Edmund J. Sutcliffe
Thoughtful Solutions, Creatively Implemented and Communicated
Edmund's SQL Quick Reference -
1
2
Click here
for a printable version of this page.
Basics of SELECT
This
section relates to the standard SQL language found in the
majority of relational databases. It is recognised that some
of the names may differ by manufacturer. It is also recognised
that some database systems my also require you to authenticate
(log in).
SIMPLE
SELECT
SELECT
4 + 2
Select
on its own provides a print statement. If you run this on
your system you will get 6.
SELECT
SYNTAX
SELECT
column1,column...
FROM tablename
[WHERE condition
HAVING condition
GROUP BY columns
ORDER BY columns
COMPUTE BY columns]
[]
optional statements.
column is any valid column name for the data in your database
tablename.
The tablename is any valid table within your active database.
The condition is a true or false (boolean) statement, used
to compare values.
SELECTING
ALL
SELECT
* FROM tablename
This
will select all columns and records from the tablename
specified.
CONDITIONAL
SELECT
SELECT
* FROM stock_cars
WHERE
price >= 5000
Conditions
allow us to specify a subset of the list of all data stored
within a table. The condition is supplied after the where
clause and uses logical operators between the column name
and the value you are looking for.
LIST
OF OPERATORS
| = |
Equal |
| <>
or != |
Not
Equal |
| < |
Less
Than |
| > |
Greater
Than |
| <= |
Less
Than or Equal To |
| >= |
Greater
Than or Equal To |
Many
conditions can be placed after the where clause by
using AND and OR logical conditioning.
Example: WHERE price > 5000 AND colour='Red'
IN
BETWEEN....
SELECT
* FROM tablename
WHERE manufacturer IN ('Audi',
'Mercedes', 'Jaguar')
SELECT
* FROM
tablename
WHERE
age BETWEEN
18
AND
30
BETWEEN
and IN are useful for ranges of values.
LIKE
This...
SELECT
firstname,surname
FROM employees
WHERE position LIKE '%[Mm]anager%'
The
use of the like command allows for wildcard substitution,
where % is any number of any characters, [list]
is a single character within the list, and _ represents
any single character.
JOINS
Relational
databases store data in tables, and are broken down into normal
form. To make a relationship between the tables a join clause
is required. This is acheived through a WHERE clause
which joins the common columns of 2 or more tables together.
SELECT
a.firstname,
b.title FROM authors a, books b
WHERE a.au_id=b.au_id
Note
that joins require the tables to be joined to follow the FROM
clause and for shortcut we have given an alias which follows
the table name, but precedes the column names. Our join happens
at the WHERE clause using author id from both tables
as being the common field.
A JOIN is normally performed across keys of tables, e.g. a
Primary key being associated with a Foreign key. A Primary
key is a column(s) in a table where all entries must be unique.
A Foreign key is a column in a table that relates to a Primary
key in another table.
|