- This way we improve error messages
on infinite recursion
- Demo:
```nix
let x = builtins.fetchTree x;
in x
```
- Before:
```bash
$ nix-instantiate --extra-experimental-features flakes --strict
error: infinite recursion encountered
```
- After:
```bash
$ nix-instantiate --extra-experimental-features flakes --strict
error: infinite recursion encountered
at /data/github/kamadorueda/nix/test.nix:1:9:
1| let x = builtins.fetchTree x;
| ^
2| in x
```
Mentions: #3505
- This way we improve error messages
on infinite recursion
- Demo:
```nix
let
x = builtins.fetchMercurial x;
in
x
```
- Before:
```bash
$ nix-instantiate --show-trace --strict
error: infinite recursion encountered
```
- After:
```bash
nix-instantiate --show-trace --strict
error: infinite recursion encountered
at /data/github/kamadorueda/test/default.nix:2:7:
1| let
2| x = builtins.fetchMercurial x;
| ^
3| in
```
Mentions: #3505
This reverts some parts of commit
8430a8f086 which was trying to rethrow
some exceptions while we weren’t in the context of a `catch` block,
causing some weird “terminate called without an active exception”
errors.
Fix#5368
We now build the context (so this has the side-effect of making
builtins.{path,filterSource} work on derivations outputs, if IFD is
enabled) and then check that the path has no references (which is what
we really care about).
The boolean is only used to determine if the formals are set to a
non-null pointer in all our cases. We can get rid of that allocation and
instead just compare the pointer value with NULL. Saving up to
sizeof(bool) + platform specific alignment per ExprLambda instace.
Probably not a lot of memory but perhaps a few kilobyte with nixpkgs?
This also gets rid of a potential issue with dereferencing formals based on
the value of the boolean that didn't have to be aligned with the formals
pointer but was in all our cases.
I had started the trend of doing `std::visit` by value (because a type
error once mislead me into thinking that was the only form that
existed). While the optomizer in principle should be able to deal with
extra coppying or extra indirection once the lambdas inlined, sticking
with by reference is the conventional default. I hope this might even
improve performance.
I found it somewhat confusing to have an error like
error: attribute 'getFlake' missing
if the required experimental-feature (`flakes`) is not enabled. Instead,
I'd expect Nix to throw an error just like it's the case when using e.g. `nix
flake` without `flakes` being enabled.
With this change, the error looks like this:
$ nix-instantiate -E 'builtins.getFlake "nixpkgs"'
error: Cannot call 'builtins.getFlake' because experimental Nix feature 'flakes' is disabled. You can enable it via '--extra-experimental-features flakes'.
at «string»:1:1:
1| builtins.getFlake "nixpkgs"
| ^
I didn't use `settings.requireExperimentalFeature` here on purpose
because this doesn't contain a position. Also, it doesn't seem as if we
need to catch the error and check for the missing feature here since
this already happens at evaluation time.
Previously, type or coercion errors for string interpolation, path
interpolation, and plus expressions were always reported at the
beginning of the outer expression. This leads to confusing evaluation
error messages making it hard to accurately diagnose and then fix the
error.
For example, errors were reported as follows.
```
cannot coerce an integer to a string
1| let foo = 7; in "bar" + foo
| ^
cannot add a string to an integer
1| let foo = "bar"; in 4 + foo
| ^
cannot coerce an integer to a string
1| let foo = 7; in "x${foo}"
| ^
```
This commit changes the ExprConcatStrings expression vector to store a
sequence of expressions *and* their expansion locations so that error
locations can be reported accurately. For interpolation, the error is
reported at the beginning of the entire `${foo}`, not at the beginning
of `foo` because I thought this was slightly clearer. The previous
errors are now reported as:
```
cannot coerce an integer to a string
1| let foo = 7; in "bar" + foo
| ^
cannot add a string to an integer
1| let foo = "bar"; in 4 + foo
| ^
cannot coerce an integer to a string
1| let foo = 7; in "x${foo}"
| ^
```
The error is reported at this kind of precise location even for
multi-line indented strings.
