Background

SQL, Structured Query Language, is a programming language designed to manage data stored in
relational databases. SQL operates through simple, declarative statements. This keeps data
accurate and secure, and it helps maintain the integrity of databases, regardless of size.

Here’s an appendix of commonly used commands.

Commands
ALTER TABLE
ALTER TABLE table_name
ADD column_name datatype;

ALTER TABLE lets you add columns to a table in a database.

AND
SELECT column_name(s)
FROM table_name
WHERE column_1 = value_1
  AND column_2 = value_2;

AND is an operator that combines two conditions. Both conditions must be true for the row to be
included in the result set.

AS
SELECT column_name AS 'Alias'
FROM table_name;

AS is a keyword in SQL that allows you to rename a column or table using an alias.

AVG()
SELECT AVG(column_name)
FROM table_name;

AVG() is an aggregate function that returns the average value for a numeric column.

BETWEEN
SELECT column_name(s)
FROM table_name
WHERE column_name BETWEEN value_1 AND value_2;
The BETWEEN operator is used to filter the result set within a certain range. The values can be
numbers, text or dates.

CASE
SELECT column_name,
  CASE
    WHEN condition THEN 'Result_1'
    WHEN condition THEN 'Result_2'
    ELSE 'Result_3'
  END
FROM table_name;

CASEstatements are used to create different outputs (usually in the SELECT statement). It is
SQL’s way of handling if-then logic.

COUNT()
SELECT COUNT(column_name)
FROM table_name;

COUNT() is a function that takes the name of a column as an argument and counts the number of
rows where the column is not NULL.

CREATE TABLE
CREATE TABLE table_name (
  column_1 datatype,
  column_2 datatype,
  column_3 datatype
);

CREATE TABLE creates a new table in the database. It allows you to specify the name of the table
and the name of each column in the table.

DELETE
DELETE FROM table_name
WHERE some_column = some_value;

DELETE statements are used to remove rows from a table.

GROUP BY
SELECT column_name, COUNT(*)
FROM table_name
GROUP BY column_name;
GROUP BY is a clause in SQL that is only used with aggregate functions. It is used in
collaboration with the SELECT statement to arrange identical data into groups.

HAVING
SELECT column_name, COUNT(*)
FROM table_name
GROUP BY column_name
HAVING COUNT(*) > value;

HAVING was added to SQL because the WHERE keyword could not be used with aggregate
functions.

INNER JOIN
SELECT column_name(s)
FROM table_1
JOIN table_2
  ON table_1.column_name = table_2.column_name;

An inner join will combine rows from different tables if the join condition is true.

INSERT
INSERT INTO table_name (column_1, column_2, column_3)
VALUES (value_1, 'value_2', value_3);

INSERT statements are used to add a new row to a table.

IS NULL / IS NOT NULL

SELECT column_name(s)
FROM table_name
WHERE column_name IS NULL;

IS NULL and IS NOT NULL are operators used with the WHERE clause to test for empty values.

LIKE
SELECT column_name(s)
FROM table_name
WHERE column_name LIKE pattern;

LIKE is a special operator used with the WHERE clause to search for a specific pattern in a column.

LIMIT
SELECT column_name(s)
FROM table_name
LIMIT number;

LIMIT is a clause that lets you specify the maximum number of rows the result set will have.

MAX()
SELECT MAX(column_name)
FROM table_name;

MAX() is a function that takes the name of a column as an argument and returns the largest value
in that column.

MIN()
SELECT MIN(column_name)
FROM table_name;

MIN() is a function that takes the name of a column as an argument and returns the smallest
value in that column.

OR
SELECT column_name
FROM table_name
WHERE column_name = value_1
   OR column_name = value_2;

OR is an operator that filters the result set to only include rows where either condition is true.

ORDER BY
SELECT column_name
FROM table_name
ORDER BY column_name ASC | DESC;

ORDER BY is a clause that indicates you want to sort the result set by a particular column either
alphabetically or numerically.

OUTER JOIN
SELECT column_name(s)
FROM table_1
LEFT JOIN table_2
  ON table_1.column_name = table_2.column_name;
An outer join will combine rows from different tables even if the join condition is not met. Every
row in the left table is returned in the result set, and if the join condition is not met, then NULL
values are used to fill in the columns from the right table.

ROUND()
SELECT ROUND(column_name, integer)
FROM table_name;

ROUND() is a function that takes a column name and an integer as arguments. It rounds the values
in the column to the number of decimal places specified by the integer.

SELECT
SELECT column_name
FROM table_name;

SELECT statements are used to fetch data from a database. Every query will begin with SELECT.

SELECT DISTINCT
SELECT DISTINCT column_name
FROM table_name;

SELECT DISTINCT specifies that the statement is going to be a query that returns unique values
in the specified column(s).

SUM
SELECT SUM(column_name)
FROM table_name;

SUM() is a function that takes the name of a column as an argument and returns the sum of all the
values in that column.

UPDATE
UPDATE table_name
SET some_column = some_value
WHERE some_column = some_value;

UPDATE statements allow you to edit rows in a table.

WHERE
SELECT column_name(s)
FROM table_name
WHERE column_name operator value;
WHERE is a clause that indicates you want to filter the result set to include only rows where the
following condition is true.

WITH

SQL is incredibly powerful, and like every well-made development tool, it has a few commands
which it’s vital for a good developer to know. Here is a list of SQL queries that are really
important for coding & optimization. Each of the queries in our SQL tutorial is consequential to
almost every system that interacts with an SQL database.

