srowen commented on a change in pull request #23548: [SPARK-26620][PYTHON] Make
`DataFrameReader.json()` and `csv()` in Python should accept DataFrame.
URL: https://github.com/apache/spark/pull/23548#discussion_r261770133
##########
File path: python/pyspark/sql/readwriter.py
##########
@@ -289,8 +294,14 @@ def func(iterator):
keyed._bypass_serializer = True
jrdd = keyed._jrdd.map(self._spark._jvm.BytesToString())
return self._df(self._jreader.json(jrdd))
+ elif isinstance(path, DataFrame):
+ if len(path.schema) != 1 or not
isinstance(path.schema[0].dataType, StringType):
+ raise TypeError("The DataFrame can contain only one String
column: %s"
+ % path.schema.simpleString())
+ jdataset =
path._jdf.__getattr__('as')(self._spark._jvm.Encoders.STRING())
+ return self._df(self._jreader.json(jdataset))
else:
- raise TypeError("path can be only string, list or RDD")
+ raise TypeError("path can be only string, list, RDD or DataFrame")
Review comment:
I'd also say this is working as intended. on the scala side it makes sense
to support `Dataset[String]` and an `RDD[String]` is the closest analog here in
Pyspark. It's easy to select a column out of a DataFrame anyway.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
