
lecturer(lecturer_id, first_name, last_name, dob, gender, address, email)
teaching(subject_id, lecturer_id)
grade(code, from_score, to_score)

1. Introduction: Constraints and Triggers

• A constraint is a relationship among data elements that the DBMS

is required to enforce
• Example: key constraints
• Triggers are only executed when a specified condition occurs,
e.g., insertion of a tuple
• Easier to implement than complex constraints

3
 2. Constraints
2.1. Keys: PRIMARY KEY vs. UNIQUE
2.2. Foreign keys
2.3. Attribute-based checks
2.4. Tuple-based checks
2.5. Assertions

2. Constraints: Kinds of Constraints

• Keys
• Foreign-key, or referential-integrity
• Value-based constraints
• Constrain values of a particular attribute.
• Tuple-based constraints
• Relationship among components
• Assertions: any SQL boolean expression

4
 2.1. Keys: PRIMARY KEY vs. UNIQUE
• Declaring: similar syntax as primary key
• Example:
CREATE TABLE student (
student_id CHAR(8) NOT NULL,
first_name VARCHAR(20) NOT NULL,
last_name VARCHAR(20) NOT NULL,
...
email varchar(50) UNIQUE,
clazz_id CHAR(8),
CONSTRAINT student_pk PRIMARY KEY (student_id));

2.1. Keys: PRIMARY KEY vs. UNIQUE

PRIMARY KEY UNIQUE KEY
Number defined on table One Multiple
Null columns allowed No Yes
Default index CLUSTERED NON-CLUSTERED
Purpose Enforce Entity Integrity Enforce Unique Data
Number of columns One or more columns One or more columns
Referenced by a Foreign Key Yes Yes
Constraint

10

10

5
 2.2. Expressing Foreign Keys

• Use keyword REFERENCES, either:

1. After an attribute (for one-attribute keys)
2. As an element of the schema:
[CONSTRAINT <name>] FOREIGN KEY (<list of attributes>)
REFERENCES <relation> (<attributes>)
• Referenced attributes must be declared PRIMARY KEY or
UNIQUE

11

11

2.2. Foreign keys: Example

CREATE TABLE clazz (

clazz_id CHAR(8) NOT NULL PRIMARY KEY,
name VARCHAR(20), … );

CREATE TABLE student (

student_id CHAR(8) NOT NULL,
... ,
clazz_id CHAR(8),
CONSTRAINT student_pk PRIMARY KEY (student_id));

ALTER TABLE student ADD CONSTRAINT student_fk_class

FOREIGN KEY (clazz_id) REFERENCES clazz(clazz_id);

12

12

6
 2.2. Foreign keys: Enforcing constraint
• An insert or update to student that introduces a non-existent
clazz_id (clazz_id value is not found in clazz)
 Reject
• A deletion or update to clazz that removes a clazz_id value found
in some tuples of student?
• Default: reject the modification
• Cascade: make the same changes in student
• Set NULL: change clazz_id in student to NULL

13

13

2.2. Foreign keys: Choosing policy

ALTER TABLE student

ADD CONSTRAINT student_fk_class FOREIGN KEY
(clazz_id) REFERENCES clazz(clazz_id)
ON DELETE SET NULL
ON UPDATE CASCADE;

14

14

7
 2.3. Attribute-based checks
• Declaring
• Constraints on the value of a particular attribute
• Add CHECK(<condition>) to the declaration for the attribute or add as
relation-schema element
• The condition may use the name of the attribute, but any other relation or
attribute name must be in a subquery
• Example:
CREATE TABLE student (
student_id CHAR(8) NOT NULL PRIMARY KEY, ...,
gender CHAR(1),
clazz_id CHAR(8) CHECK (clazz_id IN (SELECT clazz_id FROM clazz)),
CONSTRAINT student_chk_gender CHECK (gender = 'F' OR gender = 'M'));

