Adventures in Machine Learning

Maximizing the Power of SQL Server’s STRING_SPLIT() Function

SQL Server STRING_SPLIT() Function: Splitting Strings Made EasyIn SQL Server, there’s a built-in function called STRING_SPLIT() that makes it easy to split a string into separate rows of substrings based on a separator. This table-valued function was introduced in SQL Server 2016, and it has become increasingly popular among developers.

In this article, we’ll explore the features of the STRING_SPLIT() function and how to use it to split strings in SQL Server.

Syntax

To use the STRING_SPLIT() function, you need to specify two parameters: the input string and the separator that you want to use to split the input string. Here’s the basic syntax:

STRING_SPLIT ( input_string , separator )

The input_string parameter is the string that you want to split.

It can be a character string, binary string, text, or text string. The separator parameter is the character or symbol that you want to use to split the input string into separate substrings.

The output of the STRING_SPLIT() function is a table that contains the rows of substrings. Each row contains a single substring value in the value column.

The function automatically assigns a name to the value column, but you can override this default name by using the AS keyword to assign a different name. You can also use the ORDER BY clause in the STRING_SPLIT() function to sort the rows of substrings in ascending or descending order.

Examples

Comma-Separated Value String

One common use of the STRING_SPLIT() function is to split a comma-separated value string into separate rows. Let’s say you have a string variable that contains a list of values separated by commas:

DECLARE @str varchar(100) = ‘apple,banana,orange’

To split this string into separate rows of substrings, you can use the STRING_SPLIT() function:

SELECT value FROM STRING_SPLIT(@str, ‘,’);

This query will return the following result set:

value

——

apple

banana

orange

Note that the STRING_SPLIT() function automatically removes any leading or trailing spaces from each substring. If you need to include empty substrings in your result set, you can add a WHERE clause to filter out those substrings:

SELECT value FROM STRING_SPLIT(‘one,,two,,three’, ‘,’) WHERE value <> ”;

This query will return the following result set:

value

——

one

two

three

Multi-Valued Columns

Another use of the STRING_SPLIT() function is to normalize multi-valued columns in a table. Let’s say you have a sample table called contacts that contains a column for phone numbers, but some contacts have multiple phone numbers separated by semicolons:

CREATE TABLE contacts (

id INT PRIMARY KEY,

name VARCHAR(50),

phone_numbers VARCHAR(100)

);

INSERT INTO contacts (id, name, phone_numbers) VALUES

(1, ‘John Doe’, ‘555-1234’),

(2, ‘Jane Smith’, ‘555-5678;555-8765’),

(3, ‘Bob Johnson’, ‘555-4321’);

To normalize the phone_numbers column into separate rows, you can use the STRING_SPLIT() function in a CROSS APPLY subquery:

SELECT c.id, c.name, s.value AS phone_number

FROM contacts c

CROSS APPLY STRING_SPLIT(c.phone_numbers, ‘;’) s

ORDER BY c.id

This query will return the following result set:

id name phone_number

— ———- ————

1 John Doe 555-1234

2 Jane Smith 555-5678

2 Jane Smith 555-8765

3 Bob Johnson 555-4321

Aggregate Function

You can also use the STRING_SPLIT() function with aggregate functions like COUNT and SUM to get counts or total values for each substring. Let’s say you have a table called orders that contains a comma-separated list of product IDs for each order:

CREATE TABLE orders (

id INT PRIMARY KEY,

customer VARCHAR(50),

product_ids VARCHAR(100)

);

INSERT INTO orders (id, customer, product_ids) VALUES

(1, ‘Alice’, ‘1,2,3’),

(2, ‘Bob’, ‘2,3,4,5’),

(3, ‘Alice’, ‘1,3,5’);

To count how many times each product ID appears in the orders table, you can use the STRING_SPLIT() function in a CROSS APPLY subquery and then apply a GROUP BY clause:

SELECT s.value AS product_id, COUNT(*) AS order_count

FROM orders o

CROSS APPLY STRING_SPLIT(o.product_ids, ‘,’) s

GROUP BY s.value

ORDER BY order_count DESC

This query will return the following result set:

product_id order_count

———- ———–

3 3

1 2

2 2

5 2

4 1

Conclusion

The SQL Server STRING_SPLIT() function provides a simple and effective way to split strings into separate rows of substrings based on a separator. With its flexible syntax and powerful features, it has become a popular tool among SQL Server developers and DBAs. Whether you need to split comma-separated value strings, normalize multi-valued columns, or perform aggregate functions on substrings, the STRING_SPLIT() function is a valuable addition to your SQL Server toolbox.

