pyspark.sql.functions.try_to_date#

pyspark.sql.functions.try_to_date(col, format=None)[source]#

This is a special version of to_date that performs the same operation, but returns a NULL value instead of raising an error if date cannot be created.

Added in version 4.0.0.

Parameters:
colColumn or column name

input column of values to convert. A column that evaluates to a string, date, or timestamp.

format: literal string, optional

format to use to convert date values. A column that evaluates to a string.

Returns:
Column

date value as pyspark.sql.types.DateType type. Returns a column that evaluates to a date.

Examples

>>> import pyspark.sql.functions as sf
>>> df = spark.createDataFrame([('1997-02-28',)], ['ts'])
>>> df.select('*', sf.try_to_date(df.ts)).show()
+----------+---------------+
|        ts|try_to_date(ts)|
+----------+---------------+
|1997-02-28|     1997-02-28|
+----------+---------------+
>>> df.select('*', sf.try_to_date('ts', 'yyyy-MM-dd')).show()
+----------+---------------------------+
|        ts|try_to_date(ts, yyyy-MM-dd)|
+----------+---------------------------+
|1997-02-28|                 1997-02-28|
+----------+---------------------------+
>>> df = spark.createDataFrame([('foo',)], ['ts'])
>>> df.select(sf.try_to_date(df.ts)).show()
+---------------+
|try_to_date(ts)|
+---------------+
|           NULL|
+---------------+