1. Sections 4– Review of Joins, Group functions, COUNT, DISTINCT, NVL
Database Programming
Sections 4– Review of Joins, Group functions, COUNT, DISTINCT, NVL
10/3/06 revised 11/13/07 revised again 11/2/10
DP_4Lesson_Fa11.ppt
DP04_Fa11.txt

2. Cross Join
SELECT last_name, department_name FROM employees CROSS JOIN departments;
SELECT last_name, department_name FROM employees, departments;
The 2nd gets the same results as the first. Without the WHERE clause you get a Cartesian product
Both above return 160 rows.
Marge Hohly

3. Natural Join
SELECT employee_id, last_name, department_id, department_name FROM employees NATURAL JOIN departments ;
SELECT e.employee_id, e.last_name, e.department_id, d.department_name FROM employees e, departments d WHERE e.department_id = d.department_id;
SELECT last_name, department_name FROM employees NATURAL JOIN departments; returns 12 rows
Marge Hohly

4. Join…On and Join…Using
SELECT employee_id, last_name, department_name FROM employees JOIN departments USING (department_id);
Joins by column names and data types that are identical in each table but USING statement limits to one column.
Join .. On
SELECT e.employee_id, e.last_name, d.department_id, d.location_id FROM employees e JOIN departments d ON (e.department_id = d.department_id);
All employees and their work locations.
Works when column name are different
Marge Hohly

5. Join .. Using
SELECT e.employee_id, e.last_name, e.department_id, d.department_name FROM employees e JOIN departments d USING (department_id );
Compare with previous and next slide
Marge Hohly

6. Join .. On
SELECT e.employee_id, e.last_name, e.department_id, d.location_id FROM employees e JOIN departments d ON (e.department_id = d.department_id );
Marge Hohly

7. Join .. On
SELECT e.employee_id, e.last_name,e.salary FROM employees e JOIN job_grades j ON (e.salary BETWEEN j.lowest_sal AND j.highest_sal);
Marge Hohly

8. Join.. On
SELECT e.employee_id, e.last_name,e.salary, j.grade_level FROM employees e,job_grades j WHERE e.salary >= j.lowest_sal AND e.salary <= j.highest_sal;
SELECT e.employee_id, e.last_name,e.salary, j.grade_level FROM employees e,job_grades j WHERE e.salary BETWEEN j.lowest_sal AND j.highest_sal;
Note these give the same results
Marge Hohly

9. Self-Joins (Join.. ON) Joining the table to itself Not common join
SELECT e.last_name emp, m.last_name mgr FROM employees e JOIN employees m ON(e.manager_id = m.employee_id);
Joining the table to itself
Not common join
Marge Hohly

10. Outer Joins (Right Outer Join, Left Outer Join)
SELECT e.employee_id, e.last_name, e.department_id, d.department_name FROM employees e RIGHT OUTER JOIN departments d ON( e.department_id = d.department_id);
These product the same results.
Right shows all data from right table (departments) including Nulls, shows departments without employees
Left shows all data from Employees. Returns all employees with or without a department
Marge Hohly

11. Outer Joins (Right Outer Join, Left Outer Join)
SELECT e.employee_id, e.last_name, e.department_id, d.department_name FROM employees e LEFT OUTER JOIN departments d ON( e.department_id = d.department_id);
These product the same results.
Marge Hohly

12. Full Outer Join
SELECT e.employee_id, e.last_name, e.department_id, d.department_name FROM employees e FULL OUTER JOIN departments d ON(e.department_id = d.department_id);
Retrieves all data from employees and departments. Shows employees without departments and departments without employees.
Marge Hohly

13. Group Functions
Group functions operate on sets of rows to give one result per group
AVG
COUNT
MAX
MIN
SUM
STDDEV
VARIANCE
Returns as many lines as there are groups
These are "aggregate.“ functions.
They require that a group of values be treated together
Group functions, unlike single-row functions, operate on sets of rows and give one result back per group.
Some of the group functions are listed on this page, including A-V-G for average, COUNT, MAX, which stands for maximum, MIN, which stands for minimum, S-T-D-D-E-V which stands for standard deviation, SUM, and VARIANCE. They can be used in many parts of a SELECT statement, except for the WHERE clause.
Marge Hohly

14. Group functions Used in the SELECT clause
Never used in the WHERE clause
Group functions ignore NULL values
You can have more than one group function in the SELECT clause on same or different columns
Some group functions work on numeric data types only
Marge Hohly

15. The MIN and MAX functions are used to return the minimum or maximum value from a set of values. These functions may be used with columns that have NUMBER, VARCHAR2, and DATE data types. When MIN and MAX are used with numeric data types, they return the smallest or the largest values. When used with DATE, MIN returns the earliest date and MAX returns the latest or most recent date. With character values, MIN returns the character closest to upper case A, MAX returns the character at the other end of the alphabetic scale, including the numbers and other special characters. When used with character values, MIN and MAX will search beginning with the first character. If the first characters are the same in all values, it will move on to the second value, and so forth. In the example to the right, the syntax SELECT MAX(salary) FROM employees, returns the highest number in the salary column of the employees table.
The SUM function returns the sum of a set of values. It is usable only on numeric data types.
The AVG function returns the average of a set of values; it is also usable only on columns with a number type.
Marge Hohly