Using STRING_SPLIT() Function with an Aggregate Function: A Comprehensive GuideIn SQL Server, the STRING_SPLIT() function is a useful tool for splitting a string into separate rows of substrings based on a separator. In addition to splitting strings, you can also use this function in conjunction with an aggregate function to perform more complex calculations.

In this article, we’ll explore the features of the STRING_SPLIT() function and how you can use it with an aggregate function to get more meaningful insights from your data.

Using CONCAT_WS() Function

If you’re working with a string that contains multiple values that you want to concatenate into a single string, you can use the CONCAT_WS() function. This function takes a separator as the first argument and then any number of input values after that.

It returns a concatenated string with the separator between each value. Let’s say you have a contacts table with columns for first name, last name, and phone numbers:

CREATE TABLE contacts (

id INT PRIMARY KEY,

first_name VARCHAR(50),

last_name VARCHAR(50),

phone_numbers VARCHAR(100)

);

INSERT INTO contacts (id, first_name, last_name, phone_numbers) VALUES

(1, ‘John’, ‘Doe’, ‘555-1234;555-5678’),

(2, ‘Jane’, ‘Smith’, ‘555-4321’),

(3, ‘Bob’, ‘Johnson’, ‘555-8765’);

If you want to concatenate the first and last names into a single column separated by a space, you can use the CONCAT_WS() function in a SELECT statement:

SELECT id, CONCAT_WS(‘ ‘, first_name, last_name) AS full_name, phone_numbers

FROM contacts

This query will return the following result set:

id full_name phone_numbers

— ————– ——————-

1 John Doe 555-1234;555-5678

2 Jane Smith 555-4321

3 Bob Johnson 555-8765

Using COUNT() Function

If you want to count the number of substrings in a STRING_SPLIT() result set, you can use the COUNT() function. This function returns the number of rows in a result set.

Let’s say you want to count the number of phone numbers for each contact in the contacts table. You can use the STRING_SPLIT() function in a CROSS APPLY subquery to split the phone_numbers column into separate rows of substrings, and then use the COUNT() function to get the number of phone numbers for each contact:

SELECT id, CONCAT_WS(‘ ‘, first_name, last_name) AS full_name, COUNT(value) AS num_phones

FROM contacts

CROSS APPLY STRING_SPLIT(phone_numbers, ‘;’)

GROUP BY id, first_name, last_name

This query will return the following result set:

id full_name num_phones

— ————— ———-

1 John Doe 2

2 Jane Smith 1

3 Bob Johnson 1

Using SUM() Function

If you want to get a sum of the values in a STRING_SPLIT() result set, you can use the SUM() function. This function returns the sum of all values in a column.

Let’s say you have a product_sales table with a column for product IDs that contains comma-separated values:

CREATE TABLE product_sales (

id INT PRIMARY KEY,

product_ids VARCHAR(100),

sales INT

);

INSERT INTO product_sales (id, product_ids, sales) VALUES

(1, ‘1,2,3’, 100),

(2, ‘2,3,4,5’, 200),

(3, ‘1,3,5’, 50);

If you want to get the total sales for each product ID, you can use the STRING_SPLIT() function in a CROSS APPLY subquery to split the product_ids column into separate rows of substrings, and then use the SUM() function to get the total sales for each product ID:

SELECT s.value AS product_id, SUM(ps.sales) AS total_sales

FROM product_sales ps

CROSS APPLY STRING_SPLIT(ps.product_ids, ‘,’) s

GROUP BY s.value

ORDER BY total_sales DESC

This query will return the following result set:

product_id total_sales

———- ———–

3 350

2 300

1 150

5 50

4 200

Conclusion

The STRING_SPLIT() function is a powerful tool in SQL Server that makes it easy to split a string into separate rows of substrings. When used with an aggregate function such as COUNT() or SUM(), it can provide you with more meaningful insights into your data.

Whether you need to count the number of substrings or get the total sales for each value in a STRING_SPLIT() result set, the possibilities are endless. By incorporating the STRING_SPLIT() function into your SQL Server queries, you can save time and effort while gaining deeper insights into your data.

In conclusion, the SQL Server STRING_SPLIT() function is a powerful tool that allows you to split a string into separate rows of substrings based on a separator. When used with an aggregate function such as COUNT() or SUM(), it can provide you with more meaningful insights into your data.

The examples provided in this article showcase how the STRING_SPLIT() function can be used to handle various scenarios, like splitting a comma-separated list, normalizing multi-valued columns, concatenating values, counting substrings, and getting the total value of substrings. By incorporating the STRING_SPLIT() function into your SQL Server queries, you can save time and effort while gaining deeper insights into your data, thereby making informed decisions about your business.

Popular Posts