1. Retrieving Tables 31. Database Management

2. Selecting Columns from a
32. Adding Tables to Our New DB
Table
3. Outputting Data –
33. Modifying and Deleting Tables
Constraint
4. Outputting Data – ‘Order
34. Successful Indexing
By’
5. Outputting Data – ‘Group
35. Improving Performance
By’
6. Data Manipulation –
36. Conditional Subquery Results
COUNT
7. Data Manipulation Using
37. Copying Selections
SUM
8. Data Manipulation Using
38. Catching NULL Results
AVG
9. SQL Query for Listing all
39. HAVING can be Relieving!
Views
10. Creating a View 40. Tie things up with Strings!
11. Retrieving a View 41. Use COALESCE
12. Updating a View 42. Use Convert
13. Dropping a View 43. DENSE_RANK()Analytical
14. Display User Tables 44. Query_partition_clause
15. Display Primary Keys 45. Last five records from the table
16. Displaying Unique Keys 46. LAG
17. Displaying Foreign Keys 47. LEAD
18. Displaying Triggers 48. PERCENT_RANK
19. Displaying Internal Tables 49. MIN
20. Displaying a List of
50. MAX
Procedures
21. Swapping the Values 51. Top- N queries
22. Returning a Column of
52. CORR Analytic Query
Values
23. SELECT TOP Clause 53. NTILE Analytic Query
24. Searching for SQL Tables 54. VARIANCE, VAR_POP, and VAR_SAMP
 Query
25. Between Monday and 55. STDDEV, STDDEV_POP and
Tuesday STDDEV_SAMP Queries
26. Find Intersection of 2
56. Pattern Matching
Tables
27. UNION 57. FIRST_VALUE
28. Friendly Column Labels 58. LAST_VALUE
29. Always and Everywhere! 59. Prediction
30. Developer-Friendly SQL 60. CLUSTER_SET

1.   SQL Query for Retrieving Tables

This query can be run to retrieve the list of tables present in a database where the database is
“My_Schema”.

With the SELECT command, users can define the columns that they want to get in the query
output. This command is also useful to get which column users want to see as the output table.
The SELECT statement is applied to pick data from a table. The data retrieved is put in a result
table, named the result-set. The output data is saved in a result table. This output table is also
termed as the result-set.

1SELECT * FROM My_Schema.Tables;

2.   Query for Selecting Columns from a Table

This is perhaps the most widely used SQL queries examples. In the example below, we are
extracting the “Student_ID” column or attribute from the table “STUDENT”. The select
statement is used to select data from the database.

1SELECT Student_ID FROM STUDENT;

If you want to display all the attributes from a particular table, this is the right query to use:

1SELECT * FROM STUDENT;

3.   Query for Outputting Data Using a Constraint

This SQL query retrieves the specified attributes from the table on the constraint Employee ID
=0000

1SELECT EMP_ID, NAME FROM EMPLOYEE_TBL WHERE EMP_ID = '0000';

4. Query for Outputting Sorted Data Using ‘Order By’

This query orders the results with respect to the attribute which is referenced to using “Order By”
– so for example, if that attribute is an integer data type, then the result would either be sorted in
ascending or descending order; likewise if the data type is a String then the result would be
ordered in alphabetical order. The order by clause is used to sort the data from the table. The
order by clause should always be used in the last of the SQL query.

1SELECT EMP_ID, LAST_NAME FROM EMPLOYEE

2WHERE CITY = 'Seattle' ORDER BY EMP_ID;

The ordering of the result can also be set manually, using “asc ” for ascending and “desc” for
descending.

Ascending (ASC) is the default condition for the ORDER BY clause. In other words, if users
don’t specify ASC or DESC after the column name, then the result will be ordered in ascending
order only.

1SELECT EMP_ID, LAST_NAME FROM EMPLOYEE_TBL

2WHERE CITY = 'INDIANAPOLIS' ORDER BY EMP_ID asc;

5.   SQL Query for Outputting Sorted Data Using ‘Group

By’
The ‘Group By’ property groups the resulting data according to the specified attribute.

The SQL query below will select Name, Age columns from Patients table, then will filter them
by Age value to include records where Age is more than 40 and then will group records with
similar Age value and then finally will output them sorted by Name. The basic rule is that the
group by clause should always follow a where clause in a Select statement and must precede the
Order by clause.

1SELECT Name, Age FROM Patients WHERE Age > 40

2GROUP BY Name, Age ORDER BY Name;

Another sample of use of Group By: this expression will select records with a price lesser than
70 from Orders table, will group records with a similar price, will sort the output by price and
will also add the column COUNT(price) that will display how many records with similar price
were found:

1SELECT COUNT(price), price FROM orders

2WHERE price < 70 GROUP BY price ORDER BY price

Note: you should use the very same set of columns for both SELECT and GROUP BY
commands, otherwise you will get an error. Many thanks to Sachidannad for pointing out!

SQL Queries for Data Manipulation Using Math Functions

There are a lot of built-in math functions like COUNT and AVG which provide basic
functionalities of counting the number of results and averaging them respectively.

6. Data Manipulation Using COUNT

This query displays the total number of customers by counting each customer ID. In addition, it
groups the results according to the country of each customer. In count, if users define
DISTINCT, then they cal also define the query_partition_clause. This clause is a part of the
analytic clause, and  the other clauses such as order_by_clause and windowing_clause are not
permitted.