15

15

2.3. Attribute-based checks

• Timing of checks
• Only when a value for that attribute is inserted or updated

CREATE TABLE student (

student_id CHAR(8) NOT NULL PRIMARY KEY, ...,
gender CHAR(1),
clazz_id CHAR(8) CHECK (clazz_id IN (SELECT
clazz_id FROM clazz)),
CONSTRAINT student_chk_gender CHECK (gender = 'F'
OR gender = 'M') );
Not checked if a class is deleted
from clazz

16

16

8
 2.4. Tuple-based checks

• CHECK (<condition>) may be added as a relation-schema

element.
• The condition may refer to any attribute of the relation
• But other attributes or relations require a subquery
• Timing of checks: on insert or update only.
CREATE TABLE grade(
code CHAR(1) NOT NULL,
from_score DECIMAL(3,1) NOT NULL,
to_score DECIMAL(3,1) NOT NULL, ...,
CONSTRAINT grade_chk_toScore CHECK (to_score >
from_score) );

17

17

2.5. Assertions

• Declaring
• Database-schema elements, like relations or views
• Defined by:

CREATE ASSERTION <name>

CHECK (<condition>);
• Condition may refer to any relation or attribute in the database schema
• Drop an assertion:
DROP ASSERTION <assertion name>;

18

18

9
 2.5. Assertions: Example

CREATE ASSERTION teachingSubject CHECK (

(SELECT COUNT(*) FROM teaching) >=
(SELECT COUNT(*) FROM subject) );

CREATE ASSERTION numberStdInClass CHECK (

NOT EXISTS (
SELECT * FROM clazz c
WHERE number_students <>
(SELECT count(*) FROM student
WHERE clazz_id = c.clazz_id) ) );

19

19

2.5. Assertions

• Timing of Assertion Checks

• In principle, we must check every assertion after every modification to any
relation of the database
• A clever system can observe that only certain changes could cause a given
assertion to be violated
• No change to student can affect teaching Subject
• Neither can an insertion to teaching
• Very hard to implement assertions efficiently

20

20

10
 3. Triggers
3.1. Motivation
3.2. Trigger
3.3. Using trigger
3.4. Examples

21

21

3.1. Motivation

• Assertions
• powerful,
• but the DBMS often can’t tell when they need to be checked
• Attribute- and tuple-based checks
• checked at known times,
• but are not powerful
• Triggers let the user decide when to check for any condition

22

22

11
 3.1. Motivation: ECA Rules

• A trigger defines an operation that is performed when a specific event

occurs on a relation:
• inserts a new record / updates an existing record / deletes a record
• Trigger functions have access to special variables from the database
engine
• Called also ECA rules (Event-Condition-Action)
• Event: type of database modification
• Condition: Any SQL Boolean-valued expression
• Action: Any SQL statements

23

23

3.1. Motivation: Example

• Constraint: when a new student is inserted into student relation, the
number of students in his class must be increased
student(student_id, first_name,last_name, dob, gender, address, note, email, clazz_id)
clazz(clazz_id, name, lecturer_id, monitor_id, number_students)

Event
CREATE TRIGGER clazz_changes_tg
AFTER INSERT ON student
REFERENCING NEW ROW AS nnn
Condition
FOR EACH ROW
WHEN (nnn.clazz_id IS NOT NULL)
BEGIN
update clazz Action
set number_students = number_students + 1
where clazz_id = nnn.clazz_id;
END;
24

24

12
 3.2. Trigger: Definition syntax

• Creating a trigger:
CREATE [OR REPLACE] TRIGGER <trigger_name>
{BEFORE | AFTER | INSTEAD OF }
{INSERT | DELETE | UPDATE [OF <attribute_name>]}
ON <table_name>
REFERENCING {NEW | OLD} {ROW | TABLE} AS <name>
[FOR EACH ROW ]
[WHEN (<condition>) ]
BEGIN
<trigger body goes here >
END;

• Dropping a trigger:
DROP TRIGGER <trigger_name>;

25

25

3.2. Trigger: Event

• AFTER, BEFORE, INSTEAD OF:
• AFTER, BEFORE: used for tables / views
• INSTEAD OF: used only for views
• A way to execute view modifications: triggers translate them to
• appropriate modifications on the base tables
• INSERT, DELETE, UPDATE , UPDATE OF
• UPDATE OF <colums>: update on a particular column

