Commit graph

1326 commits

Author SHA1 Message Date
Eelco Dolstra dc29e9fb47
downloadCached: Return ETag
(cherry picked from commit 529add316c)
2019-06-24 21:58:33 +02:00
Eelco Dolstra 94f11d0a61
Fix abort in fromTOML
Fixes #2969.
2019-06-24 17:09:09 +02:00
Eelco Dolstra d4fe9daed6
Simplify getFlake() / fetchFlake() logic 2019-06-21 19:04:58 +02:00
Eelco Dolstra aa2846198f
Don't update the global registry when building a locked flake
It's unnecessary and slows things down (e.g. when you're on a Thalys
with super-crappy Internet).
2019-06-21 18:34:43 +02:00
Eelco Dolstra d132d057a8
Handle store symlinks in flake directories
E.g. 'nix path-info ./result' inside a flake directory now works
again.
2019-06-21 15:29:05 +02:00
Eelco Dolstra 29ccb2e969
Fix 32-bit overflow with --no-net
--no-net causes tarballTtl to be set to the largest 32-bit integer,
which causes comparison like 'time + tarballTtl < other_time' to
fail on 32-bit systems. So cast them to 64-bit first.

https://hydra.nixos.org/build/95076624
2019-06-21 12:53:52 +02:00
Eelco Dolstra e75ffbf04a
Merge pull request #2937 from CSVdB/fuzzymatching-v2
FuzzyMatching works
2019-06-18 18:03:38 +02:00
Nick Van den Broeck 59714a15e0 FuzzyMatching works
Fixes #2843
2019-06-18 17:01:57 +02:00
Eelco Dolstra 9d1207c02c
nix flake check: Check apps 2019-06-17 17:59:57 +02:00
Eelco Dolstra 06010eaf19
Fix fetchTarball with chroot stores
Fixes #2405.
2019-06-12 10:34:13 +02:00
Eelco Dolstra 69b047f4ce
writeRegistry(): Write correct version 2019-06-11 21:32:57 +02:00
Eelco Dolstra c47d2dac6c
Disable EvalCache in impure mode 2019-06-07 22:38:39 +02:00
Eelco Dolstra 6644b6099b
Add flake evaluation cache
This exploits the hermetic nature of flake evaluation to speed up
repeated evaluations of a flake output attribute.

For example (doing 'nix build' on an already present package):

  $ time nix build nixpkgs:firefox

  real    0m1.497s
  user    0m1.160s
  sys     0m0.139s

  $ time nix build nixpkgs:firefox

  real    0m0.052s
  user    0m0.038s
  sys     0m0.007s

The cache is ~/.cache/nix/eval-cache-v1.sqlite, which has entries like

  INSERT INTO Attributes VALUES(
    X'92a907d4efe933af2a46959b082cdff176aa5bfeb47a98fabd234809a67ab195',
    'packages.firefox',
    1,
    '/nix/store/pbalzf8x19hckr8cwdv62rd6g0lqgc38-firefox-67.0.drv /nix/store/g6q0gx0v6xvdnizp8lrcw7c4gdkzana0-firefox-67.0 out');

where the hash 92a9... is a fingerprint over the flake store path and
the contents of the lockfile. Because flakes are evaluated in pure
mode, this uniquely identifies the evaluation result.
2019-06-07 22:25:48 +02:00
Eelco Dolstra 54aff8430c
Move flake-related stuff to src/libexpr/flake 2019-06-05 16:51:54 +02:00
Eelco Dolstra 1b05792988
Shorter syntax for referencing flake outputs
Fixes #2819.
2019-06-04 22:35:43 +02:00
Eelco Dolstra ce225615c3
Eliminate duplicate fetching of the top-level flake 2019-06-04 21:10:53 +02:00
Eelco Dolstra 087530dec4
Add comments 2019-06-04 21:07:55 +02:00
Eelco Dolstra 1e53a07712
Make non-flake inputs lazy
Also add a proper test for non-flake inputs.
2019-06-04 20:56:42 +02:00
Eelco Dolstra 1c5067b9a7
Check hash 2019-06-04 20:35:35 +02:00
Eelco Dolstra 5fe7be2409
Rename dep -> input
Also use nlohmann::json range-based for.
2019-06-04 20:08:13 +02:00
Eelco Dolstra 9e99b5205c
Move LockFile and related types to a separate file 2019-06-04 20:01:21 +02:00
Eelco Dolstra 278114d559
Fix GC closure generation 2019-06-04 19:45:16 +02:00
Eelco Dolstra 6dbd5c26e6
Make flake input fetching lazy
As long as the flake input is locked, it is now only fetched when it
is evaluated (e.g. "nixpkgs" is fetched when
"inputs.nixpkgs.<something>" is evaluated).

This required adding an "id" attribute to the members of "inputs" in
lockfiles, e.g.

  "inputs": {
    "nixpkgs/release-19.03": {
      "id": "nixpkgs",
      "inputs": {},
      "narHash": "sha256-eYtxncIMFVmOHaHBtTdPGcs/AnJqKqA6tHCm0UmPYQU=",
      "nonFlakeInputs": {},
      "uri": "github:edolstra/nixpkgs/e9d5882bb861dc48f8d46960e7c820efdbe8f9c1"
    }
  }

because the flake ID needs to be known beforehand to construct the
"inputs" attrset.

Fixes #2913.
2019-06-04 19:17:03 +02:00
Eelco Dolstra 653c4e439b
Fix clang compilation error
https://hydra.nixos.org/build/94332344

https://stackoverflow.com/questions/46114214/lambda-implicit-capture-fails-with-variable-declared-from-structured-binding
2019-06-03 21:51:06 +02:00
Nick Van den Broeck d9a6a75ed2 Made epochs more fine-grained
Fixes #2894
2019-06-03 14:47:47 +02:00
Eelco Dolstra fb692e5f7b
Bindings: Add convenience method for requiring an attribute 2019-06-03 09:27:03 +02:00
Eelco Dolstra 8abb8647a3
Automatically determine subdir for path flakes
This means that in a flake in a subdirectory of a Git repo, you can
now do

  $ nix build

rather than the inconvenient

  $ nix build ../..?dir=foo/bar
2019-05-31 21:52:02 +02:00
Eelco Dolstra ccb1bad612
Allow bare flakerefs as installables
So now