Syntax: SELECT COUNT(colname) FROM table name;

1SELECT COUNT(CustomerID), Country FROM Customers GROUP BY Country;

7. Data Manipulation Using SUM

SUM calculates the total of the attribute that is given to it as an argument. SUM is an aggregate
function and it calculates the sum of all the distinct values. and sum of all the duplicate values.

1SELECT SUM(Salary)FROM Employee WHERE Emp_Age < 30;

8. Data Manipulation Using AVG

Simple – an average of a given attribute. Average is also an aggregate function in SQL. The
AVG() function computes the average of non-NULL values in a column. It ignores the null
values.

1SELECT AVG(Price)FROM Products;

9.   SQL Query for Listing all Views

This SQL query lists all the views available in the schema.

1SELECT * FROM My_Schema.views;

10. Query for Creating a View

A view is a tailored table that is formed as a result of a query. It has tables and rows just like any
other table. It’s usually a good idea to run queries in SQL as independent views because this
allows them to be retrieved later to view the query results, rather than computing the same
command every time for a particular set of results.

1CREATE VIEW Failing_Students AS

2SELECT S_NAME, Student_ID
3FROM STUDENT
4WHERE GPA > 40;

11. Query for Retrieving a View

The standard syntax of selecting attributes from a table is applicable to views as well.

1SELECT * FROM Failing_Students;

12. Query for Updating a View

This query updates the view named ‘Product List’ – and if this view doesn’t exist, then the
Product List view gets created as specified in this query. The view is also called as a virtual
table. In other words, view is just a mirrored copy of a table whose data is the result of a stored
query.

A view is a legitimate copy of a different table or sequence of tables. A view obtains its
information or data from the tables from previously created tables known as base tables. Base
tables are real tables. All procedures implemented on a view really modify the base table. Users
can use views just like the real or base tables. In view, users can apply various DDL, DML
commands such as update, insert into, and delete.

1CREATE OR REPLACE VIEW [ Product List] AS

2SELECT ProductID, ProductName, Category
3FROM Products
4WHERE Discontinued = No;

13. Query for Dropping a View

This query will drop or delete a view named ‘V1’. The important thing to remember here is that
the DROP VIEW is disallowed if there are any views dependent on the view you are about to
drop.

1DROP VIEW V1;

14. Query to Display User Tables

A user-defined table is a representation of defined information in a table, and they can be used
as arguments for procedures or user-defined functions. Because they’re so useful, it’s useful to
keep track of them using the following query. User tables explains the relational tables of the
current user.

1SELECT * FROM Sys.objects WHERE Type='u'

15. Query to Display Primary Keys
A primary key uniquely identifies all values within a table. A primary key imposes a NOT
NULL restriction and a unique constraint in one declaration. In other words, it prevents various
rows from having similar values or sequences of columns. It doesn’t allow null values. The
primar key can be defined on a single column or the combination of two columns in a table. It is
responsible for all the relationships between the tables.

The following SQL query lists all the fields in a table’s primary key.

1SELECT * from Sys.Objects WHERE Type='PK'

16. Query for Displaying Unique Keys

A Unique Key allows a column to ensure that all of its values are different. Unique key also
recognizes a different tuple uniquely in relation to or table. A table can have more than one
unique key. Unique key constraints can take only one NULL value for the column.

1SELECT * FROM Sys.Objects WHERE Type='uq'

17. Displaying Foreign Keys

Foreign keys link one table to another – they are attributes in one table which refer to the
primary key of another table.

1SELECT * FROM Sys.Objects WHERE Type='f'

Primary, Unique, and Foreign are part of the constraints in SQL. Constraints are essential to the
scalability, compliance, and sincerity of the data. Constraints implement particular rules,
assuring the data adheres to the conditions outlined. For example, these are the laws imposed on
the columns of the database tables. These are applied to restrict the kind of data in the table. This
assures the efficiency and authenticity of the database.

18. Displaying Triggers
A Trigger is sort of an ‘event listener’ – i.e, it’s a pre-specified set of instructions that execute
when a certain event occurs. The list of defined triggers can be viewed using the following
query.

1SELECT * FROM Sys.Objects WHERE Type='tr'

19. Displaying Internal Tables

Internal tables are formed as a by-product of a user-action and are usually not accessible. The
data in internal tables cannot be manipulated; however, the metadata of the internal tables can be
viewed using the following query.

1SELECT * FROM Sys.Objects WHERE Type='it'

20. Displaying a List of Procedures

A stored procedure is a group of advanced SQL queries that logically form a single unit and
perform a particular task. Thus, using the following query you can keep track of them:

1SELECT * FROM Sys.Objects WHERE Type='p'

Improve your skills with SQL trainer (for beginner and advanced developers)

.. and TWENTY More Advanced SQL Queries for our

Users!
21. Swapping the Values of Two Columns in a table
In this and subsequent examples, we will use a common company database including several
tables that are easily visualized. Our practice DB will include a Customer table and an Order
table. The Customers table will contain some obvious columns including ID, Name, Address,
zip, and email, for example, where we assume for now that the primary key field for indexing is
the Customer_ID field.

