Commit graph

13905 commits

Author SHA1 Message Date
Alexander Bantyev 2fc3a15861
Make nix-env less prominent in manual TOC 2023-03-01 13:54:45 +04:00
Bob van der Linden 12538605fd
nix-profile: add FIXME about using C++20 std::ranges 2023-03-01 07:40:53 +01:00
Cole Helbling 990fbb8fef WIP: pin cachix actions to 2.13.3 to see if CI will pass 2023-02-28 16:19:11 -08:00
Cole Helbling 2b801d6e3c Check if NIX_LINK_NEW exists instead of checking that NIX_LINK doesn't exist
For brand new installations, neither NIX_LINK_NEW
(`$XDG_STATE_HOME/nix/profile` or `~/.local/state/nix/profile`), nor
NIX_LINK (`~/.nix-profile`) will exist.

This restores functionality to nix-env, which is relied upon by GitHub
Actions such as https://github.com/cachix/cachix-action and the Nixpkgs
EditorConfig (and other) CI.
2023-02-28 16:10:42 -08:00
Matthew Kenigsberg f86f2b973f Always set valid in path-info --json output
Currently the valid key is only present when the path is invalid, which
makes checking path validity more complex than it should be. With this
change, the valid key can always be used to check if a path is valid
2023-02-28 16:04:41 -07:00
Valentin Gagarin d5af43cb42
Merge pull request #7600 from obsidiansystems/explicit-drv-ness 2023-02-28 23:48:17 +01:00
John Ericson ea0adfc582 Get rid of .drv special-casing for store path installables
The release notes document the change in behavior, I don't include it
here so there is no risk to it getting out of sync.

> Motivation

>> Plumbing CLI should be simple

Store derivation installations are intended as "plumbing": very simple
utilities for advanced users and scripts, and not what regular users
interact with. (Similarly, regular Git users will use branch and tag
names not explicit hashes for most things.)

The plumbing CLI should prize simplicity over convenience; that is its
raison d'etre. If the user provides a path, we should treat it the same
way not caring what sort of path it is.

>> Scripting

This is especially important for the scripting use-case. when arbitrary
paths are sent to e.g. `nix copy` and the script author wants consistent
behavior regardless of what those store paths are. Otherwise the script
author needs to be careful to filter out `.drv` ones, and then run `nix
copy` again with those paths and `--derivation`. That is not good!

>> Surprisingly low impact

Only two lines in the tests need changing, showing that the impact of
this is pretty light.

Many command, like `nix log` will continue to work with just the
derivation passed as before. This because we used to:

- Special case the drv path and replace it with it's outputs (what this
  gets rid of).

- Turn those output path *back* into the original drv path.

Now we just skip that entire round trip!

> Context

Issue #7261 lays out a broader vision for getting rid of `--derivation`,
and has this as one of its dependencies. But we can do this with or
without that.

`Installable::toDerivations` is changed to handle the case of a
`DerivedPath::Opaque` ending in `.drv`, which is new: it simply doesn't
need to do any extra work in that case. On this basis, commands like
`nix {show-derivation,log} /nix/store/...-foo.drv` still work as before,
as described above.

When testing older daemons, the post-build-hook will be run against the
old CLI, so we need the old version of the post-build-hook to support
that use-case.

Co-authored-by: Travis A. Everett <travis.a.everett@gmail.com>
Co-authored-by: Valentin Gagarin <valentin.gagarin@tweag.io>
2023-02-28 17:07:05 -05:00
Timothy DeHerrera 639659dec2
doc/manual: add release note for --stdin flag 2023-02-28 12:31:05 -07:00
Timothy DeHerrera df643051e2
nix-store: read paths from standard input
Resolves #7437 for new `nix-store` by adding a `--stdin` flag.
2023-02-28 12:29:16 -07:00
Timothy DeHerrera 269caa5317
feat: read installable paths from stdin
Resolves #7437 for new `nix` commands only by adding a `--stdin` flag.

