Compare commits

..

256 Commits

Author SHA1 Message Date
Gregory Becker
54e5439dd6 Spec.format: conditional format strings 2023-08-22 11:22:36 -07:00
Wouter Deconinck
9f1a30d3b5 veccore: new variant vc (#39542)
* veccore: new variants umesimd and vc

* veccore: remove variant umesimd again
2023-08-22 11:20:19 -04:00
Harmen Stoppels
1340995249 clingo-bootstrap: pgo, lto, allocator optimizations (#34926)
Add support for PGO and LTO for gcc, clang and apple-clang, and add a
patch to allow mimalloc as an allocator in operator new/delete, give
reduces clingo runtime by about 30%.
2023-08-22 14:44:07 +02:00
dependabot[bot]
afebc11742 Bump sphinx from 6.2.1 to 7.2.2 in /lib/spack/docs (#39502)
Bumps [sphinx](https://github.com/sphinx-doc/sphinx) from 6.2.1 to 7.2.2.
- [Release notes](https://github.com/sphinx-doc/sphinx/releases)
- [Changelog](https://github.com/sphinx-doc/sphinx/blob/master/CHANGES)
- [Commits](https://github.com/sphinx-doc/sphinx/compare/v6.2.1...v7.2.2)

---
updated-dependencies:
- dependency-name: sphinx
  dependency-type: direct:production
  update-type: version-update:semver-major
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2023-08-22 12:02:58 +02:00
Jim Edwards
34e9fc612c parallelio: add v2.6.1 (#39479) 2023-08-22 09:44:06 +02:00
dependabot[bot]
1d8ff7f742 Bump sphinx-rtd-theme from 1.2.2 to 1.3.0 in /lib/spack/docs (#39562)
Bumps [sphinx-rtd-theme](https://github.com/readthedocs/sphinx_rtd_theme) from 1.2.2 to 1.3.0.
- [Changelog](https://github.com/readthedocs/sphinx_rtd_theme/blob/master/docs/changelog.rst)
- [Commits](https://github.com/readthedocs/sphinx_rtd_theme/compare/1.2.2...1.3.0)

---
updated-dependencies:
- dependency-name: sphinx-rtd-theme
  dependency-type: direct:production
  update-type: version-update:semver-minor
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2023-08-22 09:32:39 +02:00
Axel Huebl
0e27f05611 openPMD-api: 0.15.2 (#39528)
Latest patch release.
2023-08-21 18:31:53 -07:00
Axel Huebl
19aaa97ff2 C-Blosc2: v2.10.2 (#39537)
This fixes regressions with external dependencies introduced
in 2.10.1
2023-08-21 18:29:58 -07:00
Kamil Iskra
990309355f [bats] Update to v1.10.0 (#39561)
* [bats] Update to v1.10.0
  Original bats has been abandoned for years; nowadays it's
  community-maintained.
* [bats] Fix style
2023-08-21 18:25:35 -07:00
eugeneswalker
2cb66e6e44 tasmanian@:7.9 +cuda conflicts with cuda@12 (#39541) 2023-08-21 13:08:12 -07:00
David Gardner
cfaade098a add sundials v6.6.0 (#39526) 2023-08-21 10:48:58 -07:00
Wouter Deconinck
ed65532e27 singularityce: fix after no spaces in Executable allowed (#39553) 2023-08-21 13:43:11 -04:00
Dr Marco Claudio De La Pierre
696d4a1b85 amdblis recipe: adding variant for a performance flag (#39549)
* amdblis recipe: adding variant for a performance flag

Signed-off-by: Dr Marco De La Pierre <marco.delapierre@gmail.com>

* typo fix

Signed-off-by: Dr Marco De La Pierre <marco.delapierre@gmail.com>

* style fix

Signed-off-by: Dr Marco De La Pierre <marco.delapierre@gmail.com>

* another typo fix

Signed-off-by: Dr Marco De La Pierre <marco.delapierre@gmail.com>

* one more style fix

Signed-off-by: Dr Marco De La Pierre <marco.delapierre@gmail.com>

---------

Signed-off-by: Dr Marco De La Pierre <marco.delapierre@gmail.com>
2023-08-21 10:16:39 -07:00
Taillefumier Mathieu
8def75b414 Update cp2k recipe (#39128)
Co-authored-by: Mikael Simberg <mikael.simberg@iki.fi>
Co-authored-by: Rocco Meli <r.meli@bluemail.ch>
2023-08-21 17:09:28 +02:00
Harmen Stoppels
5389db821d aws-sdk-cpp: new versions and greatly reduce install size(#39511) 2023-08-21 14:52:21 +02:00
Wouter Deconinck
0d5ae3a809 cepgen: new version 1.1.0 (#39550) 2023-08-21 08:08:05 +02:00
Wouter Deconinck
b61ad8d2a8 whizard: new versions 3.1.1 and 3.1.2 (#39540) 2023-08-21 08:07:36 +02:00
Wouter Deconinck
b35db020eb zlib: new version 1.3 (#39539) 2023-08-21 08:06:41 +02:00
Wouter Deconinck
ca1d15101e zlib: add git url (#39533) 2023-08-21 08:04:28 +02:00
Brian Spilner
c9ec5fb9ac cdo: add v2.2.2 (#39506)
* update cdo-2.2.2

* add note on hdf5
2023-08-18 15:46:50 -07:00
Todd Gamblin
71abb8c7f0 less: update version, dependencies (#39521) 2023-08-18 15:43:54 -07:00
John W. Parent
4dafae8d17 spack.bat: Fixup CL arg parsing (#39359)
Previous changes to this file stopped directly processing CL args to
stop batch `for` from interpolating batch reserved characters needed in
arguments like URLS. In doing so, we relied on `for` for an easy
"split" operation, however this incorrectly splits paths with spaces in
certain cases. Processing everything ourselves with manual looping via
`goto` statements allows for full control over CL parsing and handling
of both paths with spaces and reserved characters.
2023-08-18 14:29:47 -07:00
Jen Herting
b2b00df5cc [py-gradio-client] New package (#39496) 2023-08-18 12:03:37 -05:00
Jen Herting
114e5d4767 [py-openmim] Beginning work (#39494) 2023-08-18 12:00:48 -05:00
Jen Herting
fd70e7fb31 [py-confection] new package (#39491) 2023-08-18 11:59:47 -05:00
George Young
77760c8ea4 py-multiqc: add 1.15, correct py-matplotlib dependency (#39509)
Co-authored-by: LMS Bioinformatics <bioinformatics@lms.mrc.ac.uk>
2023-08-18 11:56:20 -05:00
Jen Herting
737a6dcc73 py-markdown-it-py: add linkify variant and v2.2.0 (#39492)
* [py-markdown-it-py] added linkify variant

* [py-markdown-it-py] added version 2.2.0
2023-08-18 11:14:15 -05:00
Seth R. Johnson
3826fe3765 fortrilinos: release 2.3.0 compatible with trilinos@14 (#39500)
* fortrilinos: release 2.3.0 compatible with trilinos@14.0
2023-08-18 12:10:00 -04:00
Benjamin Meyers
edb11941b2 New: py-alpaca-farm, py-alpaca-eval, py-tiktoken; Updated: py-accerlate, py-transformers (#39432) 2023-08-18 10:29:12 -05:00
Axel Huebl
1bd58a8026 WarpX 23.08 (#39407)
* WarpX 23.08

Update WarpX and related Python packages to the lastest releases.

* fix style

---------

Co-authored-by: eugeneswalker <eugenesunsetwalker@gmail.com>
2023-08-18 10:03:47 -05:00
Jordan Galby
f8e0c8caed Fix Spack freeze on install child process unexpected exit (#39015)
* Fix spack frozen on child process defunct

* Rename parent/child pipe to read/write to emphasize non-duplex mode
2023-08-18 09:41:02 +02:00
Vijay M
d0412c1578 Updating the versions of available tarballs and adding an eigen3 variant as well (#39498) 2023-08-17 22:23:32 -04:00
Harmen Stoppels
ec500adb50 zlib-api: use zlib-ng +compat by default (#39358)
In the HPC package manager, we want the fastest `zlib`  implementation by default.  `zlib-ng` is up to 4x faster than stock `zlib`, and it can do things like take advantage of AVX-512 instructions.  This PR makes `zlib-ng` the default `zlib-api` provider (`zlib-api` was introduced earlier, in #37372).

As far as I can see, the only issues you can encounter are:

1. Build issues with packages that heavily rely on `zlib` internals. In Gitlab CI only one out of hundreds of packages had that issue (it extended zlib with deflate stuff, and used its own copy of zlib sources).
2. Packages that like to detect `zlib-ng` separately and rely on `zlib-ng` internals. The only issue I've found with this among the hundreds of packages built in CI is `perl` trying to report more specific zlib-ng version details, and relied on some internals that got refactored. But yeah... that warrants a patch / conflict and is nothing special.

At runtime, you cannot really have any issues, given that zlib and zlib-ng export the exact same symbols (and zlib-ng tests this in their CI).

You can't really have issues with externals when using zlib-ng either. The only type of issue is when system zlib is rather new, and not marked as external; if another external uses new symbols, and Spack builds an older zlib/zlib-ng, then the external might not find the new symbols. But this is a configuration issue, and it's not an issue caused by zlib-ng, as the same would happen with older Spack zlib.

* zlib-api: use zlib-ng +compat by default
* make a trivial change to zlib-ng to trigger a rebuild
* add `haampie` as maintainer
2023-08-17 14:03:14 -07:00
Harmen Stoppels
30f5c74614 py-flake8: bump including deps (#39426) 2023-08-17 21:02:11 +02:00
Massimiliano Culpo
713eb210ac zig: add v0.11.0 (#39484) 2023-08-17 19:18:00 +02:00
Massimiliano Culpo
a022e45866 ASP-based solver: optimize key to intermediate dicts (#39471)
Computing str(spec) is faster than computing hash(spec), and
since all the abstract specs we deal with come from user configuration
they cannot cover DAG structures that are not captured by str() but
are captured by hash()
2023-08-17 14:11:49 +02:00
eugeneswalker
82685a68d9 boost %oneapi: add cxxflags -Wno-error=enum-constexpr-conversion (#39477) 2023-08-17 07:49:20 -04:00
dependabot[bot]
b19691d503 Bump mypy from 1.5.0 to 1.5.1 in /lib/spack/docs (#39478)
Bumps [mypy](https://github.com/python/mypy) from 1.5.0 to 1.5.1.
- [Commits](https://github.com/python/mypy/compare/v1.5.0...v1.5.1)

---
updated-dependencies:
- dependency-name: mypy
  dependency-type: direct:production
  update-type: version-update:semver-patch
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2023-08-17 12:35:49 +02:00
eugeneswalker
54ea860b37 gmsh %oneapi: add cflag: -Wno-error=implicit-function-declaration (#39476) 2023-08-17 06:09:07 -04:00
eugeneswalker
fb598baa53 py-ruamel-yaml-clib %oneapi: -Wno-error=incompatible-function-pointer-types (#39480) 2023-08-17 11:59:45 +02:00
eugeneswalker
02763e967a sz %oneapi: add cflag=-Wno-error=implicit-function-declaration (#39467) 2023-08-17 02:36:17 -07:00
Seth R. Johnson
2846be315b kokkos: use 'when' instead of 'conflicts' for CUDA variants (#39463) 2023-08-17 05:12:52 -04:00
eugeneswalker
4818b75814 silo %oneapi: add cflags: -Wno-error=int, -Wno-error=int-conversion (#39475) 2023-08-17 04:42:58 -04:00
eugeneswalker
b613bf3855 py-matplotlib %oneapi: add cxxflags=-Wno-error=register (#39469) 2023-08-17 04:13:09 -04:00
Adam J. Stewart
3347372a7b py-deepspeed: add new package (#39427) 2023-08-17 08:47:40 +02:00
Sergey Kosukhin
c417a77a19 snappy: patch and conflict for %nvhpc (#39063) 2023-08-17 08:44:38 +02:00
Martin Aumüller
90d0d0176c botan: version 3 requires newer GCC (#39450)
e.g. 3.1.1 produces this during configuration when trying to install:
  ERROR: This version of Botan requires at least gcc 11.0
2023-08-17 08:34:43 +02:00
Adam J. Stewart
72b9f89504 py-setuptools: document Python 3.12 support (#39449) 2023-08-17 08:31:53 +02:00
Peter Scheibel
a89f1b1bf4 Add debugging statements to file search (#39121)
Co-authored-by: Scheibel <scheibel1@ml-9983616.the-lab.llnl.gov>
2023-08-17 08:31:06 +02:00
Adam J. Stewart
c6e26251a1 py-lightning: add v2.0.7 (#39468) 2023-08-17 08:19:50 +02:00
Harmen Stoppels
190a1bf523 Delay abstract hashes lookup (#39251)
Delay lookup for abstract hashes until concretization time, instead of
until Spec comparison. This has a few advantages:

1. `satisfies` / `intersects` etc don't always know where to resolve the
   abstract hash (in some cases it's wrong to look in the current env,
   db, buildcache, ...). Better to let the call site dictate it.
2. Allows search by abstract hash without triggering a database lookup,
   causing quadratic complexity issues (accidental nested loop during
   search)
3. Simplifies queries against the buildcache, they can now use Spec
   instances instead of strings.

The rules are straightforward:

1. a satisfies b when b's hash is prefix of a's hash
2. a intersects b when either a's or b's hash is a prefix of b's or a's
   hash respectively
2023-08-17 08:08:50 +02:00
eugeneswalker
e381e166ec py-numcodecs %oneapi: add cflags -Wno-error=implicit-function-declaration (#39454) 2023-08-16 15:52:01 -05:00
eugeneswalker
2f145b2684 pruners-ninja cflags: -Wno-error=implicit-function-declaration (#39452) 2023-08-16 15:51:38 -05:00
eugeneswalker
4c7748e954 amrex+sycl: restructure constraint on %oneapi (#39451) 2023-08-16 11:49:32 -06:00
eugeneswalker
86485dea14 hdf5-vol-cache %oneapi: cflags: add -Wno-error=incompatible-function-pointer-types (#39453) 2023-08-16 09:56:09 -06:00
Todd Kordenbrock
00f8f5898a faodel: update github URL organization to sandialabs (#39446) 2023-08-16 07:25:47 -07:00
Massimiliano Culpo
f41d7a89f3 Extract Package from PackageNode for error messages 2023-08-16 06:20:57 -07:00
Harmen Stoppels
4f07205c63 Avoid sort on singleton list during edge insertion (#39458)
The median length of this list of 1. For reasons I don't know, `.sort()`
still like to call the key function.

This saves ~9% of total database read time, and the number of calls
goes from 5305 -> 1715.
2023-08-16 14:33:03 +02:00
Massimiliano Culpo
08f9c7670e Do not impose provider conditions, if the node is not a provider (#39456)
* Do not impose provider conditions, if the node is not a provider

fixes #39455

When a node can be a provider of a spec, but is not selected as
a provider, we should not be imposing provider conditions on the
virtual.