26

26

13
 3.2. Trigger: Level

• Row-level trigger:
• Indicated by option FOR EACH ROW
• Trigger executes once for each modified tuple
• Statement-level trigger:
• Without option FOR EACH ROW or with FOR EACH
STATEMENT
• Trigger execute once for a SQL statement, regardless how
many
tuples are modified

27

27

3.2. Trigger: Referencing

• INSERT statements imply a new tuple (for row-level) or new table
(for statement-level)
• The table is the set of inserted tuples
• DELETE implies an old tuple or table
• UPDATE implies both
• Refer to these by :
REFERENCING [NEW | OLD] [TUPLE | TABLE] AS <name>
• Each DBMS has its own implementation, REFERENCING may not
used:
• Access directly to special variables from the database engine:
NEW, OLD,…

28

28

14
 3.2. Trigger: Condition

• Any boolean-valued condition

• Evaluated on the database as it would exist before or after the
triggering event, depending on whether BEFORE or AFTER is
used.
• But always before the changes take effect.
• Access the new/old tuple/table through the names in the
REFERENCING clause

29

29

3.2. Trigger: Action

• Can be more than one SQL statement:

• Surrounded by BEGIN .. END
• Language:
• Simple SQL statements
• Extention of SQL: procedural languages, depends on each DBMD
• PL/SQL (Oracle), PL/pgSQL (PostgreSQL), T-SQL(SQL Server) ,..

30

30

15
 3.3. Using triggers
• Auditing data modification (keeping history of data), providing
transparent
event logging
• Validation and business security checking if so is desired
• Eg. column formatting before and after inserts into database
• Enforcing complex integrity constraints
• Enforcing complex business rules
• Maintaining replicate tables
• Building complex views that are updatable
•…

31

31

3.3. Using triggers: Guidelines for designing triggers

• Do not define triggers that duplicate database features

• do not define triggers to reject bad data if you can do the same
checking
through constraints
• Use triggers only for centralized, global operations that must fire for the
triggering statement, regardless of which user or database application
issues
the statement
• Do not create recursive triggers
• Use triggers on DATABASE judiciously (server error, logon, logoff,…):
• they are executed for every user every time the event occurs on which
the
trigger is created.

32

32

16
 3.4. Examples: Oracle

• Add a new column in clazz relation

alter table clazz
add column number_students integer not null default 0;

• Create a trigger on student relation

CREATE TRIGGER clazz_changes_tg
AFTER UPDATE ON student
FOR EACH ROW
WHEN (:NEW.clazz_id <> :OLD.clazz_id)
BEGIN
update clazz set number_students= number_students+1
where clazz_id = :NEW.clazz_id;
update clazz set number_students = number_students-1
where clazz_id = :OLD.clazz_id;
END;

33

33

3.4. Examples: PostgreSQL

CREATE FUNCTION public.tg_fnc_change_clazz()

RETURNS trigger LANGUAGE 'plpgsql' AS $$
BEGIN
update clazz set number_students = number_students+1
where clazz_id = NEW.clazz_id;
update clazz set number_students = number_students-1
where clazz_id = OLD.clazz_id;
return NEW;
END; $$

CREATE TRIGGER tg_af_update_clazz

AFTER UPDATE OF clazz_id
ON student
FOR EACH ROW
EXECUTE PROCEDURE public.tg_fnc_change_clazz();

34

34

17
 Summary
• Constraints, Assertions, Triggers:
• How to declare
• Timing of checks
• Differences
• Only use them if you really need to, especially triggers
• Each DBMS has its own variation in implementation:
• Options
• Syntax: triggers as an example
•  Reading documentation for each DBMS installed

35

35

References
• Hector Garcia-Molina, Jeffrey D. Ullman, Jennifer Widom.
Database Systems: The Complete Book. Pearson Prentice Hall.
the 2nd edition. 2008: Chapter 7
• Nguyen Kim Anh, Nguyên lý các hệ cơ sở dữ liệu, NXB Giáo
dục. 2004: Chương 7

37

37

18