This probably helps with at least some of the cases mentioned in #561
Without this, flakes within the same tree and same lock data will have
the same fingerprint and the eval cache for one flake will be
incorrectly used for another.
Alternative to #4639. You can still read flake.lock, but at least in
reproducible workflows like NixOS configurations where you require a
non-dirty tree, evaluation will fail because there is no rev.
With --no-write-lock-file, it's possible that flake.lock is out of
sync with the actual inputs used by the evaluation. So doing fromJSON
(readFile ./flake.lock) will give wrong results.
Fixes#4639.
This fixes a use-after-free bug:
1. s = new EvalState();
2. callFlake()
3. static vCallFlake now references s
4. delete s;
5. s2 = new EvalState();
6. callFlake()
7. static vCallFlake still references s
8. crash
Nix 2.3 did not have a problem with recreating EvalState.
This fixes a class of crashes and introduces ptr<T> to make the
code robust against this failure mode going forward.
Thanks regnat for the idea of a ref<T> without overhead!
Closes#4895Closes#4893Closes#5127Closes#5113
Use `$(libdir)` while installing .pc files looks like a more generic
solution. For example, it will work for distributions like RHEL or
Fedora where .pc files are installed in `/usr/lib64/pkgconfig`.
This extends https://github.com/NixOS/nix/pull/4978 with
supporting the SCP-like urls in expressions like
```nix
builtins.fetchGit {
url = "git@github.com:NixOS/nix.git";
ref = "master";
}
```
It does not operate on a derivation and does not return a
derivation path. Instead it works at the language level,
where a distinct term "package" is more appropriate to
distinguish the parent object of `meta.position`; an
attribute which doesn't even make it into the derivation.
Some people want to avoid using registries at all on their system; Instead
of having to add --no-registries to every command, this commit allows to
set use-registries = false in the config. --no-registries is still allowed
everywhere it was allowed previously, but is now deprecated.
Co-authored-by: Eelco Dolstra <edolstra@gmail.com>
We need to support it for the “old” fetch* functions for backwards
compatibility, but we don’t need it for fetchTree (as it’s a new
function).
Given that changing the `name` messes-up the content hashing, we can
just forbid passing a custom `name` argument to it
Before this commit, nixConfig.flake-registry didn't have any real effect
on the evaluation, since config was applied after inputs were evaluated.
Change this behavior: apply the config in the beginning of flake::lockFile.
Previously, the build system used uname(1) output when it wanted to
check the operating system it was being built for, which meant that it
didn't take into-account cross-compilation when the build and host
operating systems were different.
To fix this, instead of consulting uname output, we consult the host
triple, specifically the third "kernel" part.
For "kernel"s with stable ABIs, like Linux or Cygwin, we can use a
simple ifeq to test whether we're compiling for that system, but for
other platforms, like Darwin, FreeBSD, or Solaris, we have to use a
more complicated check to take into account the version numbers at the
end of the "kernel"s. I couldn't find a way to just strip these
version numbers in GNU Make without shelling out, which would be even
more ugly IMO. Because these checks differ between kernels, and the
patsubst ones are quite fiddly, I've added variables for each host OS
we might want to check to make them easier to reuse.
Linux is (as far as I know) the only mainstream operating system that
requires linking with libdl for dlopen. On BSD, libdl doesn't exist,
so on non-FreeBSD BSDs linking will currently fail. On macOS, it's
apparently just a symlink to libSystem (macOS libc), presumably
present for compatibility with things that assume Linux.
So the right thing to do here is to only add -ldl on Linux, not to add
it for everything that isn't FreeBSD.
As described in #4745 it's otherwise fairly hard to understand where
this is coming from. Say you have an expression which uses e.g.