* Adjust the integrity constraint, by using the correct atom
2023-08-16 05:15:13 -07:00
Harmen Stoppels
b451791336 json: minify by default (#39457) 2023-08-16 11:26:40 +00:00
Massimiliano Culpo
47f176d635 Add new custom markers to unit tests (#33862)
* Add "only_clingo", "only_original" and "not_on_windows" markers

* Modify tests to use the "not_on_windows" marker

* Mark tests that run only with clingo

* Mark tests that run only with the original concretizer
2023-08-16 09:04:10 +02:00
Patrick Bridges
b6ae751657 Fixed HeFFTe package spec to not do the smoke test prior to 2.2.0 (#39435)
* Fixed HeFFTe package spec to not do the smoke test prior to 2.2.0, where it breaks
* Convert test return to 'raise SkipTest'

---------

Co-authored-by: Tamara Dahlgren <35777542+tldahlgren@users.noreply.github.com>
2023-08-15 21:03:14 -04:00
Massimiliano Culpo
9bb5cffc73 Change semantic for providers
If a possible provider is not used to satisfy a vdep,
then it's not a provider of that vdep.
2023-08-15 15:54:37 -07:00
Massimiliano Culpo
135b44ca59 Change "effect_rules" for symmetry with trigger rules
This even though right now we don't have cases where
the effect is on another package.
2023-08-15 15:54:37 -07:00
Massimiliano Culpo
d3aca68e8f Rework conflicts so that "vendors" is not needed anymore 2023-08-15 15:54:37 -07:00
Massimiliano Culpo
fb83f8ef31 Add a description at the top of lp files 2023-08-15 15:54:37 -07:00
Massimiliano Culpo
f69c18a922 Remove commented out code in lp files 2023-08-15 15:54:37 -07:00
Massimiliano Culpo
b95a9d2e47 Reduce line length in lp file 2023-08-15 15:54:37 -07:00
Massimiliano Culpo
def4d19980 Demote warning to debug message 2023-08-15 15:54:37 -07:00
Massimiliano Culpo
1db91e0ccd Rename "main_node" -> "make_node" 2023-08-15 15:54:37 -07:00
Massimiliano Culpo
34ebe7f53c Rename ""*_node" -> "*_dupe" 2023-08-15 15:54:37 -07:00
Massimiliano Culpo
d07d5410f3 Rename "stringify", improve docs 2023-08-15 15:54:37 -07:00
Massimiliano Culpo
1db73eb1f2 Add vendors directive
For the time being this directive prevents the vendored package
to be in the same DAG as the one vendoring it.
2023-08-15 15:54:37 -07:00
Massimiliano Culpo
2da34de519 Add "^" automatically for named conflicts that don't refer to 'this' package
See https://github.com/spack/spack/pull/38447#discussion_r1285291520
2023-08-15 15:54:37 -07:00
Massimiliano Culpo
d237430f47 Inline a few functions that are not needed anymore 2023-08-15 15:54:37 -07:00
Massimiliano Culpo
3f0adae9ef Remove the need for "node_regex" 2023-08-15 15:54:37 -07:00
Massimiliano Culpo
3b4d7bf119 Rename method: "root_node" -> "main_node" 2023-08-15 15:54:37 -07:00
Massimiliano Culpo
b3087b32c6 Rename const: "root_node_id" -> "main_node_id" 2023-08-15 15:54:37 -07:00
Massimiliano Culpo
ad9c90cb2e Rename atom: "special_case" -> "multiple_nodes_attribute" 2023-08-15 15:54:37 -07:00
Massimiliano Culpo
1b0e113a9d Rename atom: "facts" -> "pkg_fact" 2023-08-15 15:54:37 -07:00
Massimiliano Culpo
6df5738482 Simplify "node_has_variant" internal atom. 2023-08-15 15:54:37 -07:00
Massimiliano Culpo
927d831612 Removed leftover TODOs 2023-08-15 15:54:37 -07:00
Massimiliano Culpo
3f3c75e56a ecp-data-viz-sdk: fix building with new encoding 2023-08-15 15:54:37 -07:00
Massimiliano Culpo
9733bb3da8 We cannot require "mpich" as an mpi provider and ask for openmpi 2023-08-15 15:54:37 -07:00
Massimiliano Culpo
1de5117ef1 Improve handling of cases with cycles
To avoid paying the cost of setup and of a full grounding again,
move cycle detection into a separate program and check first if
the solution has cycles.

If it has, ground only the integrity constraint preventing cycles
and solve again.
2023-08-15 15:54:37 -07:00
Massimiliano Culpo
cf8f44ae5a Fix ecp-data-vis-sdk: a package cannot impose constraints on transitive deps 2023-08-15 15:54:37 -07:00
Massimiliano Culpo
006e69265e Optimize grounding of "can_inherit_flags" 2023-08-15 15:54:37 -07:00
Massimiliano Culpo
eaec3062a1 Fix computation of max nodes 2023-08-15 15:54:37 -07:00
Massimiliano Culpo
d5eb5106b0 Add unit-tests for use cases requiring separate concretization of build deps 2023-08-15 15:54:37 -07:00
Massimiliano Culpo
9f8edbf6bf Add a new configuration option to select among different concretization modes
The "concretizer" section has been extended with a "duplicates:strategy"
attribute, that can take three values:

- "none": only 1 node per package
- "minimal": allow multiple nodes opf specific packages
- "full": allow full duplication for a build tool
2023-08-15 15:54:37 -07:00
Massimiliano Culpo
a4301badef Fix a few bugs in the encoding when imposing constraints on build deps only
These bugs would show up when we try to split nodes by
imposing different targets or different compilers to all
build dependencies.
2023-08-15 15:54:37 -07:00
Massimiliano Culpo
4565811556 Construct unification sets on demand, improve heuristic 2023-08-15 15:54:37 -07:00
Massimiliano Culpo
b94d54e4d9 Reduce the number of unification sets to only two 2023-08-15 15:54:37 -07:00
Massimiliano Culpo
a410b22098 Make cycle detection optional, to speed-up grounding and solving 2023-08-15 15:54:37 -07:00
Massimiliano Culpo
c1a73878ea Deduplicate trigger and effect conditions in packages
This refactor introduces extra indices for triggers and
effect of a condition, so that the corresponding clauses
are evaluated once for every condition they apply to.
2023-08-15 15:54:37 -07:00
Massimiliano Culpo
ae553051c8 Extract a function to emit variant rules 2023-08-15 15:54:37 -07:00
Massimiliano Culpo
b94e22b284 ASP-based solver: do not optimize on known dimensions
All the solution modes we use imply that we have to solve for all
the literals, except for "when possible".

Here we remove a minimization on the number of literals not
solved, and emit directly a fact when a literal *has* to be
solved.
2023-08-15 15:54:37 -07:00
Massimiliano Culpo
e25dcf73cd Tweak a unit test by allowing a different type of exception to be raised 2023-08-15 15:54:37 -07:00
Massimiliano Culpo
b7cc4bd247 Reduce the dependency types in a solve
Count the maximum number of nodes based on dependency types
2023-08-15 15:54:37 -07:00
Massimiliano Culpo
22c95923e3 Parametrize all the logic program for multiple nodes
Introduce the concept of "condition sets", i.e. the set of packages on which
a package can require / impose conditions. This currently maps to the link/run
sub-dag of each package + its direct build dependencies.

Parametrize the "condition" and "requirement" logic to multiple nodes.
2023-08-15 15:54:37 -07:00
Massimiliano Culpo
c050b99a06 Introduce unification sets
Unification sets are possibly overlapping sets of nodes that
might contain at most a single configuration for any package.
2023-08-15 15:54:37 -07:00
Massimiliano Culpo
60f82685ae Allow clingo to generate edges 2023-08-15 15:54:37 -07:00
Massimiliano Culpo
27ab53b68a Rework the encoding to introduce node(ID, Package) nested facts
So far the encoding has a single ID per package, i.e. all the
facts will be node(0, Package). This will prepare the stage for
extending this logic and having multiple nodes from the same
package in a DAG.
2023-08-15 15:54:37 -07:00
Massimiliano Culpo
907a80ca71 Remove unneeded #defined directives 2023-08-15 15:54:37 -07:00
Massimiliano Culpo
a53cc93016 Remove useless rule
The version_equivalent fact was deleted in #36347,
but the corresponding rule was not removed.
2023-08-15 15:54:37 -07:00
Massimiliano Culpo
6ad0dc3722 Transform many package related facts to use a nested function
Each fact that is deduced from package rules, and start with
a bare package atom, is transformed into a "facts" atom containing
a nested function.

For instance we transformed

  version_declared(Package, ...) -> facts(Package, version_declared(...))

This allows us to clearly mark facts that represent a rule on the package,
and will be of help later when we'll have to distinguish the cases where
the atom "Package" is being used referred to package rules and not to a
node in the DAG.
2023-08-15 15:54:37 -07:00
afzpatel
87d4bdaa02 adding new rpp package (#38942)
* initial commit for adding new rpp package
* fix styling
2023-08-15 14:04:35 -07:00
Mikael Simberg
36394aab2f Add mold 2.1.0 (#39443) 2023-08-15 09:50:47 -07:00
Larry Knox
358947fc03 Add version HDF5 1.14.2 (#39409)
* Add version HDF5 1.14.2
2023-08-15 11:21:21 -05:00
Andrey Prokopenko
477a3c0ef6 arborx: new version and patch for 1.4 with Trilinos (#39438)
* arborx: patch 1.4 for Trilinos 14.0

* arborx: version 1.4.1
2023-08-15 06:46:41 -07:00
Mikael Simberg
c6c5e11353 Add gperftools 2.11 (#39440) 2023-08-15 10:41:12 +02:00
Massimiliano Culpo
29e2997bd5 spack.caches: uppercase global variables (#39428) 2023-08-15 09:59:02 +02:00
Bryan Herman
41bd6a75d5 squashfs: Add static variant (#39433) 2023-08-15 09:54:27 +02:00
Harmen Stoppels
0976ad3184 spack.config: use all caps for globals (#39424) 2023-08-15 08:19:36 +02:00
Mark W. Krentel
fc1d9ba550 intel-xed: add version 2023.07.09 (#39436) 2023-08-14 22:33:10 -04:00
Richard Berger
61f0088a27 Spiner and ports-of-call updates (#39437)
* ports-of-call: new version 1.5.2
* spiner: new version 1.6.2
2023-08-14 22:22:52 -04:00
markus-ferrell
c202a045e6 Windows: executable/path handling (#37762)
Windows executable paths can have spaces in them, which was leading to
errors when constructing Executable objects: the parser was intended
to handle cases where users could provide an executable along with one
or more space-delimited arguments.

* Executable now assumes that it is constructed with a string argument
  that represents the path to the executable, but no additional arguments.
* Invocations of Executable.__init__ that depended on this have been
  updated (this includes the core, tests, and one instance of builtin
  repository package).
* The error handling for failed invocations of Executable.__call__ now
  includes a check for whether the executable name/path contains a
  space, to help users debug cases where they (now incorrectly)
  concatenate the path and the arguments.
2023-08-14 23:29:12 +00:00
Bryan Herman
843e1e80f0 fuse-overlayfs needs pkgconfig to build (#39431) 2023-08-14 19:19:07 -04:00
Paul Kuberry
643c028308 xyce: Disable CMake test for OneAPI (#39429) 2023-08-14 13:27:33 -07:00
markus-ferrell
d823037c40 Windows: enable "spack install" tests (#34696)
* The module-level skip for tests in `cmd.install` on Windows is removed.
  A few classes of errors still persist:

  * Cdash tests are not working on Windows
  * Tests for failed installs are also not working (this will require
    investigating bugs in output redirection)
  * Environments are not yet supported on Windows

  overall though, this enables testing of most basic uses of "spack install"
* Git repositories cached for version lookups were using a layout that
  mimicked the URL as much as possible. This was useful for listing the
  cache directory and understanding what was present at a glance, but
  the paths were overly long on Windows. On all systems, the layout is
  now a single directory based on a hash of the Git URL and is shortened
  (which ensures a consistent and acceptable length, and also avoids
  special characters).
  * In particular, this removes util.url.parse_git_url and its associated
    test, which were used exclusively for generating the git cache layout
* Bootstrapping is now enabled for unit tests on Windows
2023-08-14 13:15:40 -07:00
Bryan Herman
4d945be955 fixed build of libfuse need werror (#39430) 2023-08-14 14:39:40 -04:00
eugeneswalker
a4ac3f2767 trilinos@14.4.0 +kokkos: use external kokkos (#39397) 2023-08-14 14:29:13 -04:00
Massimiliano Culpo
6e31676b29 Fix style issues with latest versions of tools (#39422) 2023-08-14 12:38:59 -04:00
Mikael Simberg
1fff0241f2 Add Boost 1.83.0 (#39415) 2023-08-14 12:28:44 -04:00
Harmen Stoppels
a2a52dfb21 Fix containerize view symlink issue (#39419) 2023-08-14 16:02:48 +00:00
Mikael Simberg
f0ed159a1b Add patch to build silo on nixos (#39375)
Change the shebang in mkinc from /usr/bin/perl to /usr/bin/env perl for
portability to systems that don't necessarily have perl in /usr/bin.
Also adds perl as a build-time dependency.
2023-08-14 17:03:13 +02:00
Mikael Simberg
9bf7fa0067 Add fmt 10.1.0 (#39413) 2023-08-14 17:01:34 +02:00
Mikael Simberg
fbaea0336e mold: add v2.0.0 (#39416) 2023-08-14 10:53:34 -04:00
Brian Van Essen
1673d3e322 Added package for the hipTT library (#39388) 2023-08-14 16:11:30 +02:00
Wouter Deconinck
c7cca3aa8d dd4hep: new version 1.26 (#39230)
Release notes: https://github.com/AIDASoft/DD4hep/releases/tag/v01-26

Diff: https://github.com/AIDASoft/DD4hep/compare/v01-25-01...v01-26

Relevant changes:
- updated requirements on cmake,
- can now support compressed hepmc3.

---------

Co-authored-by: wdconinc <wdconinc@users.noreply.github.com>
Co-authored-by: Massimiliano Culpo <massimiliano.culpo@gmail.com>
2023-08-14 15:56:40 +02:00
Harmen Stoppels
da46b63a34 Fix broken semver regex (#39414) 2023-08-14 15:38:06 +02:00
Harmen Stoppels
c882214273 spack bootstrap dev: detect git as an external (#39417)
#36770 added git as a dependency to `setuptools-scm`. This in turn makes `git` a
transitive dependency for our bootstrapping process. 

