In this article, we are going to check if the Pyspark DataFrame or Dataset is Empty or Not.
At first, let’s create a dataframe
# import modules from pyspark.sql import SparkSession
from pyspark.sql.types import StructType, StructField, StringType
# defining schema schema = StructType([
StructField( 'COUNTRY' , StringType(), True ),
StructField( 'CITY' , StringType(), True ),
StructField( 'CAPITAL' , StringType(), True )
]) # Create Spark Object spark = SparkSession.builder.appName( "TestApp" ).getOrCreate()
# Create Empty DataFrame with Schema. df = spark.createDataFrame([], schema)
# Show schema and data df.printSchema() df.show(truncate = False )
|
Output:
Checking dataframe is empty or not
We have Multiple Ways by which we can Check :
Method 1: isEmpty()
The isEmpty function of the DataFrame or Dataset returns true when the DataFrame is empty and false when it’s not empty. If the dataframe is empty, invoking “isEmpty” might result in NullPointerException.
Note : calling df.head() and df.first() on empty DataFrame returns java.util.NoSuchElementException: next on empty iterator exception.
print (df.head( 1 ).isEmpty)
print (df.first( 1 ).isEmpty)
print (df.rdd.isEmpty())
|
Output:
True True True
Method 2: count()
It calculates the count from all partitions from all nodes
Code:
print (df.count() > 0 )
print (df.count() = = 0 )
|
False True