Adds support for parsing multiple netencode values from stdin.
This is overly complicated for my tastes, but I don’t see a better way
of writing this logic that does not read all of stdin before starting
to parse the first value.
A kingdom for a conduit.
Change-Id: Ia4f849d4096c43e887756b756d2a85d7f9cd380a
Reviewed-on: https://cl.tvl.fyi/c/depot/+/6631
Autosubmit: Profpatsch <mail@profpatsch.de>
Reviewed-by: Profpatsch <mail@profpatsch.de>
Tested-by: BuildkiteCI
* //users/sterni/emacs: fix for the bqn-mode issue is in channels now.
* //third_party/buzz: pin openssl to 1.1, as nixpkgs now defaults to 3.0
Change-Id: I4b8410cbeb2d6ac210789b0b5687209d50e6572e
Reviewed-on: https://cl.tvl.fyi/c/depot/+/6628
Tested-by: BuildkiteCI
Reviewed-by: sterni <sternenseemann@systemli.org>
Reviewed-by: grfn <grfn@gws.fyi>
ical, but smol
to fit in qr
Change-Id: I37f99a20cfc96b85778a097b7c4f70923f026cd4
Reviewed-on: https://cl.tvl.fyi/c/depot/+/6617
Autosubmit: Profpatsch <mail@profpatsch.de>
Tested-by: BuildkiteCI
Reviewed-by: Profpatsch <mail@profpatsch.de>
* //users/wpcarro/emacs: rls was removed from nixpkgs and is hopefully
no longer necessary in this emacs setup.
Change-Id: Ib8371da652c434e74eb67a2030b212a6423a1891
Reviewed-on: https://cl.tvl.fyi/c/depot/+/6577
Autosubmit: sterni <sternenseemann@systemli.org>
Reviewed-by: tazjin <tazjin@tvl.su>
Reviewed-by: wpcarro <wpcarro@gmail.com>
Tested-by: BuildkiteCI
This shortcut is so unbelievably annoying on some keyboards.
Change-Id: I5c96d490fd42aa8422fa48be87079b60af81fcd3
Reviewed-on: https://cl.tvl.fyi/c/depot/+/6527
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
Work machines should prefer `/hadrian` and personal machines should prefer
`/depot`.
Change-Id: I03b605cc5ce233767146d83f14f368863897cf8b
Reviewed-on: https://cl.tvl.fyi/c/depot/+/6368
Tested-by: BuildkiteCI
Reviewed-by: wpcarro <wpcarro@gmail.com>
Autosubmit: wpcarro <wpcarro@gmail.com>
This means it'll no longer run under Xwayland. Requires applying small
fix to bqn-mode which stopped compiling with emacs HEAD.
Change-Id: I998a303a0b98bcd45c1de430462b1069bef8718e
Reviewed-on: https://cl.tvl.fyi/c/depot/+/6203
Autosubmit: sterni <sternenseemann@systemli.org>
Tested-by: BuildkiteCI
Reviewed-by: sterni <sternenseemann@systemli.org>
* //users/Profpatsch/writers: adjust for API change in
makePythonWriter: Since it has become cross-aware, we also
need to provide the buildPackages python set we want to use.
Change-Id: I4a201647281827ad983eb2dda6f6305f1ed824e7
Reviewed-on: https://cl.tvl.fyi/c/depot/+/6196
Tested-by: BuildkiteCI
Autosubmit: sterni <sternenseemann@systemli.org>
Reviewed-by: Profpatsch <mail@profpatsch.de>
There might be a better way to do this using `direnv`'s `use_nix` and shlevy's
`nix-buffer` Elisp package, but I'm doing this for now.
Change-Id: I4246c519f39eea1bde3191762b184b37d30fbeed
Reviewed-on: https://cl.tvl.fyi/c/depot/+/6049
Tested-by: BuildkiteCI
Reviewed-by: wpcarro <wpcarro@gmail.com>
Autosubmit: wpcarro <wpcarro@gmail.com>
This is useful for eglot launching rust-analyzer when I'm outside of a
depot project.
Change-Id: I1fa1dc11e3eabe4bdedbd6389b55411641391c7e
Reviewed-on: https://cl.tvl.fyi/c/depot/+/6176
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
the upstream Nix test suite uses the pattern of having `.nix` and
`.exp` files for input/expected output, and with this shitty function
navigating between them is a lot simpler
Change-Id: I9d91290057521fe1e1599f69fd6b0f35e1b59960
Reviewed-on: https://cl.tvl.fyi/c/depot/+/6058
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
I wonder if there's an Elisp linter that can catch these errors at nix-build
time.
Change-Id: Ib54e1e57e46ef81021fc373ceecab5a729646472
Reviewed-on: https://cl.tvl.fyi/c/depot/+/6048
Reviewed-by: wpcarro <wpcarro@gmail.com>
Autosubmit: wpcarro <wpcarro@gmail.com>
Tested-by: BuildkiteCI
This only somewhat works, and I don't really need it.
Change-Id: I4e78e5fd49ec07e6cb90a3feef8a4c87cb7b7707
Reviewed-on: https://cl.tvl.fyi/c/depot/+/6047
Reviewed-by: wpcarro <wpcarro@gmail.com>
Autosubmit: wpcarro <wpcarro@gmail.com>
Tested-by: BuildkiteCI
CI couldn't catch this because the dependency is discovered at runtime, and
there were no tests consuming `list-concat`. Added the tests and removed the
usage of `-concat`.
Change-Id: I5bcce9aa8170001a9d35c104da09374ab30ffbc5
Reviewed-on: https://cl.tvl.fyi/c/depot/+/6043
Reviewed-by: wpcarro <wpcarro@gmail.com>
Autosubmit: wpcarro <wpcarro@gmail.com>
Tested-by: BuildkiteCI
new functions:
- duplicate
- last
- delete
- wrap
also:
- drop support for `list-head` (in favor of `list-first`)
- add optional arg to first
Change-Id: If3c48d4749a3bc5a853995996fa02a65a4076a10
Reviewed-on: https://cl.tvl.fyi/c/depot/+/6041
Reviewed-by: wpcarro <wpcarro@gmail.com>
Autosubmit: wpcarro <wpcarro@gmail.com>
Tested-by: BuildkiteCI
* //users/grfn/system/home: build rain with Go 1.17, as it fails to
build with Go 1.18 which introduces new compile-time errors
* //3p/nixpkgs: pick ntfy from stable channel as it does not build on
unstable
* //users/sterni/emacs: make sure use-package is available before
org-tracker can be loaded dynamically from $HOME/src. Interestingly
this only became a problem with this channel bump.
Change-Id: Id7d23b66bc3ba0202a01b4f8d670854e1f31175e
Reviewed-on: https://cl.tvl.fyi/c/depot/+/5988
Tested-by: BuildkiteCI
Reviewed-by: sterni <sternenseemann@systemli.org>
Reviewed-by: grfn <grfn@gws.fyi>
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: sterni <sternenseemann@systemli.org>
Let's just roll with the TVL conventions
Change-Id: I1b90421ca7d9ad0d1ebbe922afdc62f80f0d5e97
Reviewed-on: https://cl.tvl.fyi/c/depot/+/6033
Reviewed-by: wpcarro <wpcarro@gmail.com>
Autosubmit: wpcarro <wpcarro@gmail.com>
Tested-by: BuildkiteCI
