Count Substring In String Column Using Spark Dataframe

I have a Spark dataframe with a column (assigned_products) of type string that contains values such as the following: 'POWER BI PRO+Power BI (free)+AUDIO CONFERENCING+OFFICE 365 EN

Solution 1:

Here's a non-udf solution. Split your string on the character you are trying to count and the value you want is the length of the resultant array minus 1:

from pyspark.sql.functions import col, size, split
DF.withColumn('Number_Products_Assigned', size(split(col("assigned_products"), r"\+")) - 1)

You have to escape the + because it's a special regex character.

+--------------------+------------------------+
|   assigned_products|Number_Products_Assigned|
+--------------------+------------------------+
|POWER BI PRO+Powe...|                       3|
+--------------------+------------------------+

Solution 2:

from pyspark.sql.functions import col,udf

@udf(returnType='int')defcnt(s):
    return s.count('+')

DF.withColumn('Number_Products_Assigned', cnt(col("assigned_products")))

Solution 3:

Replace will replace the occurrence of the sub-string with null string. So we can count the occurrences by comparing the lengths before and after the replacement as follows:

Using SparkSQL:

SELECT length(x) - length(replace(x,'+')) as substring_count
FROM  (select'abc+def+ghi++aaa'as x) -- Sample data

Output:

substring_count
---------------
4

Using PySpark functions:

import pyspark.sql.functions as F

df1 = spark.sql("select 'abc+def+ghi++aaa' as x") # Sample data
df1.withColumn('substring_count', 
                F.length(col('x')) 
               - F.length(F.regexp_replace(col('x'), '\+', '')) 
              ).show()

Output:

+----------------+---------------+
|               x|substring_count|
+----------------+---------------+
|abc+def+ghi++aaa|              4|
+----------------+---------------+