With this in mind, we can easily imagine an Orders table which likewise contains the indexed
customer ID field, along with details of each order placed by the customer. This table will
include the order Number, Quantity, Date, Item, and Price. In our first one of SQL examples,
imagine a situation where the zip and phone fields were transposed and all the phone numbers
were erroneously entered into the zip code field. We can easily fix this problem with the
following SQL statement:

1UPDATE Customers SET Zip=Phone, Phone=Zip

22. Returning a Column of Unique Values

Now, suppose that our data entry operator added the same Customers to the Customers table
more than once by mistake. As you know, proper indexing requires that the key field contains
only unique values. To fix the problem, we will use SELECT DISTINCT to create an indexable
list of unique customers:
1SELECT DISTINCT ID FROM Customers

23. Making a Top 25 with the SELECT TOP Clause

Next, imagine that our Customers table has grown to include thousands of records, but we just
want to show a sample of 25 of these records to demonstrate the column headings and The
SELECT TOP clause allows us to specify the number of records to return, like a Top-25 list. In
this example we will return the top 25 from our Customers table:

1SELECT TOP 25 FROM Customers WHERE Customer_ID<>NULL;

24. Searching for SQL Tables with Wildcards

Wildcard characters or operators like “%” make it easy to find particular strings in a large table
of thousands of records. Suppose we want to find all of our customers who have names
beginning with “Herb” including Herberts, and Herbertson. The % wildcard symbol can be used
to achieve such a result. The following SQL query will return all rows from the Customer table
where the Customer_name field begins with “Herb”:

1SELECT * From Customers WHERE Name LIKE 'Herb%'

25. Between Monday and Tuesday

Today is Wednesday, and we arrive at work and discover that our new data entry clerk in
training has entered all new orders incorrectly on Monday and Tuesday. We wish to teach our
new trainee to find and correct all erroneous records. What’s the easiest way to get all the records
from the Orders table entered on Monday and Tuesday? The Between clause makes the task a
breeze:

1SELECT ID FROM Orders WHERE

2Date BETWEEN ‘01/12/2018’ AND ‘01/13/2018’

26. Finding the Intersection of Two Tables

Undoubtedly the whole reason that a relational database exists in the first place is to find
matching records in two tables! The JOIN statement accomplishes this core objective of SQL
and makes the task easy. Here we are going to fetch a list of all records which have matches in
the Customers and Orders tables:

1SELECT ID FROM Customers INNER

2JOIN Orders ON Customers.ID = Orders.ID

The point of INNER JOIN, in this case, is to select records in the Customers table which have a
matching customer ID values in the Orders table and return only those records. Of course, there
are many types of JOIN, such as FULL, SELF, and LEFT, but for now, let’s keep things
interesting and move on to more diverse types of SQL queries.

27. Doubling the Power with UNION

We can combine the results of two SQL queries examples into one naturally with the UNION
keyword. Suppose we want to create a new table by combining the Customer_name and phone
from Customers with a list of that customer’s recent orders so that we can look for patterns and
perhaps suggest future purchases. Here is a quick way to accomplish the task:

1SELECT phone FROM Customers

2UNION SELECT item FROM Orders

The UNION keyword makes it possible to combine JOINS and other criteria to achieve very
powerful new table generation potential.

28. Making Column Labels More Friendly

Aliasing column labels give us the convenience of renaming a column label to something more
readable. There is a tradeoff when naming columns to make them succinct results in reduced
readability in subsequent daily use. In our Orders table, the item column contains the description
of purchased products. Let’s see how to alias the item column to temporarily rename it for
greater user-friendliness:

1SELECT Item AS item_description FROM Orders

29. Always and Everywhere!

Wouldn’t it be great if there were a set of conditions you could depend on every time? The SQL
queries using ANY and ALL can make this ideal a reality! Let’s look at how the ALL keyword
is used to include records only when a set of conditions is true for ALL records. In the following
example, we will return records from the Orders table where the idea is to get a list of high
volume orders for a given item, in this case for customers who ordered more than 50 of the
product:

1SELECT Item FROM Orders

2WHERE id = ALL
3(SELECT ID FROM Orders
4WHERE quantity > 50)

30. Writing Developer Friendly SQL

An often overlooked but very important element of SQL scripting is adding comments to a script
of queries to explain what it’s doing for the benefit of future developers who may need to revise
and update your SQL queries.

A SQL script is a collection of SQL elements and commands accumulated as a file in SQL
Scripts. This script file can include many SQL commands or PL/SQL codes. One can
utilize SQL Scripts to build, edit, design, execute, and delete files.

The — single line and the /* .. */ multi-line delimiters empower us to add useful comments to
scripts, but this is also used in another valuable way. Sometimes a section of code may not be in
use, but we don’t want to delete it, because we anticipate using it again. Here we can simply add
the comment delimiter to deactivate it momentarily:

1
2
3
4
5
/* This query below is commented so it won't execute*/
6 /*
7 SELECT item FROM Orders
8 WHERE date ALL = (SELECT Order_ID FROM Orders
9 WHERE quantity > 50)
1 */
0  
/* the SQL query below the will be executed
1 ignoring the text after "--"
1 */
1  
2 SELECT item -- single comment 
1 FROM Orders -- another single comment
WHERE id
3 ALL = (SELECT ID FROM Orders
1 WHERE quantity > 25)
4
1
5
1
6

31.  SQL queries for Database Management

So far we have explored SQL query examples for querying tables and combining records from
multiple queries. Now it’s time to take a step upward and look at the database on a structural
level. Let’s start with the easiest SQL statement of all which creates a new database. Here, we
are going to create the DB as a container for our Customers and Orders tables used in the
previous ten examples above:

1CREATE DATABASE AllSales

32. Adding Tables to Our New DB
Next, we will actually add the Customers table which we’ve been using in previous examples,
and then add some of the column labels which we are already familiar with:

1CREATE TABLE Customers (

2ID varchar(80),
3Name varchar(80),
4Phone varchar(20),
5....
);
6

Although most databases are created using a UI such as Access or OpenOffice, it is important to
know how to create and delete databases and tables programmatically via code with SQL
statements. This is especially so when installing a new web app and the UI asks new users to
enter names for DBs to be added during installation.