Since `git` may take a long time to build, and is found on most systems, try to 
detect it as an external.
2023-08-14 10:29:15 +00:00
Pariksheet Nanda
2bacab0402 arrow: fix conditional spec expression (#38212) (#38213) 2023-08-14 09:27:50 +02:00
Hao Lyu
0681d9a157 wrf: add v4.5.1 (#39291) 2023-08-14 09:20:56 +02:00
Audrius Kalpokas
887847610e gromacs: add plumed 2.8.3-2.9.0 variants (#39411) 2023-08-14 09:17:16 +02:00
Pariksheet Nanda
282a01ef76 py-scikits-odes: add a new package (#39061) 2023-08-14 02:00:03 +00:00
Adam J. Stewart
151c551781 Python: fix Apple Clang +optimizations build (#39412) 2023-08-13 21:47:47 -04:00
Hariharan Devarajan
abbd1abc1a release 0.0.1 py-dlio-profiler-py (#39377) 2023-08-13 12:20:58 -05:00
Wouter Deconinck
49c505cc14 py-pyliblzma: remove python2 package (#39399)
Last release 13 years ago, before python 3.2 was released. No depedendents or next of kin could be located.
2023-08-13 11:57:52 -05:00
Sergey Kosukhin
237a56a305 autotools: set 'ldlibs' as 'LIBS' (#17254) 2023-08-13 00:38:10 -07:00
Harmen Stoppels
7e7e6c2797 bazel: pretty print is all but pretty (#35963)
* bazel: pretty print is all but pretty
* Don't ask bazel to explain what it's up to
2023-08-12 20:43:09 -04:00
Alex Richert
e67c61aac0 Allow static-only openssl on Linux (#38098) 2023-08-12 20:42:53 +02:00
Scott Wittenburg
1b1ed1b1fa ci: continue to support SPACK_SIGNING_KEY (#39170) 2023-08-12 08:52:46 -07:00
snehring
ec0e51316b usalign: update checksum (#39400) 2023-08-12 08:40:44 -07:00
Vicente Bolea
533821e46f adios2: add mgard variant (#39405) 2023-08-12 08:38:33 -07:00
Wouter Deconinck
6c5d125cb0 cernlib: new variant shared (#39406)
This allows users to install cernlib with static libraries only, instead of both static and shared.
2023-08-12 08:34:48 -07:00
Richard Berger
668fb1201f lammps: new stable versions, deprecate older patch versions (#39317) 2023-08-12 09:57:38 +02:00
Adam J. Stewart
f7918fd8ab Python: remove maintainer (#39384) 2023-08-12 09:08:25 +02:00
Axel Huebl
fc1996e0fa WarpX 23.07 (#39108)
* WarpX 23.07

Update WarpX and related Python packages to the lastest releases.

* py-picmistandard: refresh hashes

Keeping even a single `git` version in here confuses the fetcher.
Remove broken old versions.

* Remove `py-warpx` versions with broken PICMI
2023-08-11 17:35:04 -07:00
eugeneswalker
ed3aaafd73 e4s cray sles ci: expand spec list (#39081)
* e4s cray sles ci: expand spec list

* flux-core: disable until python+gary-sles issue is resolved
2023-08-11 23:42:10 +00:00
Wouter Deconinck
63bb2c9bad py-cryptography: does not run-depend on py-setuptools-rust (#39386)
* py-cryptography: does not run-depend on py-setuptools-rust

* py-cryptography: depens_on py-setuptools-rust when @3.4.2:

Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>

* py-cryptography: re-add depends_on type=run for narrow range

---------

Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
2023-08-11 17:53:25 +00:00
Alex Richert
a67455707a ncl: add support for grib (#39277)
Co-authored-by: Alex Richert <alexander.richert@noaa.gov>
2023-08-11 08:42:08 +00:00
Cristian Di Pietrantonio
09ca71dbe0 kokkos: rename 'std' variant to 'cxxstd'. (#39319)
Co-authored-by: Cristian Di Pietrantonio <cdipietrantonio@pawsey.org.au>
2023-08-11 07:18:02 +00:00
Todd Gamblin
ea082539e4 Revert "package import: remove magic import line (#39183)" (#39380)
This reverts commit 8e7c53a8ba.
2023-08-11 09:04:40 +02:00
Massimiliano Culpo
143146f4f3 spack.repo: uppercase the global PATH variable (#39372)
This makes the name of the global variable representing
the repository currently in use uppercase. Doing so is advised
by pylint rules, and helps to identify where the global is used.
2023-08-11 09:04:16 +02:00
Melven Roehrig-Zoellner
ee6ae402aa qt: fedora patch missing includes in qtlocation (#38987) 2023-08-11 08:10:06 +02:00
Adam J. Stewart
0b26b26821 GDAL: fix supported versions of HDF5 (#39378) 2023-08-11 08:02:15 +02:00
dependabot[bot]
c764f9b1ab build(deps): bump mypy from 1.4.1 to 1.5.0 in /lib/spack/docs (#39383)
Bumps [mypy](https://github.com/python/mypy) from 1.4.1 to 1.5.0.
- [Commits](https://github.com/python/mypy/compare/v1.4.1...v1.5.0)

---
updated-dependencies:
- dependency-name: mypy
  dependency-type: direct:production
  update-type: version-update:semver-minor
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2023-08-11 08:01:34 +02:00
William R Tobin
db19d83ea7 essl: add +lapackforessl variant (#39362) 2023-08-11 08:01:02 +02:00
Gregor Daiß
24256be6d6 sgpp: fix build issues (#39086)
* fix(sgpp): Fix installation phase scons args

* fix(sgpp): Workaround for distutils deprecation

The distutils deprecation warning in Python 3.10 - 3.11 caused
problems within the SGpp SConfigure checks by causing failures
when looking for Python.h. This commit works around this by adding a
patch that simply disables the warning. It also puts limits on the
python dependency version until distutils is removed from SGpp.

* fix(sgpp): cleanup and simplify

* fix(sgpp): Fix style
2023-08-11 07:55:44 +02:00
Sergey Kosukhin
633723236e mpich: fix building with NVHPC (#39082) 2023-08-11 07:29:16 +02:00
Yan
381f31e69e py-flit: add 3.8 and 3.9 (#39307) 2023-08-10 20:23:58 -04:00
Massimiliano Culpo
9438cac219 Fix conflicts on direct dependencies that don't start with ^ (#39369) 2023-08-10 16:58:05 -04:00
Harmen Stoppels
85cf66f650 git: add new versions, deprecated / drop versions (#39368)
Remove git versions deprecated in Spack 0.19
2023-08-10 21:12:23 +02:00
markus-ferrell
f3c080e546 Windows build systems: use ninja and enable tests (#33589)
* Set default CMake generator is ninja on Windows
* Enable build systems tests (except for autotools/make)
2023-08-10 17:16:23 +00:00
dependabot[bot]
37634f8b08 build(deps): bump pygments from 2.15.1 to 2.16.1 in /lib/spack/docs (#39365)
Bumps [pygments](https://github.com/pygments/pygments) from 2.15.1 to 2.16.1.
- [Release notes](https://github.com/pygments/pygments/releases)
- [Changelog](https://github.com/pygments/pygments/blob/master/CHANGES)
- [Commits](https://github.com/pygments/pygments/compare/2.15.1...2.16.1)

---
updated-dependencies:
- dependency-name: pygments
  dependency-type: direct:production
  update-type: version-update:semver-minor
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2023-08-10 18:10:32 +02:00
ddeeptimahanti
2ae8bbce9e trilinos: fix @13.2: +python (#39370) 2023-08-10 11:50:33 -04:00
William R Tobin
b8bfaf65bf intel-oneapi-mkl: linking with the +cluster variant is broken for external mkl (#39343)
* Update package.py

Adding `mpi` variant to deal with external oneapi-mkl usage per #38238.

* style conformance

* accept both mpi specification mechanisms

* style conformance

* Update package.py

* Update var/spack/repos/builtin/packages/intel-oneapi-mkl/package.py

Co-authored-by: Robert Cohn <rscohn2@gmail.com>

* rename mpi variant mpi_family

* style conformance

* update help message for mpi_family

---------

Co-authored-by: Robert Cohn <rscohn2@gmail.com>
2023-08-10 10:33:29 -04:00
Alberto Invernizzi
7968cb7fa2 mpich: fix macos problem with -flat-namespace (#35611)
Co-authored-by: Harmen Stoppels <harmenstoppels@gmail.com>
2023-08-10 08:34:33 -04:00
Nicholas Sly
ebc2efdfd2 Add LLVM_ENABLE_RTTI option to llvm-amdgpu package, required by mesa. (#35655) 2023-08-10 07:14:02 -04:00
Harmen Stoppels
ff07fd5ccb Remove qt variant in cmake (#39360)
`cmake+qt` depends on `qt`, which depends on `libmng`, which is a CMake
package, and has been for 4 years. Nobody ever complained about
`cmake+qt` not concretizing... so why pay the solve cost.


Before:

```
    setup          3.779s
    load           0.018s
    ground         2.625s
    solve          4.511s
    total         11.236s
```

After:

```
    setup          3.734s
    load           0.018s
    ground         0.468s
    solve          0.560s
    total          5.080s
```
2023-08-10 05:59:22 -04:00
dependabot[bot]
3f83ef6566 build(deps): bump flake8 from 6.0.0 to 6.1.0 in /lib/spack/docs (#39366)
Bumps [flake8](https://github.com/pycqa/flake8) from 6.0.0 to 6.1.0.
- [Commits](https://github.com/pycqa/flake8/compare/6.0.0...6.1.0)

---
updated-dependencies:
- dependency-name: flake8
  dependency-type: direct:production
  update-type: version-update:semver-minor
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2023-08-10 05:34:33 -04:00
dependabot[bot]
554ce7f063 build(deps): bump sphinx-design from 0.4.1 to 0.5.0 in /lib/spack/docs (#39367)
Bumps [sphinx-design](https://github.com/executablebooks/sphinx-design) from 0.4.1 to 0.5.0.
- [Release notes](https://github.com/executablebooks/sphinx-design/releases)
- [Changelog](https://github.com/executablebooks/sphinx-design/blob/main/CHANGELOG.md)
- [Commits](https://github.com/executablebooks/sphinx-design/compare/v0.4.1...v0.5.0)

---
updated-dependencies:
- dependency-name: sphinx-design
  dependency-type: direct:production
  update-type: version-update:semver-minor
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2023-08-10 05:29:52 -04:00
Aiden Grossman
23963779f4 Prefix conflict messages with package name (#39106)
* Prefix conflict messages with package name

This patch prefixes all conflict messages with the package name to
alleviate what was otherwise a very manual process. Note that this patch
is a one line change but has a fairly outsized impact.

* same for requires directive

---------

Co-authored-by: Harmen Stoppels <me@harmenstoppels.nl>
2023-08-10 08:09:00 +00:00
Ricard Zarco Badia
45c5af10c3 Fixed broken calls to _if_ninja_target_execute (#38992)
Co-authored-by: Tamara Dahlgren <35777542+tldahlgren@users.noreply.github.com>
2023-08-10 09:34:02 +02:00
Harmen Stoppels
532a37e7ba Revert "Spec versions: allow git. references for branches with / (#38239)" (#39354)
This reverts commit 3453259c98.
2023-08-10 08:56:39 +02:00
Cristian Di Pietrantonio
aeb9a92845 ncview: change URL for source code (#39321)
Co-authored-by: Cristian Di Pietrantonio <cdipietrantonio@pawsey.org.au>
2023-08-10 08:38:58 +02:00
dependabot[bot]
a3c7ad7669 build(deps): bump urllib3 from 2.0.3 to 2.0.4 in /lib/spack/docs (#39002)
Bumps [urllib3](https://github.com/urllib3/urllib3) from 2.0.3 to 2.0.4.
- [Release notes](https://github.com/urllib3/urllib3/releases)
- [Changelog](https://github.com/urllib3/urllib3/blob/main/CHANGES.rst)
- [Commits](https://github.com/urllib3/urllib3/compare/2.0.3...2.0.4)

---
updated-dependencies:
- dependency-name: urllib3
  dependency-type: direct:production
  update-type: version-update:semver-patch
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2023-08-10 05:04:58 +00:00
Juan Miguel Carceller
b99288dcae edm4hep: add edm4hep to PYTHONPATH (#38407)
Co-authored-by: jmcarcell <jmcarcell@users.noreply.github.com>
2023-08-09 18:22:17 -05:00
markus-ferrell
01b7cc5106 Windows: enable more stage tests (#36834)
Enable some already-supported tests (no changes to staging logic).
2023-08-09 14:24:30 -07:00
pabloaledo
f5888d8127 picard: add variant with jvm arguments (#39204)
* picard: add variant with jvm arguments
* fix variant and update wrapper script

---------

Signed-off-by: Pablo <pablo.aledo@seqera.io>
2023-08-09 13:24:34 -07:00
Jennifer Green
77c838ca93 libhio: package update for new sha256 checksum for v1.4.1.5 (#38696) 2023-08-09 12:10:37 -07:00
Alec Scott
11e538d962 emacs: Add v29.1 (#39147) 2023-08-09 11:32:04 -07:00
Axel Huebl
7d444038ee AMReX: 23.06+ Multi-Dim Support (#37695)
* AMReX: 23.06+ Multi-Dim Support

This updated the Spack package to allow to install AMReX, modules of
AMReX in E4S deployments and dependent packages with support for
multiple dimensions. Due to an upstream change in AMReX, we do not
longer need to ship three, binary incompatible package variants.

* [E4S] oneAPI AMReX < 23.06 Variant

Work-around the auto-concretization to the multi-dim of `dimensions`,
which only in 23.06+ became a multi-variant.

* e4s cray rhel ci: temporarily disable amrex build until spurious ci failure can be resolved

---------

Co-authored-by: eugeneswalker <eugenesunsetwalker@gmail.com>
2023-08-09 11:25:43 -07:00
Hariharan Devarajan
c24471834b Release brahma 0.0.1 (#39345)
* release brahma 0.0.1
2023-08-09 11:18:52 -07:00
Andrey Parfenov
b1e33ae37b add info about devito and qe for intel env (#39357)
Signed-off-by: Andrey Parfenov <andrey.parfenov@intel.com>
2023-08-09 12:47:35 -04:00
Jordan Galby
c36617f9da Fix package.py error handling bug (#39017) 2023-08-09 17:47:43 +02:00
dependabot[bot]
deadb64206 build(deps): bump black from 23.1.0 to 23.7.0 in /lib/spack/docs (#38982)
Bumps [black](https://github.com/psf/black) from 23.1.0 to 23.7.0.
- [Release notes](https://github.com/psf/black/releases)
- [Changelog](https://github.com/psf/black/blob/main/CHANGES.md)
- [Commits](https://github.com/psf/black/compare/23.1.0...23.7.0)

---
updated-dependencies:
- dependency-name: black
  dependency-type: direct:production
  update-type: version-update:semver-minor
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2023-08-09 17:19:49 +02:00
Massimiliano Culpo
9eaa88e467 caliper: deprecate old versions (#39352) 2023-08-09 17:11:57 +02:00
Juan Miguel Carceller
bd58801415 dd4hep: fix setting LD_LIBRARY_PATH (#38015)
Co-authored-by: jmcarcell <jmcarcell@users.noreply.github.com>
2023-08-09 16:42:05 +02:00
Juan Miguel Carceller
548a9de671 py-setuptools-scm: add a git dependency (#36770)
* Add a git dependency

* Add the dependency only for building

* Change dependency to build and run

Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>

---------

Co-authored-by: jmcarcell <jmcarcell@users.noreply.github.com>
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
2023-08-09 10:38:46 -04:00
Harmen Stoppels
8e7c53a8ba package import: remove magic import line (#39183) 2023-08-09 14:28:06 +00:00
Tamara Dahlgren
5e630174a1 cpmd: convert to new stand-alone test process (#35744) 2023-08-09 15:59:51 +02:00
Cristian Di Pietrantonio
175a65dfba libzmq: do not treat warnings as errors during compilation. (#39325)
The package won't compile with newer compilers because warnings
are converted to errors. Hence, disable such conversion.

