<discocaml>
<octachron> @yawaramin , to answer your initial question on the ocaml repository, as far as I see the choice was mostly to not lose valuable history by default. Note that people and me use merge-squash when the pull request maps to a single well-identified feature while the commit history meandered during the review and the PR author doesn't wish to clean up the history.
alfiee has quit [Ping timeout: 252 seconds]
<dh`>
octachron: yawaramin apparently doesn't accept the idea of history being valuable, or something
alfiee has joined #ocaml
alfiee has quit [Ping timeout: 272 seconds]
adrien_ is now known as adrien
mal`` has quit [Read error: Connection reset by peer]
dhil has joined #ocaml
alfiee has joined #ocaml
infinity0 has quit [Ping timeout: 260 seconds]
alfiee has quit [Ping timeout: 268 seconds]
infinity0 has joined #ocaml
mal`` has joined #ocaml
alfiee has joined #ocaml
alfiee has quit [Ping timeout: 244 seconds]
cross has joined #ocaml
adrien has quit [Ping timeout: 260 seconds]
alfiee has joined #ocaml
alfiee has quit [Ping timeout: 246 seconds]
Putonlalla has quit [Ping timeout: 252 seconds]
<discocaml>
<gooby_diatonic> There is a widespread sentiment against git merges apparently
adrien has joined #ocaml
Haudegen has quit [Quit: Bin weg.]
alfiee has joined #ocaml
<discocaml>
<contificate> yeah, I'm no fan
alfiee has quit [Ping timeout: 268 seconds]
j0lol has quit [Remote host closed the connection]
j0lol has joined #ocaml
<discocaml>
<gooby_diatonic> I like them conceptually although I do accept that they complicate the log, usually with no benefit
chiselfuse has quit [Ping timeout: 264 seconds]
chiselfuse has joined #ocaml
j0lol has quit [Remote host closed the connection]
j0lol has joined #ocaml
infinity0 has quit [Ping timeout: 245 seconds]
alfiee has joined #ocaml
alfiee has quit [Ping timeout: 252 seconds]
Haudegen has joined #ocaml
Putonlalla has joined #ocaml
infinity0 has joined #ocaml
trillion_exabyte has quit [Ping timeout: 260 seconds]
trillion_exabyte has joined #ocaml
<discocaml>
<diligentclerk> I have my local machine set to rebase by default but it seems like whenever you make a PR on github the default option is "merge into master?" or "merge master branch?" so I often accidentally merge when I meant to rebase and then I have to do a force push to fix things which is certainly not ideal practice on a public github repo
<discocaml>
<yawaramin> thanks for the answer but it doesn't really make sense imho 🙂 when a merge commit is added on top of the actual authored commits, it is pure noise on top of the signal. i don't think anything valuable would be lost by not adding a merge commit
alfiee has joined #ocaml
trillion_exabyte has quit [Ping timeout: 244 seconds]
<rustyne>
to have a merge commit allows to delimit the series of commits that’s part of the PR
trillion_exabyte has joined #ocaml
Guest52 has joined #ocaml
Guest52 has quit [Quit: Client closed]
<discocaml>
<Joe> shoulda used fossil-sc,
<discocaml>
<Joe> shoulda used fossil-scm
<discocaml>
<Joe> shoulda used fossil-scm.
<discocaml>
<Joe> shoulda used fossil-scm. 😄
<qu1j0t3>
<chorus>
alfiee has joined #ocaml
alfiee has quit [Ping timeout: 244 seconds]
alfiee has joined #ocaml
alfiee has quit [Ping timeout: 244 seconds]
Haudegen has quit [Quit: Bin weg.]
