SQL Tutorial

SQL the Language

The aim of this tutorial is to give a hands-on knowledge of this powerfull but still so simple
language to the beginers as well as to the experienced programmers who likes to revise their
understanding periodically. The totorial includes lots of example which are used by developers
and managers alike for their day to day functions in developing and supporting commercial
databases.

A few words on Oracle

Oracle stores data in an object called Table. There are different types of objects in oracle
such as

Table
Index
Contraint
Procedure
Function
Package
Trigger
Sequence

All these objects are logically stored in Oracle logical structures such as Tables, Segments, Extents
and Blobks.

Physically all these logical data are stored in o/s files called Oracle Data files. These files are
stored in o/s level in some directory which varies in different implementations.

Basic Requirement

Create sample Tables and Data used in this guide using the link below:

<< Tutorial Sample table and data >>

Data Types

CHAR – fixed length character data
VARCHAR2 – variable length character data
NUMBER (p,s) – Number with with precision p and scale s.
DATE – a valid date
TIMESTAMP – date and time inclusing hours, minutes, seconds and mil. seconds

There are other data types which are are widely used such as

TIMESTAMP WITH TIME ZONE
TIMESTAMP WITH LOCAL TIME ZONE
RAW
LONG
LONG RAW
ROWID
BLOB
CBLOB
BFILE
and
more…

Relational Operators

SET Operators

Oracle provides facilities for mathmatical SET operators in it’s SQL constructs.

UPDATE

INSERT

SELECT Statements- Basic

As the syntax suggests there are three main clause in a SELECT statement which are

SELECT – What are the information that you need to select

FROM – From where do you want to select these information

WHERE – What are the condtion that you want to use to filter query results

SELECT is the most frequently used statements in SQL. Whatever you want to do with your data you will need
to use this statement to look for them and check the values of the columns in your record.

Executing the above query will return all the records from Strudents table in your database.

Selecting selected columns from a table

SELECT student_id, f_name, l_name FROM students;

The above query will select only those students whose major_subject_id = 30. The query will returns
records for student_ids 106, 107 and 113.

Example 2: Using < operator

Example 3: Using > operator

Example 17: Using LIKE operator.

Example 18: Using OR operator.

SET Operators by example

2. Example of UNION ALL

SELECT * FROM students WHERE major_subject_id IN (30, 40)
UNION ALL
SELECT * FROM students WHERE major_subject_id = 40;

Returns all recors combined from first and second query. From the above query you will notice
that you will get the records with major_subject_id twice.

3. Example of INTERSECT

SELECT * FROM students WHERE major_subject_id IN (30, 40)
INTERSECT
SELECT * FROM students WHERE major_subject_id = 40;

Returns all records which must be returned from both queries. In the above example you will get
records for major_subject_id = 40 as the satisfies both first and second query.

4. Example of MINUS

SELECT * FROM students WHERE major_subject_id IN (30, 40)
MINUS
SELECT * FROM students WHERE major_subject_id = 40;

Reutrns all those records fro the first query which are not there in the second query. AS we can see above
major_subject_id = 40 is returned by both queries, so they will nto be selected. After removing
major_subject_id = 40 from the first query we still have records for major_subject_id = 30;

So the result from the above query will be all records which satisfies major_subject_id = 30;

NULL Values

For example if you want to select all records from Students table who do not have any major assigned to the
(meaning they have NULL in major_subject_id column) your query should look like this:

SELECT * FROM students WHERE major_subject_id IS NULL;

This will return only one record, that of David Armstrong who do not have any major subject ID.

Now if you want to select all those reocrds who definitely have a major subject ID when you can run the query

SELECT * FROM students WHERE major_subject_id IS NOT NULL;

The above record returns all records from Students table but David Armstrong. Because his major subject ID is
NULL and the above query clearly said to return only those records where major subject id IS NOT NULL.

SELECT Statements – Advanced

Data can be ordered by Ascending Order or Descending Order. The default is Ascending Order. In the above
query I did not specify any order. So the result came in Ascending order.

To Specify the order I can use the following query:

SELECT * FROM students
WHERE major_subject_id = 10 OR major_subject_id= 40
ORDER BY dob asc;

ASC stands for Ascending.