Co-authored-by: Cristian Di Pietrantonio <cdipietrantonio@pawsey.org.au>
2023-08-09 15:57:43 +02:00
Adam J. Stewart
39d4c402d5 Fix suffix of tab completion scripts (#39154) 2023-08-09 08:28:55 -05:00
Harmen Stoppels
e51748ee8f zlib-api: new virtual with zlib/zlib-ng as providers (#37372)
Introduces a new virtual zlib-api, which replaces zlib in most packages.

This allows users to switch to zlib-ng by default for better performance.
2023-08-09 09:22:58 -04:00
Alec Scott
f9457fa80b cxxopts: add v3.1.1 (#37977) 2023-08-09 14:31:10 +02:00
Alec Scott
4cc2ca3e2e rclone: add v1.63.1 and remove previously deprecated versions (#39148) 2023-08-09 14:30:33 +02:00
Massimiliano Culpo
3843001004 Fixed bugs discovered in conflicts directives (#39338) 2023-08-09 14:20:49 +02:00
Cristian Di Pietrantonio
e24bb5dd1c Changes the way MPICC is set (#39327) 2023-08-09 07:04:04 -05:00
Cristian Di Pietrantonio
f6013114eb nwchem: make sure to link both single and double precision fftw libraries. (#39333)
With the previous version of the recipe I would get linking errors due
to missing `-lfftw3f`.

Co-authored-by: Cristian Di Pietrantonio <cdipietrantonio@pawsey.org.au>
2023-08-09 14:03:15 +02:00
Martin Aumüller
bdca875eb3 protobuf: fix build for 3.21 on Centos 8 (#39162)
apply upstream patch
2023-08-09 13:44:05 +02:00
pabloaledo
af8c392de2 fq: add new package (#39205)
Signed-off-by: Pablo <pablo.aledo@seqera.io>
2023-08-09 06:23:44 -04:00
Jonathon Anderson
9aa3b4619b containerize: ensure bootstrap images contain all system dependencies (#36818)
This also makes `spack bootstrap status` exit 1 if some dependency is missing
2023-08-09 09:46:59 +02:00
Cristian Di Pietrantonio
3d733da70a idg: add new package (#39316)
Co-authored-by: Cristian Di Pietrantonio <cdipietrantonio@pawsey.org.au>
2023-08-09 02:45:34 -05:00
Cristian Di Pietrantonio
cda99b792c miniocli: add new package (#39318)
Co-authored-by: Cristian Di Pietrantonio <cdipietrantonio@pawsey.org.au>
Co-authored-by: Massimiliano Culpo <massimiliano.culpo@gmail.com>
2023-08-09 03:43:29 -04:00
Gerhard Theurich
9834bad82e esmf: specify build dependency on CMake when using internal ParallelIO (#39347) 2023-08-09 09:14:28 +02:00
Peter Scheibel
3453259c98 Spec versions: allow git. references for branches with / (#38239) 2023-08-09 09:07:04 +02:00
Adam J. Stewart
ee243b84eb py-lightly: add v1.4.15 (#39348) 2023-08-09 09:01:41 +02:00
Adam J. Stewart
5080e2cb45 py-torchmetrics: add v1.0.3 (#39349) 2023-08-09 09:01:23 +02:00
Cristian Di Pietrantonio
f42ef7aea7 ior: add lustre variant (#39320)
Co-authored-by: Cristian Di Pietrantonio <cdipietrantonio@pawsey.org.au>
2023-08-09 08:59:21 +02:00
Jon Rood
41793673d9 hypre: fix typo in --with-sycl option. (#39351) 2023-08-09 08:37:15 +02:00
Cristian Di Pietrantonio
8b6a6982ee caliper: add KOKKOS variant. (#39312)
Co-authored-by: Cristian Di Pietrantonio <cdipietrantonio@pawsey.org.au>
Co-authored-by: Massimiliano Culpo <massimiliano.culpo@gmail.com>
2023-08-09 08:35:56 +02:00
Cristian Di Pietrantonio
ee74ca6391 beast2: add dependency on libbeagle. (#39311)
Co-authored-by: Cristian Di Pietrantonio <cdipietrantonio@pawsey.org.au>
2023-08-09 08:33:53 +02:00
Cristian Di Pietrantonio
7165e70186 chgcentre: fix compilation error by using explicit casting (#39315)
Co-authored-by: Cristian Di Pietrantonio <cdipietrantonio@pawsey.org.au>
2023-08-09 08:32:58 +02:00
Massimiliano Culpo
97d632a161 Push conflict between rocm and blt down to packages that actually use blt (#39339) 2023-08-09 07:53:17 +02:00
Alec Scott
571919992d go: add v1.20.6 and v1.19.11, deprecate previous versions due to CVE-2023-29405 (#39149) 2023-08-08 16:38:21 -07:00
snehring
99112ad2ad Update py-gtdbtk to 2.3.2 (#39296)
* fastani: adding version 1.34 and changing build_system

* py-gtdbtk: adding version 2.3.2

* fastani: adding requested changes

* py-gtdbtk: adding requested changes

* fastani: remove conditional
2023-08-08 18:33:56 -05:00
Paul Kuberry
75c70c395d trilinos: Add version 14.4.0 (#39340) 2023-08-08 18:48:23 -04:00
Jen Herting
960bdfe612 [py-model-index] new package (#39301) 2023-08-08 17:20:37 -05:00
Jen Herting
97892bda18 [py-opendatalab] New package (#39302) 2023-08-08 17:19:18 -05:00
Jen Herting
cead6ef98d [py-linkify-it-py] New package (#39305)
* [py-linkify-it-py] New package

* [py-linkify-it-py] Added version 1.0.3
2023-08-08 17:13:11 -05:00
eugeneswalker
5d70c0f100 e4s ci: add libpressio (#34222)
* e4s ci: add libpressio

* libpressio ~cuda, +cuda: add +mgard
2023-08-08 12:03:23 -07:00
Adam J. Stewart
361632fc4b Ensure that all variants have a description (#39025)
* Ensure that all variants have a description

* Update mock packages too

* Fix test invocations

* Black fix

* mgard: update variant descriptions

* flake8 fix

* black fix

* Add to audit tests

* Relax type hints

* Older Python support

* Undo all changes to mock packages

* Flake8 fix
2023-08-08 09:29:49 -05:00
simonLeary42
6576655137 Name clash error message include hash of clashing packages (#39234)
Co-authored-by: Simon <simonleary@umass.edu>
2023-08-08 14:36:49 +02:00
Mikael Simberg
feb26efecd hpx: add v1.9.1 (#39336) 2023-08-08 05:28:04 -04:00
Mikael Simberg
4752d1cde3 pika-algorithms: add v0.1.4 (#39335) 2023-08-08 05:23:10 -04:00
Paul Kuberry
a07afa6e1a OpenMPI: require compilers for F77 and "modern" Fortran (#39266) 2023-08-08 11:10:27 +02:00
Adam J. Stewart
7327d2913a ML CI: get more packages working on macOS (#39199) 2023-08-08 10:59:24 +02:00
Axel Huebl
8f8a1f7f52 C-Blosc2: add v2.10.1 (#39297)
This adds the latest C-Blosc2 release, the first to add CMake
CONFIG install files for easier downstream usage of the installed
library and CMake targets.
2023-08-08 10:08:12 +02:00
Carlos Bederián
eb8d836e76 aocc: add v4.1.0 (#39280) 2023-08-08 10:07:12 +02:00
Jen Herting
bad8495e16 parallel-hashmap: add new package (#39299) 2023-08-08 10:03:01 +02:00
Cristian Di Pietrantonio
43de7f4881 plumed: add v2.9.0, v2.8.3 (#39324)
Co-authored-by: Cristian Di Pietrantonio <cdipietrantonio@pawsey.org.au>
2023-08-08 09:59:03 +02:00
Hariharan Devarajan
84585ac575 cpp-logger: add new package (#39303) 2023-08-08 09:27:44 +02:00
Harmen Stoppels
86f9d3865b Fix broken inode assertion (#39188) 2023-08-08 09:21:23 +02:00
Adam J. Stewart
834e7b2b0a py-cartopy: add v0.22.0 (#39309) 2023-08-08 08:53:04 +02:00
shanedsnyder
c14f23ddaa Add darshan 3.4.4 release (#39310) 2023-08-08 08:49:27 +02:00
Sinan
49f3681a12 package:py-wand add build env var, add new version (#39281)
* package:py-wand add build env var, add new version

* fix style

* [@spackbot] updating style on behalf of Sinan81

* improve

* improve

---------

Co-authored-by: sbulut <sbulut@3vgeomatics.com>
Co-authored-by: Sinan81 <Sinan81@users.noreply.github.com>
2023-08-07 19:45:48 -05:00
Jen Herting
19e1d10cdf [py-marshmallow] added version 3.19.0 (#38972) 2023-08-07 19:44:44 -05:00
Jen Herting
7caf2a512d [py-srsly] added version 2.4.6 (#38970)
* [py-srsly] added version 2.4.6

* [py-srsly] added dependency on py-catalogue

* [py-srsly] added conflict on newer versions of py-cython
2023-08-07 19:44:34 -05:00
Jen Herting
1f6e3cc8cb [py-pathy] new package (#38968)
* [py-pathy] new package

* [py-pathy] removed lines for unsupported versions of python
2023-08-07 19:41:21 -05:00
Jen Herting
169c4245e0 [py-gensim] added version 4.3.1 (#38272)
* [py-gensim] added version 4.3.1

* [py-gensim] flake8

* [py-gensim] removed version limitation on py-cython
2023-08-07 18:56:53 -05:00
Jen Herting
ee1982010f [py-clean-text] added version 0.6.0 (#38271)
* [py-clean-text] added missing dependency on py-poetry

* [py-clean-text] added version 0.6.0

* [py-clean-text] Made required changes
2023-08-07 18:52:22 -05:00
Melven Roehrig-Zoellner
dd396c4a76 freeglut: set correct library name (#39300) 2023-08-07 19:08:08 -04:00
Jean Luca Bez
235802013d h5bench: include tests for E4S and new version (#37830)
* include tests for h5bench
* update to new test format
* update path
* improve description and detect MPI runner
* fix test name
* Include new release
* update with corrections
* include SLURM dependency and reduce test to one process
* fixes on filters
* fix on oversubscribe
* suggested fixes

---------

Co-authored-by: Tamara Dahlgren <35777542+tldahlgren@users.noreply.github.com>
2023-08-07 11:43:32 -07:00
Harmen Stoppels
e1f07e98ae libdeflate: new versions, switch to cmake (#39062) 2023-08-07 13:48:12 -04:00
Adam J. Stewart
60aee6f535 py-torchmetrics: add v1.0.2 (#39274) 2023-08-07 10:43:43 -07:00
Adam J. Stewart
2510dc9e6e py-timm: add v0.9.5 (#39275) 2023-08-07 10:42:14 -07:00
Christopher Christofi
5607dd259b samtools: add 1.17 (#39287) 2023-08-07 10:29:03 -07:00
Harmen Stoppels
7bd5d1fd3c use {%compiler.name}{@compiler.version} in matching_specs (#39247) 2023-08-07 18:54:10 +02:00
Harmen Stoppels
f6104cc3cb spack: bump package (#39184) 2023-08-07 18:52:14 +02:00
kwryankrattiger
b54d286b4a CI: remove redundant sections (#38514)
Refactor gitlab ci configs so that mac and cray jobs can reuse as much higher level
configuration as possible.

* CI: remove redundant sections
* CI: Include base linux CI configs in cray stacks

Relocation and runner mapping is consistent between cray and linux runners.

* Export user cache path in before script
* CI: add GPG root for mac runners
* Disable user configs

Metal runners share a ~ directory

* Disable user config and add configs in activate env
2023-08-07 10:23:16 -06:00
Seth R. Johnson
ea9c488897 celeritas: new version 0.3.1 (#39288) 2023-08-07 12:16:30 +01:00
Massimiliano Culpo
ba1d295023 Extract prefix locks and failure markers from Database (#39024)
This PR extracts two responsibilities from the `Database` class:
1. Managing locks for prefixes during an installation
2. Marking installation failures

and pushes them into their own class (`SpecLocker` and `FailureMarker`). These responsibilities are also pushed up into the `Store`, leaving to `Database` only the duty to manage `index.json` files.

`SpecLocker` classes no longer share a global list of locks, but locks are per instance. Their identifier is simply `(dag hash, package name)`, and not the spec prefix path, to avoid circular dependencies across Store / Database / Spec.
2023-08-07 06:47:52 -04:00
Harmen Stoppels
27f04b3544 Picklable HTTPError (#39285) 2023-08-07 10:48:35 +02:00
Brian Van Essen
8cd9497522 Bugfix spack module files (#39282)
* Fixed the LD_LIBRARY_PATH to use the lib64 directory.
* On AMD systems the llvm/lib directory is not properly put into the LD_LIBRARY_PATH.
* Added both lib and lib64 paths for libfabric.
* Split the prepend statements.
2023-08-05 20:26:00 -07:00
935 changed files with 7005 additions and 4856 deletions

View File

@@ -165,6 +165,7 @@ jobs:
- name: Install Python packages
run: |
pip install --upgrade pip setuptools pytest coverage[toml] pytest-cov clingo pytest-xdist
pip install --upgrade flake8 "isort>=4.3.5" "mypy>=0.900" "click" "black"
- name: Setup git configuration
run: |
# Need this for the git tests to succeed.

