Hey folks, I agree that patches must be as small as possible. I believe it will significantly increase our review experience - more fast review, and, therefore, landing to master.
However, I don't agree that we should introduce criteria based on LOC, because of mentioned reasons above. I believe that patches must be atomic, no matter how much LOC it has. In the same time, we must not have the whole feature as atomic unit here. So basically my points are: * Let's do not go with strict LOC. Decision it's ok to go with one patch or not, should be up to code reviewers. * If reviewer thinks that patch could and should be splitted into few, then he/she set -1 and ask contributor to split it. * Reviewers shouldn't hesitate to set -1 and ask to split patch. __________________________________________________________________________ OpenStack Development Mailing List (not for usage questions) Unsubscribe: [email protected]?subject:unsubscribe http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