  $ nix build blender-bin

works and builds the default package from that flake. You don't need
to add a colon at the end anymore.
2019-05-31 21:42:23 +02:00
Eelco Dolstra 7adb10d29b
Fix reading the lockfile of a flake in a subdirectory 2019-05-31 20:12:59 +02:00
Eelco Dolstra 9169046e64
Add operator << for LockFile
Useful for debugging.
2019-05-31 20:10:56 +02:00
Eelco Dolstra b971e406de
Support 'dir' and other parameters in path flakerefs 2019-05-31 19:01:11 +02:00
Eelco Dolstra 094539ef4a
Rename requires -> inputs, provides -> outputs
Issue #2828.
2019-05-31 09:59:48 +02:00
Eelco Dolstra 6ae4437acb
Remove makeFlakeValue() 2019-05-29 15:44:48 +02:00
Eelco Dolstra 6e4a8c47f4
Put flake-related stuff in its own namespace 2019-05-29 15:31:07 +02:00
Eelco Dolstra c356d034f3
Make unsupported flake attributes a fatal error 2019-05-29 15:12:22 +02:00
Eelco Dolstra 6636808e90
Merge remote-tracking branch 'origin/master' into flakes 2019-05-29 12:36:44 +02:00
Eelco Dolstra ae7b56cd9a
Get last commit time of github flakes 2019-05-29 10:10:40 +02:00
Eelco Dolstra 0f840483c7
Add date of last commit to SourceInfo
This is primarily useful for version string generation, where we need
a monotonically increasing number. The revcount is the preferred thing
to use, but isn't available for GitHub flakes (since it requires
fetching the entire history). The last commit timestamp OTOH can be
extracted from GitHub tarballs.
2019-05-29 10:10:36 +02:00
Eelco Dolstra 22f2744afd
Iterate over references 2019-05-28 23:05:08 +02:00
Eelco Dolstra 9eaebbf575
Merge branch 'attrPaths' of https://github.com/NinjaTrappeur/nix 2019-05-28 22:59:05 +02:00
Eelco Dolstra ecee759b80
callFlake(): Emit source info attributes for non-flake dependencies 2019-05-28 14:01:08 +02:00
Eelco Dolstra dda4f7167b
Remove redundant resolvedRef fields since they're already in SourceInfo 2019-05-28 13:12:43 +02:00
Eelco Dolstra 4846304541
Rename contentHash -> narHash for consistency 2019-05-28 13:08:40 +02:00
Eelco Dolstra 894e007445
Move hash into SourceInfo and rename to narHash to avoid ambiguity 2019-05-28 13:07:15 +02:00
Eelco Dolstra 6d7efcfaeb
Store SourceInfo in Flake and NonFlake
This deduplicates some shared fields. Factoring out the commonality is
useful in places like makeFlakeValue().
2019-05-28 13:06:43 +02:00
Eelco Dolstra de36cf3db9
Merge branch 'nonFlakeRequiresTest' of https://github.com/CSVdB/nix into flakes 2019-05-28 12:05:11 +02:00
Nick Van den Broeck 4d030a8d96 Added nonFlakeRequires test
Fixes #2888
2019-05-28 10:51:45 +02:00
Nick Van den Broeck 6fb7545fa1 Fixed relative path parsing
Fixed #2821
2019-05-28 10:17:28 +02:00
Eelco Dolstra 90fe1dfd2f Register flake source trees as GC roots
This ensures that flakes don't get garbage-collected, which is
important to get nix-channel-like behaviour.

For example, running

  $ nix build hydra:

will create a GC root

  ~/.cache/nix/flake-closures/hydra -> /nix/store/xarfiqcwa4w8r4qpz1a769xxs8c3phgn-flake-closure

where the contents/references of the linked file in the store are the
flake source trees used by the 'hydra' flake:

  /nix/store/n6d5f5lkpfjbmkyby0nlg8y1wbkmbc7i-source
  /nix/store/vbkg4zy1qd29fnhflsv9k2j9jnbqd5m2-source
  /nix/store/z46xni7d47s5wk694359mq9ay353ar94-source

Note that this in itself is not enough to allow offline use; the
fetcher for the flakeref (e.g. fetchGit or downloadCached) must not
fail if it cannot fetch the latest version of the file, so long as it
knows a cached version.

Issue #2868.
2019-05-23 23:53:25 +02:00
Eelco Dolstra 6b77bfc28d FlakeRef::to_string(): Check round trip 2019-05-23 23:39:58 +02:00
Eelco Dolstra 6e984431dd fetchGit: Don't barf if we can't update our Git clone
Instead print a warning that we're continuing with the most recently
fetched version.
2019-05-23 23:38:40 +02:00
Eelco Dolstra f0d6d67af9 Prevent the global registry from being GC'ed
Issue #2868.
2019-05-22 23:43:58 +02:00
Eelco Dolstra df3f5a78d5 Refactor downloadCached() interface 2019-05-22 23:36:29 +02:00
Eelco Dolstra 66f1d7ee95 Fetch the flake registry from the NixOS/flake-registry repo 2019-05-22 22:56:46 +02:00
Eelco Dolstra e414bde6f9 Check the flake epoch
Closes #2883.
2019-05-22 14:31:40 +02:00
Eelco Dolstra 70136a9bf4 Move flake-related flags into a separate class
Also, rename --dont-save-lock-file to --no-save-lock-file and change
noRegistries to useRegistries.
2019-05-22 14:04:18 +02:00
Eelco Dolstra 2468672e30 Improve FlakeCommand
It now handles commonality like calling getFlake() and resolving
relative local flake refs.

Fixes #2822.
2019-05-22 14:03:28 +02:00
Eelco Dolstra 5990b86391 Use warn(), tweak messages 2019-05-21 15:03:54 +02:00
Eelco Dolstra 20a1a65d37 Only rewrite the lockfile if it changed
This removes spurious warnings about failure to write the lockfile.
2019-05-21 14:55:43 +02:00
Nick Van den Broeck ef6ae61503 Lockfile handling in resolveFlake is fixed 2019-05-17 14:50:10 +02:00
Nick Van den Broeck 98f20dee41 Give errors in resolveFlake
If DontUpdate but the lockfile isn't correct
2019-05-16 15:02:25 +02:00
Nick Van den Broeck d9ad3723d5 Fixed issue 65
lockfile updating
2019-05-16 15:02:05 +02:00
Eelco Dolstra 5c34d66538 Make flakes work with 'nix build --store ...'
It was getting confused between logical and real store paths.