View File

@@ -51,65 +51,43 @@ setlocal enabledelayedexpansion
:: subcommands will never start with '-'
:: everything after the subcommand is an arg
:: we cannot allow batch "for" loop to directly process CL args
:: a number of batch reserved characters are commonly passed to
:: spack and allowing batch's "for" method to process the raw inputs
:: results in a large number of formatting issues
:: instead, treat the entire CLI as one string
:: and split by space manually
:: capture cl args in variable named cl_args
set cl_args=%*
:process_cl_args
rem tokens=1* returns the first processed token produced
rem by tokenizing the input string cl_args on spaces into
rem the named variable %%g
rem While this make look like a for loop, it only
rem executes a single time for each of the cl args
rem the actual iterative loop is performed by the
rem goto process_cl_args stanza
rem we are simply leveraging the "for" method's string
rem tokenization
for /f "tokens=1*" %%g in ("%cl_args%") do (
set t=%%~g
rem remainder of string is composed into %%h
rem these are the cl args yet to be processed
rem assign cl_args var to only the args to be processed
rem effectively discarding the current arg %%g
rem this will be nul when we have no further tokens to process
set cl_args=%%h
rem process the first space delineated cl arg
rem of this iteration
if "!t:~0,1!" == "-" (
if defined _sp_subcommand (
rem We already have a subcommand, processing args now
if not defined _sp_args (
set "_sp_args=!t!"
) else (
set "_sp_args=!_sp_args! !t!"
)
) else (
if not defined _sp_flags (
set "_sp_flags=!t!"
shift
) else (
set "_sp_flags=!_sp_flags! !t!"
shift
)
)
) else if not defined _sp_subcommand (
set "_sp_subcommand=!t!"
shift
) else (
rem Set first cl argument (denoted by %1) to be processed
set t=%1
rem shift moves all cl positional arguments left by one
rem meaning %2 is now %1, this allows us to iterate over each
rem argument
shift
rem assign next "first" cl argument to cl_args, will be null when
rem there are now further arguments to process
set cl_args=%1
if "!t:~0,1!" == "-" (
if defined _sp_subcommand (
rem We already have a subcommand, processing args now
if not defined _sp_args (
set "_sp_args=!t!"
shift
) else (
set "_sp_args=!_sp_args! !t!"
shift
)
) else (
if not defined _sp_flags (
set "_sp_flags=!t!"
) else (
set "_sp_flags=!_sp_flags! !t!"
)
)
) else if not defined _sp_subcommand (
set "_sp_subcommand=!t!"
) else (
if not defined _sp_args (
set "_sp_args=!t!"
) else (
set "_sp_args=!_sp_args! !t!"
)
)
rem if this is not nil, we have more tokens to process
rem if this is not nu;ll, we have more tokens to process
rem start above process again with remaining unprocessed cl args
if defined cl_args goto :process_cl_args

View File

@@ -36,3 +36,9 @@ concretizer:
# on each root spec, allowing different versions and variants of the same package in
# an environment.
unify: true
# Option to deal with possible duplicate nodes (i.e. different nodes from the same package) in the DAG.
duplicates:
# "none": allows a single node for any package in the DAG.
# "minimal": allows the duplication of 'build-tools' nodes only (e.g. py-setuptools, cmake etc.)
# "full" (experimental): allows separation of the entire build-tool stack (e.g. the entire "cmake" subDAG)
strategy: none

View File

@@ -60,6 +60,7 @@ packages:
xxd: [xxd-standalone, vim]
yacc: [bison, byacc]
ziglang: [zig]
zlib-api: [zlib-ng+compat, zlib]
permissions:
read: world
write: user

View File

@@ -32,9 +32,14 @@ can't be found. You can readily check if any prerequisite for using Spack is mis
Spack will take care of bootstrapping any missing dependency marked as [B]. Dependencies marked as [-] are instead required to be found on the system.
% echo $?
1
In the case of the output shown above Spack detected that both ``clingo`` and ``gnupg``
are missing and it's giving detailed information on why they are needed and whether
they can be bootstrapped. Running a command that concretize a spec, like:
they can be bootstrapped. The return code of this command summarizes the results, if any
dependencies are missing the return code is ``1``, otherwise ``0``. Running a command that
concretizes a spec, like:
.. code-block:: console
@@ -44,7 +49,7 @@ they can be bootstrapped. Running a command that concretize a spec, like:
==> Installing "clingo-bootstrap@spack%apple-clang@12.0.0~docs~ipo+python build_type=Release arch=darwin-catalina-x86_64" from a buildcache
[ ... ]
triggers the bootstrapping of clingo from pre-built binaries as expected.
automatically triggers the bootstrapping of clingo from pre-built binaries as expected.
Users can also bootstrap all the dependencies needed by Spack in a single command, which
might be useful to setup containers or other similar environments:

View File

@@ -104,11 +104,13 @@ Clone `spack-configs <https://github.com/spack/spack-configs>`_ repo and activat
`Intel oneAPI CPU environment <https://github.com/spack/spack-configs/blob/main/INTEL/CPU/spack.yaml>`_ contains applications tested and validated by Intel, this list is constantly extended. And currently it supports:
- `Devito <https://www.devitoproject.org/>`_
- `GROMACS <https://www.gromacs.org/>`_
- `HPCG <https://www.hpcg-benchmark.org/>`_
- `HPL <https://netlib.org/benchmark/hpl/>`_
- `LAMMPS <https://www.lammps.org/#gsc.tab=0>`_
- `OpenFOAM <https://www.openfoam.com/>`_
- `Quantum Espresso <https://www.quantum-espresso.org/>`_
- `STREAM <https://www.cs.virginia.edu/stream/>`_
- `WRF <https://github.com/wrf-model/WRF>`_

View File

@@ -2243,7 +2243,7 @@ looks like this:
url = "http://www.openssl.org/source/openssl-1.0.1h.tar.gz"
version("1.0.1h", md5="8d6d684a9430d5cc98a62a5d8fbda8cf")
depends_on("zlib")
depends_on("zlib-api")
parallel = False

View File

@@ -1,13 +1,13 @@
sphinx==6.2.1
sphinx==7.2.2
sphinxcontrib-programoutput==0.17
sphinx_design==0.4.1
sphinx-rtd-theme==1.2.2
sphinx_design==0.5.0
sphinx-rtd-theme==1.3.0
python-levenshtein==0.21.1
docutils==0.18.1
pygments==2.15.1
urllib3==2.0.3
pygments==2.16.1
urllib3==2.0.4
pytest==7.4.0
isort==5.12.0
black==23.1.0
flake8==6.0.0
mypy==1.4.1
black==23.7.0
flake8==6.1.0
mypy==1.5.1

View File

@@ -1754,9 +1754,14 @@ def find(root, files, recursive=True):
files = [files]
if recursive:
return _find_recursive(root, files)
tty.debug(f"Find (recursive): {root} {str(files)}")
result = _find_recursive(root, files)
else:
return _find_non_recursive(root, files)
tty.debug(f"Find (not recursive): {root} {str(files)}")
result = _find_non_recursive(root, files)
tty.debug(f"Find complete: {root} {str(files)}")
return result
@system_path_filter

View File

@@ -143,7 +143,7 @@ def get_fh(self, path: str) -> IO:
def release_by_stat(self, stat):
key = (stat.st_dev, stat.st_ino, os.getpid())
open_file = self._descriptors.get(key)
assert open_file, "Attempted to close non-existing inode: %s" % stat.st_inode
assert open_file, "Attempted to close non-existing inode: %s" % stat.st_ino
open_file.refs -= 1
if not open_file.refs:

View File

@@ -780,7 +780,7 @@ def __enter__(self):
raise RuntimeError("file argument must be set by __init__ ")
# Open both write and reading on logfile
if type(self.logfile) == io.StringIO:
if isinstance(self.logfile, io.StringIO):
self._ioflag = True
# cannot have two streams on tempfile, so we must make our own
sys.stdout = self.logfile

View File

@@ -286,7 +286,7 @@ def _check_build_test_callbacks(pkgs, error_cls):
"""Ensure stand-alone test method is not included in build-time callbacks"""
errors = []
for pkg_name in pkgs:
pkg_cls = spack.repo.path.get_pkg_class(pkg_name)
pkg_cls = spack.repo.PATH.get_pkg_class(pkg_name)
test_callbacks = getattr(pkg_cls, "build_time_test_callbacks", None)
# TODO (post-34236): "test*"->"test_*" once remove deprecated methods
@@ -312,7 +312,7 @@ def _check_patch_urls(pkgs, error_cls):
errors = []
for pkg_name in pkgs:
pkg_cls = spack.repo.path.get_pkg_class(pkg_name)
pkg_cls = spack.repo.PATH.get_pkg_class(pkg_name)
for condition, patches in pkg_cls.patches.items():
for patch in patches:
if not isinstance(patch, spack.patch.UrlPatch):
@@ -342,7 +342,7 @@ def _search_for_reserved_attributes_names_in_packages(pkgs, error_cls):
errors = []
for pkg_name in pkgs:
name_definitions = collections.defaultdict(list)
pkg_cls = spack.repo.path.get_pkg_class(pkg_name)
pkg_cls = spack.repo.PATH.get_pkg_class(pkg_name)
for cls_item in inspect.getmro(pkg_cls):
for name in RESERVED_NAMES:
@@ -383,7 +383,7 @@ def _ensure_packages_are_pickeleable(pkgs, error_cls):
"""Ensure that package objects are pickleable"""
errors = []
for pkg_name in pkgs:
pkg_cls = spack.repo.path.get_pkg_class(pkg_name)
pkg_cls = spack.repo.PATH.get_pkg_class(pkg_name)
pkg = pkg_cls(spack.spec.Spec(pkg_name))
try:
pickle.dumps(pkg)
@@ -424,7 +424,7 @@ def _ensure_all_versions_can_produce_a_fetcher(pkgs, error_cls):
"""Ensure all versions in a package can produce a fetcher"""
errors = []
for pkg_name in pkgs:
pkg_cls = spack.repo.path.get_pkg_class(pkg_name)
pkg_cls = spack.repo.PATH.get_pkg_class(pkg_name)
pkg = pkg_cls(spack.spec.Spec(pkg_name))
try:
spack.fetch_strategy.check_pkg_attributes(pkg)
@@ -449,7 +449,7 @@ def _ensure_docstring_and_no_fixme(pkgs, error_cls):
]
for pkg_name in pkgs:
details = []
filename = spack.repo.path.filename_for_package_name(pkg_name)
filename = spack.repo.PATH.filename_for_package_name(pkg_name)
with open(filename, "r") as package_file:
for i, line in enumerate(package_file):
pattern = next((r for r in fixme_regexes if r.search(line)), None)
@@ -461,7 +461,7 @@ def _ensure_docstring_and_no_fixme(pkgs, error_cls):
error_msg = "Package '{}' contains boilerplate that need to be removed"
errors.append(error_cls(error_msg.format(pkg_name), details))
pkg_cls = spack.repo.path.get_pkg_class(pkg_name)
pkg_cls = spack.repo.PATH.get_pkg_class(pkg_name)
if not pkg_cls.__doc__:
error_msg = "Package '{}' miss a docstring"
errors.append(error_cls(error_msg.format(pkg_name), []))
@@ -474,7 +474,7 @@ def _ensure_all_packages_use_sha256_checksums(pkgs, error_cls):
"""Ensure no packages use md5 checksums"""
errors = []
for pkg_name in pkgs:
pkg_cls = spack.repo.path.get_pkg_class(pkg_name)
pkg_cls = spack.repo.PATH.get_pkg_class(pkg_name)
if pkg_cls.manual_download:
continue
@@ -511,7 +511,7 @@ def _ensure_env_methods_are_ported_to_builders(pkgs, error_cls):
"""Ensure that methods modifying the build environment are ported to builder classes."""
errors = []
for pkg_name in pkgs:
pkg_cls = spack.repo.path.get_pkg_class(pkg_name)
pkg_cls = spack.repo.PATH.get_pkg_class(pkg_name)
buildsystem_variant, _ = pkg_cls.variants["build_system"]
buildsystem_names = [getattr(x, "value", x) for x in buildsystem_variant.values]
builder_cls_names = [spack.builder.BUILDER_CLS[x].__name__ for x in buildsystem_names]
@@ -538,7 +538,7 @@ def _linting_package_file(pkgs, error_cls):
"""Check for correctness of links"""
errors = []
for pkg_name in pkgs:
pkg_cls = spack.repo.path.get_pkg_class(pkg_name)
pkg_cls = spack.repo.PATH.get_pkg_class(pkg_name)
# Does the homepage have http, and if so, does https work?
if pkg_cls.homepage.startswith("http://"):
@@ -562,7 +562,7 @@ def _unknown_variants_in_directives(pkgs, error_cls):
"""Report unknown or wrong variants in directives for this package"""
errors = []
for pkg_name in pkgs:
pkg_cls = spack.repo.path.get_pkg_class(pkg_name)
pkg_cls = spack.repo.PATH.get_pkg_class(pkg_name)
# Check "conflicts" directive
for conflict, triggers in pkg_cls.conflicts.items():
@@ -628,15 +628,15 @@ def _unknown_variants_in_dependencies(pkgs, error_cls):
"""Report unknown dependencies and wrong variants for dependencies"""
errors = []
for pkg_name in pkgs:
pkg_cls = spack.repo.path.get_pkg_class(pkg_name)
filename = spack.repo.path.filename_for_package_name(pkg_name)
pkg_cls = spack.repo.PATH.get_pkg_class(pkg_name)
filename = spack.repo.PATH.filename_for_package_name(pkg_name)
for dependency_name, dependency_data in pkg_cls.dependencies.items():
# No need to analyze virtual packages
if spack.repo.path.is_virtual(dependency_name):
if spack.repo.PATH.is_virtual(dependency_name):
continue
try:
dependency_pkg_cls = spack.repo.path.get_pkg_class(dependency_name)
dependency_pkg_cls = spack.repo.PATH.get_pkg_class(dependency_name)
except spack.repo.UnknownPackageError:
# This dependency is completely missing, so report
# and continue the analysis
@@ -675,7 +675,7 @@ def _ensure_variant_defaults_are_parsable(pkgs, error_cls):
"""Ensures that variant defaults are present and parsable from cli"""
errors = []
for pkg_name in pkgs:
pkg_cls = spack.repo.path.get_pkg_class(pkg_name)
pkg_cls = spack.repo.PATH.get_pkg_class(pkg_name)
for variant_name, entry in pkg_cls.variants.items():
variant, _ = entry
default_is_parsable = (
@@ -709,18 +709,33 @@ def _ensure_variant_defaults_are_parsable(pkgs, error_cls):
return errors
@package_directives
def _ensure_variants_have_descriptions(pkgs, error_cls):
"""Ensures that all variants have a description."""
errors = []
for pkg_name in pkgs:
pkg_cls = spack.repo.PATH.get_pkg_class(pkg_name)
for variant_name, entry in pkg_cls.variants.items():
variant, _ = entry
if not variant.description:
error_msg = "Variant '{}' in package '{}' is missing a description"
errors.append(error_cls(error_msg.format(variant_name, pkg_name), []))
return errors
@package_directives
def _version_constraints_are_satisfiable_by_some_version_in_repo(pkgs, error_cls):
"""Report if version constraints used in directives are not satisfiable"""
errors = []
for pkg_name in pkgs:
pkg_cls = spack.repo.path.get_pkg_class(pkg_name)
filename = spack.repo.path.filename_for_package_name(pkg_name)
pkg_cls = spack.repo.PATH.get_pkg_class(pkg_name)
filename = spack.repo.PATH.filename_for_package_name(pkg_name)
dependencies_to_check = []
for dependency_name, dependency_data in pkg_cls.dependencies.items():
# Skip virtual dependencies for the time being, check on
# their versions can be added later
if spack.repo.path.is_virtual(dependency_name):
if spack.repo.PATH.is_virtual(dependency_name):
continue
dependencies_to_check.extend([edge.spec for edge in dependency_data.values()])
@@ -729,7 +744,7 @@ def _version_constraints_are_satisfiable_by_some_version_in_repo(pkgs, error_cls
for s in dependencies_to_check:
dependency_pkg_cls = None
try:
dependency_pkg_cls = spack.repo.path.get_pkg_class(s.name)
dependency_pkg_cls = spack.repo.PATH.get_pkg_class(s.name)
# Some packages have hacks that might cause failures on some platform
# Allow to explicitly set conditions to skip version checks in that case
skip_conditions = getattr(dependency_pkg_cls, "skip_version_audit", [])
@@ -772,7 +787,7 @@ def _analyze_variants_in_directive(pkg, constraint, directive, error_cls):
except variant_exceptions as e:
summary = pkg.name + ': wrong variant in "{0}" directive'
summary = summary.format(directive)
filename = spack.repo.path.filename_for_package_name(pkg.name)
filename = spack.repo.PATH.filename_for_package_name(pkg.name)
error_msg = str(e).strip()
if isinstance(e, KeyError):