To order data in Descending order I can use the query:

SELECT * FROM students
WHERE major_subject_id = 10 OR major_subject_id= 40
ORDER BY dob desc;

I can use Ascending Order and Descending Order together in the same query on different columns as shown
below

SELECT * FROM students
WHERE major_subject_id = 10 OR major_subject_id= 40
ORDER BY f_name asc, l_name desc;

SELECT Statement with GROUP BY Clause:

I can use the GROUP BY clause with a select statement to group all the returned data together to
display divided into groups. For example if I want to count the number of students and group them by their
major subjects I can use the query:

SELECT major_subject_id, count(*)
FROM students
WHERE major_subject_id IS NOT NULL
GROUP BY major_subject_id;

SELECT Statement with GROUP BY and HAVING Clause:

Another clause that is usefull is the HAVING clause. In the example below the query reutrns only thos
subject_id’s where the number of studfent enrolled us greater than 4

It is possible to use GROUP BY and ORDER BY clauses in the same query as show below:

SELECT major_subject_id, COUNT (*)
FROM students
WHERE major_subject_id IS NOT NULL
GROUP BY major_subject_id
ORDER BY major_subject_id;

The above query will display the each major_subject_id’s and how many students are enrolled in those
subjects.

SELECT Statement with GROUP BY and ORDER BY Clause:

Example of GROUP BY, HAVING and ORDER BY Clauses all in the same query:

SELECT major_subject_id, COUNT (*)
FROM students
WHERE major_subject_id IS NOT NULL
GROUP BY major_subject_id
HAVING COUNT (*) >= 4
ORDER BY major_subject_id;

SELECT Statement USING subquery:

A SELECT Statement with a subquery is a query which uses the result of another query (the sub query)
as values in its WHERE clause. In the example below the subquery first finds the subject_id for the Subject
Biology, in this case the value is 10. The outer SELECT statement then uses the value from the subquery
to filter the data from the students table in its WHERE clause.

SELECT * FROM students
WHERE major_subject_id
IN ( SELECT subject_id
FROM majors
WHERE subject_name=’Biology’
);

It is possible to use more than one column in a subquery as shown below:

SELECT * FROM students
WHERE major_subject_id in (
SELECT subject_id
FROM majors
WHERE subject_name = (SELECT subject_name
FROM majors
WHERE subject_teacher = ‘Dr. Elms Somerfield’))

Co-Related Sub Queries

A co-related subquery is a query where the outer query is executed first. Then for each record returned by the
outer query, the inner query is executed. In the example below the outer query first gets student record from
students table and then it executes the inner query once for each record returned by the outer query. In the
sub query the value of major_subject_id of the outer record is checked against the value of subject_id of
majors table. If the values match then only that record will be displayed to us. That is why the record below
will not return the student record which does not have a subject id assigned to it.

SELECT * FROM students s
WHERE EXISTS
(SELECT 1
FROM majors m
WHERE s.major_subject_id = m.subject_id);

In a SQL query’s FROM clause you can use the return values of another query as
if they are a table. This technique helps to simplify SQL queries.

An example is:

SELECT *
FROM students a, (SELECT * FROM majors) b
WHERE a.major_subject_id = b.subject_id
AND a.major_subject_id=10;

JOINS

Inner Join or Equi Joins

Example 1: Joining two tables to display results from both tables.

SELECT subject_name, f_name, l_name
FROM students s, majors m
WHERE s.major_subject_id = m.subject_id
and m.subject_name=’History’;

In the above example students and majors tables are joined together. The joining condition is
specified in where clause as:

“WHERE s.major_subject_id = m.subject_id
and m.subject_name=’History’”

It means that display all the students whose major subject is History. We can get the same information
except the subject name from students atble if we query it with the major_subject_id. But if you need
the subject name in your displayed result then you must join the two tables.

Example 2: The query in example 1 can be re-written using the new ANSI JOIN as

SELECT subject_name, f_name, l_name
FROM students s
INNER JOIN majors m on
s.major_subject_id = m.subject_id
and m.subject_name=’History’;

Which will give the same result.

Outer Join

This Join is used to return data where the primary data source is there and there are no secondary data
source. In the Students and Majors example you will have noticed that the subject Social Science was not taken
by any student. So when you join Students and Majors tables with a simple equi join or inner join you will
get only those major subjects and their students which were taken by students with the condition

students.major_subject_id = major.subject_id.

So to get all the Majors and the students who has taken those majors you will need to use a SPEACIAL join
called OUTER JOIN. In Oracle OUTER JOIN is denoted by “+”.

The query below will return all the Majors and students

SELECT subject_name, f_name, l_name
FROM students s, majors m
WHERE s.major_subject_id(+) = m.subject_id;

NOTE: It is important to note that you will need to put the “+” sign to the column where there are no values.
In the example above I have said s.major_subject_id(+) because I know that there are no record which matches
the subject_id for Social Science in Students table.