Also, make fetchGit and fetchMercurial update allowedPaths properly.

(Maybe the evaluator, rather than the caller of the evaluator, should
apply toRealPath(), but that's a bigger change.)
2019-05-15 15:38:24 +02:00
Nick Van den Broeck 4c9ebd20d7 One FIXME was already fixed 2019-05-15 08:10:46 +02:00
Nick Van den Broeck d209bdcd08 Fixed issue #47
Content hashes
2019-05-09 14:27:55 +02:00
Eelco Dolstra 8fc1c3f413
Fix assertion failure in FlakeRef::to_string()
Fixes

  $ nix build
  nix: src/libexpr/primops/flakeref.cc:169: std::__cxx11::string nix::FlakeRef::to_string() const: Assertion `!rev' failed.
  Aborted

e.g. when flake.nix doesn't exist.

Also use gitRev().
2019-05-08 22:09:18 +02:00
Eelco Dolstra 455aa8d9ea
Add newline at end of lockfile
Suggested by @grahamc.
2019-05-08 18:28:01 +02:00
Eelco Dolstra 2bc55aba1e
Merge remote-tracking branch 'origin/master' into flakes 2019-05-08 14:30:27 +02:00
Eelco Dolstra 54e54db2e2
Merge remote-tracking branch 'tweag/flake-test' into flakes 2019-05-08 13:55:55 +02:00
Eelco Dolstra 77e1f9010c
Export missing rev/shortRev attributes 2019-05-08 13:38:32 +02:00
Eelco Dolstra 9d07c3717b
updateLockFile: Do "git add" in a slightly nicer way
"--intent-to-add" ensures the change shows up in "git diff".
2019-05-08 13:26:20 +02:00
Eelco Dolstra ddd42b7e94 Fix immutable flakeref construction
We were appending ref/revs incorrectly for the IsGit case (by
appending /<ref>/<rev> rather than ?ref=<ref>&rev=<rev).
2019-05-07 23:32:09 +02:00
Eelco Dolstra 2a41a567e2 Improve FlakeRef::to_string()
We were incorrectly using path syntax (i.e. /<ref>/<rev>) for Git
repositories. This is only valid for GitHub flakerefs.
2019-05-07 23:32:09 +02:00
Eelco Dolstra 3c171851a8 Make the URL/path of the global flake registry configurable 2019-05-07 22:49:13 +02:00
Nick Van den Broeck 201f92e02c Fixed Flake data type and flake fetching 2019-05-07 05:57:08 +02:00
Daniel Schaefer 3f192ac80c Add builtins.hashFile
For text files it is possible to do it like so:
`builtins.hashString "sha256" (builtins.readFile /tmp/a)`
but that doesn't work for binary files.

With builtins.hashFile any kind of file can be conveniently hashed.
2019-05-03 17:23:36 +02:00
Eelco Dolstra f8c4742c2f Fix 'git add' when subdir is empty 2019-05-03 13:15:13 +02:00
Eelco Dolstra 2aafa6901e Merge remote-tracking branch 'tweag/subdir' into flakes 2019-05-03 12:54:59 +02:00
Nick Van den Broeck e0d4aa75fc Fixed compile errors 2019-05-02 08:40:00 +02:00
Eelco Dolstra 5d6e8c008b Allow 'dir' parameter in github: URIs
E.g. 'github:edolstra/dwarffs/flake?dir=foo/bar'.
2019-05-01 20:44:30 +02:00
Eelco Dolstra a37436d792 Accept empty directories 2019-05-01 20:44:30 +02:00
Eelco Dolstra fa88f71520 Validate 'dir=' parameters
We reject any path element starting with a '.' (mostly to reject '.'
and '..').
2019-05-01 20:44:30 +02:00
Eelco Dolstra ab9e47284a Improve error message 2019-05-01 20:44:30 +02:00
Nick Van den Broeck 43408d3cd6 flake.lock now uses flakeRef.subdir 2019-05-01 16:24:33 +02:00
Eelco Dolstra 00db8d4549 Support 'dir' parameters in https and ssh flake URIs 2019-05-01 14:24:22 +02:00
Nick Van den Broeck 9b3069a88c Fuzzymatching
Fixed issue #61
2019-05-01 12:59:35 +02:00
Nick Van den Broeck eba85e2367 WIP: still need to adapt flakeref parsing 2019-05-01 12:59:12 +02:00
Eelco Dolstra 4588a6ff3c Merge remote-tracking branch 'tweag/flakeFlags' into flakes 2019-04-30 12:46:54 +02:00
Nick Van den Broeck 24b35bf9e7 Fixed issue #13 2019-04-30 11:26:45 +02:00
Nick Van den Broeck 35d1c95f7f Fix flag registry order 2019-04-30 11:04:31 +02:00
Nick Van den Broeck e9c42c06ef Fixed lookupFlake bug 2019-04-25 10:50:59 +02:00
Eelco Dolstra 21d5abfc14 Merge remote-tracking branch 'tweag/flake-clone' into flakes 2019-04-24 12:47:35 +02:00
Eelco Dolstra 4fb594a375 Merge remote-tracking branch 'tweag/resolvedFlake' into flakes 2019-04-19 15:56:17 +02:00
Nick Van den Broeck e51abb6631 Changed some names 2019-04-19 14:23:35 +02:00
Eelco Dolstra bc259192b4 fetchGit: Return revCount for dirty working trees 2019-04-19 14:15:51 +02:00
Eelco Dolstra 0cbda84f5b exportGit: Don't clone local repositories
This ensures that commands like 'nix flake info /my/nixpkgs' don't
copy a gigabyte of crap to ~/.cache/nix.

