Although I haven't touched MediaWiki code for a year or so, based on my experience with large codebases with tons of contributors, I would be very much PRO.
I understand it is a pain, but as Legoktm points out, it is a manageable pain. Having a consistent and higher-quality code base is worth the migration pain. Three more advantages: * future changesets are cleaner, as one does not have to do the clean up in addition to the actual change they wanted to do * automatic testing tools can capture issues with a higher confidence if it doesn't have to take historical exceptions into account * most developers code by copy-and-paste of style, structures, and ideas. So even if a new styleguide is in place, it can often be the case that a developer will start building off the old styleguide as they simply keep their code consistent with the code that they are looking at
hth
On Fri, Feb 12, 2016 at 11:57 AM Brad Jorsch (Anomie) bjorsch@wikimedia.org wrote:
On Fri, Feb 12, 2016 at 2:26 PM, Legoktm legoktm.wikipedia@gmail.com wrote:
I think you're going to end up in rebase hell regardless, so we should rip off the bandaid quickly and get it over with, and use the automated tools we have to our advantage.
This. Just get it over with, and then it's only one patch screwing up rebases and blames instead of lots.
-- Brad Jorsch (Anomie) Senior Software Engineer Wikimedia Foundation _______________________________________________ Wikitech-l mailing list Wikitech-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikitech-l