`types.package`:
``` nix
{ outputs = { self, nixpkgs }: {
packages.x86_64-linux.hello = let
foo = nixpkgs.lib.evalModules {
modules = [
{
options.foo.bar = with nixpkgs.lib; mkOption { type = types.package; };
}
{
foo.bar = ./.;
}
];
};
in builtins.trace foo.config.foo.bar.outPath nixpkgs.legacyPackages.x86_64-linux.hello;
defaultPackage.x86_64-linux = self.packages.x86_64-linux.hello;
};
}
```
Then you'll get an error trace like this:
```
error: 'builtins.storePath' is not allowed in pure evaluation mode
at /nix/store/p4h2x6r80njkb0j2rc1xjhhl99yri3zb-source/lib/attrsets.nix:328:15:
327| let
328| path' = builtins.storePath path;
| ^
329| res =
… while evaluating the attribute 'config.foo.bar.outPath'
at /nix/store/p4h2x6r80njkb0j2rc1xjhhl99yri3zb-source/lib/attrsets.nix:332:11:
331| name = sanitizeDerivationName (builtins.substring 33 (-1) (baseNameOf path'));
332| outPath = path';
| ^
333| outputs = [ "out" ];
… while evaluating the attribute 'packages.x86_64-linux.hello'
at /nix/store/6c1rfsqzrhjw1235palzjmf5vihcpci7-source/flake.nix:3:5:
2| { outputs = { self, nixpkgs }: {
3| packages.x86_64-linux.hello = let
| ^
4| foo = nixpkgs.lib.evalModules {
```
Fixes#4745
They are equivalent according to
<https://spec.commonmark.org/0.29/#hard-line-breaks>,
and the trailing spaces tend to be a pain (because the make git
complain, editors tend to want to remove them − the `.editorconfig`
actually specifies that − etc..).
I think that it's not very helpful to get "cached failures" in a wrong
`flake.nix`. This can be very confusing when debugging a Nix expression.
See for instance NixOS/nixpkgs#118115.
In fact, the eval cache allows a forced reevaluation which is used for
e.g. `nix eval`.
This change makes sure that this is the case for `nix build` as well. So
rather than
λ ma27 [~/Projects/exp] → ../nix/outputs/out/bin/nix build -L --rebuild --experimental-features 'nix-command flakes'
error: cached failure of attribute 'defaultPackage.x86_64-linux'
the evaluation of already-evaluated (and failed) attributes looks like
this now:
λ ma27 [~/Projects/exp] → ../nix/outputs/out/bin/nix build -L --rebuild --experimental-features 'nix-command flakes'
error: attribute 'hell' missing
at /nix/store/mrnvi9ss8zn5wj6gpn4bcd68vbh42mfh-source/flake.nix:6:35:
5|
6| packages.x86_64-linux.hello = nixpkgs.legacyPackages.x86_64-linux.hell;
| ^
7|
(use '--show-trace' to show detailed location information)
When working on some more complex Nix code, there are sometimes rather
unhelpful or misleading error messages, especially if coerce-errors are
thrown.
This patch is a first steps towards improving that. I'm happy to file
more changes after that, but I'd like to gather some feedback first.
To summarize, this patch does the following things:
* Attrsets (a.k.a. `Bindings` in `libexpr`) now have a `Pos`. This is
helpful e.g. to identify which attribute-set in `listToAttrs` is
invalid.
* The `Value`-struct has a new method named `determinePos` which tries
to guess the position of a value and falls back to a default if that's
not possible.
This can be used to provide better messages if a coercion fails.
* The new `determinePos`-API is used by `builtins.concatMap` now. With
that change, Nix shows the exact position in the error where a wrong
value was returned by the lambda.
To make sure it's still obvious that `concatMap` is the problem,
another stack-frame was added.
* The changes described above can be added to every other `primop`, but
first I'd like to get some feedback about the overall approach.
* The position of the `name`-attribute appears in the trace.
* If e.g. `meta` has no `outPath`-attribute, a `cannot coerce set to
string` error will be thrown where `pos` points to `name =` which is
highly misleading.
This avoids an ambiguity where the `StorePathWithOutputs { drvPath, {}
}` could mean "build `brvPath`" or "substitute `drvPath`" depending on
context.