Fixes #60.
2019-04-19 14:10:57 +02:00
Eelco Dolstra 160ce18a0e Improve missing flake.nix error message 2019-04-19 11:43:56 +02:00
Eelco Dolstra 6960ee929d Clean up exportGit argument handling 2019-04-19 11:34:23 +02:00
Eelco Dolstra 46cb15df9b Fix assertion failure in FlakeRef::to_string() 2019-04-19 11:16:14 +02:00
Eelco Dolstra 6e4210d8ce Fix assertion failure 2019-04-19 10:58:08 +02:00
Eelco Dolstra 160b974fb0 Fix mutability check 2019-04-17 13:54:06 +02:00
Nick Van den Broeck b42ba08fc8 Add command flake clone 2019-04-17 13:31:09 +02:00
Eelco Dolstra 939bee06cd Pass a flake to itself as "self" 2019-04-16 16:29:44 +02:00
Eelco Dolstra 3ddb6d1833 Allow refs to start with a digit
E.g. we want to accept "19.03" as a ref.
2019-04-16 16:24:51 +02:00
Eelco Dolstra 260527a90c Use the lock file 2019-04-16 16:18:47 +02:00
Eelco Dolstra 3d0e81051f Fix lock file generation
Before:

   "requires": {
        "nixpkgs": {
            "uri": "nixpkgs"
        }
    },

After:

    "requires": {
        "nixpkgs": {
            "uri": "github:edolstra/nixpkgs/f10e8a02eb7fa2b4a070f30cf87f4efcc7f3186d"
        }
    },
2019-04-16 15:43:55 +02:00
Eelco Dolstra 8c4e759efd updateLockFile(): Make sure Git can see flake.lock 2019-04-16 15:11:17 +02:00
Eelco Dolstra 60834492ae Update lock files from InstallableFlake::toValue()
This ensures that the lock file is updated *before* evaluating it, and
that it gets updated for any nix command, not just 'nix build'.

Also, while computing the lock file, allow arbitrary registry lookups,
not just at top-level.

Also, improve some error messages slightly.
2019-04-16 15:02:02 +02:00
Eelco Dolstra 7b312a8762 Pass stuff by reference 2019-04-16 14:27:54 +02:00
Eelco Dolstra e1d73edb10 writeLockFile(): Emit empty objects rather than null 2019-04-16 14:23:10 +02:00
Eelco Dolstra ed9d725392 getFlake(): Use impureIsAllowed
This fixes 'nix build nixpkgs:hello' without --impure.
2019-04-16 14:16:20 +02:00
Eelco Dolstra aecf07b1d6 Remove dead function 2019-04-16 14:08:14 +02:00
Eelco Dolstra 035ac44354 Fix makeFlakeValue() 2019-04-16 13:56:08 +02:00
Eelco Dolstra 3a5493bfe8 Merge remote-tracking branch 'tweag/privateGithub' into flakes 2019-04-16 12:23:56 +02:00
Nick Van den Broeck 7587d62d02 Fixed flake pin issues 2019-04-16 08:38:17 +02:00
Nick Van den Broeck b3d33b02e3 Added support for private github repositories 2019-04-16 08:14:46 +02:00
Eelco Dolstra b4e367bf4a FlakeRef::to_string(): Drop the "flake:" prefix
This is unnecessary in most contexts and makes 'nix flake list' output
less readable.
2019-04-15 14:13:10 +02:00
Eelco Dolstra c179f668e5 Slight cleanup 2019-04-15 14:08:18 +02:00
Eelco Dolstra f6d684b5e2 getFlakeRegistries(): Return registries regardless of pureEval
This makes e.g. 'nix flake list' work.
2019-04-15 13:48:56 +02:00
Eelco Dolstra 84c12dbd7c Move --impure to MixEvalArgs 2019-04-15 13:45:51 +02:00
Eelco Dolstra bb6e6923f2 Add environment variable NIX_SHOW_SYMBOLS for dumping the symbol table 2019-04-11 23:04:13 +02:00
Nick Van den Broeck 4ad4e48668 FlakeRegistry = FlakeRef -> FlakeRef 2019-04-11 07:08:04 +02:00
Nick Van den Broeck c64f98b883 FlakeAlias is implemented 2019-04-09 10:16:02 +02:00
Nick Van den Broeck f39670c631 Took ref and rev out of FlakeRef 2019-04-09 09:51:42 +02:00
Nick Van den Broeck 3ec0c82fab Fixed dependency resolution 2019-04-09 09:51:42 +02:00
Nick Van den Broeck 641db127be FlakeIds are now properly looked up in registries 2019-04-09 09:51:36 +02:00
Nick Van den Broeck 18c019b616 Added nonFlakeRequires and the command nix flake deps 2019-04-09 08:38:09 +02:00
Eelco Dolstra c996e04aca Allow relative paths in flakerefs
Also allow "." as an installable to refer to the flake in the current
directory. E.g.

  $ nix build .

will build 'provides.defaultPackage' in the flake in the current
directory.
2019-04-08 23:52:09 +02:00
Eelco Dolstra ee1254d4f5 nix: Add --impure as a shorter alias of --no-pure-eval 2019-04-08 23:19:19 +02:00
Eelco Dolstra 6a4c7fb975 Add path flakeref variant
Unlike file://<path>, this allows the path to be a dirty Git tree, so

  nix build /path/to/flake:attr

is a convenient way to test building a local flake.
2019-04-08 23:09:18 +02:00
Eelco Dolstra 9b7eac332b Merge remote-tracking branch 'origin/master' into flakes 2019-04-08 14:23:48 +02:00
Eelco Dolstra 154244adc6 nix: New installables syntax
The general syntax for an installable is now
<flakeref>:<attrpath>. The attrpath is relative to the flake's
'provides.packages' or 'provides' if the former doesn't yield a
result. E.g.

  $ nix build nixpkgs:hello

is equivalent to

  $ nix build nixpkgs:packages.hello

Also, '<flakeref>:' can be omitted, in which case it defaults to
'nixpkgs', e.g.

  $ nix build hello
2019-04-08 14:21:13 +02:00
Eelco Dolstra 6b0ca8e803 findAlongAttrPath(): Throw AttrPathNotFound 2019-04-08 14:20:58 +02:00
Eelco Dolstra f32fbf952d Fix Bison 3.3 warning 2019-03-27 21:09:31 +01:00
Eelco Dolstra c101b29133 Merge remote-tracking branch 'tweag/flake-registries' into flakes 2019-03-27 14:40:36 +01:00
Eelco Dolstra f9c7176a87 nix flake add: Handle ~/.config/nix not existing
Fixes

