Conversation

Notices

1. Embed this notice
   :umu: :umu: (a1ba@suya.place)'s status on Saturday, 25-May-2024 10:52:41 JST :umu: :umu:
   how to send a oneline patch on GitLab through web interface
   
   1. Open file you want to edit on GitLab.
   2. Wait until all JavaScript bloatware loads.
   3. Find out that GitLab logged you out.
   4. Press Sign in button.
   5. Wait until Cloudflare checks your belonging to human race.
   6. Enter credentials.
   7. Wait until Cloudflare checks your belonging to human race AGAIN.
   8. Edit the file.
   9. Fill out PR.
   10. Find out that GitLab instead of commiting your changes on top of upstream branch, committed them on top of 3 year old fork master branch, mixing-in some old and unrelated commits to the PR.
   11. Remove your fork.
   12. Edit the file AGAIN.
   13. Fill out PR.
   
   I know GitLab is Microsoft and therefore bad and evil but GitLab isn't any better.
   
   Also I knew I had to do this through CLI, I thought it would be fun to do it slightly unusual way and oh boy it was. :)
   • Haelwenn /элвэн/ :triskell: likes this.
   • Embed this notice
     Haelwenn /элвэн/ :triskell: (lanodan@queer.hacktivis.me)'s status on Saturday, 25-May-2024 10:54:29 JST Haelwenn /элвэн/ :triskell:

     in reply to
     @a1ba s;I know GitLab is Microsoft;I know GitHub is Microsoft;
     
     And the worst is glab (gitlab-cli) is all kinds of broken as well, like it ends up creating branchless MRs if you didn't push beforehand.
     Doughnut Lollipop 【記録係】:blobfoxgooglymlem: likes this.
   • Embed this notice
     Haelwenn /элвэн/ :triskell: (lanodan@queer.hacktivis.me)'s status on Saturday, 25-May-2024 11:00:31 JST Haelwenn /элвэн/ :triskell:

     in reply to

     • Haelwenn /элвэн/ :triskell:
     @a1ba Also I love how gitlab fucked up the diff at some point on https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/63007 so it tagged all the previously merged MRs.
     And even as I fixed it right after, the links are still active, so I'm receiving noisy notifications, even today…
   • Embed this notice
     Haelwenn /элвэн/ :triskell: (lanodan@queer.hacktivis.me)'s status on Saturday, 25-May-2024 11:08:18 JST Haelwenn /элвэн/ :triskell:

     in reply to
     @a1ba I think it was because I did a bad rebase or something… kind of a mess to see the difference in terms of commits.
     
     https://gitlab.alpinelinux.org/lanodan/aports/-/commits/60cb0600c4f1d1e2fe2362fdaf6020d5ade82108 v1
     https://gitlab.alpinelinux.org/lanodan/aports/-/commits/5decb36b0400c535ee1c6cfce499c8e8e2ea3a7e v2
   • Embed this notice
     :umu: :umu: (a1ba@suya.place)'s status on Saturday, 25-May-2024 11:08:19 JST :umu: :umu:

     in reply to

     • Haelwenn /элвэн/ :triskell:
     @lanodan lol wut
   • Embed this notice
     Haelwenn /элвэн/ :triskell: (lanodan@queer.hacktivis.me)'s status on Saturday, 25-May-2024 11:15:27 JST Haelwenn /элвэн/ :triskell:

     in reply to

     • Haelwenn /элвэн/ :triskell:
     @a1ba At least I would say this is worse than the time I accidentally sent like a dozen of emails instead of one because OpenSMTPd sendmail(1) (the de-facto standard command) fails to properly distinguish between a temporary failure and a permanent one.
     So what I thought wasn't in the queue… in fact was.
   • Embed this notice
     Haelwenn /элвэн/ :triskell: (lanodan@queer.hacktivis.me)'s status on Saturday, 25-May-2024 11:18:35 JST Haelwenn /элвэн/ :triskell:

     in reply to
     @a1ba At least this time the reason it was sent multiple time was between the keyboard and the chair, I should have checked the queue before re-sending.
   • Embed this notice
     :umu: :umu: (a1ba@suya.place)'s status on Saturday, 25-May-2024 11:18:36 JST :umu: :umu:

     in reply to

     • Haelwenn /элвэн/ :triskell:
     @lanodan classic network issue when data was sent multiple times because of dubious error handling somewhere deep in the code
   • Embed this notice
     Haelwenn /элвэн/ :triskell: (lanodan@queer.hacktivis.me)'s status on Saturday, 25-May-2024 11:21:11 JST Haelwenn /элвэн/ :triskell:

     in reply to

     • 1A 🢒 λC
     @iacore @a1ba You know how an SDK is like re-providing at least half of an operating system?
     Forges re-provide the other half. ^^
   • Embed this notice
     1A 🢒 λC (iacore@mastodon.de)'s status on Saturday, 25-May-2024 11:21:12 JST 1A 🢒 λC

     in reply to

     • Haelwenn /элвэн/ :triskell:

     @lanodan @a1ba why does gitlab have so many features why

   • Embed this notice
     Haelwenn /элвэн/ :triskell: (lanodan@queer.hacktivis.me)'s status on Saturday, 25-May-2024 11:29:35 JST Haelwenn /элвэн/ :triskell:

     in reply to

     • 1A 🢒 λC
     @iacore @a1ba I forgot if GitLab has this but I think it can provide a dev environment now…
     
     Also I've used SailfishOS sdk in gitlab-ci, because I didn't want that crap on my machines, so dev setup on phone running SailfishOS (integrated SSH FTW), prod done in docker.
     https://git.pleroma.social/pleroma/kyclos/-/blob/develop/.gitlab-ci.yml?ref_type=heads
   • Embed this notice
     1A 🢒 λC (iacore@mastodon.de)'s status on Saturday, 25-May-2024 11:29:36 JST 1A 🢒 λC

     in reply to

     • Haelwenn /элвэн/ :triskell:

     @lanodan @a1ba can we have an (SDK name here)-Gitlab OS?