View File

@@ -2383,22 +2383,12 @@ def __init__(self, all_architectures):
self.possible_specs = specs
def __call__(self, spec, **kwargs):
def __call__(self, spec: Spec, **kwargs):
"""
Args:
spec (str): The spec being searched for in its string representation or hash.
spec: The spec being searched for
"""
matches = []
if spec.startswith("/"):
# Matching a DAG hash
query_hash = spec.replace("/", "")
for candidate_spec in self.possible_specs:
if candidate_spec.dag_hash().startswith(query_hash):
matches.append(candidate_spec)
else:
# Matching a spec constraint
matches = [s for s in self.possible_specs if s.satisfies(spec)]
return matches
return [s for s in self.possible_specs if s.satisfies(spec)]
class FetchIndexError(Exception):

View File

@@ -124,9 +124,9 @@ def _read_and_sanitize_configuration() -> Dict[str, Any]:
def _bootstrap_config_scopes() -> Sequence["spack.config.ConfigScope"]:
tty.debug("[BOOTSTRAP CONFIG SCOPE] name=_builtin")
config_scopes: MutableSequence["spack.config.ConfigScope"] = [
spack.config.InternalConfigScope("_builtin", spack.config.config_defaults)
spack.config.InternalConfigScope("_builtin", spack.config.CONFIG_DEFAULTS)
]
configuration_paths = (spack.config.configuration_defaults_path, ("bootstrap", _config_path()))
configuration_paths = (spack.config.CONFIGURATION_DEFAULTS_PATH, ("bootstrap", _config_path()))
for name, path in configuration_paths:
platform = spack.platforms.host().name
platform_scope = spack.config.ConfigScope(

View File

@@ -476,15 +476,22 @@ def ensure_executables_in_path_or_raise(
def _add_externals_if_missing() -> None:
search_list = [
# clingo
spack.repo.path.get_pkg_class("cmake"),
spack.repo.path.get_pkg_class("bison"),
spack.repo.PATH.get_pkg_class("cmake"),
spack.repo.PATH.get_pkg_class("bison"),
# GnuPG
spack.repo.path.get_pkg_class("gawk"),
spack.repo.PATH.get_pkg_class("gawk"),
# develop deps
spack.repo.PATH.get_pkg_class("git"),
]
if IS_WINDOWS:
search_list.append(spack.repo.path.get_pkg_class("winbison"))
detected_packages = spack.detection.by_executable(search_list)
spack.detection.update_configuration(detected_packages, scope="bootstrap")
search_list.append(spack.repo.PATH.get_pkg_class("winbison"))
externals = spack.detection.by_executable(search_list)
# System git is typically deprecated, so mark as non-buildable to force it as external
non_buildable_externals = {k: externals.pop(k) for k in ("git",) if k in externals}
spack.detection.update_configuration(externals, scope="bootstrap", buildable=True)
spack.detection.update_configuration(
non_buildable_externals, scope="bootstrap", buildable=False
)
def clingo_root_spec() -> str:

View File

@@ -23,6 +23,7 @@
from ._common import _root_spec
from .config import root_path, spec_for_current_python, store_path
from .core import _add_externals_if_missing
class BootstrapEnvironment(spack.environment.Environment):
@@ -185,6 +186,7 @@ def pytest_root_spec() -> str:
def ensure_environment_dependencies() -> None:
"""Ensure Spack dependencies from the bootstrap environment are installed and ready to use"""
_add_externals_if_missing()
with BootstrapEnvironment() as env:
env.update_installations()
env.update_syspath_and_environ()

View File

@@ -1027,7 +1027,7 @@ def get_cmake_prefix_path(pkg):
def _setup_pkg_and_run(
serialized_pkg, function, kwargs, child_pipe, input_multiprocess_fd, jsfd1, jsfd2
serialized_pkg, function, kwargs, write_pipe, input_multiprocess_fd, jsfd1, jsfd2
):
context = kwargs.get("context", "build")
@@ -1048,12 +1048,12 @@ def _setup_pkg_and_run(
pkg, dirty=kwargs.get("dirty", False), context=context
)
return_value = function(pkg, kwargs)
child_pipe.send(return_value)
write_pipe.send(return_value)
except StopPhase as e:
# Do not create a full ChildError from this, it's not an error
# it's a control statement.
child_pipe.send(e)
write_pipe.send(e)
except BaseException:
# catch ANYTHING that goes wrong in the child process
exc_type, exc, tb = sys.exc_info()
@@ -1102,10 +1102,10 @@ def _setup_pkg_and_run(
context,
package_context,
)
child_pipe.send(ce)
write_pipe.send(ce)
finally:
child_pipe.close()
write_pipe.close()
if input_multiprocess_fd is not None:
input_multiprocess_fd.close()
@@ -1149,7 +1149,7 @@ def child_fun():
For more information on `multiprocessing` child process creation
mechanisms, see https://docs.python.org/3/library/multiprocessing.html#contexts-and-start-methods
"""
parent_pipe, child_pipe = multiprocessing.Pipe()
read_pipe, write_pipe = multiprocessing.Pipe(duplex=False)
input_multiprocess_fd = None
jobserver_fd1 = None
jobserver_fd2 = None
@@ -1174,7 +1174,7 @@ def child_fun():
serialized_pkg,
function,
kwargs,
child_pipe,
write_pipe,
input_multiprocess_fd,
jobserver_fd1,
jobserver_fd2,
@@ -1183,6 +1183,12 @@ def child_fun():
p.start()
# We close the writable end of the pipe now to be sure that p is the
# only process which owns a handle for it. This ensures that when p
# closes its handle for the writable end, read_pipe.recv() will
# promptly report the readable end as being ready.
write_pipe.close()
except InstallError as e:
e.pkg = pkg
raise
@@ -1192,7 +1198,16 @@ def child_fun():
if input_multiprocess_fd is not None:
input_multiprocess_fd.close()
child_result = parent_pipe.recv()
def exitcode_msg(p):
typ = "exit" if p.exitcode >= 0 else "signal"
return f"{typ} {abs(p.exitcode)}"
try:
child_result = read_pipe.recv()
except EOFError:
p.join()
raise InstallError(f"The process has stopped unexpectedly ({exitcode_msg(p)})")
p.join()
# If returns a StopPhase, raise it
@@ -1212,6 +1227,10 @@ def child_fun():
child_result.print_context()
raise child_result
# Fallback. Usually caught beforehand in EOFError above.
if p.exitcode != 0:
raise InstallError(f"The process failed unexpectedly ({exitcode_msg(p)})")
return child_result
@@ -1256,9 +1275,8 @@ def make_stack(tb, stack=None):
func = getattr(obj, tb.tb_frame.f_code.co_name, "")
if func:
typename, *_ = func.__qualname__.partition(".")
if isinstance(obj, CONTEXT_BASES) and typename not in basenames:
break
if isinstance(obj, CONTEXT_BASES) and typename not in basenames:
break
else:
return None

View File

@@ -55,7 +55,8 @@ def flags_to_build_system_args(self, flags):
setattr(self, "configure_flag_args", [])
for flag, values in flags.items():
if values:
values_str = "{0}={1}".format(flag.upper(), " ".join(values))
var_name = "LIBS" if flag == "ldlibs" else flag.upper()
values_str = "{0}={1}".format(var_name, " ".join(values))
self.configure_flag_args.append(values_str)
# Spack's fflags are meant for both F77 and FC, therefore we
# additionaly set FCFLAGS if required.

View File

@@ -248,7 +248,8 @@ def std_cmake_args(self):
@staticmethod
def std_args(pkg, generator=None):
"""Computes the standard cmake arguments for a generic package"""
generator = generator or "Unix Makefiles"
default_generator = "Ninja" if sys.platform == "win32" else "Unix Makefiles"
generator = generator or default_generator
valid_primary_generators = ["Unix Makefiles", "Ninja"]
primary_generator = _extract_primary_generator(generator)
if primary_generator not in valid_primary_generators:

View File

@@ -209,5 +209,5 @@ def install(self, pkg, spec, prefix):
def check(self):
"""Search Meson-generated files for the target ``test`` and run it if found."""
with fs.working_dir(self.build_directory):
self._if_ninja_target_execute("test")
self._if_ninja_target_execute("check")
self.pkg._if_ninja_target_execute("test")
self.pkg._if_ninja_target_execute("check")

View File

@@ -30,7 +30,7 @@
class PythonExtension(spack.package_base.PackageBase):
maintainers("adamjstewart", "pradyunsg")
maintainers("adamjstewart")
@property
def import_modules(self):
@@ -201,7 +201,7 @@ def update_external_dependencies(self, extendee_spec=None):
else:
python = self.get_external_python_for_prefix()
if not python.concrete:
repo = spack.repo.path.repo_for_pkg(python)
repo = spack.repo.PATH.repo_for_pkg(python)
python.namespace = repo.namespace
# Ensure architecture information is present
@@ -301,7 +301,7 @@ def get_external_python_for_prefix(self):
return python_externals_configured[0]
python_externals_detection = spack.detection.by_executable(
[spack.repo.path.get_pkg_class("python")], path_hints=[self.spec.external_path]
[spack.repo.PATH.get_pkg_class("python")], path_hints=[self.spec.external_path]
)
python_externals_detected = [

View File

@@ -140,8 +140,6 @@ class ROCmPackage(PackageBase):
depends_on("hsa-rocr-dev", when="+rocm")
depends_on("hip +rocm", when="+rocm")
conflicts("^blt@:0.3.6", when="+rocm")
# need amd gpu type for rocm builds
conflicts("amdgpu_target=none", when="+rocm")

View File

@@ -20,9 +20,9 @@
def misc_cache_location():
"""The ``misc_cache`` is Spack's cache for small data.
"""The ``MISC_CACHE`` is Spack's cache for small data.
Currently the ``misc_cache`` stores indexes for virtual dependency
Currently the ``MISC_CACHE`` stores indexes for virtual dependency
providers and for which packages provide which tags.
"""
path = spack.config.get("config:misc_cache", spack.paths.default_misc_cache_path)
@@ -35,7 +35,7 @@ def _misc_cache():
#: Spack's cache for small data
misc_cache: Union[
MISC_CACHE: Union[
spack.util.file_cache.FileCache, llnl.util.lang.Singleton
] = llnl.util.lang.Singleton(_misc_cache)
@@ -91,6 +91,6 @@ def symlink(self, mirror_ref):
#: Spack's local cache for downloaded source archives
fetch_cache: Union[
FETCH_CACHE: Union[
spack.fetch_strategy.FsCache, llnl.util.lang.Singleton
] = llnl.util.lang.Singleton(_fetch_cache)

View File

@@ -535,7 +535,7 @@ def __job_name(name, suffix=""):
"""Compute the name of a named job with appropriate suffix.
Valid suffixes are either '-remove' or empty string or None
"""
assert type(name) == str
assert isinstance(name, str)
jname = name
if suffix:
@@ -885,7 +885,7 @@ def generate_gitlab_ci_yaml(
cli_scopes = [
os.path.relpath(s.path, concrete_env_dir)
for s in cfg.scopes().values()
if type(s) == cfg.ImmutableConfigScope
if isinstance(s, cfg.ImmutableConfigScope)
and s.path not in env_includes
and os.path.exists(s.path)
]
@@ -1504,7 +1504,7 @@ def copy_stage_logs_to_artifacts(job_spec: spack.spec.Spec, job_log_dir: str) ->
return
try:
pkg_cls = spack.repo.path.get_pkg_class(job_spec.name)
pkg_cls = spack.repo.PATH.get_pkg_class(job_spec.name)
job_pkg = pkg_cls(job_spec)
tty.debug("job package: {0}".format(job_pkg))
except AssertionError:

View File

@@ -291,7 +291,7 @@ def ensure_single_spec_or_die(spec, matching_specs):
if len(matching_specs) <= 1:
return
format_string = "{name}{@version}{%compiler}{arch=architecture}"
format_string = "{name}{@version}{%compiler.name}{@compiler.version}{arch=architecture}"
args = ["%s matches multiple packages." % spec, "Matching packages:"]
args += [
colorize(" @K{%s} " % s.dag_hash(7)) + s.cformat(format_string) for s in matching_specs

View File

@@ -47,7 +47,7 @@ def configs(parser, args):
def packages(parser, args):
pkgs = args.name or spack.repo.path.all_package_names()
pkgs = args.name or spack.repo.PATH.all_package_names()
reports = spack.audit.run_group(args.subcommand, pkgs=pkgs)
_process_reports(reports)
@@ -57,7 +57,7 @@ def packages_https(parser, args):
if not args.check_all and not args.name:
tty.die("Please specify one or more packages to audit, or --all.")
pkgs = args.name or spack.repo.path.all_package_names()
pkgs = args.name or spack.repo.PATH.all_package_names()
reports = spack.audit.run_group(args.subcommand, pkgs=pkgs)
_process_reports(reports)

View File

@@ -126,7 +126,7 @@ def blame(parser, args):
blame_file = path
if not blame_file:
pkg_cls = spack.repo.path.get_pkg_class(args.package_or_file)
pkg_cls = spack.repo.PATH.get_pkg_class(args.package_or_file)
blame_file = pkg_cls.module.__file__.rstrip("c") # .pyc -> .py
# get git blame for the package

View File

@@ -4,6 +4,7 @@
# SPDX-License-Identifier: (Apache-2.0 OR MIT)
import os.path
import shutil
import sys
import tempfile
import llnl.util.filesystem
@@ -68,11 +69,10 @@
def _add_scope_option(parser):
scopes = spack.config.scopes()
scopes_metavar = spack.config.scopes_metavar
parser.add_argument(
"--scope",
choices=scopes,
metavar=scopes_metavar,
metavar=spack.config.SCOPES_METAVAR,
help="configuration scope to read/modify",
)
@@ -169,7 +169,7 @@ def _reset(args):
if not ok_to_continue:
raise RuntimeError("Aborting")
for scope in spack.config.config.file_scopes:
for scope in spack.config.CONFIG.file_scopes:
# The default scope should stay untouched
if scope.name == "defaults":
continue
@@ -186,7 +186,7 @@ def _reset(args):
if os.path.exists(bootstrap_yaml):
shutil.move(bootstrap_yaml, backup_file)
spack.config.config.clear_caches()
spack.config.CONFIG.clear_caches()
def _root(args):
@@ -326,6 +326,7 @@ def _status(args):
if missing:
print(llnl.util.tty.color.colorize(legend))
print()
sys.exit(1)
def _add(args):