33. Modifying and Deleting Tables with SQL

The ALTER statement is used to modify or change the meaning of a table. In the case of the
relational tables with columns, ALTER statement is used to update the table to the new or
modified rules or definition. Alter belongs to the DDL category of Commands. Data definition
language can be described as a pattern for commands through which data structures are
represented.

Imagine that you decide to send a birthday card to your customers to show your appreciation for
their business, and so you want to add a birthday field to the Customers table. In these SQL
examples, you see how easy it is to modify existing tables with the ALTER statement:

1ALTER TABLE Customers ADD Birthday varchar(80)

If a table becomes corrupted with bad data you can quickly delete it like this:

1DROP TABLE table_name

34. The Key to Successful Indexing

An index is a schema element that includes a record for each content that arrives in the indexed
column of the database table or cluster and gives a high-speed path to rows. There are many
types of indexes such as Bitmap indexes, Partitioned indexes, Function-based indexes, and
Domain indexes.

Accurate indexing requires that the Primary Key column contains only unique values for this
purpose. This guarantees that JOIN statements will maintain integrity and produce valid matches.
Let’s create our Customers table again and establish the ID column as the Primary Key:
1CREATE TABLE Customers (
2ID int NOT NULL,
3Name varchar(80) NOT NULL,
4PRIMARY KEY (ID)
);
5

We can extend the functionality of the Primary Key so that it automatically increments from a
base. Change the ID entry above to add the AUTO_INCREMENT keyword as in the following
statement:

1ID int NOT NULL AUTO_INCREMENT

35. Advanced Concepts For Improving Performance

Whenever practical, is always better to write the column name list into a SELECT statement
rather than using the * delimiter as a wildcard to select all columns. SQL Server has to do a
search and replace operation to find all the columns in your table and write them into the
statement for you (every time the SELECT is executed). For example:

1SELECT * FROM Customers

Would actually execute much faster on our database as:

1SELECT Name, Birthday, Phone,

2Address, Zip FROM Customers

Performance pitfalls can be avoided in many ways. For example, avoid the time sinkhole of
forcing SQL Server to check the system/master database every time by using only a stored
procedure name, and never prefix it with SP_. Also setting NOCOUNT ON reduces the time
required for SQL Server to count rows affected by INSERT, DELETE, and other commands.
Using INNER JOIN with a condition is much faster than using WHERE clauses with conditions.
We advise developers to learn SQL server queries to an advanced level for this purpose. For
production purposes, these tips may be crucial to adequate performance. Notice that our tutorial
examples tend to favor the INNER JOIN.

36. Conditional Subquery Results

The SQL operator EXISTS tests for the existence of records in a subquery and returns a value
TRUE if a subquery returns one or more records. Have a look at this query with a subquery
condition:

1SELECT Name FROM Customers WHERE EXISTS

2(SELECT Item FROM Orders
3WHERE Customers.ID = Orders.ID AND Price < 50)
In this example above, the SELECT returns a value of TRUE when a customer has orders valued
at less than $50.

37. Copying Selections from Table to Table

There are a hundred and one uses for this SQL tool. Suppose you want to archive your yearly
Orders table into a larger archive table. This next example shows how to do it.

1INSERT INTO Yearly_Orders

2SELECT * FROM Orders
3WHERE Date<=1/1/2018

This example will add any records from the year 2018 to the archive.

38. Catching NULL Results

The NULL is the terminology applied to describe an absent value. Null does not mean zero. A
NULL value in a column of a table is a condition in a domain that seems to be empty. A column
with a NULL value is a domain with absent value. It is essential to recognize that a NULL value
is distinct from a zero.

In cases where NULL values are allowed in a field, calculations on those values will produce
NULL results as well. This can be avoided by the use of the IFNULL operator. In this next
example, a value of zero is returned rather than a value of NULL when the calculation
encounters a field with a NULL value:

1SELECT Item, Price *

2(QtyInStock + IFNULL(QtyOnOrder, 0))
3FROM Orders

39. HAVING can be Relieving!

The problem was that the SQL WHERE clause could not operate on aggregate functions. The
problem was solved by using the HAVING clause. As an example, this next query fetches a list
of customers by the region where there is at least one customer per region:

1SELECT COUNT(ID), Region

2FROM Customers
3GROUP BY Region
4HAVING COUNT(ID) > 0;

40. Tie things up with Strings!

Let’s have a look at processing the contents of field data using functions. Substring is probably
the most valuable of all built-in functions. It gives you some of the power of Regex, but it’s not
so complicated as Regex. Suppose you want to find the substring left of the dots in a web
address. Here’s how to do it with an SQL Select query:

1SELECT SUBSTRING_INDEX("www.bytescout.com", ".", 2);

This line will return everything to the left of the second occurrence of “. ” and so, in this case, it
will return

1<a href="https://bytescout.com">www.bytescout.com</a>

Check this video to learn about every SQL query:

.. and 20 more useful SQL Queries examples!!

41. Use COALESCE to return the first non-null expression
The SQL Coalesce is used to manage the NULL values of the database. In this method, the
NULL values are substituted with the user-defined value. The SQL Coalesce function assesses
the parameters in series and always delivers first non-null value from the specified argument
record.