In new ANSI JOIN format it will look like this

SELECT subject_name, f_name, l_name
FROM students s
RIGHT JOIN majors m on
s.major_subject_id = m.subject_id;

The “RIGHT JOIN” means that return all records from the table on the right hand side of the join condition
(in this case Majors denoted as m) even if there are no corresponding values in the LEFT table
(in this case Students denoted as s).

Similarly there is a LEFT JOIN which means retrun everything from the right hand side of the condition
even if no corresponding value exists in the left hand side of the conditon. For example:

SELECT m.subject_name, f_name, l_name
FROM students s
LEFT JOIN majors m on
s.major_subject_id = m.subject_id;

The query above will David Armstrong’s record even though it does not have a major subject ID.

USING Oracle FUNCTIONS in SQL

In our Students table we know that David Armstrong has a NULL major subject ID. If run the query

SELECT student_id, f_name, l_name, major_subject_id FROM students;

Then it will return all records from students table and will display a NULL for Davis’ major subject id.
But we can replace the NULL value in the result and instead show a more value such as ’999′. To do that
we will use NVL function as show below:

SELECT student_id, f_name, l_name, NVL(major_subject_id,’999′) FROM students;

In the above query the NVL function will check each returned value of major_subject_id. If if finds the
major_subject_id is NULL then it will replace it with 999 or whatever value you would like to assign.

In the example below I am assigning major_subject_id = 20 to any record which has a NULL major_subject_id.

DECODE

This function is used to replace a value from the returned result to make it for usefull for your purpose. It
is like an if-then-else construct.

Eor example If I run the query

SELECT f_name, l_name, major_subject_id
FROM students;

I’ll get all records from students table which will display major_subject_id as 10, 20, 30 etc as they are
in the table.

But If I want to replace the value of major_subject_id 10 by “Biological Sciences” and the rest of the values
to be displayed as they are I can issue the command:

SELECT f_name, l_name,  DECODE(major_subject_id,10,’Biological Sciences’, major_subject_id)
FROM students;

It simply means that

if major_subject_id = 10 then
display Biological Sciences
else
display whatever value is returned.
end if;

The above query will display major_subject_id as Biological Sciences for all records where major_subject_id=10.
The “major_subject_id” after “‘Biological Sciences’,” in the above statement means display whatever is there
in the table (statement in the ELSE condition).

It is possible to large number of values in decode statement. For example

SELECT f_name, l_name, DECODE(major_subject_id,10,’Biological Sciences’, 20, ‘Chemical Sciences’, major_subject_id)
FROM students.

It simply means that

if major_subject_id = 10 then
display Biological Sciences
elsif major_subject_id = 20 then
display Chemical Sciences
else
display whatever value is returned.
end if;

CASE

Case is another form of if-then-else construct.

SELECT CASE major_subject_id
WHEN 10 THEN
‘Biological Sciences’
WHEN 20 THEN
‘Chemical Sciences’
ELSE
‘Other Subject’
end Subject
FROM students;

The above query means that

if major_subject_id = 10 then
display Biological Sciences
elsif major_subject_id = 20 then
display Chemical Sciences
else
display Other Subject
end if;

TO_DATE

This is a very powerfull robust Oracle function that you will find yourself using in your day to day work.
So it is very important you understand this function as much as possible.

In the basic form it’s syntax looks like this:

TO_DATE(DateInString, Format).

Example of TO_DATE Function

SELECT TO_DATE(’01-12-2009′,’DD-MM-YYYY’) from dual;

SELECT TO_DATE(’12-01-2009′,’MM-DD-YYYY’) from dual;

SELECT TO_DATE(’01-DEC-2009 10:30.11′,’DD-MON-YYYY HH:MI:SS’) from dual;

SELECT TO_DATE(’01-DEC-2009 14:30.11′,’DD-MON-YYYY HH24:MI:SS’) from dual;

In the above explample notice that I am using a table called “dual”. This is a special table provided
in Oracle to satisfy the need for a table in a SELECT statement. By now is is clear that any SELECT query
we write uses a table from where to select the data. But there are some situation where you don’t have
the table. For example you want to display the Oracle system date. There is no such table which stored
this information. But you can use SYSDATE to do that in your select statement. But without a table how can
you write your SELECT statement. This is where the special table called “dual” comes in.

If you say

SELECT sysdate FROM dual;

It will display the Oracle system date and you did not used any actal table in your SELECT statement. In
the examples below you will see many examples of the usefullness of this table.

TO_CHAR

This is also a very powerfull robust Oracle function that you will find yourself using in your day to day work.
The function is generally used for DatE formatting and Number formatting.

