programming, video games, dadding. I happen to work for Xbox, but don't represent them.
after like 3 weeks of work i managed to finally get out the PR i've been working on. 67 files touched, ouch. this is not a brag. that's a bad thing. but this was also a "rearchitect how the main messaging component works", which is tied in with a lot of other parts, so I could not figure out how to do this in small, incremental pieces without ripping off the bandaid.
rip my team for having to review this thing. it does have some killer unit tests with it, though.
oh speaking of lol, here's one that's in progress 😎
noah@Noahs-MacBook-Pro ~/personal/netrunner [(7a573ce...) ? +0 ~37 -0 ! (1)]
$ git diff --shortstat
37 files changed, 1194 insertions(+), 1163 deletions(-)
Yeah that's pretty significant, haha. I mean, I'd do it if I had to, but I'm not gonna love it. The worst part of mine is I renamed a lot of files and changed just enough that git isn't counting it as a move+modify
oh god I hate when git does that. That breaks git blame, makes it harder to follow changes across time. 
godspeed to you and your reviewers
For big PRs I like to rewrite git history to split the PR into atomically reviewable commits, but it can be a lot of work and idk how much reviewers make use of it.
I don't think I could manage atomically reviewable here--everything is intertwined--but I like this idea and I especially like the spirit of it. You've inspired me to write a reviewer's guide to outline the changes... I can meet them halfway on this