Syntax

1SELECT COALESCE(NULL,NULL,'ByteScout',NULL,'Byte')

Output
ByteScout

42. Use Convert to transform any value into a particular

datatype
This is used to convert a value into a defined datatype. For example, if you want to convert a
particular value into int datatype then convert function can be used to achieve this. For example,

Syntax

1SELECT CONVERT(int, 27.64)

Output
27

43. DENSE_RANK()Analytical query

It is an analytic query that computes the rank of a row in an arranged collection of rows. An
output rank is a number starting from 1. DENSE_RANK is one of the most important analytic
SQL queries. It returns rank preferences as sequential numbers. It does not jump rank in event of
relations. For example, the following query will give the sequential ranks to the employee.

1
2
3
4
5
6
7
8
9 SELECT eno,
1 dno,
0 salary,
DENSE_RANK() OVER (PARTITION BY dno ORDER BY salary) AS ranking
1
FROM employee;
1
 
1 ENO  DNO SALARY RANKING
2 ---------- ---------- ---------- ----------
1 7933  10 1500   1
3 7788  10 2650   2
7831  10 6000   3
1 7362  20 900    1
4 7870  20 1200   2
1 7564  20 2575   3
5 7784  20 4000   4
1 7903  20 4000   4
7901  30 550    1
6 7655  30 1450   2
1 7522  30 1450   2
7 7844  30 1700   3
1 7493  30 1500   4
7698  30 2850   5
8
1
9
2
0
2
1
2
2

44. Query_partition_clause
The query_partition_clause breaks the output set into distributions, or collections, of data. The
development of the analytic query is limited to the confines forced by these partitions, related to
the process a GROUP BY clause modifies the performance of an aggregate function. If the
query_partition_clause is eliminated, the entire output collection is interpreted as a separate
partition.

The following query applies an OVER clause, so the average displayed is based on all the
records of the output set.

1
2
3
4
5
6
7
8 SELECT eno, dno, salary,
9 AVG(salary) OVER () AS avg_sal
FROM employee;
1
0  
EO  DNO SALARY AVG_SAL
1 ---------- ---------- ---------- ----------
1 7364 20 900 2173.21428
1 7494 30 1700 2173.21428
2 7522 30 1350 2173.21428
7567 20 3075 2173.21428
1 7652 30 1350 2173.21428
3 7699 30 2950 2173.21428
1 7783 10 2550 2173.21428
4 7789 20 3100 2173.21428
1 7838 10 5100 2173.21428
7845 30 1600 2173.21428
5 7877 20 1200 2173.21428
1 7901 30 1050 2173.21428
6 7903 20 3100 2173.21428
1 7935 10 1400 2173.21428
7
1
8
1
9
2
0

45. Finding the last five records from the table

Now, if you want to fetch the last eight records from the table then it is always difficult to get
such data if your table contains huge information. For example, you want to get the last 8
records fr om the employee table then you can use rownum and a union clause. The rownum is
temporary in SQL.

For example,

1Select * from Employee A where rownum <=8

2union
3select * from (Select * from Employee A order by rowid desc) where rownum <=8;

The above SQL query will give you the last eight records from the employee table where
rownum is a pseudo column. It indexes the data in an output set.

46. LAG
The LAG is applied to get data from a prior row. This is an analytical function. For example, the
following query gives the salary from the prior row to compute the difference between the salary
of the current row and that of the prior row. In this query, the ORDER BY of the LAG
function is applied. The default is 1 if you do not define offset. The arbitrary default condition
is given if the offset moves past the range of the window. The default is null if you do not define
default.

Syntax

1SELECT dtno,
2       eno,
3       emname,
4       job,
5       salary,
       LAG(sal, 1, 0) OVER (PARTITION BY dtno ORDER BY salary) AS salary_prev
6
FROM   employee;
7

Output

1 DTNO ENO ENAME JOB SAL SAL_PREV

---------- ---------- ---------- --------- ---------- ----------
2
10 7931 STEVE CLERK 1300 0
3 10 7783 JOHN MANAGER 2450 1300
4 10 7834 KING PRESIDENT 5000 2450
5 20 7364 ROBIN CLERK 800 0
6 20 7876 BRIAN CLERK 1100 800
20 7567 SHANE MANAGER 2975 1100
7 20 7784 SCOTT ANALYST 3000 2975
8 20 7908 KANE ANALYST 3000 3000
9 30 7900 JAMES CLERK 950 0
1 30 7651 CONNER SALESMAN 1250 950
0 30 7522 MATTHEW SALESMAN 1250 1250
30 7843 VIVIAN SALESMAN 1500 1250
1 30 7494 ALLEN SALESMAN 1600 1500
1
1
2
1
3
1 30 7695 GLEN MANAGER 2850 1600
4
1
5
1
6

47. LEAD
The LEAD is also an analytical query that is applied to get data from rows extra down the output
set. The following query gives the salary from the next row to compute the deviation between the
salary of the prevailing row and the subsequent row. The default is 1 if you do not define offset.
The arbitrary default condition is given if the offset moves past the range of the window. The
default is null if you do not define default.