16. The COUNT function returns the number of rows counted with not null values for the expression specified. COUNT star returns the number of rows in a table. This is used to ensure all rows are counted, in case particular columns have null values.
VARIANCE is a statistical function. This returns the variance and, again, ignores null values. It can also only be used with number data types.
S-T-D-D-E-V is the statistical function that returns the standard deviation. It ignores null values and it can only be used with numeric data types. A standard deviation is a measure of the spread or dispersion of a set of data. In the example on the right, SELECT STDDEV (salary) FROM employees measures the standard deviation across all employee salaries.
Marge Hohly

17. Marge Hohly

18. AVG function
Returns the average of a set of values – usable only on columns of number type
Syntax: SELECT AVG(salary) FROM employees WHERE job_id LIKE ‘%REP%’;
Marge Hohly

19. SUM function
Returns the sum of a set of values – usable only on columns of number type
Syntax: SELECT SUM(salary) FROM employees WHERE job_id LIKE ‘%REP%’;
SELECT SUM(salary) FROM employees WHERE department_id = 80;
Without GROUP BY clause returns 1 line.
SELECT SUM(salary) FROM employees GROUP BY department_id;
SELECT department_id, SUM(salary) FROM employees GROUP BY department_id;
SELECT type_code
FROM d_songs
WHERE SUM(duration) = 100;
Marge Hohly

20. MIN and MAX functions
Return the minimum and maximum value from a set of values
May be used with columns of NUMBERS, VARCHAR2, and DATE datatypes SELECT MIN(department_id) FROM departments; SELECT MAX(last_name) FROM employees; SELECT MIN(hire_date), MAX(hire_date) FROM employees WHERE job_id LIKE ‘%REP%’;
Marge Hohly

21. COUNT function
Returns the number of rows counted with non-null values for the expression specified SELECT COUNT(commission_pct) FROM employees; SELECT COUNT(year) FROM d_cds WHERE year < 2001; SELECT COUNT(DISTINCT year) FROM d_cds WHERE year < 2001;
Null values are ignored if the Null are not specified with another value.
Note example above counts the 4 non-null values.
SELECT COUNT(commission_pct),SUM(commission_pct),AVG(commission_pct), Count(NVL(commission_pct,0)), AVG(NVL(commission_pct,0))
FROM employees;
Marge Hohly

22. COUNT function cont’d
COUNT(*) returns the number of rows in the table SELECT COUNT(*) FROM employees;
Shows that there are 20 records in the Employees table
Marge Hohly

23. STDDEV function
A statistical function that returns the standard deviation ignoring null values for expressions of NUMBER type SELECT STDDEV(salary) FROM employees;
SELECT COUNT(salary),ROUND(STDDEV(salary),1)
FROM employees;
Marge Hohly

24. VARIANCE function
A statistical function that returns the variance ignoring null values for expressions NUMBER type SELECT VARIANCE(salary) FROM employees;
Run command.
SELECT ROUND(VARIANCE(salary),1)
FROM employees;
Marge Hohly

25. Multiple group functions
Can use more than one group function in the SELECT clause
Same or different columns
SELECT MAX(salary), MIN(salary), MIN(employee_id FROM employees WHERE department_id = 60;
Let’s review a few facts about using group functions.
Group functions are written in the SELECT clause and cannot be used in the WHERE clause. They ignore NULL values.
You can use more than one group function in the SELECT clause and even on different columns. Consider adding a WHERE condition to restrict the number of rows returned as in the example on this slide.
SUM, AVG, STDDEV, VARIANCE – numeric data type only
Marge Hohly

26. DISTINCT keyword
The DISTINCT keyword can be used with all group functions
In forces the group function to consider only non-duplicate values SELECT COUNT(DISTINCT(last_name)) FROM employees;
SELECT COUNT(DISTINCT(department_id))
FROM employees;
Results 7 rows rather than 19 rows without DISTINCT
SELECT COUNT(Year)
FROM d_cds
WHERE year < 2001;
SELECT COUNT(DISTINCT Year)
We’ve seen the use of the DISTINCT keyword in the SELECT statement to eliminate duplicate values. You can also use the DISTINCT keyword with group functions. It forces the group function to consider only non-duplicate values. The keyword DISTINCT, when used in a query selecting more than one column, will return non-duplicate combinations of the columns. Place the DISTINCT keyword inside the GROUP function. Take time to execute several queries using the DISTINCT keyword in Application Express. Note that if DISTINCT is placed outside of the COUNT function, it will have no effect on the result of the query.
Marge Hohly

27. NVL function This is used to replace a NULL with a given value
The value must be of the same datatype as the colunm
NO! SELECT commission_pct, NVL(commission_pct, ‘not eligible’) FROM employees; YES! SELECT commission_pct, NVL(commission_pct, 0) FROM employees;
Unlike single-row functions, group functions ignore null values. Remember, if a null value is encountered with a single-row function, it returns null as an answer.
The null value function, which we encountered before, can be used inside of group functions. Look at the SELECT clause below. The null value is used to change the values of customer_orders from null values to zero. Take time to use the null value function in Application Express and be sure to note the differences when executing the same query with and without the null value function
SELECT AVG(NVL(customer_order, 0))
FROM
Marge Hohly

28. Using NVL with group functions
The NVL function is nested inside the group function When you want to include rows with null values, use NVL function to add a value to the null rows. SELECT AVG(commission_pct), AVG(NVL(commission_pct,0)) FROM employees; Which column will have the lowest value?
AVG(commission_pct) AVG(NVL(commission_pct,0))
(this one averaged all values and where there was no value a 0 was put in.)
Marge Hohly