  $ nix flake add fnord github:edolstra/fnord
  error: opening file '/home/eelco/.config/nix/registry.json': No such file or directory
2019-03-26 12:48:57 +01:00
Eelco Dolstra 42be60c6af Merge remote-tracking branch 'tweag/flake-add' into flakes 2019-03-26 12:35:27 +01:00
Eelco Dolstra edb3836696 Merge remote-tracking branch 'tweag/auto-update-flake' into flakes 2019-03-26 11:50:22 +01:00
Eelco Dolstra b5565a7081 Merge remote-tracking branch 'origin/master' into flakes 2019-03-26 11:44:14 +01:00
Jonas Chevalier 514b3c7f83
Add isPath primop
this is added for completeness' sake since all the other possible
`builtins.typeOf` results have a corresponding `builtins.is<Type>`
2019-03-24 11:36:49 +01:00
Nick Van den Broeck a554f523db Combining registries properly 2019-03-23 13:42:48 +01:00
Nick Van den Broeck 5e4d92d267 Issue #15 is finished 2019-03-22 15:22:18 +01:00
Nick Van den Broeck e007f367bd Fixed minor things 2019-03-22 11:21:49 +01:00
Nick Van den Broeck d4ee8afd59 Implemented --flake flag for nix build
Also fixed Eelco's PR comments
2019-03-22 11:16:20 +01:00
Linus Heckemann 2aa89daab3 eval: improve type description for primops and applied primops
This can make type errors a little easier to understand.
2019-03-21 15:31:46 +01:00
Eelco Dolstra 1f64f4c7c8
pkg-config files: Use c++17 2019-03-14 14:11:12 +01:00
Eelco Dolstra ef52ccf035
experimental/optional -> optional 2019-03-14 14:10:52 +01:00
Nick Van den Broeck 6542de98c2 Implemented writeRegistry 2019-03-07 12:00:10 +01:00
Nick Van den Broeck 9ff1a9ea65 Implemented json flag for nix flake info 2019-03-07 11:56:43 +01:00
Nick Van den Broeck cfb6ab80ce Implemented "nix flake info" 2019-02-26 13:12:51 +01:00
Eelco Dolstra d342de02b9
fetchFlake: Use infinite TTL when the revision is specified 2019-02-25 23:23:45 +08:00
Eelco Dolstra 529add316c
downloadCached: Return ETag
This allows fetchFlake() to return the Git revision of a GitHub
archive.
2019-02-25 23:20:50 +08:00
Eelco Dolstra 6e9182fbc2
Add basic flake lock file support 2019-02-12 22:43:22 +01:00
Eelco Dolstra e38ec77de8
Interpret all file:// URIs as Git repositories 2019-02-12 22:06:19 +01:00
Eelco Dolstra beab05851b
nix: Add --flake flag
This allows using an arbitrary "provides" attribute from the specified
flake. For example:

  nix build --flake nixpkgs packages.hello

(Maybe provides.packages should be used for consistency...)
2019-02-12 21:55:43 +01:00
Eelco Dolstra 272b58220d
Enforce use of immutable flakes in pure mode
... plus a temporary hack to allow impure flakes at top-level for the
default installation source.
2019-02-12 21:07:48 +01:00
Eelco Dolstra ba05f29838
nix: Enable pure mode by default
We want to encourage a brave new world of hermetic evaluation for
source-level reproducibility, so flakes should not poke around in the
filesystem outside of their explicit dependencies.

Note that the default installation source remains impure in that it
can refer to mutable flakes, so "nix build nixpkgs.hello" still works
(and fetches the latest nixpkgs, unless it has been pinned by the
user).

A problem with pure evaluation is that builtins.currentSystem is
unavailable. For the moment, I've hard-coded "x86_64-linux" in the
nixpkgs flake. Eventually, "system" should be a flake function
argument.
2019-02-12 20:35:03 +01:00
Eelco Dolstra 91a6a47b0e
Improve flake references 2019-02-12 18:23:11 +01:00
Eelco Dolstra 0cd7f2cd8d
pkg-config files: Use c++17 2019-02-12 13:44:20 +01:00
Eelco Dolstra c8a0b9d5cb
experimental/optional -> optional 2019-02-12 13:43:32 +01:00
Eelco Dolstra 52419f8db3
Add flake registry
This will eventually be moved to nixos.org.
2019-02-11 13:01:39 +01:00
Eelco Dolstra dcae46ab14
Add github URIs
For example, github:edolstra/dwarffs is more-or-less equivalent to
https://github.com/edolstra/dwarffs.git. It's a much faster way to get
GitHub repositories: it fetches tarballs rather than entire Git
repositories. It also allows fetching specific revisions by hash
without specifying a ref (e.g. a branch name):

  github:edolstra/dwarffs/41c0c1bf292ea3ac3858ff393b49ca1123dbd553
2019-02-11 12:05:17 +01:00
Eelco Dolstra ef4cf4e681
Introduce flake URIs 2019-02-11 12:05:13 +01:00
Eelco Dolstra 7a5cf31060
Initial flake support 2019-02-11 12:00:13 +01:00
Eelco Dolstra f216c76c56
Bindings::get(): Add convenience method
This allows writing attribute lookups as

    if (auto name = value.attrs->get(state.sName))
      ...
2019-02-11 11:55:35 +01:00
Eelco Dolstra c02da99757
EvalState::allocAttr(): Add convenience method 2019-02-11 11:55:29 +01:00
Eelco Dolstra aa0e2a2e70
Make constant primops lazy 2019-02-11 11:52:07 +01:00
Shea Levy b30be6b450
Add builtins.appendContext.
A partner of builtins.getContext, useful for the same reasons.
2019-01-31 08:52:23 -05:00
Shea Levy 1d757292d0
Add builtins.getContext.
This can be very helpful when debugging, as well as enabling complex
black magic like surgically removing a single dependency from a
string's context.
2019-01-14 11:27:10 -05:00
Shea Levy 087be7281a
Treat plain derivation paths in context as normal paths.
Previously, plain derivation paths in the string context (e.g. those
that arose from builtins.storePath on a drv file, not those that arose
from accessing .drvPath of a derivation) were treated somewhat like
derivaiton paths derived from .drvPath, except their dependencies
weren't recursively added to the input set. With this change, such
plain derivation paths are simply treated as paths and added to the
source inputs set accordingly, simplifying context handling code and
removing the inconsistency. If drvPath-like behavior is desired, the
.drv file can be imported and then .drvPath can be accessed.