1 SELECT eno,
       empname,
2
       job,
3        salary,
4        LEAD(salary, 1, 0) OVER (ORDER BY salary) AS salary_next,
5        LEAD(salary, 1, 0) OVER (ORDER BY salary) - salary AS salary_diff
6 FROM   employee;
7  
8 ENO EMPNAME JOB SALARY SALARY_NEXT SALARY_DIFF
---------- ---------- --------- ---------- ---------- ----------
9 7369 STEVE CLERK 800 950 150
1 7900 JEFF CLERK 950 1100 150
0 7876 ADAMS CLERK 1100 1250 150
1 7521 JOHN SALESMAN 1250 1250 0
1 7654 MARK SALESMAN 1250 1300 50
7934 TANTO CLERK 1300 1500 200
1 7844 MATT SALESMAN 1500 1600 100
2 7499 ALEX SALESMAN 1600 2450 850
1 7782 BOON MANAGER 2450 2850 400
3 7698 BLAKE MANAGER 2850 2975 125
7566 JONES MANAGER 2975 3000 25
1
7788 SCOTT ANALYST 3000 3000 0
4 7902 FORD ANALYST 3000 5000 2000
1 7839 KING PRESIDENT 5000 0 -5000
5
1
6
1
7
1
8
1
9
2
0
2
1
2
2
2
3
2
4

48. PERCENT_RANK
The PERCENT_RANK analytic query. The ORDER BY clause is necessary for this query.
Excluding a partitioning clause from the OVER clause determines the entire output set is
interpreted as a separate partition. The first row of the standardized set is indicated 0 and the
last row of the set are indicated 1. For example, the SQL query example gives the following
output.

Syntax

1SELECT
2     prdid, SUM(amount),
3     PERCENT_RANK() OVER (ORDER BY SUM(amount) DESC) AS percent_rank
4     FROM sales
5     GROUP BY prdid
     ORDER BY prdid;
6

Output

1PRDID        SUM(AMOUNT)  PERCENT_RANK

2----------- ----------- ------------
3          1    22623.5            0
4          2   223927.08           1

49. MIN
Utilizing a blank OVER clause converts the MIN into an analytic function. This is also an
analytical query. In this, the entire result set is interpreted as a single partition. It gives you the
minimum salary for all employees and their original data. For example, the following query is
displaying the use of MIN in the Select query.

1 SELECT eno,
2
3
4
5
6
7
8
9
1
       empname,
0        dtno,
1        salary,
1        MIN(salary) OVER (PARTITION BY dtno) AS min_result
1 FROM   employee;
2  
1      ENO   EMPNAME          DTNO     SALARY MIN_RESULT
3 ---------- ---------- ---------- ---------- ---------------
      7782 CLARK              10       2450            1300
1       7839 KING               10       5000            1300
4       7934 MILLER             10       1300            1300
1       7566 JONES              20       2975             800
5       7902 FORD               20       3000             800
      7876 ADAMS              20       1100             800
1
      7369 SMITH              20        800             800
6       7788 SCOTT              20       3000             800
1       7521 WARD               30       1250             950
7       7844 TURNER             30       1500             950
1       7499 ALLEN              30       1600             950
      7900 JAMES              30        950             950
8       7698 BLAKE              30       2850             950
1       7654 MARTIN             30       1250             950
9
2
0
2
1
2
2
2
3

50. MAX
Using a blank row OVER clause converts the MAX into an analytic function. The lack of a
partitioning clause indicates the entire output set is interpreted as a separate partition. This
gives the maximum salary for all employees and their original data. For example, the following
query displays the use of MAX in the select query.

1 SELECT eno,
2        empname,
3
4
5
6
7
8
9
1
0
       dtno,
1        salary,
1        MAX(salary) OVER () AS max_result
1 FROM   employee;
2  
1      ENO   EMPNAME          DTNO       SALARY    MAX_RESULT
3 ---------- ---------- ---------- ---------- ----------
      7369 SMITH              20        800       3000
1       7499 ALLEN              30       1600       3000
4       7521 WARD               30       1250       3000
1       7566 JONES              20       2975       3000
5       7654 MARTIN             30       1250       3000
1       7698 BLAKE              30       2850       3000
      7782 CLARK              10       2450       3000
6       7788 SCOTT              20       3000       3000
1       7839 KING               10       5000       3000
7       7844 TURNER             30       1500       3000
1       7876 ADAMS              20       1100       3000
8       7900 JAMES              30        950       3000
      7902 FORD               20       3000       3000
1       7934 MILLER             10       1300       3000
9
2
0
2
1
2
2
2
3

51. Top- N queries

Top-N queries give a process for restricting the number of rows delivered from organized
assemblages of data. They are remarkably beneficial when users want to give the top or bottom
number of rows from a table.

For example, the following query gives the 20 rows with 10 different values:

1 SELECT price
2 FROM   sales_order
3 ORDER BY price;
4  
PRICE
5
----------
6 100
7 100
8 200
9 200
300
1 300
0 400
1 400
1 500
1 500
600
2
 
1 PRICE
3 ----------
1 600
4 700
700
1 800
5 800
1 900
6 900
1 1000
1000
7
 
1 20 rows selected.
8
1
9
2
0
2
1
2
2
2
3
2
4
2
5
2
6
2
7
2
8
2
9
3
0
3
1

52. CORR Analytic Query

The CORR analytic function is utilized to determine the coefficient of correlation. This query is
also used to calculate the Pearson correlation coefficient. The function calculates the following
on rows in the table with no null values. This query always returns the values between +1 and -1,
which describe the following:

Syntax: CORR(exp1, exp2) [ OVER (analytic_clause) ]

Example

1SELECT empid,
2       name,
3       dno,
4       salary,
5       job,
       CORR(SYSDATE - joiningdate, salary) OVER () AS my_corr_val
6FROM   employee;
7