It also brings the internals closer in line to the new CLI, by
generalizing the `Buildable` type is used there and makes that
distinction already.
In doing so, relegate `StorePathWithOutputs` to being a type just for
backwards compatibility (CLI and RPC).
These are by no means part of the notion of a store, but rather are
things that happen to use stores. (Or put another way, there's no way
we'd make them virtual methods any time soon.) It's better to move them
out of that too-big class then.
Also, this helps us remove StorePathWithOutputs from the Store interface
altogether next commit.
The PR #4240 changed messag of the error that was thrown when an auto-called
function was missing an argument.
However this change also changed the type of the error, from `EvalError`
to a new `MissingArgumentError`. This broke hydra which was relying on
an `EvalError` being thrown.
Make `MissingArgumentError` a subclass of `EvalError` to un-break hydra.
Example:
error: builder for '/nix/store/9ysqfidhipyzfiy54mh77iqn29j6cpsb-failing.drv' failed with exit code 1;
last 1 log lines:
> FAIL
For full logs, run 'nix log /nix/store/9ysqfidhipyzfiy54mh77iqn29j6cpsb-failing.drv'.
… while importing '/nix/store/pfp4a4bjh642ylxyipncqs03z6kkgfvy-failing'
at /nix/store/25wgzr2qrqqiqfbdb1chpiry221cjglc-source/flake.nix:58:15:
57|
58| ifd = import self.hydraJobs.broken;
| ^
59|
Changes:
* The divider lines are gone. These were in practice a bit confusing,
in particular with --show-trace or --keep-going, since then there
were multiple lines, suggesting a start/end which wasn't the case.
* Instead, multi-line error messages are now indented to align with
the prefix (e.g. "error: ").
* The 'description' field is gone since we weren't really using it.
* 'hint' is renamed to 'msg' since it really wasn't a hint.
* The error is now printed *before* the location info.
* The 'name' field is no longer printed since most of the time it
wasn't very useful since it was just the name of the exception (like
EvalError). Ideally in the future this would be a unique, easily
googleable error ID (like rustc).
* "trace:" is now just "…". This assumes error contexts start with
something like "while doing X".
Example before:
error: --- AssertionError ---------------------------------------------------------------------------------------- nix
at: (7:7) in file: /home/eelco/Dev/nixpkgs/pkgs/applications/misc/hello/default.nix
6|
7| x = assert false; 1;
| ^
8|
assertion 'false' failed
----------------------------------------------------- show-trace -----------------------------------------------------
trace: while evaluating the attribute 'x' of the derivation 'hello-2.10'
at: (192:11) in file: /home/eelco/Dev/nixpkgs/pkgs/stdenv/generic/make-derivation.nix
191| // (lib.optionalAttrs (!(attrs ? name) && attrs ? pname && attrs ? version)) {
192| name = "${attrs.pname}-${attrs.version}";
| ^
193| } // (lib.optionalAttrs (stdenv.hostPlatform != stdenv.buildPlatform && !dontAddHostSuffix && (attrs ? name || (attrs ? pname && attrs ? version)))) {
Example after:
error: assertion 'false' failed
at: (7:7) in file: /home/eelco/Dev/nixpkgs/pkgs/applications/misc/hello/default.nix
6|
7| x = assert false; 1;
| ^
8|
… while evaluating the attribute 'x' of the derivation 'hello-2.10'
at: (192:11) in file: /home/eelco/Dev/nixpkgs/pkgs/stdenv/generic/make-derivation.nix
191| // (lib.optionalAttrs (!(attrs ? name) && attrs ? pname && attrs ? version)) {
192| name = "${attrs.pname}-${attrs.version}";
| ^
193| } // (lib.optionalAttrs (stdenv.hostPlatform != stdenv.buildPlatform && !dontAddHostSuffix && (attrs ? name || (attrs ? pname && attrs ? version)))) {
libc++10 seems to be stricter on what it allows in variant conversion.
I'm not sure what the rules are here, but this is the minimal change
needed to get through the compilation errors.