This is a backwards-incompatibility, but storePath is never used on
drv files within nixpkgs and almost never used elsewhere.
2019-01-13 11:29:55 -05:00
Eelco Dolstra c7bf1cdb4e
Merge pull request #2608 from dtzWill/fix/issue-2546
EvalState::resetFileCache: clear parse cache as well as eval cache
2019-01-10 20:56:31 +01:00
John Ericson fef9f5653b Remove mentions of libformat, it no longer exists 2019-01-05 14:31:29 -05:00
Will Dietz 21ea00d3ec EvalState::resetFileCache: clear parse cache as well as eval cache
Fixes #2546.

(at least the basic reproduction I've been testing)
2018-12-31 10:18:28 -06:00
Daiderd Jordan 7e35e914c1
fetchGit: allow fetching explicit refs
Trying to fetch refs that are not in refs/heads currently fails because
it looks for refs/heads/refs/foo instead of refs/foo.

eg.

	builtins.fetchGit {
	  url = https://github.com/NixOS/nixpkgs.git;
	  ref = "refs/pull/1024/head;
	}
2018-12-14 20:12:22 +01:00
Eelco Dolstra 6024dc1d97
Support SRI hashes
SRI hashes (https://www.w3.org/TR/SRI/) combine the hash algorithm and
a base-64 hash. This allows more concise and standard hash
specifications. For example, instead of

  import <nix/fetchurl.nl> {
    url = https://nixos.org/releases/nix/nix-2.1.3/nix-2.1.3.tar.xz;
    sha256 = "5d22dad058d5c800d65a115f919da22938c50dd6ba98c5e3a183172d149840a4";
  };

you can write

  import <nix/fetchurl.nl> {
    url = https://nixos.org/releases/nix/nix-2.1.3/nix-2.1.3.tar.xz;
    hash = "sha256-XSLa0FjVyADWWhFfkZ2iKTjFDda6mMXjoYMXLRSYQKQ=";
  };

In fixed-output derivations, the outputHashAlgo is no longer mandatory
if outputHash specifies the hash (either as an SRI or in the old
"<type>:<hash>" format).

'nix hash-{file,path}' now print hashes in SRI format by default. I
also reverted them to use SHA-256 by default because that's what we're
using most of the time in Nixpkgs.

Suggested by @zimbatm.
2018-12-13 14:30:52 +01:00
CHEIKH Chawki fa5143c722 Solve hg "abandoned transaction" issue 2018-12-06 13:57:59 +01:00
Eelco Dolstra 4aee93d5ce
fetchGit: Drop unnecessary localRef 2018-11-20 20:59:44 +01:00
Eelco Dolstra 3f4de91d80
Merge branch 'better-git-cache' of https://github.com/graham-at-target/nix 2018-11-20 20:41:19 +01:00
Jan Path d1b049c4ea Fix typo in comments 2018-10-31 20:50:18 +01:00
Guillaume Maudoux 6a5bf9b143 simplify handling of extra '}' 2018-10-27 00:14:51 +02:00
Eelco Dolstra 8327a7a8fa
Merge branch 'dirOf-relative' of https://github.com/lheckemann/nix 2018-09-13 14:33:12 +02:00
Eelco Dolstra 91405986f4
Convert NIX_COUNT_CALLS to JSON too 2018-09-05 21:57:54 +02:00
Eelco Dolstra 0a2545f95c
Log stats to stderr
We shouldn't pollute stdout.
2018-09-05 21:35:58 +02:00
Michael Bishop 4b034f390c remove the old text format output 2018-09-02 18:25:23 -03:00
Michael Bishop 2fd1008c70 add JSON to NIX_SHOW_STATS 2018-09-01 20:05:06 -03:00
Michael Bishop c29e5fbb13 improve the stats when profiling 2018-09-01 17:11:56 -03:00
Eelco Dolstra 475a0a54a9
fetchGit/fetchMercurial: Don't absolutize paths
This is already done by coerceToString(), provided that the argument
is a path (e.g. 'fetchGit ./bla'). It fixes the handling of URLs like
git@github.com:owner/repo.git. It breaks 'fetchGit "./bla"', but that
was never intended to work anyway and is inconsistent with other
builtin functions (e.g. 'readFile "./bla"' fails).
2018-09-01 00:19:49 +02:00
aszlig 0ad643ed5c
libexpr: Use int64_t for NixInt
Using a 64bit integer on 32bit systems will come with a bit of a
performance overhead, but given that Nix doesn't use a lot of integers
compared to other types, I think the overhead is negligible also
considering that 32bit systems are in decline.

The biggest advantage however is that when we use a consistent integer
size across all platforms it's less likely that we miss things that we
break due to that. One example would be:

https://github.com/NixOS/nixpkgs/pull/44233

On Hydra it will evaluate, because the evaluator runs on a 64bit
machine, but when evaluating the same on a 32bit machine it will fail,
so using 64bit integers should make that consistent.

While the change of the type in value.hh is rather easy to do, we have a
few more options available for doing the conversion in the lexer:

  * Via an #ifdef on the architecture and using strtol() or strtoll()
    accordingly depending on which architecture we are. For the #ifdef
    we would need another AX_COMPILE_CHECK_SIZEOF in configure.ac.
  * Using istringstream, which would involve copying the value.
  * As we're already using boost, lexical_cast might be a good idea.

Spoiler: I went for the latter, first of all because lexical_cast does
have an overload for const char* and second of all, because it doesn't
involve copying around the input string. Also, because istringstream
seems to come with a bigger overhead than boost::lexical_cast:

https://www.boost.org/doc/libs/release/doc/html/boost_lexical_cast/performance.html

The first method (still using strtol/strtoll) also wasn't something I
pursued further, because it is also locale-aware which I doubt is what
we want, given that the regex for int is [0-9]+.

Signed-off-by: aszlig <aszlig@nix.build>
Fixes: #2339
2018-08-29 01:05:52 +02:00
Eelco Dolstra c651b7bdc9
Revert "Fix parser/lexer generation with parallel make"
This reverts commit d277442df5.

Make sucks.
2018-08-23 00:23:35 +02:00
Eelco Dolstra 9b1bdf2db8
FIx floating point evaluation
Fixes #2361.
2018-08-19 11:59:49 +02:00
Graham Christensen 02098d2073 fetchGit: use a better caching scheme
The current usage technically works by putting multiple different
repos in to the same git directory. However, it is very slow as
Git tries very hard to find common commits between the two
repositories. If the two repositories are large (like Nixpkgs and
another long-running project,) it is maddeningly slow.

