the-distro/nix-forgejo
1
1
Fork 0
Code Issues 1 Pull requests Projects Releases Packages Wiki Activity

feat: add package for afnix deployment #4

Merged
emilylange merged 1 commit from forgejo-afnix into main 2025-07-29 18:01:42 +00:00
Conversation 3 Commits 1 Files changed 2 +19
delroth commented 2025-07-29 04:20:56 +00:00
Owner
Copy link

This currently follows the same version as Lix but without the custom
patches / themes / emojis. We might adopt some of the patches later -
for now, starting fresh so we can figure out exactly what we want.

Tested: nix-build . -A packages.forgejo-afnix tests.vm-afnix

This currently follows the same version as Lix but without the custom patches / themes / emojis. We might adopt some of the patches later - for now, starting fresh so we can figure out exactly what we want. Tested: nix-build . -A packages.forgejo-afnix tests.vm-afnix
This currently follows the same version as Lix but without the custom
patches / themes / emojis. We might adopt some of the patches later -
for now, starting fresh so we can figure out exactly what we want.

Tested: nix-build . -A packages.forgejo-afnix tests.vm-afnix
requested review from emilylange 2025-07-29 04:21:12 +00:00
emilylange requested changes 2025-07-29 13:15:29 +00:00
emilylange left a comment
Owner
Copy link

I did not expect you to not want a single patch from here.
Please just use pkgs.forgejo from nixpkgs in your configuration then.
With that you will at least get a cache hit from cache.nixos.org.

I see no value in adding this vanilla version to this repository here, just to open up the possibility of using some of the patches in the future.
We can reopen this when the time for that comes, but until then, please just use pkgs.forgejo from nixpkgs, which I am also a maintainer of.

I did not expect you to not want a single patch from here. Please just use `pkgs.forgejo` from nixpkgs in your configuration then. With that you will at least get a cache hit from cache.nixos.org. I see no value in adding this vanilla version to this repository here, just to open up the possibility of using some of the patches in the future. We can reopen this when the time for that comes, but until then, please just use `pkgs.forgejo` from nixpkgs, which I am also a maintainer of.
delroth force-pushed forgejo-afnix from 1dc9dfada4 to 766c7f2843 2025-07-29 14:38:05 +00:00 Compare
delroth commented 2025-07-29 14:38:54 +00:00
Author
Owner
Copy link

I did not expect you to not want a single patch from here.

No, me neither, and I didn't expect anyone to care about the intermediate state of this repository for the few hours until I figured out which patches from the list we want and which ones we don't want.

We can reopen this when the time for that comes

I've added one which seems like we'll obviously want too given the similarity in the AFNix and Lix setups.

FWIW in the future I'd appreciate if you sought clarification / confirmation instead of straight up rejecting changes in such a way. Like most people, I usually have some justification behind my actions, and if I didn't think we'd need patches reasonably soon with a very high probability I'd indeed have just used pkgs.forgejo. I don't have so much time and hatred for other contributors that I go make random changes to projects if I don't expect to need them.

I appreciate that might and later are imprecise words, but that doesn't mean you have to immediately assume they mean will not and never when the alternate meanings of likely will and once I've gotten the initial changes out of the way are not only equally as valid but also more consistent with the fact that I'm proposing this PR in the first place. how soon? and do you really need none of the patches are two questions I could have trivially answered, leaving us probably both significantly happier with the rest of our day.

> I did not expect you to not want a single patch from here. No, me neither, and I didn't expect anyone to care about the intermediate state of this repository for the few hours until I figured out which patches from the list we want and which ones we don't want. > We can reopen this when the time for that comes I've added one which seems like we'll obviously want too given the similarity in the AFNix and Lix setups. ---- FWIW in the future I'd appreciate if you sought clarification / confirmation instead of straight up rejecting changes in such a way. Like most people, I usually have some justification behind my actions, and if I didn't think we'd need patches reasonably soon with a very high probability I'd indeed have just used `pkgs.forgejo`. I don't have so much time and hatred for other contributors that I go make random changes to projects if I don't expect to need them. I appreciate that `might` and `later` are imprecise words, but that doesn't mean you have to immediately assume they mean `will not` and `never` when the alternate meanings of `likely will` and `once I've gotten the initial changes out of the way` are not only equally as valid but also more consistent with the fact that I'm proposing this PR in the first place. `how soon?` and `do you really need none of the patches` are two questions I could have trivially answered, leaving us probably both significantly happier with the rest of our day.
requested review from emilylange 2025-07-29 14:40:51 +00:00
emilylange commented 2025-07-29 17:52:15 +00:00
Owner
Copy link

I would have appreciated some communication leading up to this. We share lots of rooms on Matrix where we could have coordinated this in more detail.

I will try to acknowledge at least some of your points and share my side of things. A priority of mine, if you look at past conversations, PRs and probably too many private DM chats, is to unblock or work around a given thing as soon as possible until I have more time for the given thing.

For example #3 had a bunch of communication outside that PR. Notably, I asked how urgent this is, provided multiple immediate workarounds to substitute that broken FOD, and shared how long I thought it would take me to have to time to review that.
At the time, #3 was blocking https://git.lix.systems/the-distro/infra/pulls/238.

