paulk-asert commented on PR #2210: URL: https://github.com/apache/groovy/pull/2210#issuecomment-2852386998
I was thinking a bit more about usage overnight. For arrays, the "clean insertion" point argument is less compelling but somewhat useful none-the-less. Given that it isn't easy to partition primitive arrays, I think partitionPoint is slightly less compelling. We could get by with both bisectLeft and bisectRight for instance. But even for natural ordering scenarios, the convenience of a single method is nicer than the two separate methods in my view. I looked at the grades example from Python bisect: https://docs.python.org/3/library/bisect.html#examples ``` def scores = [33, 99, 77, 70, 89, 90, 100] // convert these to A grade (90 and above), // B grade (80 up to but not including 90), etc. ``` With binary search we have something like this: ``` def grade(score) { int[] breakpoints = [59, 69, 79, 89] def idx = Arrays.binarySearch(breakpoints, score) if (idx < 0) { idx = (idx + 1).abs() } 'FDCBA'[idx] } assert scores.collect(this.&grade) == ['F', 'A', 'C', 'C', 'B', 'A', 'A'] ``` We have to artificially subtract one from each of the breakpoints compared to the Python bisect example. With partitionPoint it would become: ``` int[] breakpoints = [60, 70, 80, 90] assert scores2.collect { sc -> 'FDCBA'[breakpoints.partitionPoint{ it <= sc }] } == ['F', 'A', 'C', 'C', 'B', 'A', 'A'] ``` This is much cleaner. -- 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. To unsubscribe, e-mail: notifications-unsubscr...@groovy.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