This change busts the cache for existing deployments, but users
will be promptly repaid in per-repository performance.
2018-08-17 11:27:34 -04:00
Eelco Dolstra d277442df5
Fix parser/lexer generation with parallel make
Fun fact: rules with multiple targets don't work properly with 'make
-j'. For example, a rule like

  a b: c
    touch a b

is equivalent to

  a: c
    touch a b

  b: c
    touch a b

so with 'make -j', the 'touch' command will be run twice. See
e.g. https://stackoverflow.com/questions/2973445/gnu-makefile-rule-generating-a-few-targets-from-a-single-source-file.
2018-08-17 12:59:23 +02:00
Linus Heckemann d7402c9cd5 dirOf: allow use on non-absolute paths 2018-08-13 11:27:35 +02:00
Eelco Dolstra bc65e02d96
Merge pull request #2326 from aszlig/fix-symlink-leak
Fix symlink leak in restricted eval mode
2018-08-03 17:01:34 +02:00
aszlig 43e28a1b75
Fix symlink leak in restricted eval mode
In EvalState::checkSourcePath, the path is checked against the list of
allowed paths first and later it's checked again *after* resolving
symlinks.

The resolving of the symlinks is done via canonPath, which also strips
out "../" and "./". However after the canonicalisation the error message
pointing out that the path is not allowed prints the symlink target in
the error message.

Even if we'd suppress the message, symlink targets could still be leaked
if the symlink target doesn't exist (in this case the error is thrown in
canonPath).

So instead, we now do canonPath() without symlink resolving first before
even checking against the list of allowed paths and then later do the
symlink resolving and checking the allowed paths again.

The first call to canonPath() should get rid of all the "../" and "./",
so in theory the only way to leak a symlink if the attacker is able to
put a symlink in one of the paths allowed by restricted evaluation mode.

For the latter I don't think this is part of the threat model, because
if the attacker can write to that path, the attack vector is even
larger.

Signed-off-by: aszlig <aszlig@nix.build>
2018-08-03 06:46:43 +02:00
Eelco Dolstra 45bcf5416a
Merge branch 'prim_mapAttr-fix' of https://github.com/volth/nix 2018-07-31 20:05:07 +02:00
Eelco Dolstra a7fb7d3cde
Merge pull request #2303 from volth/patch-4
parser.y: fix assoc of -> and < > <= >=
2018-07-23 11:38:15 +02:00
volth deaa6e9a34
parser.y: right-associativity of -> 2018-07-23 07:28:48 +00:00
volth 85fe4a819c
parser.y: fix assoc of -> and < > <= >=
The parser allowed senseless `a > b > c` but disallowed `a -> b -> c` which seems valid
It might be a typo
2018-07-21 15:24:51 +00:00
volth e2b114cfe1
prim_foldlStrict: call forceValue() before value is copied
forceValue() were called after a value is copied effectively forcing only one of the copies keeping another copy not evaluated.
This resulted in its evaluation of the same lazy value more than once (the number of hits is not big though)
2018-07-21 06:44:42 +00:00
Eelco Dolstra c2de2ff385
Fix build on 32-bit systems and macOS
Apparently, on macOS, 'long' != 'int64_t'.

https://hydra.nixos.org/build/77100756
2018-07-11 21:12:18 +02:00
Eelco Dolstra efd04888ca
Shup up a warning 2018-07-11 21:05:09 +02:00
Eelco Dolstra 875cd9da2b
Remove unused function printStats2()
Closes #2282.
2018-07-11 20:29:18 +02:00
volth c03d73c1cf
prim_map, prim_genList: no need to force function argument which will be stored in mkApp 2018-07-07 23:30:17 +00:00
volth 627e28ba33 prim_mapAttrs: `f' must be evaluated lazily to avoid infinite recursion 2018-07-06 21:52:54 +00:00
volth 1515c65616 prim_concatMap: no need to force value 2018-07-05 15:33:33 +00:00
volth e6bf1a79d7 prim_mapAttrs: must be lazy to avoid infinite recursion 2018-07-05 15:33:12 +00:00
volth 841747b0e6
prim_concatMap: allocate intermediate list on stack 2018-07-05 12:37:37 +00:00
volth ee218f99ca
primops.cc: fix comment 2018-07-05 11:58:15 +00:00
volth 403a76a18f lib.concatMap and lib.mapAttrs to be builtins 2018-07-05 02:54:09 +00:00
Eelco Dolstra ea3c9dab5f
Include cpptoml for build simplicity 2018-07-03 18:39:36 +02:00
Eelco Dolstra 3b1f54cf06
Add a fromTOML primop
This is primarily useful for processing Cargo.lock files.
2018-07-03 18:39:36 +02:00
Eelco Dolstra a92ed973e5
Store floating point numbers in double precision
Even on 32-bit systems, Value has enough space to hold a double.
2018-07-03 18:39:32 +02:00
Eelco Dolstra c905d8b0a8
GC_malloc -> GC_MALLOC
This makes it possible to build with -DGC_DEBUG.
2018-06-12 17:49:55 +02:00
Eelco Dolstra 455d1f01d0
Don't scan for roots in dynamic libraries
This reduces the risk of object liveness misdetection. For example,
Glibc has an internal variable "mp_" that often points to a Boehm
object, keeping it alive unnecessarily. Since we don't store any
actual roots in global variables, we can just disable data segment
scanning.