Date Formating <TO_CHAR(date)>

Basic Syntax:

TO_CHAR(date1, format_mask).

The TO_CHAR (date) function takes a date as input parameter and convert it into CHARACTER and display the
result as specified in the market mask.

Acceptable format masks:

D
DD
DDD
DY
DAY

MM
MON
MONTH

Y
YY
YYY
YYYY
RR
RRRR
HH
HH24

MI
SS
SSSSS
FF
AM
PM

Example using these format masks.

SELECT TO_CHAR(sysdate, ‘ddd’ ) from dual;

SELECT TO_CHAR(sysdate, ‘Dy’ ) from dual;

SELECT TO_CHAR(sysdate, ‘Day’ ) from dual;

SELECT TO_CHAR(sysdate, ‘mm’ ) from dual;

SELECT TO_CHAR(sysdate, ‘YY’ ) from dual;

SELECT TO_CHAR(sysdate, ‘RR’ ) from dual;

SELECT TO_CHAR(sysdate, ‘YYYY’ ) from dual;

SELECT TO_CHAR(sysdate, ‘RRRR’ ) from dual;

SELECT TO_CHAR(sysdate,’DD-MON-YYYY’) from dual;

SELECT TO_CHAR(sysdate,’DD-Mon-YYYY’) from dual;

SELECT TO_CHAR(sysdate,’DD-MONTH-YYYY’) from dual;

SELECT TO_CHAR(sysdate,’DD-MON-YYYY hh24′) from dual;

SELECT TO_CHAR(sysdate,’DD-MON-YYYY hh24:mi’) from dual;

SELECT TO_CHAR(sysdate,’DD-MON-YYYY hh24:mi:ss’) from dual;

SELECT TO_CHAR(sysdate,’MON-DD-YYYY’) from dual;

SELECT TO_CHAR(sysdate,’MM-DD-YYYY’) from dual;

SELECT TO_CHAR(sysdate,’MONTH-DD-YYYY’) from dual;

SELECT TO_CHAR(sysdate,’YYYY-Mon-DD’) from dual;

SELECT TO_CHAR(sysdate,’YYYY-MON-DD’) from dual;

SELECT TO_CHAR(sysdate,’YYYY-MM-DD’) from dual;

TO_CHAR(number1, format_mask).

The TO_CHAR (number) function takes a number as input parameter and convert it into CHARACTER and display the
result as specified in the market mask.

Some example are here:

select to_char(’1234560.789′,’$99,99,9999.9′) from dual;

select to_char(’1234560.789′,’$99,99,9999.99′) from dual;

select to_char(’1234560.789′,’$99,99,9999.990′) from dual;

select to_char(’1234560.789′,’$99,99,9999.9900′) from dual;

select to_char(’1234560.789′,’$99,99,9999.99000′) from dual;

UPPER

Basic Syntax:

UPPER (char);

This function accept a character string as input parameter and convert the string letters
into upper case charcters. In will ignore numbers if there are any in the input string.

Example;

select upper(‘test’) from dual;

LOWER

Basic Syntax:

LOWER (char);

This function accept a character string as input parameter and convert the string letters
into lower case charcters. In will ignore numbers if there are any in the input string.

Example;

select upper(‘Test’) from dual;

SUBSTR

Basic Syntax:
SUBSTR (strin1, pos1, pos2);

pos1 – is the start of the string
pos2 – is the number of characters starting from pos1. If no pos2 is specified then the function
will return the all the remaining characters starting from pos1.

Some example:

SELECT SUBSTR(‘ABCDEFGH’, 1, 5) from dual; will return ABCDE.

SELECT SUBSTR(‘ABCDEFGH’, 3, 15) from dual; will return CDEFGH.

SELECT SUBSTR(‘ABCDEFGH’, 5) from dual; will return EFGH.

INSTR

Basic Syntax:

Example 1:

SELECT INSTR(‘CORRUPTION CONTROL COORDINATOR’,'CO’, 1, 1) from dual;

Example 1:

SELECT INSTR(‘CORRUPTION CONTROL COORDINATOR’,'CO’, 1, 2) from dual; will return 12.

It looks for the string2 value “CO” in string1 – starting at position 1. It is lokking for the 2nd occurance
which is at position 12 (CONTROL). So the return value is 12.

Example 2:
SELECT INSTR(‘CORRUPTION CONTROL COORDINATOR’,'CO’, 8, 2) from dual;will return 20.

It looks for the string2 value “CO” in string1 – starting at position 8. It is lokking for the 2nd occurance
which is at position 20 (COORDINATOR). So the return value is 20.