If paths are also passed on the cli they will be combined with the ones
from standard input.
2023-02-28 12:29:15 -07:00
Eelco Dolstra db14e1d4ae
Merge pull request #7920 from cole-h/fix-profile-XDG_STATE_HOME-unbound
nix-profile{,-daemon}.sh.in: Allow XDG_STATE_HOME to be unset
2023-02-28 19:49:04 +01:00
Cole Helbling 24eaa086f0 nix-profile{,-daemon}.sh.in: Allow XDG_STATE_HOME to be unset
One of our CI machines installs Nix via the official script and then
sources the nix-profile.sh script to setup the environment. However, it
doesn't have XDG_STATE_HOME set, which causes sourcing the script to
fail.
2023-02-28 10:10:20 -08:00
Robert Hensing 4489def1b3
Merge pull request #7883 from hercules-ci/json-guidelines
cli-guideline: Add JSON guideline
2023-02-28 17:14:42 +01:00
Robert Hensing d0d0b9a748 doc/cli-guideline: Improve examples
Turns out that the settings themselves have a bad data model anyway, so we cut that. They do still occur in the first example, but not in focus.
2023-02-28 16:35:47 +01:00
Robert Hensing 17f70b10bf doc/cli-guideline: Apply suggestions from code review
Thanks Valentin!
2023-02-28 16:35:42 +01:00
Robert Hensing f09ccd8ea9 doc/cli-guideline: Add JSON guideline 2023-02-28 16:35:36 +01:00
Robert Hensing 892d46adbb
flake.nix: Force the ./configure tests setting
This always forces the setting, rather than relying on its default, and cleans up the code a bit.

Co-authored-by: John Ericson <git@JohnEricson.me>
2023-02-28 15:56:46 +01:00
Eelco Dolstra b799425c4a
Merge pull request #7916 from NixOS/release-notes
2.14 release notes
2023-02-28 14:21:42 +01:00
Eelco Dolstra 1e07102937 Release notes 2023-02-28 13:44:14 +01:00
Eelco Dolstra a4a5d828e2
Merge pull request #7793 from layus/interrupt_downloads
Check interrupts even when download stalled
2023-02-28 13:29:29 +01:00
Eelco Dolstra da1f49c4e3
Merge pull request #7904 from anatol/patch-1
Example uses gitlab.com thus clarify the comment
2023-02-28 13:29:10 +01:00
Bob van der Linden 0167862e8e
fixup! tests: nix-profile: test install error message upon conflicting files 2023-02-28 10:53:42 +01:00
Bob van der Linden 3efa476c54
tests: nix-profile: test install error message upon conflicting files 2023-02-28 09:28:05 +01:00
Bob van der Linden 872cdb4346
nix-profile-install: show helpful error upon package conflict
Whenever a file conflict happens during "nix profile install" an error
is shown that was previously thrown inside builtins.buildEnv.

We catch BuildProfileConflictError here so that we can provide the user
with more useful instructions on what to do next.

Most notably, we give the user concrete commands to use with all
parameters  already filled in. This avoids the need for the user to look
up these commands in manual pages.
2023-02-28 09:28:05 +01:00
Eelco Dolstra 3d15dbadc2
Merge pull request #7911 from edolstra/revert-7689
Revert #7689
2023-02-28 08:46:55 +01:00
Eelco Dolstra 85a68a5580
Merge pull request #7915 from NixOS/dependabot/github_actions/zeebe-io/backport-action-1.2.0
Bump zeebe-io/backport-action from 1.1.0 to 1.2.0
2023-02-28 08:45:48 +01:00
dependabot[bot] c403303a9a
Bump zeebe-io/backport-action from 1.1.0 to 1.2.0
Bumps [zeebe-io/backport-action](https://github.com/zeebe-io/backport-action) from 1.1.0 to 1.2.0.
- [Release notes](https://github.com/zeebe-io/backport-action/releases)
- [Commits](https://github.com/zeebe-io/backport-action/compare/v1.1.0...v1.2.0)

---
updated-dependencies:
- dependency-name: zeebe-io/backport-action
  dependency-type: direct:production
  update-type: version-update:semver-minor
...

Signed-off-by: dependabot[bot] <support@github.com>
2023-02-27 22:57:30 +00:00
Bob van der Linden 3113b13df9
buildenv: throw BuildEnvFileConflictError with more context
At the moment an Error is thrown that only holds an error message
regarding `nix-env` and `nix profile`. These tools make use of
builtins.buildEnv, but buildEnv is also used in other places. These
places are unrelated to Nix profiles, so the error shouldn't mention
these tools.

