Re:How I review code
by kripke on Wednesday January 24, 2018 @04:29PM
(#16226502)
1000 should really be handled by automated tools. Takes useless burden from the reviewer, and emotionally easier for both sides too.
[ Parent
| on HN ]
It can be, especially at a company. (But then watch the bikeshed discussion on the tools. And PEP8 is a guideline, not a set of hard and fast rules. Beautiful is better than ugly and all that. ).
What I see as one of 5 maintainers of an open source project is that when a review comes back with a bunch of formatting comments, it's because the reviewer didn't step back and see the other parts. Raymond Hettinger has a talk where he discusses it, but it's the sort of feedback that can be given in almost any case and can obscure the more fundamental issues with the code.