CMSC325

SQL Functions
DR. RICHA SHARMA
L O C K H AV E N U N I V E R S I T Y

1
 Introduction
 Retrieving information from tables often requires data
manipulations, such as decomposing any attribute into small
units (e.g. date) or working with a particular case or part of text.
 SQL functions are very useful to do such data manipulations
without having the need to add any additional attribute!

 There are several SQL functions, such as string, data

conversion, arithmetic, date, and time functions that we can
use in our select statements to produce results as well as
format our output.
 We can test these functions without having to set up test tables
in SQL Server! [Oracle requires us to use a built-in virtual table
called dual]

2
 Few Examples
 Mathematical calculations:
SELECT 2*5-4

 Changing case of text:

SELECT UPPER(string functions')

 Joining/concatenating strings:
SELECT ‘Good' + ' ' + ‘Morning!’

 Getting current date:

SELECT getdate()
SELECT getdate() + 2

3
 String Functions
Function Name Function Desription
Concatenation function [ + in SQL Server, Join two or more strings
|| in Oracle, concat() in MySQL ]

Case change [ upper() / lower() in SQL To convert upper case string to

Server, Oracle, MySQL ; ucase()/lcase() in lower and vice-versa
some versions of MySQL ]
Substring [ SUBSTRING(value, start, len) in To return substring of a string
SQL Server, SUBSTR(value, p, len) in Oracle
& MySQL ]
Length [ LEN(value) in SQL Server, Length of string
LENGTH(value) in Oracle and MySQL ]
Replace [ REPLACE(value, X, Y) ] Replace X by Y in value string.

Reverse [ REVERSE(value) ] Reverse a string

Trim [ TRIM() ] { LTRIM() and RTRIM() too! ] Remove leading and trailing
spaces from a string

Source: https://www.w3schools.com/sql/sql_ref_sqlserver.asp 4
 Examples
 Write a query to select employee title, first name and last name
in upper case from ‘Emp’ table – all these three attributes
combined as one column named as ‘Employee Name’.
 select Emp_title + EMP_FNAME + ' ' + EMP_LNAME as
EMP_NAME from EMP;
 Write a query to get a substring of length 10 starting from
position 1 on product description attribute from the product table.
The extracted substrings should be unique!
 select distinct substring(product.P_DESCRIPT, 1, 10)
from PRODUCT;
 Write a query to find maximum size or length of the customer last
name from customer table.
 select max(len(CUS_LNAME)) from customer;

5
 Numeric Functions
Function Name Function Desription
Basic numeric operators [ + , - , *, / , %, PI() ]
Abs(x) Returns absolute value of a number

Round(x) Rounds a value to a specified precision

Sqrt(x) Returns square-root of a number

Exp(X) / Log(X) Returns e / log of an input value.

Ceiling(X) Return the smallest integer value that is >= a

Ceil(X) in Oracle, MySQL number
Floor(X) Return the largest integer value that is <= X

Max / min Returns maximum or minimum value in a set of

values.
Trigonometry functions sin, cos, tan, asin, acos, atan

Source: https://www.w3schools.com/sql/sql_ref_sqlserver.asp 6
 Examples
 Let's say we have a table of racer ids and times in minutes for a
boat ‘race’ . How can we convert minutes to hours and minutes?
Boat_id Elapsed_time
123 234
111 321
333 275

 Write a query to find the total price of the products ordered from
‘line’ table rounding the price to two decimal places.
 select round(sum(line.LINE_PRICE),2) from line;

7
 Date/Time Functions
 Date is stored internally as an integer value equal to the number
of days since December 31, 1899. Because DATE values are
stored as integers, we can use them in arithmetic expressions
using date values!
 Expressed as string
 Dates are specified as quoted character strings!
 Few date functions next!!
Select emp_num, year(emp_hire_date) as ‘Year of Hiring’
from emp where year(emp_hire_date) > 1990
Select emp_num, emp_hire_date, dateadd(month, 6,
emp_hire_date) from emp
Select emp_num, datediff(year, Emp_hire_date, getdate())
from employee where year(emp_hire_date) >=‘1991’

8
 Date/Time Functions
Function Name Function Desription
getdate() or Returns today’s date
current_timestamp in MySQL
Sysdate() - Oracle

year(val), month(val), Returns year, month, day of the input date

day(val)
Adds a number of dateparts to a given date
DATEADD(datepart, number, where dateparts can be days, weeks, months,
date) quarters, years, minutes, or hours
Returns the difference between two dates
DATEDIFF(datepart, startdate, expressed in a selected datepart, eg. year,
enddate) month etc.

