fediversity/fediversity
13
2
Fork 6
Code Issues 157 Pull requests 29 Projects 1 Activity Actions

unify versioning (minus around nixops, for now) #311

Merged
fricklerhandwerk merged 1 commit from kiara/fediversity:unify-versioning into main 2025-05-05 09:22:17 +02:00
Conversation 9 Commits 1 Files changed 15 +184 -201
kiara commented 2025-04-17 20:35:02 +02:00
Owner
Copy link

closes #279

not sure how to port the nixops4 bits, so that bit i put off to #309.

closes #279 not sure how to port the nixops4 bits, so that bit i put off to #309.
kiara changed title from WIP: unify versioning - not sure how to port nixops4 bits so wait until #309? to WIP: unify versioning 2025-04-18 19:14:32 +02:00
fricklerhandwerk reviewed 2025-04-21 13:58:42 +02:00
flake.nix Outdated
@ -17,0 +2,4 @@
outputs = { self, ... }: let
inherit sources = import ./npins;
inherit (sources) nixpkgs flake-parts nixops4 nixops4-nixos git-hooks;
pkgs = import nixpkgs;
fricklerhandwerk commented 2025-04-21 13:58:42 +02:00
Owner
Copy link

This can't work, you're missing the argument to nixpkgs.

Also to make it unambiguous that you're just taking the pure library, do

lib = import "${nixpkgs}/lib";
This can't work, you're missing the argument to `nixpkgs`. Also to make it unambiguous that you're just taking the pure library, do ```nix lib = import "${nixpkgs}/lib"; ```
kiara marked this conversation as resolved
fricklerhandwerk reviewed 2025-04-21 14:00:34 +02:00
infra/common/resource.nix Outdated
@ -9,6 +8,9 @@ let
inherit (lib) attrValues elem mkDefault;
inherit (lib.attrsets) concatMapAttrs optionalAttrs;
inherit (lib.strings) removeSuffix;
# TODO: use thru nix path after #309
fricklerhandwerk commented 2025-04-21 14:00:34 +02:00
Owner
Copy link

Why would we use nix path anywhere?

Why would we use nix path anywhere?
kiara commented 2025-04-23 15:54:34 +02:00
Author
Owner
Copy link

(fwiw i removed this comment here now)

(fwiw i removed this comment here now)
kiara marked this conversation as resolved
fricklerhandwerk reviewed 2025-04-21 14:02:54 +02:00
infra/common/resource.nix Outdated
@ -10,2 +9,4 @@
inherit (lib.attrsets) concatMapAttrs optionalAttrs;
inherit (lib.strings) removeSuffix;
# TODO: use thru nix path after #309
sources = import ../../npins;
fricklerhandwerk commented 2025-04-21 14:02:54 +02:00
Owner
Copy link

At this point we may set up our own little module system class to pass around sources. But I see this is trying to work around flake-parts, so either we rip out flake-parts or embrace it.

At this point we may set up our own little module system class to pass around `sources`. But I see this is trying to work around flake-parts, so either we rip out flake-parts or embrace it.
kiara commented 2025-04-23 15:55:29 +02:00
Author
Owner
Copy link

i see little point ripping out flake-parts while nixops4 needs flakes as our external interface. it creates a diff moving infra/flake-part.nix into flake.nix, without meaningfully helping address unifying how we manage dependencies.

i see little point ripping out flake-parts while nixops4 needs flakes as our external interface. it creates a diff moving `infra/flake-part.nix` into `flake.nix`, without meaningfully helping address unifying how we manage dependencies.
fricklerhandwerk marked this conversation as resolved
kiara commented 2025-04-21 16:54:01 +02:00
Author
Owner
Copy link

@fricklerhandwerk thanks for your feedback - i'll close in favor of #309, but yeah let's further discuss the dependency management still (as per matrix).

@fricklerhandwerk thanks for your feedback - i'll close in favor of #309, but yeah let's further discuss the dependency management still (as per matrix).
kiara closed this pull request 2025-04-21 16:54:01 +02:00
kiara reopened this pull request 2025-04-23 15:56:02 +02:00
kiara changed title from WIP: unify versioning to unify versioning (minus around nixops, for now) 2025-04-23 15:56:43 +02:00
fricklerhandwerk commented 2025-04-23 16:46:58 +02:00
Owner
Copy link

Looks good generally (we can keep trimming it down later), but I don't understand why we would need multiple shell.nix or .envrc

Looks good generally (we can keep trimming it down later), but I don't understand why we would need multiple `shell.nix` or `.envrc`
kiara commented 2025-04-23 17:04:32 +02:00
Author
Owner
Copy link

@fricklerhandwerk thanks for reviewing!