Compare that to how I viewed and experienced your PR.

You pinged and asked me about Forgejo two weeks ago in a mutual room, I responded and offered to add a package variant for you if you tell me what you want and need.

Fast-forward two weeks, you appear to have made up your mind, have a vision, and opened this PR without a heads-up and all I see is this resulting in more work for me in this repository, for something that could be pkgs.forgejo until at least a single patch is used.

You need to realize that adding more package variants comes with a real cost.
And I read "later" in this context as weeks, potentially months, as that is how I experienced the actual meaning of it over the last year or so here.
So with that and without knowing where you stand nor a pointer of how much you actually looked into everything surrounding Forgejo here, I reviewed this the way I did.

  • I pointed you to pkgs.forgejo to unblock https://git.lix.systems/the-distro/infra/pulls/242, in case you were somehow laser-focused thinking you had to use nix-forgejo with the module in infra. Not because I think you are stupid, but because some things are that way.
  • I mentioned why pkgs.forgejo is better for you as consumer, in that it is at the very least on cache.nixos.org, while also hinting that it is equally well maintained.
  • I made clear I don't want this vanilla package, but open to revisit this once you picked a patch.
  • I did not provide insights into my reasoning, which I probably should have.
  • I did not ask for clarifications on the timeline, which would have been good.
  • I did not suggest patches to you, because I thought you made up your mind for at least the next few weeks and what felt to me like you wanting to do this alone, not sharing where you are going with this.

What particularly bothers me is that I did not ask for clarification on the timeline. But there are many things that led to up this and that could have prevented it.

I do stand by the fact that I only see downsides weighing out the upsides of adding this as a vanilla variant, even as a first step, to then maybe pick patches weeks or months later, because of cost and overhead.

And since you mentioned it, I do genuinely care a lot about the state of this repository.

I personally would approach this nix code differently to reduce some aspects of the inevitable overhead, but that would likely result in a lot of back and forth, thus even more delay, so let's not.

And finally, I hope we can get along with fewer misunderstandings in the future, as we probably learn to understand each other some more in the future. I don't think this will be the last instance of something like this. And I certainly feel like I have an image of you that is that does not quite match reality.

I would have appreciated some communication leading up to this. We share lots of rooms on Matrix where we could have coordinated this in more detail. I will try to acknowledge at least some of your points and share my side of things. A priority of mine, if you look at past conversations, PRs and probably too many private DM chats, is to unblock or work around a given thing as soon as possible until I have more time for the given thing. For example #3 had a bunch of communication outside that PR. Notably, I asked how urgent this is, provided multiple immediate workarounds to substitute that broken FOD, and shared how long I thought it would take me to have to time to review that. At the time, #3 was blocking <https://git.lix.systems/the-distro/infra/pulls/238>. Compare that to how I viewed and experienced your PR. You pinged and asked me about Forgejo two weeks ago in a mutual room, I responded and offered to add a package variant for you if you tell me what you want and need. Fast-forward two weeks, you appear to have made up your mind, have a vision, and opened this PR without a heads-up and all I see is this resulting in more work for me in this repository, for something that could be `pkgs.forgejo` until at least a single patch is used. You need to realize that adding more package variants comes with a real cost. And I read "later" in this context as weeks, potentially months, as that is how I experienced the actual meaning of it over the last year or so here. So with that and without knowing where you stand nor a pointer of how much you actually looked into everything surrounding Forgejo here, I reviewed this the way I did. - I pointed you to `pkgs.forgejo` to unblock <https://git.lix.systems/the-distro/infra/pulls/242>, in case you were somehow laser-focused thinking you had to use `nix-forgejo` with the module in `infra`. Not because I think you are stupid, but because some things are that way. - I mentioned why `pkgs.forgejo` is better for you as consumer, in that it is at the very least on cache.nixos.org, while also hinting that it is equally well maintained. - I made clear I don't want this vanilla package, but open to revisit this once you picked a patch. - I did not provide insights into my reasoning, which I probably should have. - I did not ask for clarifications on the timeline, which would have been good. - I did not suggest patches to you, because I thought you made up your mind for at least the next few weeks and what felt to me like you wanting to do this alone, not sharing where you are going with this. What particularly bothers me is that I did not ask for clarification on the timeline. But there are many things that led to up this and that could have prevented it. I do stand by the fact that I only see downsides weighing out the upsides of adding this as a vanilla variant, even as a first step, to then maybe pick patches weeks or months later, because of cost and overhead. And since you mentioned it, I do genuinely care a lot about the state of this repository. I personally would approach this nix code differently to reduce some aspects of the inevitable overhead, but that would likely result in a lot of back and forth, thus even more delay, so let's not. And finally, I hope we can get along with fewer misunderstandings in the future, as we probably learn to understand each other some more in the future. I don't think this will be the last instance of something like this. And I certainly feel like I have an image of you that is that does not quite match reality.
👍 1
emilylange deleted branch forgejo-afnix 2025-07-29 18:01:42 +00:00
Sign in to join this conversation.
Reviewers
No reviewers
emilylange
Labels
Clear labels
No items
No labels
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
2 participants
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference: the-distro/nix-forgejo#4
No description provided.
Delete branch "forgejo-afnix"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?