CONVERT(varchar(length), Format code can be any number from 100 to

dateVal, fmt_code) 127

CONVERT(datetime, str_val) Converts str_val to date/time

ISDATE(val) Return true if val is a valid date.

Source: https://www.w3schools.com/sql/sql_ref_sqlserver.asp 9
 Conversion Functions
 Conversion function in SQL Server is ‘convert’ function specified
as:
CONVERT(data_type(length), expression, format)

 On this function, length is optional and is used with string data.

Format is also optional and is used with date type data!

 Example:
SELECT CONVERT(int, 55.75)
SELECT CONVERT(float, '55.75')

 Oracle has specific functions for conversion such as: to_char(),

to_date(), to_number() etc.

10
 Cast Functions
 Cast function in SQL Server converts a value (of any type) into a
specified datatype:

CAST(expression AS datatype(length))

 Example:
SELECT CAST(25.65 AS int)
SELECT CAST(25.65 AS varchar)
SELECT CAST('25.65' AS float)
SELECT CAST('2023-10-02' AS datetime)

11
 Regular Expressions
 We use LIKE to perform some simple pattern matching.
Example:
SELECT emp_num, emp_lname, emp_phone
FROM Emp WHERE emp_lname LIKE 'W%'

 Regular Expressions allow us to represent a broad range of

patterns and use them to search for matches more precisely!
 For example, the pattern : ^192[5-9]$
is used to find strings of digits that start with 192 and end in
either a 5, 6, 7, 8 or 9.
The carat (^) is a special character indicating the start of the
string.
The $ indicates the end of the string.
The range [5-9] matches characters from 5 to 9.
12
 Regular Expressions (ctd.)
 Sample patterns:
 \ - matches a special character or literal
 \( - matches a single left parenthesis
 \. - matches an actual period
 \d - matches a digit
 \D - matches a non-digit
 \w - matches a "word" character - alphanumeric or underscore
 \W - matches a non-word character
 \s - matches a whitespace character (e.g. space, tab, newline)
 \S – matches a non-space character
 \d\.\d matches numbers of the form 1.2, 3.0, etc.
 \d{3} matches any 3 digits
 For example, SSN pattern would be \d{3}\-\d{2}\-\d{4}
 A product code pattern may be like: [ABC][a-z]*\d{3}

https://learn.microsoft.com/en-us/sql/ssms/scripting/search-text-with-regular-expressions?
13
view=sql-server-ver16
 Regular Expressions (ctd.)
 Few more patterns:

 [abc] – matches "a" or "b" or "c"

 [a-z] – matches any lowercase letter from "a" to "z"
 a|b – matches an "a" or a "b" matches
 * indicates any number of repetitions

14
 Relational Set Operations
 Relational databases support set operations too – these
operations are equivalent to logical AND / OR / NOT
operators!

 UNION ALL: Returns all rows retrieved by either query including

duplicate rows.

 UNION (OR): Returns all non-duplicate rows retrieved by all

queries

 INTERSECT (AND): Returns all rows retrieved by both queries

 EXCEPT (MINUS): Returns the remaining rows when rows

retrieved by second query are subtracted from rows returned by
first query
15
 Example: set operations
 Example of union operation:

SELECT * FROM Vendor WHERE v_state = 'TN' OR v_state = 'FL’

is equivalent to the query:

SELECT * FROM Vendor WHERE v_state = 'TN'

union
SELECT * FROM Vendor WHERE v_state = 'FL’

 Let’s try one more example:

SELECT * FROM Vendor WHERE v_contact like 'Smith%'

union all
SELECT * FROM Vendor WHERE v_contact = 'Smith'

16
 Relational Set Operations (ctd.)
 If we are working with the same table, use of the logical
operators is more convenient!

 On different tables with the same structure (same number of

columns and same data types, regardless of column name)
we can use the set operators – gives us some tools to
perform logical operations on the resultant query.

 For example: Assume we have two tables university

Employees and Students, we can find those who appear in
both tables as show in the figure:

17
 Relational Set Operations (ctd.)

 Let’s assume we have two tables: ‘US_Products’ for those

products that are sold in the US and ‘International_Products’ are
those that are sold internationally, and that these tables have the
same underlying schema. We may have products that appear in
both lists!
 We can use INTERSECT operation to find common products and
EXCEPT operation to find the ones sold either in US or
internationally!
18