replaced the implementation of Table method dropDuplicateRows() with …

[lwhite1]

…one that uses less memory

Thanks for contributing.

• [x ] Tick to sign-off your agreement to the Developer Certificate of Origin (DCO) 1.1

Description

The method dropDuplicateRows was very inefficient in its use of memory. There was, by the end of the method:

• the original table
• a sorted copy of the original
• a copy of the sorted copy, without the duplicates

Plus, the comparison method to test equality value-by-value used a generic method that auto boxed all the primitive values.

This was replaced by a method that computes a hash function for the rows which is used to test for equality, eliminating the need for the sorted table copy. Unfortunately, the auto boxing equality test remains at least for now, however. it is only called when duplicate rows are encountered, rather than for every row in the table.

Testing

Did you add a unit test?
yes