.envrc i guess prob seems like it should match our number of shells, as any would only help enter 'their' shell i guess.

on the multiple shells, the intent was to reflect decoupling components (and already segregating the proxmox bits to nuke after #116), but fair enough, i deduplicated them again now.

@fricklerhandwerk thanks for reviewing! `.envrc` i guess prob seems like it should match our number of shells, as any would only help enter 'their' shell i guess. on the multiple shells, the intent was to reflect decoupling components (and already segregating the proxmox bits to nuke after #116), but fair enough, i deduplicated them again now.
👍 1
fricklerhandwerk commented 2025-04-23 17:06:58 +02:00
Owner
Copy link

We currently have no need to emphasize composition, and if we need to we can still follow through with the module system idea. Right now it's just a monorepo with stuff in it.

We currently have no need to emphasize composition, and if we need to we can still follow through with the module system idea. Right now it's just a monorepo with stuff in it.
👍 1
fricklerhandwerk reviewed 2025-04-23 17:11:20 +02:00
default.nix Outdated
@ -0,0 +24,4 @@
## Add a directory here if pre-commit hooks shouldn't apply to it.
optout = [
"npins"
".terraform"
fricklerhandwerk commented 2025-04-23 17:07:41 +02:00
Owner
Copy link

Not using that yet

Not using that yet
kiara marked this conversation as resolved
default.nix Outdated
@ -0,0 +41,4 @@
# shell for testing TF directly
shell = pkgs.mkShellNoCC {
inherit (pre-commit-check) shellHook;
buildInputs = pre-commit-check.enabledPackages;
fricklerhandwerk commented 2025-04-23 17:08:08 +02:00
Owner
Copy link

++ [ pkgs.npins ] (we had a schema divergence multiple times now, please always ship npins with the project)

`++ [ pkgs.npins ]` (we had a schema divergence multiple times now, please always ship npins with the project)
kiara marked this conversation as resolved
flake.nix Outdated
@ -54,4 +38,1 @@
shellcheck.enable = true;
};
devShells.default = pkgs.mkShell {
fricklerhandwerk commented 2025-04-23 17:08:52 +02:00
Owner
Copy link

We can just re-export that from default.nix no? Why keep the split?

We can just re-export that from `default.nix` no? Why keep the split?
kiara commented 2025-04-23 20:35:39 +02:00
Author
Owner
Copy link

i didn't know how to get non-flake nixops. i thought we imported flake inputs into default.nix before, yet now i don't even see it anymore.

i didn't know how to get non-flake nixops. i thought we imported flake inputs into `default.nix` before, yet now i don't even see it anymore.
fricklerhandwerk commented 2025-04-25 09:56:42 +02:00
Owner
Copy link

We may have talked about it in a side channel, but this is how one can do it: https://github.com/fricklerhandwerk/flake-inputs (real-world example: bbf2cdaa6b/default.nix (L2-L3))

But in this case I would have thought we wanted it the other way round, and just not use the flake inputs?

We may have talked about it in a side channel, but this is how one can do it: https://github.com/fricklerhandwerk/flake-inputs (real-world example: https://github.com/ngi-nix/ngipkgs/blob/bbf2cdaa6ba3d6c6f576d188df34cc8c508292ab/default.nix#L2-L3) But in this case I would have thought we wanted it the other way round, and just not use the flake inputs?
kiara commented 2025-04-25 19:59:07 +02:00
Author
Owner
Copy link

npins seems harder to use in flakes, in the sense they disallow builtins.currentSystem.
even with flake-inputs, i'm not sure how to proceed given that's aimed at getting inputs, while with nixops4 i don't know how to manually translate these config bits from the flake since they seemingly point nowhere.

npins seems harder to use in flakes, in the sense they disallow `builtins.currentSystem`. even with `flake-inputs`, i'm not sure how to proceed given that's aimed at getting `inputs`, while with nixops4 i don't know how to manually translate [these `config` bits](https://github.com/nixops4/nixops4/blob/main/flake.nix#L32-L35) from the flake since they seemingly point nowhere.
fricklerhandwerk commented 2025-04-29 10:26:02 +02:00
Owner
Copy link

That's the config of the packages submodule from flake-parts. And we don't need builtins.currentSystem -- just pass system from the forAllSystems into that call, such as here: 02ce465c91/flake.nix (L90-L96)

That's the `config` of the `packages` submodule from `flake-parts`. And we don't need `builtins.currentSystem` -- just pass `system` from the `forAllSystems` into that call, such as here: https://github.com/ngi-nix/ngipkgs/blob/02ce465c911a3c61791a6f15be2f6001c33317d9/flake.nix#L90-L96
kiara commented 2025-04-30 15:10:32 +02:00
Author
Owner
Copy link

thanks for the hint about system!

nixops4 tho feels like a pain to decouple from flakes, given the self-referential nature of flakes followed by flake-parts 😭

thanks for the hint about `system`! nixops4 tho feels like a pain to decouple from flakes, given the self-referential nature of flakes followed by `flake-parts` 😭
😕 1
fricklerhandwerk commented 2025-04-30 15:25:52 +02:00
Owner
Copy link

Yeah but that's fine, nixops can keep living in its flake world.

