IRCaBot 2.1.0
GPLv3 © acetone, 2021-2022
#i2p-dev
/2021/12/14
eyedeekay I see branch permissions are working
zzz yeah a few false starts on my side
zzz had to checkout from the same rev that you started at to avoid the dreaded diverged/detached head state
zzz not sure if I'm getting better at git but I am getting braver
dr|z3d_ you'll be a master before none too long, zzz.
dr|z3d_ here's a git command you might want to have a go at, to prune your remote dead branches: git push origin <deadbranch> --delete
dr|z3d unlike mtn where they tended to stick around forever, in git you can just nuke them. not much point keeping them around when they've been merged with master branch.
zzz yeah, not too concerned about housekeeping yet
dr|z3d can't do much with that iprestrict commit as is, zzz. fetching the commit and then cherry-picking it in a new branch results in the entire codebase being pulled :|
dr|z3d it looks like it's orphaned from the master branch despite having a commit ref.
zzz did the best I could, dunno what I did wrong
dr|z3d it's fine if you just want to keep it attached to the merge request commentry, but if you want to push it to master or another branch you probably want to cherry-pick the commit from whatever branch you've got it in locally from master or a dev branch.
dr|z3d (and then push to the chosen branch)
dr|z3d something like: git checkout master && git checkout iprestrict -b && git cherry-pick -x d3af2af3 (assuming the commit's local)
dr|z3d then you should be able to push the new branch with: git push origin iprestrict
dr|z3d obviously you can do the same on the master branch.
zzz not clear there's any issue within our gitlab if we did merge it, but it's often easier just to close the MR and checkin the diff directly
zzz you can do the same if you want to test
dr|z3d I don't think the issue's with gitlab per se, just the way you've done that commit. It should be showing up as a commit on the master branch, no?
zzz no, it's on top of idk's ip-restrictions branch that he started
dr|z3d and yeah, probably just closing the MR and committing to the master branch would fix.
zzz theres nothing in master yet
zzz and nothing in the zzz or i2p-hackers accounts yet
dr|z3d you'll probably need to remove the draft (WIP prefix) to do so.
dr|z3d maybe there's an issue with the way eyedeekay committed the patch, that's also possible.
dr|z3d I mean, it all looks vaguely ok, but when I attempt to cherry-pick that commit from idk's branch, it wants to pull in every single file from the repo. Could also be something I'm doing wrong.
dr|z3d probably the safest way as you said, zzz, is to close the merge request and just push the commit to master.
eyedeekay git fetch "git@i2pgit.org:idk/i2p.i2p.git" 'ip-restrictions'
eyedeekay git checkout -b 'idk/i2p.i2p-ip-restrictions' FETCH_HEAD
eyedeekay seems to work for me, as long as I don't already have an ip-restrictions branch on the fork I'm using, if I do, then I can just
eyedeekay git pull idk ip-restrictions
eyedeekay with the branch checked out
dr|z3d I was trying to pull your branch on top of the master branch
dr|z3d I also tried creating an orphan branch and pulling to that and then cherry-picking the commit.
dr|z3d let me try git checkout (upstream)master and then git pull your_branch one more time.
dr|z3d or rather checkout master, create a new branch from that, then pull.
zzz ip-restrictions is one rev behind master, so trying to put it on top is what causes divergence
dr|z3d > fatal: refusing to merge unrelated histories
dr|z3d sounds about right, zzz.
dr|z3d let's see if I can checkout an earlier revision and patch that instead.
dr|z3d I could, but hassle. I'll wait for the commit to appear in master :|