Move clearValue inside Value
mkInt instead of setInt
mkBool instead of setBool
mkString instead of setString
mkPath instead of setPath
mkNull instead of setNull
mkAttrs instead of setAttrs
mkList instead of setList*
mkThunk instead of setThunk
mkApp instead of setApp
mkLambda instead of setLambda
mkBlackhole instead of setBlackhole
mkPrimOp instead of setPrimOp
mkPrimOpApp instead of setPrimOpApp
mkExternal instead of setExternal
mkFloat instead of setFloat
Add note that the static mk* function should be removed eventually
Rather than storing the derivation outputs as `drvPath!outputName` internally,
store them as `drvHashModulo!outputName` (or `outputHash!outputName` for
fixed-output derivations).
This makes the storage slightly more opaque, but enables an earlier
cutoff in cases where a fixed-output dependency changes (but keeps the
same output hash) − same as what we already do for input-addressed
derivations.
Crucially this introduces BoehmGCStackAllocator, but it also
adds a bunch of wiring to avoid making libutil depend on bdw-gc.
Part of the solutions for #4178, #4200
This makes it even clearer which of the two hashes was specified in the
nix files. Some may think that "wanted" and "got" is obvious, but:
"got" could mean "got in nix file" and "wanted" could mean "want to see in nix file".
This makes it possible to have per-project configuration in flake.nix,
e.g. binary caches and other stuff:
nixConfig.bash-prompt-suffix = "[1;35mngi# [0m";
nixConfig.substituters = [ "https://cache.ngi0.nixos.org/" ];
The registry targets generally follow a URL formatting schema with
support for a query parameter of "?dir=subpath" to specify a sub-path
location below the URL root.
Alternatively, an absolute path can be specified. This specification
mode accepts the query parameter but ignores/drops it. It would
probably be better to either (a) disallow the query parameter for the
path form, or (b) recognize the query parameter and add to the path.
This patch implements (b) for consistency, and to make it easier for
tooling that might switch between a remote git reference and a local
path reference.
See also issue #4050.
std::optional had redundant checks for whether it had a value.
An object is emplaced either way so it can be dereferenced
without repeating a value check
`nix flake info` calls the github 'commits' API, which requires
authorization when the repository is private. Currently this request
fails with a 404.
This commit adds an authorization header when calling the 'commits' API.
It also changes the way that the 'tarball' API authenticates, moving the
user's token from a query parameter into the Authorization header.
The query parameter method is recently deprecated and will be disallowed
in November 2020. Using them today triggers a warning email.
Directly register the store classes rather than a function to build an
instance of them.
This gives the possibility to introspect static members of the class or
choose different ways of instantiating them.
Otherwise the result of the printing can't be parsed back correctly by
Nix (because the unescaped `${` will be parsed as the begining of an
anti-quotation).
Fix#3989
The change in 626200713b didn't account
for when the number of auto arguments is bigger than the number of
formal arguments. This causes the following:
$ nix-instantiate --eval -E '{ ... }@args: args.foo' --argstr foo foo
nix-instantiate: src/libexpr/attr-set.hh:55: void nix::Bindings::push_back(const nix::Attr&): Assertion `size_ < capacity_' failed.
Aborted (core dumped)
This was broken in 50f13b06fb. Once
again it turns out that putting a bool in a std::variant is a bad
idea, since pointers get silently cast to them...
The command line options --arg and --argstr that are used by a bunch of
CLI commands to pass arguments to top-level functions in files go
through the same code-path as auto-calling top-level functions with
their default arguments - this, however, was only passing the arguments
that were *explicitly* mentioned in the formals of the function - in the
case of an as-pattern with an ellipsis (eg args @ { ... }) extra passed
arguments would get omitted. This fixes that to instead pass *all*
specified auto args in the case that our function has an ellipsis.
Fixes#598
Fixes#3872.
This is a bit hacky. Ideally we would automatically re-evaluate the
failed attribute iff we need to print the error message (so in
commands like 'nix search' we wouldn't re-evaluate because we're
suppressing errors).
