John Snow <js...@redhat.com> writes: > On 8/5/20 3:36 AM, Markus Armbruster wrote: >> John Snow <js...@redhat.com> writes: >> >>> On 8/4/20 4:03 AM, Markus Armbruster wrote: >>>> The pain of tweaking the parser is likely dwarved several times over by >>>> the pain of the flag day. >>> >>> You mention this often; I wonder if I misunderstand the critique, >>> because the pain of a "flag day" for a new file format seems >>> negligible to me. >>> >>> I don't think we edit these .json files very often. Generally, we add >>> a new command when we need one. The edits are usually one or two lines >>> plus docstrings. >>> >>> If anyone has patches in-flight, I genuinely doubt it will take more >>> than a few minutes to rewrite for the new file format. >>> >>> No? >> >> You describe the the flag day's one-time pain. >> >> There's also the longer term pain of having to work around git-blame >> unable to see beyond the flag day. >> > > So it's not really the "flag day" we're worried about anymore, it's > the ongoing ease-of-use for vcs history.
Feel free to call that pain however you want. I'm going to call it "lasting aftereffects of the flag day" :) >> I'm not claiming the pain is prohibitive (if I thought it was, I >> would've tried to strange this thread in its crib), I am claiming it'll >> be much more painful (read: expensive) than a parser tweak. >> > > I do use `git blame` quite a lot, but with a project as old as QEMU, > most of my trips through history do involve jumping across a few > refactor gaps as a normal part of that process. > > As Dan points out, I often have to back out and add refactorSHA^ to my > invocation, and just keep hopping backwards until I find what I am > truly after. It just feels like a fact of programmer life for me at > this point. The fact that we all need to cope with this class of issue doesn't mean we should create more instances unthinkingly. We should only when we believe the benefits are worth it, and can't find a cheaper way to get them. We've discussed "is it really that bad" at some length. What I'm missing so far is a clear writeup of the benefits beyond "editor works out of the box" (which is quite a desirable benefit, but can also be had without a flag day). > I've not used Paolo's invocation before, but it looks like it might be > useful. I'll try to remember to try it out.
