qudade commented on a change in pull request #26118: [SPARK-24915][PySpark] Fix
Row handling with Schema.
URL: https://github.com/apache/spark/pull/26118#discussion_r335021018
##########
File path: python/pyspark/sql/types.py
##########
@@ -599,6 +599,8 @@ def toInternal(self, obj):
if isinstance(obj, dict):
return tuple(f.toInternal(obj.get(n)) if c else obj.get(n)
for n, f, c in zip(self.names, self.fields,
self._needConversion))
+ elif isinstance(obj, Row):
+ return self.toInternal(obj.asDict())
Review comment:
yes, but because `Row` also behaves like a `dict` (as in `row[colName]`) the
tuple handling can be off. e.g. if you have `row = Row(name='Alice',
date=date(2019,10,1))` the underlying tuple reverses the order of fields (see
examples above).
How about
`elif isinstance(obj, Row) and getattr(obj, "__from_dict__", False):`
as in
https://github.com/apache/spark/blob/master/python/pyspark/sql/types.py#L614
?
----------------------------------------------------------------
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:
us...@infra.apache.org
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