Yeah but that's fine, nixops can keep living in its flake world.
kiara commented 2025-05-01 11:21:52 +02:00
Author
Owner
Copy link

i squashed a version with just nixops flakes-based now

i squashed a version with just nixops flakes-based now
services/.envrc Outdated
@ -0,0 +1,10 @@
#!/usr/bin/env bash
fricklerhandwerk commented 2025-04-23 17:11:11 +02:00
Owner
Copy link

Don't think that file is needed

Don't think that file is needed
kiara marked this conversation as resolved
services/default.nix Outdated
@ -0,0 +1,14 @@
{
fricklerhandwerk commented 2025-04-23 17:10:34 +02:00
Owner
Copy link

Is that imported from anywhere else except the top level? Because if not, we could just leave it with { pkgs }: for now.

Is that imported from anywhere else except the top level? Because if not, we could just leave it with `{ pkgs }:` for now.
kiara marked this conversation as resolved
kiara force-pushed unify-versioning from ad0ee3cff5 to edd35361a1 2025-04-23 20:34:31 +02:00 Compare
kiara force-pushed unify-versioning from b378cc5292 to aec599821f 2025-05-01 11:17:52 +02:00 Compare
requested review from fricklerhandwerk 2025-05-01 11:21:58 +02:00
kiara force-pushed unify-versioning from aec599821f to 56522e6125 2025-05-01 11:45:38 +02:00 Compare
kiara force-pushed unify-versioning from 56522e6125 to f4ebf55a15 2025-05-01 13:24:56 +02:00 Compare
fricklerhandwerk approved these changes 2025-05-05 09:22:06 +02:00
fricklerhandwerk left a comment
Owner
Copy link

Let's keep fixing it up piecemeal

Let's keep fixing it up piecemeal
❤️ 1
fricklerhandwerk deleted branch unify-versioning 2025-05-05 09:22:18 +02:00
Sign in to join this conversation.
Reviewers
No reviewers
fricklerhandwerk
Labels
Clear labels   
0 points

fibonacci complexity: 0 points

  
0.5 points

fibonacci complexity: 0.5 points

  
1 point

fibonacci complexity: 1 point

  
13 points

fibonacci complexity: 13 points

  
2 points

fibonacci complexity: 2 points

  
21 points

fibonacci complexity: 21 points

  
3 points

fibonacci complexity: 3 points

  
34 points

fibonacci complexity: 34 points

  
5 points

fibonacci complexity: 5 points

  
55 points

fibonacci complexity: 55 points

  
8 points

fibonacci complexity: 8 points

  
api service

features that may relate more to a (to be created) API service

  
blocked

   
component: fediversity panel

   
component: nixops4

   
documentation

   
estimation high: >3d

   
estimation low: <2h

   
estimation mid: <8h

   
infinite points

fibonacci complexity: infinite points

  
productisation

features out-scoped with the decoupling of the productisation effort

  
project-management

Project management related tasks

  
question

   
role: application developer

   
role: application operator

   
role: hosting provider

   
role: maintainer

   
security

   
technical debt

   
testing

   
type unclear

   
type: bug

Anything that doesn't work as advertised

  
type: deliverable

   
type: key result

   
type: objective

   
type: task

   
type: user story

   
user experience
No labels
0 points
0.5 points
1 point
13 points
2 points
21 points
3 points
34 points
5 points
55 points
8 points
api service
blocked
component: fediversity panel
component: nixops4
documentation
estimation high: >3d
estimation low: <2h
estimation mid: <8h
infinite points
productisation
project-management
question
role: application developer
role: application operator
role: hosting provider
role: maintainer
security
technical debt
testing
type unclear
type: bug
type: deliverable
type: key result
type: objective
type: task
type: user story
user experience
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.

Blocks
#65 Automated dependency updates
fediversity/fediversity
Reference: fediversity/fediversity#311
No description provided.
Delete branch "kiara/fediversity:unify-versioning"

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?