alfiee has joined #ocaml
alfiee has quit [Ping timeout: 260 seconds]
bartholin has joined #ocaml
Haudegen has joined #ocaml
<dh`>
extra vacuous commits when merging pull requests are a github problem
<dh`>
if you don't like it, don't use pull requests
alfiee has joined #ocaml
alfiee has quit [Ping timeout: 245 seconds]
<discocaml>
<gooby_diatonic> You mean Github creates merge commit as well as its own PR commit or
<dh`>
GH has three settings for merging pull requests: squash, rebase, and merge
<dh`>
in all cases you get a GH-generated commit at the head of the trunk afterwards
<dh`>
in the squash case, that has all the changes in it and you lose the original detailed history (well, mostly, it's still in the pull request history)
<dh`>
in the rebase case, it's AFAIK always vacuous
<dh`>
in the merge case, if your branch isn't up to date relative to the trunk it's a merge commit with the prior head of the trunk
<dh`>
if your branch _is_ up to date, it's vacuous
<dh`>
there's a certain tidiness to merge-squash but it's really only a good idea for projects in minor maintenance mode where the changes are small and one at a time
<dh`>
and as far as merge commits making a mess, that's a git bug, you can always use something else :-)
<dh`>
(actually tbh it's a whole family of git bugs)
alfiee has joined #ocaml
alfiee has quit [Ping timeout: 272 seconds]
Haudegen has quit [Quit: No Ping reply in 180 seconds.]
Anarchos has quit [Quit: Vision[]: i've been blurred!]
Anarchos has joined #ocaml
<discocaml>
<contificate> when not working with others, I truly believe the correct way is to rebase and then `git reset --hard` to update the `main` branch, then force push, prefer a linear history of me staging changes, not some awful merge nonsense github wants
<discocaml>
<contificate> when working with others, whatever minimises the tears - but please squash fixup commits, I always address review comments using `git rebase -i`
alfiee has joined #ocaml
<dh`>
working by yourself there's no reason to bother with pull requests at all
<discocaml>
<contificate> yeah, hence my reset-based workflow 🙏
<discocaml>
<contificate> just don't like github interspersing its own verified commits
<dh`>
I just commit to master
<discocaml>
<contificate> fair, but I do like maintaining separate branches because I check them out using `git worktree`
<dh`>
no need to frob anything except the occasional git rebase -i origin/master to fix up goofs
Tuplanolla has joined #ocaml
<dh`>
or, I don't use git for things that don't need to hang out on github
alfiee has quit [Ping timeout: 244 seconds]
<discocaml>
<shawnfrostx> i use sapling it’s pretty good for solo development
<discocaml>
<contificate> if you haven't got a `git.` subdomain hosting a cgit instance, you're probably ngmi
<discocaml>
<shawnfrostx> + sourcehut instead of gh
<discocaml>
<shawnfrostx> havent had a problem yet and if i really need sth to be on github it’s easy to get sth auto pushed from sourcehut to gh
<discocaml>
<contificate> github is really a portfolio website more than anything else
<discocaml>
<shawnfrostx> one time i came across a guy who had some script that added and removed a new line on alternating days so he could get his commit calendar all green
infohazards has quit [Remote host closed the connection]
<discocaml>
<contificate> think you can do that retrospectively, no?
<discocaml>
<shawnfrostx> could be. if i wind up on that same gh profile somehow i’ll see if he’s still doing it
infohazards has joined #ocaml
infohazards has quit [Remote host closed the connection]
infohazards has joined #ocaml
infohazards has quit [Remote host closed the connection]
<companion_cube>
Alright, I really like caqti now
infohazards has joined #ocaml
infohazards has quit [Remote host closed the connection]
infohazards has joined #ocaml
Anarchos has quit [Quit: Vision[]: i've been blurred!]
alfiee has joined #ocaml
alfiee has quit [Ping timeout: 244 seconds]
chrisz has quit [Ping timeout: 268 seconds]
chrisz has joined #ocaml
alfiee has joined #ocaml
Anarchos has joined #ocaml
alfiee has quit [Ping timeout: 272 seconds]
ygrek has quit [Remote host closed the connection]
malte has quit [Ping timeout: 252 seconds]
infohazards has quit [Remote host closed the connection]
infohazards has joined #ocaml
alfiee has joined #ocaml
Serpent7776 has quit [Ping timeout: 252 seconds]
dhil has quit [Ping timeout: 268 seconds]
malte has joined #ocaml
malte has quit [Ping timeout: 260 seconds]
malte has joined #ocaml
bartholin has quit [Remote host closed the connection]
<Anarchos>
I am running the testsuite of ocaml on haikuOS (with native compilation activated) and i run into this strange error : seems a parallel random error in double formatting !!