This generic error is now BuildEnvFileConflictError, which holds more
contextual information about the files that were conflicting while
building the environment.
2023-02-27 21:39:34 +01:00
Théophane Hufschmitt eae89aca1b
Merge pull request #7776 from yorickvP/fix-path-escapes-7707
Properly escape local paths into URLs in fetchTree
2023-02-27 21:10:25 +01:00
Eelco Dolstra f08ad5bdba
Merge pull request #7913 from fricklerhandwerk/master
add information on the `build-hook` setting
2023-02-27 17:33:25 +01:00
Valentin Gagarin fd0e21475c add information on the build-hook setting
add a warning that you probably never want to change this.
2023-02-27 16:27:56 +01:00
Yorick ae5082bbba
tests/fetchGit: add regression test for #7707 2023-02-27 15:33:54 +01:00
Yorick 2c0866fc3f
fetchTree: convert fs path to url via ParsedURL::to_string 2023-02-27 15:30:04 +01:00
Yorick 0844856c84
url: make percentEncode stricter, expose and unit test it 2023-02-27 15:30:00 +01:00
Eelco Dolstra dd93c12c6a Revert "getDefaultNixPath: actually respect {restrict,pure}-eval"
This reverts commit 1cba5984a6.
2023-02-27 15:11:36 +01:00
Eelco Dolstra e928c72cf9 Revert "Document default nix-path value"
This reverts commit dba9173a1d.
2023-02-27 14:16:49 +01:00
Valentin Gagarin 1b49e6fea9
use the term reference correctly 2023-02-27 13:04:54 +01:00
Valentin Gagarin b0c23999d2
add link to definition 2023-02-27 12:56:22 +01:00
Valentin Gagarin ec317949f9
remove "references" as distinct term 2023-02-27 12:55:14 +01:00
Théophane Hufschmitt 995bfeef3b
Merge pull request #7796 from hercules-ci/fix-7263
Ensure that `self.outPath == ./.`
2023-02-27 10:26:02 +01:00
Robert Hensing e76619a402 rl-next: Describe fixed flake outPath semantics
The reference documentation already implies the correct semantics.
2023-02-26 14:41:23 +01:00
Anatol Pomozov d731235f6b
Example uses gitlab.com thus clarify the comment 2023-02-24 07:17:47 -08:00
Valentin Gagarin 164a5e2629 incorporate feedback from maintainers 2023-02-24 15:05:31 +01:00
Robert Hensing c6051cac6f doc: Add test dependencies to prerequisites 2023-02-24 09:59:25 +01:00
Robert Hensing 8648ebc2cc Add ./configure --disable-tests option
Building without tests is useful for bootstrapping with a smaller footprint
or running the tests in a separate derivation. Otherwise, we do compile and
run them.

This isn't fine grained as to allow picking `check` but not `installcheck`
or vice versa, but it's good enough for now.

I've tried to use Nixpkgs' `checkInputs`, but those inputs weren't discovered
properly by the configure script. We can emulate its behavior very well though.
2023-02-24 09:50:21 +01:00
Valentin Gagarin a556143dc1 add objectives and tasks for Nix team
so far there were no even remotely measurable objectives, only a general
purpose statement.

this change is intended to focus the team's work on what I (and many
others I talked to) perceive to be the main pain point in the
development process.

Co-authored-by: solene.rapenne@tweag.io
2023-02-24 09:50:06 +01:00
Eelco Dolstra 92611e6e4c
Merge pull request #7412 from edolstra/release-process
Document the release process
2023-02-24 09:13:46 +01:00
Robert Hensing b5bbf14b87
Merge pull request #5753 from obsidiansystems/init-twice-works
Clean up daemon handling in the tests
2023-02-23 18:13:14 +01:00
John Ericson 87da941348 Clean up daemon handling
Split `common.sh` into the vars and functions definitions vs starting
the daemon (and possibly other initialization logic). This way,
`init.sh` can just `source` the former. Trying to start the daemon
before `nix.conf` is written will fail because `nix daemon` requires
`--experimental-features 'nix-command'`.

`killDaemon` is idempotent, so it's safe to call when no daemon is
running.

`startDaemon` and `killDaemon` use the PID (which is now exported to
subshells) to decide whether there is work to be done, rather than
`NIX_REMOTE`, which might conceivably be set differently even if a
daemon is running.

`startDaemon` and `killDaemon` can save/restore the old `NIX_REMOTE` as
`NIX_REMOTE_OLD`.

`init.sh` kills daemon before deleting everything (including the daemon
socket).
2023-02-23 11:31:44 -05:00
John Ericson 5dbbf23332 Make init.sh safe to run twice
`init.sh` is tested on its own. We used to do that. I deleted it in
4720853129 but I am not sure why. Better
to just restore it; at one point working on this every other test
passed, so seems good to check whether `init.sh` can be run twice.

We don't *need* to run `init.sh` twice, but I want to try to make our
tests as robust as possible so that manual debugging (where tests for
better or worse might be run ways that we didn't expect) is less
fragile.
2023-02-23 11:31:44 -05:00