> Hi Joel, every commit in this MR is below 500 lines (if we don't consider test > code), so I can easily create separate MRs if you prefer, although that > doesn't really alter the complexity of the changeset.
It may not alter the complexity of the changeset, however it does alter the complexity of the review. The problem with large reviews is that they have significantly higher cognitive overhead for the reviewer, plus there is only one large feedback loop - if there are issues that we need to provide feedback on, it will typically be repeated multiple times throughout the change, rather than a single comment on a much smaller diff. > In any case I have noticed that I committed the resource zip by mistake so > this MR has to be redone anyway. > Oh and although I tried to separate the change in logically separate commits, > I cannot guarantee that not merging all the commits together would still yield > a bug-free master branch. I have run tests on the complete changeset after > all. Sure. There are two options here - one is to merge into an intermediate branch/repo while the code is being changed, then merge that back into the master once complete. The other is to have stacked merge proposals, go through the review process, then land them all together. -- https://code.launchpad.net/~aieri/hw-health-charm/+git/hw-health-charm/+merge/364694 Your team Nagios Charm developers is subscribed to branch hw-health-charm:master. -- Mailing list: https://launchpad.net/~nagios-charmers Post to : [email protected] Unsubscribe : https://launchpad.net/~nagios-charmers More help : https://help.launchpad.net/ListHelp
