Details
-
Improvement
-
Status: Open
-
Minor
-
Resolution: Unresolved
-
2.3.0, 2.3.1, 2.3.2
-
None
-
None
Description
This is for continuous discussion with a requirement added in https://github.com/apache/spark/pull/22455#discussion_r223197863.
Summary:
SparkDataFrame is a S4 object, `show()` is the default method to display the data frame to screen output. Currently the output is simply in string format returned by `showString()` call which pre-formats the data frame and displays as a table. This lacks the flexibility to re-format the output with a more user friendly and pretty fashion, as has been seen in 1) S3 object's `print()` method allows to specify arguments like `quote` etc to control the output; 2) external tools such as `Jupyter` R notebook implement their own customized way of display.
This Jira aims to explore a feasible solution to improve the screen output experience by both supporting a pretty print from with the SparkR package and also offering a common hook for external tools to customize the display function.