try_to_time

Converts a column into TimeType using the optionally specified format. Specify formats according to datetime pattern. By default, it follows casting rules to TimeType if the format is omitted. Equivalent to col.cast("time"). The function always returns null on an invalid input.

Syntax

import pyspark.sql.functions as sf

sf.try_to_time(str=<str>)

# With format
sf.try_to_time(str=<str>, format=<format>)

Parameters

Parameter	Type	Description
str	pyspark.sql.Column or str	String to be parsed to time.
format	pyspark.sql.Column or str	Optional. Time format pattern to follow.

Returns

pyspark.sql.Column: time value as pyspark.sql.types.TimeType type.

Examples

Example 1: Convert string to a time.

import pyspark.sql.functions as sf
df = spark.createDataFrame([("10:30:00",)], ["str"])
df.select(sf.try_to_time(df.str).alias("time")).show()

+--------+
|    time|
+--------+
|10:30:00|
+--------+

Example 2: Convert string to a time with a format.

import pyspark.sql.functions as sf
df = spark.createDataFrame([("10:30:00", "HH:mm:ss")], ["str", "format"])
df.select(sf.try_to_time(df.str, df.format).alias("time")).show()

+--------+
|    time|
+--------+
|10:30:00|
+--------+

Example 3: Conversion failure results in NULL.

import pyspark.sql.functions as sf
df = spark.createDataFrame([("malformed",)], ["str"])
df.select(sf.try_to_time(df.str).alias("time")).show()

+----+
|time|
+----+
|NULL|
+----+