This allows interactively inspecting the state of the evaluator at the
point of failure.
Example:
$ nix eval path:///home/eelco/Dev/nix/flake2#modules.hello-closure._final --start-repl-on-eval-errors
error: --- TypeError -------------------------------------------------------------------------------------------------------------------------------------------------------------------- nix
at: (20:53) in file: /nix/store/4264z41dxfdiqr95svmpnxxxwhfplhy0-source/flake.nix
19|
20| _final = builtins.foldl' (xs: mod: xs // (mod._module.config { config = _final; })) _defaults _allModules;
| ^
21| };
attempt to call something which is not a function but a set
Starting REPL to allow you to inspect the current state of the evaluator.
The following extra variables are in scope: arg, fun
Welcome to Nix version 2.4. Type :? for help.
nix-repl> fun
error: --- EvalError -------------------------------------------------------------------------------------------------------------------------------------------------------------------- nix
at: (150:28) in file: /nix/store/4264z41dxfdiqr95svmpnxxxwhfplhy0-source/flake.nix
149|
150| tarballClosure = (module {
| ^
151| extends = [ self.modules.derivation ];
attribute 'derivation' missing
nix-repl> :t fun
a set
nix-repl> builtins.attrNames fun
[ "tarballClosure" ]
nix-repl>
This fixes an error found in builtins.path that looks like:
store path mismatch in (possibly filtered) path added from '/private/tmp/nix-shell.CyXViH/nix-test/filter-source/filterin'
when no hash is specified
If a repo is dirty, it used to return a `rev` object with an "empty"
sha1 (0000000000000000000000000000000000000000). Please note that this
only applies for `builtins.fetchGit` and *not* for `builtins.fetchTree{
type = "git"; }`.
The original idea was to implement a git-fetcher in Nix's core that
supports content hashes[1]. In #3549[2] it has been suggested to
actually use `fetchTree` for this since it's a fairly generic wrapper
over the new fetcher-API[3] and already supports content-hashes.
This patch implements a new git-fetcher based on `fetchTree` by
incorporating the following changes:
* Removed the original `fetchGit`-implementation and replaced it with an
alias on the `fetchTree` implementation.
* Ensured that the `git`-fetcher from `libfetchers` always computes a
content-hash and returns an "empty" revision on dirty trees (the
latter one is needed to retain backwards-compatibility).
* The hash-mismatch error in the fetcher-API exits with code 102 as it
usually happens whenever a hash-mismatch is detected by Nix.
* Removed the `flakes`-feature-flag: I didn't see a reason why this API
is so tightly coupled to the flakes-API and at least `fetchGit` should
remain usable without any feature-flags.
* It's only possible to specify a `narHash` for a `git`-tree if either a
`ref` or a `rev` is given[4].
* It's now possible to specify an URL without a protocol. If it's missing,
`file://` is automatically added as it was the case in the original
`fetchGit`-implementation.
[1] https://github.com/NixOS/nix/pull/3216
[2] https://github.com/NixOS/nix/pull/3549#issuecomment-625194383
[3] https://github.com/NixOS/nix/pull/3459
[4] https://github.com/NixOS/nix/pull/3216#issuecomment-553956703
This further continues with the dependency inverstion. Also I just went
ahead and exposed `parseDerivation`: it seems like the more proper
building block, and not a bad thing to expose if we are trying to be
less wedded to drv files on disk anywas.
On nix-env -qa -f '<nixpkgs>', this reduces maximum RSS by 20970 KiB
and runtime by 0.8%. This is mostly because we're not parsing the hash
part as a hash anymore (just validating that it consists of base-32
characters).
Also, replace storePathToHash() by StorePath::hashPart().
This function was used in only one place, where it could easily be
replaced by readDerivation() since it's not
performance-critical. (This function appears to have been modelled
after queryDerivationOutputs(), which exists only to make the garbage
collector faster.)
This fixes an issue where lockfile generation was not idempotent:
after updating a lockfile, a "follows" node would end up pointing to a
new copy of the node, rather than to the original node.