With this, the max RSS doing 100 evaluations of
nixos.tests.firefox.x86_64-linux.drvPath went from 718 MiB to 455 MiB.
2018-06-12 17:49:55 +02:00
Eelco Dolstra 30964103dc
Add temporary stats 2018-06-12 17:49:55 +02:00
Eelco Dolstra 24c6806994
Cache parse trees
This prevents EvalState::resetFileCache() from parsing everything all
over again.
2018-06-12 17:49:55 +02:00
Eelco Dolstra 6ad0a2f749
Remove duplicate definition of allocBytes() 2018-06-12 17:49:51 +02:00
Eelco Dolstra 5d4a9d5677
Fix static assertion failure on 32-bit systems 2018-05-30 17:47:30 +02:00
Eelco Dolstra 1672bcd230
Move evaluator-specific settings out of libstore 2018-05-30 13:29:50 +02:00
Eelco Dolstra bd56b5fe3f
Merge pull request #2187 from bgamari/stoi-exceptions
json-to-value: Use strtol instead of strtoi
2018-05-28 14:54:49 +02:00
Ben Gamari 7f560b81ea json-to-value: Use strtol instead of strtoi
NixInt is long, so strtoi is too restrictive.
2018-05-26 18:43:46 -04:00
Will Dietz 93ae90de0f eval.cc: add message to static_assert, message can be omitted w/c++17 2018-05-26 14:04:41 -05:00
Eelco Dolstra 743359bc8a
Merge pull request #2157 from volth/bitwise
add builtins: __bitAnd,  __bitOr,  __bitXor
2018-05-24 15:00:39 +02:00
Eelco Dolstra 4bb8741b98
Make Env self-describing
If the Env denotes a 'with', then values[0] may be an Expr* cast to a
Value*. For code that generically traverses Values/Envs, it's useful
to know this.
2018-05-22 16:02:32 +02:00
Eelco Dolstra 9fd7cf98db
Memoise checkSourcePath()
This prevents hydra-eval-jobs from statting the same files over and
over again.
2018-05-22 13:02:14 +02:00
volth 6cc28c0589
add `mod' and bitwise builtins: camel-case function names 2018-05-16 10:52:19 +00:00
volth f3c090f91c
add mod' and bitwise builtins: remove mod' and shifts 2018-05-16 06:57:11 +00:00
volth 49b7cf1813
add `mod' and bitwise builtins: remove infix functions 2018-05-16 06:55:24 +00:00
volth 8a6a14e1f5 add `mod' and bitwise builtins 2018-05-12 16:50:39 +00:00
Eelco Dolstra 1ad19232c4
Don't return negative numbers from the flex tokenizer
Fixes #1374.
Closes #2129.
2018-05-11 12:05:12 +02:00
Eelco Dolstra f3c85f9eb3
Revert "Throw a specific error for incomplete parse errors."
This reverts commit 6498adb002. We don't
actually use IncompleteParseError in 'nix repl'.
2018-05-11 11:40:50 +02:00
Eelco Dolstra a91c4ca01f
In restricted eval mode, allow access to the closure of store paths
E.g. this makes

  nix eval --restrict-eval -I /nix/store/foo '(builtins.readFile "/nix/store/foo/symlink/bla")'

(where /nix/store/foo/symlink is a symlink to another path in the
closure of /nix/store/foo) succeed.

This fixes a regression in Hydra compared to Nix 1.x (where there were
no restrictions at all on access to the Nix store).
2018-05-09 15:45:05 +02:00
Eelco Dolstra 53ec5ac69f
Fix some random -Wconversion warnings 2018-05-02 13:56:34 +02:00
Shea Levy e2b028353b
Merge branch 'pos-crash-fix' of git://github.com/dezgeg/nix 2018-04-23 08:43:54 -04:00
Félix Baylac-Jacqué 00584bb091
parser: Allow mixed nested and top-level attrpaths
Fixes #2077.
2018-04-18 18:44:38 +02:00
Shea Levy b37f5ae31d
isFunction: True on primops.
Fixes #2073
2018-04-17 14:33:12 -04:00
Shea Levy a38fe5c1a8
Export required C++ version in pkgconfig. 2018-04-09 11:32:43 -04:00
Shea Levy fd98fca7bb
Make prim_exec and prim_importNative available to plugins 2018-04-09 10:26:50 -04:00
Tuomas Tynkkynen af86132e1a libexpr: Make unsafeGetAttrPos not crash on noPos
Currently e.g. `builtins.unsafeGetAttrPos "abort" builtins` will
eventually segfault because pos->file is an unset Symbol.

Found by afl-fuzz.
2018-04-03 15:54:42 +03:00
Eelco Dolstra 138af2e554
Shut up signedness warning 2018-03-19 11:57:15 +01:00
Eelco Dolstra 64441f0551
Merge pull request #1939 from dezgeg/lexer-fix
libexpr: Recognize newline in more places in lexer
2018-03-16 12:14:58 +01:00
Shea Levy cfdbfa6b2c
Catch more possible instances of passing NULL to memcpy.
Actually fixes #1976.
2018-03-14 23:53:43 -04:00
Shea Levy e2088febf3
concatLists: Don't pass NULL pointers to memcpy.
This is UB, even if the size is 0. See #1976.

Fixes #1976.
2018-03-14 23:44:02 -04:00
Eelco Dolstra 55aa622fb1
Fix compatibility with latest boost::format 2018-03-14 19:25:09 +01:00
Guillaume Maudoux 80735c4cc9
fetchGit: Fix debug message 2018-03-13 10:28:23 +01:00
Tim Engler 383afab9ad Modified MakeBinOp to no longer produce its name using concatenation and "##".
Doing so prevents emacs tags from working, as well as makes the code extremely
confusing for a newbie.

In the prior state, if someone wants to find the definition of "ExprApp" for
example, a grep through the code reveals nothing. Since the definition could be
hiding in numerous ".h" files, it's really difficult to find. This personally
took me several hours to figure out.
2018-03-09 03:16:33 +00:00
Tuomas Tynkkynen a0e38c16bc libexpr: Recognize newline in more places in lexer
Flex's regexes have an annoying feature: the dot matches everything
except a newline. This causes problems for expressions like:

"${0}\
"

where the backslash-newline combination matches this rule instead of the
intended one mentioned in the comment:

    <STRING>\$|\\|\$\\ {
                    /* This can only occur when we reach EOF, otherwise the above
                    (...|\$[^\{\"\\]|\\.|\$\\.)+ would have triggered.
                    This is technically invalid, but we leave the problem to the
                    parser who fails with exact location. */
                    return STR;
                }
However, the parser actually accepts the resulting token sequence
('"' DOLLAR_CURLY 0 '}' STR '"'), which is a problem because the lexer
rule didn't assign anything to yylval. Ultimately this leads to a crash
when dereferencing a NULL pointer in ExprConcatStrings::bindVars().

The fix does change the syntax of the language in some corner cases
but I think it's only turning previously invalid (or crashing) syntax
to valid syntax. E.g.

"a\
b"

and

''a''\
b''

were previously syntax errors but now both result in "a\nb".

Found by afl-fuzz.
2018-03-02 17:30:48 +02:00
Will Dietz e89d02bf03 fetchGit: use "HEAD" as default ref 2018-02-28 16:34:34 -06:00
Shea Levy 14ca85688c
Actually fix nixDataDir in non-canonical path 2018-02-28 06:19:40 -05:00