| 18 Dec 2025 |
helle (just a stray cat girl) | it is silly and very hardcode brained | 12:13:32 |
helle (just a stray cat girl) | and if you aren't using nixpkgs for various reasons, being able to use libarchive to handle well, any archive extraction within nix itself saves a bunch of extra steps and handling | 12:17:19 |
KFears (burnt out) | Here's my favorite bootstrapping-adjacent back to date: https://git.auxolotl.org/auxolotl/foundation/src/commit/f7d7ed8f507d30e40a1e543207b332a46336c7b9/src/fetchTarball.nix | 12:22:12 |
KFears (burnt out) | (builtins are very silly) | 12:22:35 |
KFears (burnt out) | * Here's my favorite bootstrapping-adjacent hack to date: https://git.auxolotl.org/auxolotl/foundation/src/commit/f7d7ed8f507d30e40a1e543207b332a46336c7b9/src/fetchTarball.nix | 12:23:24 |
helle (just a stray cat girl) | yeah, we need to fix and document a bunch of the builtins..... | 12:23:26 |
0x4fbb09 it/its ⛯✇ΘΔ | oh my god i forgot about that | 12:23:38 |
0x4fbb09 it/its ⛯✇ΘΔ | that's so funny | 12:23:42 |
helle (just a stray cat girl) | I do like the idea of somehow making libfetchers non-blocking, but uh, rearchitecturing that is going to be fun | 12:24:12 |
KFears (burnt out) | In reply to @helle:tacobelllabs.net yeah, we need to fix and document a bunch of the builtins..... I have an open (and unfinished) CL on builtin builders, want to take a look? | 12:24:13 |
helle (just a stray cat girl) | not at the moment, spoons low, discussing things and reading docs is doable, reading code requires keeping too much weird context | 12:24:54 |
helle (just a stray cat girl) | yes, me reading libfetchers source was a little silly | 12:25:36 |
helle (just a stray cat girl) | I've done sillier fixes on Lix while in a bad state honestly, but uh | 12:26:33 |
helle (just a stray cat girl) | trivial doc fixes because I like reading how users interact with stuff, now those I can make in nearly any state | 12:39:00 |
kloenk | the let me see right now as sitting is good right now to eehh devshell needs so long pipeline :p | 14:09:02 |
Rutile (Commentator2.0) feel free to ping | what exactly are these two lines testing? (i esp don't get the first one)
nix-instantiate --strict --eval -E "!((import (fetchTree { type = \"tarball\"; url = \"file://$tarball\"; narHash = \"$hash\"; })) ? submodules)" >&2
nix-instantiate --strict --eval -E "!((import (fetchTree { type = \"tarball\"; url = \"file://$tarball\"; narHash = \"$hash\"; })) ? submodules)" 2>&1 | grep 'true'
| 17:15:28 |
Rutile (Commentator2.0) feel free to ping | cc: ma27 (you added those lines) | 17:18:41 |
ma27 | don't remember adding those, but the first makes sure that there's no 'submodules' key in the attrset from the fetchtree call, no? | 17:21:21 |
Rutile (Commentator2.0) feel free to ping | tarball.sh lines 50 and 51 in a commit called "merge legacy fetchGit-builtin with generic fetchTree-function" | 17:22:57 |
Rutile (Commentator2.0) feel free to ping | am mostly confused as to why the stdout is piped into the stderr, when no check / grep is being done in the first place | 17:23:40 |
Rutile (Commentator2.0) feel free to ping | * tarball.sh lines 50 and 51 in a commit called "merge legacy fetchGit-builtin with generic fetchTree-function" rev f7424384 | 17:24:10 |
ma27 | oh no, sins from my youth🫠 | 17:28:04 |
ma27 | maybe it was just a debig statement to get output that I dodn't removed and passed through review? | 17:28:28 |
Rutile (Commentator2.0) feel free to ping | https://gerrit.lix.systems/c/lix/+/4790
the complaining about TMPDIR not being set and tarball.sh failing have an end! | 17:49:37 |
Rutile (Commentator2.0) feel free to ping | * https://gerrit.lix.systems/c/lix/+/4790
the days of complaining about TMPDIR not being set and tarball.sh failing have an end! | 17:49:45 |
Rutile (Commentator2.0) feel free to ping | https://buildkite.com/lix-project/lix/builds/7143/steps/canvas?jid=019b329a-3fa3-4367-a2bd-8eb464bed0fd
uhhh what? | 18:28:12 |