[SPARK-16651] Document no exception using DataFrame.withColumnRenamed when existing column doesn't exist

• Type: Improvement
• Status: Resolved
• Priority: Minor
• Resolution: Fixed
• Affects Version/s: 1.5.0
• Fix Version/s: 2.0.1
• Component/s: Documentation, PySpark
• Assignee: Dongjoon Hyun
• Reporter: Tom Phillips
• Created: 20/Jul/16 14:12
• Updated: 22/Jul/16 12:20
• Resolved: 22/Jul/16 12:20

Description

The withColumnRenamed method does not raise an exception when the existing column does not exist in the dataframe.

Example:

In [4]: df.show()
+---+-----+
|age| name|
+---+-----+
|  1|Alice|
+---+-----+


In [5]: df = df.withColumnRenamed('dob', 'date_of_birth')

In [6]: df.show()
+---+-----+
|age| name|
+---+-----+
|  1|Alice|
+---+-----+

Issue Links

• links to: [Github] Pull Request #14288 (dongjoon-hyun)