53. NTILE Analytic Query

The NTILE enables users to split a sequence set into a detailed number of relatively similar
groups, or containers, rows sanctioning. If the number of rows in the collection is less than the
number of containers defined, the number of containers will be decreased. The basic syntax is as
displayed below:

NTILE(exp) OVER ([ partition_clause ] order_by)

Example

1SELECT empid,
2       name,
3       dno,
4       salary,
5       NTILE(6) OVER (ORDER BY salary) AS container_no
FROM   employee;
6

54. VARIANCE, VAR_POP, and VAR_SAMP Query

The VARIANCE, VAR_POP, and VAR_SAMP are aggregate functions. These are utilized to
determine the variance, group variance, and sample variance of a collection of data individually.
As aggregate queries or functions, they decrease the number of rows, therefore the expression
“aggregate”. If the data isn’t arranged we change the total rows in the Employee table to a
separate row with the aggregated values. For example, the following query is displaying the use
of these functions:

1SELECT VARIANCE(salary) AS var_salary,

2       VAR_POP(salary) AS pop_salary,
3       VAR_SAMP(salary) AS samp_salary
4FROM   employee;
5 
6VAR_SALARY   POP_SALARY   SAMP_SALARY
------------ ----------- ------------
71479414.97  1588574.81   1388717.27
8

55. STDDEV, STDDEV_POP and STDDEV_SAMP Queries

The STDDEV, STDDEV_POP, and STDDEV_SAMP aggregate queries or functions are applied
to determine the standard deviation, population standard deviation, and cumulative sample
standard deviation individually. As aggregate queries, they decrease the number of rows,
therefore the expression “aggregate”. If the data isn’t arranged we convert all the rows in the
EMPLOYEE table to a separate row. For example, the following query is displaying the use of
all these functions.

1SELECT STDDEV(salary) AS stddev_salary,

2       STDDEV_POP(salary) AS pop_salary,
3       STDDEV_SAMP(salary) AS samp_salary
4FROM   employee;
5 
6STDDEV_SALARY POP_SALARY SAMP_SALARY
---------- -------------- ---------------
71193.50     1159.588      1193.603
8

If there is more than one account after dropping nulls, the STDDEV function gives the result of
the STDDEV_SAMP. Using an empty OVER clause converts the STDDEV query result into an
analytic query. The absence of a partitioning indicates the entire output set is interpreted as a
particular partition, so we accept the standard deviation of the salary and the primary data.

56. Pattern Matching

The pattern matching syntax adds various alternatives. Data must be treated precisely and in a
proper form. The PARTITION BY and ORDER BY conditions of all SQL analytic queries is
applied to split the data into accumulations and within each group. If no partitions are specified,
it is considered the entire sequence set is one huge partition.
For example,

The MEASURES clause specifies the column result that will be provided for each match.

Syntax

1MEASURES  STRT.tstamp AS initial_tstamp,

2          LAST(UP.tstamp) AS first_tstamp,
3          LAST(DOWN.tstamp) AS finished_tstamp

Example

1DEFINE
2  UP AS UP.products_sold > PREV(UP.products_sold),
3  FLAT AS FLAT.products_sold = PREV(FLAT.products_sold),
4  DOWN AS DOWN.products_sold < PREV(DOWN.products_sold)

57. FIRST_VALUE
The simplest way to get analytic functions is to begin by studying aggregate functions. An
aggregate function collects or gathers data from numerous rows into a unique result row. For
instance, users might apply the AVG function to get an average of all the salaries in the
EMPLOYEE table. Let’s take a look at how First_Value can be used. The primary explanation
for the FIRST_VALUE analytic function is displayed below.

Syntax:

1FIRST_VALUE
2  { (expr) [NULLS ]
3  | (expr [NULLS ])
4  }
  OVER (analytic clause)
5

Example

1SELECT eno,
2       dno,
3       salary,
4       FIRST_VALUE(salary) IGNORE NULLS
5         OVER (PARTITION BY dno ORDER BY salary) AS lowest_salary_in_dept
FROM   employee;
6

The above query will ignore null values.

58. LAST_VALUE
The primary explanation for the LAST_VALUE analytic query or function is displayed below.

1Syntax: LAST_VALUE
2  { (expr) [ { NULLS ]
3  | (expr [ NULLS ])
4  OVER (analytic clause)

The LAST_VALUE analytic query is related to the LAST analytic function. The function
enables users to get the last output from an organized column. Applying the default windowing
the output can be surprising. For example,

1SELECT eno,
2       dno,
3       salary,
4       LAST_VALUE(salary) IGNORE NULLS
5         OVER (PARTITION BY dno ORDER BY salary) AS highest_salary_in_dept
FROM   employee;
6

59. Prediction
The design sample foretells the gender and age of clients who are most expected to adopt an
agreement card (target = 1). The PREDICTION function takes the price matrix correlated with
the design and applies for marital status, and house size as predictors. The syntax of the
PREDICTION function can also apply a piece of arbitrary GROUPING information when
getting a partitioned model.

1
2
3
SELECT client_gender, COUNT(*) AS ct, ROUND(AVG(age)) AS average_age
4    FROM mining_data_shop
5    WHERE PREDICTION(sample COST MODEL
6       USING client_marital_status, house_size) = 1
7    GROUP BY client_gender
   ORDER BY client_gender;
8
    
9
CUST_GENDER         CNT    AVG_AGE
1 ------------ ---------- --------
0
1
1