View File

@@ -149,12 +149,11 @@ def setup_parser(subparser: argparse.ArgumentParser):
# used to construct scope arguments below
scopes = spack.config.scopes()
scopes_metavar = spack.config.scopes_metavar
check.add_argument(
"--scope",
choices=scopes,
metavar=scopes_metavar,
metavar=spack.config.SCOPES_METAVAR,
default=spack.config.default_modify_scope(),
help="configuration scope containing mirrors to check",
)

View File

@@ -83,7 +83,7 @@ def checksum(parser, args):
tty.die("`spack checksum` accepts package names, not URLs.")
# Get the package we're going to generate checksums for
pkg_cls = spack.repo.path.get_pkg_class(args.package)
pkg_cls = spack.repo.PATH.get_pkg_class(args.package)
pkg = pkg_cls(spack.spec.Spec(args.package))
# Build a list of versions to checksum
@@ -210,7 +210,7 @@ def add_versions_to_package(pkg: PackageBase, version_lines: str):
"""
# Get filename and path for package
filename = spack.repo.path.filename_for_package_name(pkg.name)
filename = spack.repo.PATH.filename_for_package_name(pkg.name)
num_versions_added = 0
version_statement_re = re.compile(r"([\t ]+version\([^\)]*\))")

View File

@@ -289,6 +289,10 @@ def ci_rebuild(args):
rebuild_everything = os.environ.get("SPACK_REBUILD_EVERYTHING")
require_signing = os.environ.get("SPACK_REQUIRE_SIGNING")
# If signing key was provided via "SPACK_SIGNING_KEY", then try to import it.
if signing_key:
spack_ci.import_signing_key(signing_key)
# Fail early if signing is required but we don't have a signing key
sign_binaries = require_signing is not None and require_signing.lower() == "true"
if sign_binaries and not spack_ci.can_sign_binaries():
@@ -418,11 +422,6 @@ def ci_rebuild(args):
dst_file = os.path.join(repro_dir, file_name)
shutil.copyfile(src_file, dst_file)
# If signing key was provided via "SPACK_SIGNING_KEY", then try to
# import it.
if signing_key:
spack_ci.import_signing_key(signing_key)
# Write this job's spec json into the reproduction directory, and it will
# also be used in the generated "spack install" command to install the spec
tty.debug("job concrete spec path: {0}".format(job_spec_json_path))
@@ -679,7 +678,7 @@ def ci_rebuild(args):
input_spec=job_spec,
buildcache_mirror_url=buildcache_mirror_url,
pipeline_mirror_url=pipeline_mirror_url,
sign_binaries=sign_binaries,
sign_binaries=spack_ci.can_sign_binaries(),
):
msg = tty.msg if result.success else tty.warn
msg(

View File

@@ -17,6 +17,7 @@
import spack.config
import spack.repo
import spack.stage
import spack.store
import spack.util.path
from spack.paths import lib_path, var_path
@@ -117,15 +118,15 @@ def clean(parser, args):
if args.downloads:
tty.msg("Removing cached downloads")
spack.caches.fetch_cache.destroy()
spack.caches.FETCH_CACHE.destroy()
if args.failures:
tty.msg("Removing install failure marks")
spack.installer.clear_failures()
spack.store.STORE.failure_tracker.clear_all()
if args.misc_cache:
tty.msg("Removing cached information on repositories")
spack.caches.misc_cache.destroy()
spack.caches.MISC_CACHE.destroy()
if args.python_cache:
tty.msg("Removing python cache files")

View File

@@ -36,13 +36,13 @@
"bash": {
"aliases": True,
"format": "bash",
"header": os.path.join(spack.paths.share_path, "bash", "spack-completion.in"),
"header": os.path.join(spack.paths.share_path, "bash", "spack-completion.bash"),
"update": os.path.join(spack.paths.share_path, "spack-completion.bash"),
},
"fish": {
"aliases": True,
"format": "fish",
"header": os.path.join(spack.paths.share_path, "fish", "spack-completion.in"),
"header": os.path.join(spack.paths.share_path, "fish", "spack-completion.fish"),
"update": os.path.join(spack.paths.share_path, "spack-completion.fish"),
},
}

View File

@@ -24,7 +24,6 @@ def setup_parser(subparser):
sp = subparser.add_subparsers(metavar="SUBCOMMAND", dest="compiler_command")
scopes = spack.config.scopes()
scopes_metavar = spack.config.scopes_metavar
# Find
find_parser = sp.add_parser(
@@ -36,7 +35,7 @@ def setup_parser(subparser):
find_parser.add_argument(
"--scope",
choices=scopes,
metavar=scopes_metavar,
metavar=spack.config.SCOPES_METAVAR,
default=spack.config.default_modify_scope("compilers"),
help="configuration scope to modify",
)
@@ -50,7 +49,7 @@ def setup_parser(subparser):
remove_parser.add_argument(
"--scope",
choices=scopes,
metavar=scopes_metavar,
metavar=spack.config.SCOPES_METAVAR,
default=None,
help="configuration scope to modify",
)
@@ -60,7 +59,7 @@ def setup_parser(subparser):
list_parser.add_argument(
"--scope",
choices=scopes,
metavar=scopes_metavar,
metavar=spack.config.SCOPES_METAVAR,
default=spack.config.default_list_scope(),
help="configuration scope to read from",
)
@@ -71,7 +70,7 @@ def setup_parser(subparser):
info_parser.add_argument(
"--scope",
choices=scopes,
metavar=scopes_metavar,
metavar=spack.config.SCOPES_METAVAR,
default=spack.config.default_list_scope(),
help="configuration scope to read from",
)
@@ -93,7 +92,7 @@ def compiler_find(args):
n = len(new_compilers)
s = "s" if n > 1 else ""
config = spack.config.config
config = spack.config.CONFIG
filename = config.get_config_filename(args.scope, "compilers")
tty.msg("Added %d new compiler%s to %s" % (n, s, filename))
colify(reversed(sorted(c.spec.display_str for c in new_compilers)), indent=4)

View File

@@ -13,12 +13,11 @@
def setup_parser(subparser):
scopes = spack.config.scopes()
scopes_metavar = spack.config.scopes_metavar
subparser.add_argument(
"--scope",
choices=scopes,
metavar=scopes_metavar,
metavar=spack.config.SCOPES_METAVAR,
help="configuration scope to read/modify",
)

View File

@@ -27,13 +27,12 @@
def setup_parser(subparser):
scopes = spack.config.scopes()
scopes_metavar = spack.config.scopes_metavar
# User can only choose one
subparser.add_argument(
"--scope",
choices=scopes,
metavar=scopes_metavar,
metavar=spack.config.SCOPES_METAVAR,
help="configuration scope to read/modify",
)
@@ -45,7 +44,7 @@ def setup_parser(subparser):
help="configuration section to print\n\noptions: %(choices)s",
nargs="?",
metavar="section",
choices=spack.config.section_schemas,
choices=spack.config.SECTION_SCHEMAS,
)
blame_parser = sp.add_parser(
@@ -55,7 +54,7 @@ def setup_parser(subparser):
"section",
help="configuration section to print\n\noptions: %(choices)s",
metavar="section",
choices=spack.config.section_schemas,
choices=spack.config.SECTION_SCHEMAS,
)
edit_parser = sp.add_parser("edit", help="edit configuration file")
@@ -64,7 +63,7 @@ def setup_parser(subparser):
help="configuration section to edit\n\noptions: %(choices)s",
metavar="section",
nargs="?",
choices=spack.config.section_schemas,
choices=spack.config.SECTION_SCHEMAS,
)
edit_parser.add_argument(
"--print-file", action="store_true", help="print the file name that would be edited"
@@ -146,10 +145,10 @@ def config_get(args):
scope, section = _get_scope_and_section(args)
if section is not None:
spack.config.config.print_section(section)
spack.config.CONFIG.print_section(section)
elif scope and scope.startswith("env:"):
config_file = spack.config.config.get_config_filename(scope, section)
config_file = spack.config.CONFIG.get_config_filename(scope, section)
if os.path.exists(config_file):
with open(config_file) as f:
print(f.read())
@@ -162,7 +161,7 @@ def config_get(args):
def config_blame(args):
"""Print out line-by-line blame of merged YAML."""
spack.config.config.print_section(args.section, blame=True)
spack.config.CONFIG.print_section(args.section, blame=True)
def config_edit(args):
@@ -181,7 +180,7 @@ def config_edit(args):
scope, section = _get_scope_and_section(args)
if not scope and not section:
tty.die("`spack config edit` requires a section argument or an active environment.")
config_file = spack.config.config.get_config_filename(scope, section)
config_file = spack.config.CONFIG.get_config_filename(scope, section)
if args.print_file:
print(config_file)
@@ -194,7 +193,7 @@ def config_list(args):
Used primarily for shell tab completion scripts.
"""
print(" ".join(list(spack.config.section_schemas)))
print(" ".join(list(spack.config.SECTION_SCHEMAS)))
def config_add(args):
@@ -251,19 +250,19 @@ def _can_update_config_file(scope: spack.config.ConfigScope, cfg_file):
def config_update(args):
# Read the configuration files
spack.config.config.get_config(args.section, scope=args.scope)
spack.config.CONFIG.get_config(args.section, scope=args.scope)
updates: List[spack.config.ConfigScope] = list(
filter(
lambda s: not isinstance(
s, (spack.config.InternalConfigScope, spack.config.ImmutableConfigScope)
),
spack.config.config.format_updates[args.section],
spack.config.CONFIG.format_updates[args.section],
)
)
cannot_overwrite, skip_system_scope = [], False
for scope in updates:
cfg_file = spack.config.config.get_config_filename(scope.name, args.section)
cfg_file = spack.config.CONFIG.get_config_filename(scope.name, args.section)
can_be_updated = _can_update_config_file(scope, cfg_file)
if not can_be_updated:
if scope.name == "system":
@@ -302,7 +301,7 @@ def config_update(args):
" the latest schema format:\n\n"
)
for scope in updates:
cfg_file = spack.config.config.get_config_filename(scope.name, args.section)
cfg_file = spack.config.CONFIG.get_config_filename(scope.name, args.section)
msg += "\t[scope={0}, file={1}]\n".format(scope.name, cfg_file)
msg += (
"\nIf the configuration files are updated, versions of Spack "
@@ -325,7 +324,7 @@ def config_update(args):
# Make a backup copy and rewrite the file
bkp_file = cfg_file + ".bkp"
shutil.copy(cfg_file, bkp_file)
spack.config.config.update_config(args.section, data, scope=scope.name, force=True)
spack.config.CONFIG.update_config(args.section, data, scope=scope.name, force=True)
tty.msg(f'File "{cfg_file}" update [backup={bkp_file}]')
@@ -337,13 +336,13 @@ def _can_revert_update(scope_dir, cfg_file, bkp_file):
def config_revert(args):
scopes = [args.scope] if args.scope else [x.name for x in spack.config.config.file_scopes]
scopes = [args.scope] if args.scope else [x.name for x in spack.config.CONFIG.file_scopes]
# Search for backup files in the configuration scopes
Entry = collections.namedtuple("Entry", ["scope", "cfg", "bkp"])
to_be_restored, cannot_overwrite = [], []
for scope in scopes:
cfg_file = spack.config.config.get_config_filename(scope, args.section)
cfg_file = spack.config.CONFIG.get_config_filename(scope, args.section)
bkp_file = cfg_file + ".bkp"
# If the backup files doesn't exist move to the next scope
@@ -457,7 +456,7 @@ def config_prefer_upstream(args):
existing = spack.config.get("packages", scope=scope)
new = spack.config.merge_yaml(existing, pkgs)
spack.config.set("packages", new, scope)
config_file = spack.config.config.get_config_filename(scope, section)
config_file = spack.config.CONFIG.get_config_filename(scope, section)
tty.msg("Updated config at {0}".format(config_file))

View File

@@ -915,11 +915,11 @@ def get_repository(args, name):
)
else:
if spec.namespace:
repo = spack.repo.path.get_repo(spec.namespace, None)
repo = spack.repo.PATH.get_repo(spec.namespace, None)
if not repo:
tty.die("Unknown namespace: '{0}'".format(spec.namespace))
else:
repo = spack.repo.path.first_repo()
repo = spack.repo.PATH.first_repo()
# Set the namespace on the spec if it's not there already
if not spec.namespace:

View File

@@ -47,14 +47,14 @@ def inverted_dependencies():
actual dependents.
"""
dag = {}
for pkg_cls in spack.repo.path.all_package_classes():
for pkg_cls in spack.repo.PATH.all_package_classes():
dag.setdefault(pkg_cls.name, set())
for dep in pkg_cls.dependencies:
deps = [dep]
# expand virtuals if necessary
if spack.repo.path.is_virtual(dep):
deps += [s.name for s in spack.repo.path.providers_for(dep)]
if spack.repo.PATH.is_virtual(dep):
deps += [s.name for s in spack.repo.PATH.providers_for(dep)]
for d in deps:
dag.setdefault(d, set()).add(pkg_cls.name)

View File

@@ -98,7 +98,7 @@ def dev_build(self, args):
tty.die("spack dev-build only takes one spec.")
spec = specs[0]
if not spack.repo.path.exists(spec.name):
if not spack.repo.PATH.exists(spec.name):
tty.die(
"No package for '{0}' was found.".format(spec.name),
" Use `spack create` to create a new package",

View File

@@ -31,9 +31,9 @@ def edit_package(name, repo_path, namespace):
if repo_path:
repo = spack.repo.Repo(repo_path)
elif namespace:
repo = spack.repo.path.get_repo(namespace)
repo = spack.repo.PATH.get_repo(namespace)
else:
repo = spack.repo.path
repo = spack.repo.PATH
path = repo.filename_for_package_name(name)
spec = Spec(name)

View File

@@ -58,7 +58,7 @@ def extensions(parser, args):
extendable_pkgs = []
for name in spack.repo.all_package_names():
pkg_cls = spack.repo.path.get_pkg_class(name)
pkg_cls = spack.repo.PATH.get_pkg_class(name)
if pkg_cls.extendable:
extendable_pkgs.append(name)
@@ -81,7 +81,7 @@ def extensions(parser, args):
if args.show in ("packages", "all"):
# List package names of extensions
extensions = spack.repo.path.extensions_for(spec)
extensions = spack.repo.PATH.extensions_for(spec)
if not extensions:
tty.msg("%s has no extensions." % spec.cshort_spec)
else:

View File

@@ -13,6 +13,7 @@
import spack
import spack.cmd
import spack.cmd.common.arguments
import spack.config
import spack.cray_manifest as cray_manifest
import spack.detection
import spack.error
@@ -27,7 +28,6 @@ def setup_parser(subparser):
sp = subparser.add_subparsers(metavar="SUBCOMMAND", dest="external_command")
scopes = spack.config.scopes()
scopes_metavar = spack.config.scopes_metavar
find_parser = sp.add_parser("find", help="add external packages to packages.yaml")
find_parser.add_argument(
@@ -47,7 +47,7 @@ def setup_parser(subparser):
find_parser.add_argument(
"--scope",
choices=scopes,
metavar=scopes_metavar,
metavar=spack.config.SCOPES_METAVAR,
default=spack.config.default_modify_scope("packages"),
help="configuration scope to modify",
)
@@ -133,9 +133,9 @@ def external_find(args):
# Add the packages that have been required explicitly
if args.packages:
pkg_cls_to_check = [spack.repo.path.get_pkg_class(pkg) for pkg in args.packages]
pkg_cls_to_check = [spack.repo.PATH.get_pkg_class(pkg) for pkg in args.packages]
if args.tags:
allowed = set(spack.repo.path.packages_with_tags(*args.tags))
allowed = set(spack.repo.PATH.packages_with_tags(*args.tags))
pkg_cls_to_check = [x for x in pkg_cls_to_check if x.name in allowed]
if args.tags and not pkg_cls_to_check:
@@ -144,15 +144,15 @@ def external_find(args):
# Since tags are cached it's much faster to construct what we need
# to search directly, rather than filtering after the fact
pkg_cls_to_check = [
spack.repo.path.get_pkg_class(pkg_name)
spack.repo.PATH.get_pkg_class(pkg_name)
for tag in args.tags
for pkg_name in spack.repo.path.packages_with_tags(tag)
for pkg_name in spack.repo.PATH.packages_with_tags(tag)
]
pkg_cls_to_check = list(set(pkg_cls_to_check))
# If the list of packages is empty, search for every possible package
if not args.tags and not pkg_cls_to_check:
pkg_cls_to_check = list(spack.repo.path.all_package_classes())
pkg_cls_to_check = list(spack.repo.PATH.all_package_classes())
# If the user specified any packages to exclude from external find, add them here
if args.exclude:
@@ -165,7 +165,7 @@ def external_find(args):
detected_packages, scope=args.scope, buildable=not args.not_buildable
)
if new_entries:
path = spack.config.config.get_config_filename(args.scope, "packages")
path = spack.config.CONFIG.get_config_filename(args.scope, "packages")
msg = "The following specs have been detected on this system and added to {0}"
tty.msg(msg.format(path))
spack.cmd.display_specs(new_entries)
@@ -239,7 +239,7 @@ def _collect_and_consume_cray_manifest_files(
def external_list(args):
# Trigger a read of all packages, might take a long time.
list(spack.repo.path.all_package_classes())
list(spack.repo.PATH.all_package_classes())
# Print all the detectable packages
tty.msg("Detectable packages per repository")
for namespace, pkgs in sorted(spack.package_base.detectable_packages.items()):

View File

@@ -268,7 +268,7 @@ def find(parser, args):
# If tags have been specified on the command line, filter by tags
if args.tags:
packages_with_tags = spack.repo.path.packages_with_tags(*args.tags)
packages_with_tags = spack.repo.PATH.packages_with_tags(*args.tags)
results = [x for x in results if x.name in packages_with_tags]
if args.loaded:

View File

@@ -349,7 +349,7 @@ def print_virtuals(pkg):
def info(parser, args):
spec = spack.spec.Spec(args.package)
pkg_cls = spack.repo.path.get_pkg_class(spec.name)
pkg_cls = spack.repo.PATH.get_pkg_class(spec.name)
pkg = pkg_cls(spec)
# Output core package information

View File

@@ -107,7 +107,7 @@ def match(p, f):
if f.match(p):
return True
pkg_cls = spack.repo.path.get_pkg_class(p)
pkg_cls = spack.repo.PATH.get_pkg_class(p)
if pkg_cls.__doc__:
return f.match(pkg_cls.__doc__)
return False
@@ -159,7 +159,7 @@ def get_dependencies(pkg):
@formatter
def version_json(pkg_names, out):
"""Print all packages with their latest versions."""
pkg_classes = [spack.repo.path.get_pkg_class(name) for name in pkg_names]
pkg_classes = [spack.repo.PATH.get_pkg_class(name) for name in pkg_names]
out.write("[\n")
@@ -201,7 +201,7 @@ def html(pkg_names, out):
"""
# Read in all packages
pkg_classes = [spack.repo.path.get_pkg_class(name) for name in pkg_names]
pkg_classes = [spack.repo.PATH.get_pkg_class(name) for name in pkg_names]
# Start at 2 because the title of the page from Sphinx is id1.
span_id = 2
@@ -313,13 +313,13 @@ def list(parser, args):
# If tags have been specified on the command line, filter by tags
if args.tags:
packages_with_tags = spack.repo.path.packages_with_tags(*args.tags)
packages_with_tags = spack.repo.PATH.packages_with_tags(*args.tags)
sorted_packages = [p for p in sorted_packages if p in packages_with_tags]
if args.update:
# change output stream if user asked for update
if os.path.exists(args.update):
if os.path.getmtime(args.update) > spack.repo.path.last_mtime():
if os.path.getmtime(args.update) > spack.repo.PATH.last_mtime():
tty.msg("File is up to date: %s" % args.update)
return

View File

@@ -109,7 +109,7 @@ def location(parser, args):
return
if args.packages:
print(spack.repo.path.first_repo().root)
print(spack.repo.PATH.first_repo().root)
return
if args.stages:
@@ -135,7 +135,7 @@ def location(parser, args):
# Package dir just needs the spec name
if args.package_dir:
print(spack.repo.path.dirname_for_package_name(spec.name))
print(spack.repo.PATH.dirname_for_package_name(spec.name))
return
# Either concretize or filter from already concretized environment

View File

@@ -54,11 +54,11 @@ def setup_parser(subparser):
def packages_to_maintainers(package_names=None):
if not package_names:
package_names = spack.repo.path.all_package_names()
package_names = spack.repo.PATH.all_package_names()
pkg_to_users = defaultdict(lambda: set())
for name in package_names:
cls = spack.repo.path.get_pkg_class(name)
cls = spack.repo.PATH.get_pkg_class(name)
for user in cls.maintainers:
pkg_to_users[name].add(user)
@@ -67,8 +67,8 @@ def packages_to_maintainers(package_names=None):
def maintainers_to_packages(users=None):
user_to_pkgs = defaultdict(lambda: [])
for name in spack.repo.path.all_package_names():
cls = spack.repo.path.get_pkg_class(name)
for name in spack.repo.PATH.all_package_names():
cls = spack.repo.PATH.get_pkg_class(name)
for user in cls.maintainers:
lower_users = [u.lower() for u in users]
if not users or user.lower() in lower_users:
@@ -80,8 +80,8 @@ def maintainers_to_packages(users=None):
def maintained_packages():
maintained = []
unmaintained = []
for name in spack.repo.path.all_package_names():
cls = spack.repo.path.get_pkg_class(name)
for name in spack.repo.PATH.all_package_names():
cls = spack.repo.PATH.get_pkg_class(name)
if cls.maintainers:
maintained.append(name)
else:

View File

@@ -90,7 +90,6 @@ def setup_parser(subparser):
# used to construct scope arguments below
scopes = spack.config.scopes()
scopes_metavar = spack.config.scopes_metavar
# Add
add_parser = sp.add_parser("add", help=mirror_add.__doc__)
@@ -99,7 +98,7 @@ def setup_parser(subparser):
add_parser.add_argument(
"--scope",
choices=scopes,
metavar=scopes_metavar,
metavar=spack.config.SCOPES_METAVAR,
default=spack.config.default_modify_scope(),
help="configuration scope to modify",
)
@@ -119,7 +118,7 @@ def setup_parser(subparser):
remove_parser.add_argument(
"--scope",
choices=scopes,
metavar=scopes_metavar,
metavar=spack.config.SCOPES_METAVAR,
default=spack.config.default_modify_scope(),
help="configuration scope to modify",
)
@@ -138,7 +137,7 @@ def setup_parser(subparser):
set_url_parser.add_argument(
"--scope",
choices=scopes,
metavar=scopes_metavar,
metavar=spack.config.SCOPES_METAVAR,
default=spack.config.default_modify_scope(),
help="configuration scope to modify",
)
@@ -167,7 +166,7 @@ def setup_parser(subparser):
set_parser.add_argument(
"--scope",
choices=scopes,
metavar=scopes_metavar,
metavar=spack.config.SCOPES_METAVAR,
default=spack.config.default_modify_scope(),
help="configuration scope to modify",
)
@@ -178,7 +177,7 @@ def setup_parser(subparser):
list_parser.add_argument(
"--scope",
choices=scopes,
metavar=scopes_metavar,
metavar=spack.config.SCOPES_METAVAR,
default=spack.config.default_list_scope(),
help="configuration scope to read from",
)
@@ -444,7 +443,7 @@ def mirror_create(args):
)
# When no directory is provided, the source dir is used
path = args.directory or spack.caches.fetch_cache_location()
path = args.directory or spack.caches.FETCH_CACHE_location()
if args.all and not ev.active_environment():
create_mirror_for_all_specs(
@@ -474,7 +473,7 @@ def create_mirror_for_all_specs(path, skip_unstable_versions, selection_fn):
path, skip_unstable_versions=skip_unstable_versions
)
for candidate in mirror_specs:
pkg_cls = spack.repo.path.get_pkg_class(candidate.name)
pkg_cls = spack.repo.PATH.get_pkg_class(candidate.name)
pkg_obj = pkg_cls(spack.spec.Spec(candidate))
mirror_stats.next_spec(pkg_obj.spec)
spack.mirror.create_mirror_from_package_object(pkg_obj, mirror_cache, mirror_stats)

View File

@@ -309,7 +309,7 @@ def refresh(module_type, specs, args):
# Skip unknown packages.
writers = [
cls(spec, args.module_set_name) for spec in specs if spack.repo.path.exists(spec.name)
cls(spec, args.module_set_name) for spec in specs if spack.repo.PATH.exists(spec.name)
]
# Filter excluded packages early
@@ -321,12 +321,13 @@ def refresh(module_type, specs, args):
file2writer[item.layout.filename].append(item)
if len(file2writer) != len(writers):
spec_fmt_str = "{name}@={version}%{compiler}/{hash:7} {variants} arch={arch}"
message = "Name clashes detected in module files:\n"
for filename, writer_list in file2writer.items():
if len(writer_list) > 1:
message += "\nfile: {0}\n".format(filename)
for x in writer_list:
message += "spec: {0}\n".format(x.spec.format())
message += "spec: {0}\n".format(x.spec.format(spec_fmt_str))
tty.error(message)
tty.error("Operation aborted")
raise SystemExit(1)
@@ -376,7 +377,7 @@ def refresh(module_type, specs, args):
def modules_cmd(parser, args, module_type, callbacks=callbacks):
# Qualifiers to be used when querying the db for specs
constraint_qualifiers = {
"refresh": {"installed": True, "known": lambda x: not spack.repo.path.exists(x)}
"refresh": {"installed": True, "known": lambda x: not spack.repo.PATH.exists(x)}
}
query_args = constraint_qualifiers.get(args.subparser_name, {})

View File

@@ -143,7 +143,7 @@ def pkg_source(args):
tty.die("spack pkg source requires exactly one spec")
spec = specs[0]
filename = spack.repo.path.filename_for_package_name(spec.name)
filename = spack.repo.PATH.filename_for_package_name(spec.name)
# regular source dump -- just get the package and print its contents
if args.canonical:
@@ -184,7 +184,7 @@ def pkg_grep(args, unknown_args):
grouper = lambda e: e[0] // 500
# set up iterator and save the first group to ensure we don't end up with a group of size 1
groups = itertools.groupby(enumerate(spack.repo.path.all_package_paths()), grouper)
groups = itertools.groupby(enumerate(spack.repo.PATH.all_package_paths()), grouper)
if not groups:
return 0 # no packages to search

View File

@@ -24,7 +24,7 @@ def setup_parser(subparser):
def providers(parser, args):
valid_virtuals = sorted(spack.repo.path.provider_index.providers.keys())
valid_virtuals = sorted(spack.repo.PATH.provider_index.providers.keys())
buffer = io.StringIO()
isatty = sys.stdout.isatty()
@@ -53,5 +53,5 @@ def providers(parser, args):
for spec in specs:
if sys.stdout.isatty():
print("{0}:".format(spec))
spack.cmd.display_specs(sorted(spack.repo.path.providers_for(spec)))
spack.cmd.display_specs(sorted(spack.repo.PATH.providers_for(spec)))
print("")

View File

@@ -20,7 +20,6 @@
def setup_parser(subparser):
sp = subparser.add_subparsers(metavar="SUBCOMMAND", dest="repo_command")
scopes = spack.config.scopes()
scopes_metavar = spack.config.scopes_metavar
# Create
create_parser = sp.add_parser("create", help=repo_create.__doc__)
@@ -45,7 +44,7 @@ def setup_parser(subparser):
list_parser.add_argument(
"--scope",
choices=scopes,
metavar=scopes_metavar,
metavar=spack.config.SCOPES_METAVAR,
default=spack.config.default_list_scope(),
help="configuration scope to read from",
)
@@ -56,7 +55,7 @@ def setup_parser(subparser):
add_parser.add_argument(
"--scope",
choices=scopes,
metavar=scopes_metavar,
metavar=spack.config.SCOPES_METAVAR,
default=spack.config.default_modify_scope(),
help="configuration scope to modify",
)
@@ -69,7 +68,7 @@ def setup_parser(subparser):
remove_parser.add_argument(
"--scope",
choices=scopes,
metavar=scopes_metavar,
metavar=spack.config.SCOPES_METAVAR,
default=spack.config.default_modify_scope(),
help="configuration scope to modify",
)

View File

@@ -29,7 +29,7 @@ def setup_parser(subparser):
def _show_patch(sha256):
"""Show a record from the patch index."""
patches = spack.repo.path.patch_index.index
patches = spack.repo.PATH.patch_index.index
data = patches.get(sha256)
if not data:
@@ -47,7 +47,7 @@ def _show_patch(sha256):
owner = rec["owner"]
if "relative_path" in rec:
pkg_dir = spack.repo.path.get_pkg_class(owner).package_dir
pkg_dir = spack.repo.PATH.get_pkg_class(owner).package_dir
path = os.path.join(pkg_dir, rec["relative_path"])
print(" path: %s" % path)
else:
@@ -60,7 +60,7 @@ def _show_patch(sha256):
def resource_list(args):
"""list all resources known to spack (currently just patches)"""
patches = spack.repo.path.patch_index.index
patches = spack.repo.PATH.patch_index.index
for sha256 in patches:
if args.only_hashes:
print(sha256)

View File

@@ -68,7 +68,7 @@ def tags(parser, args):
return
# unique list of available tags
available_tags = sorted(spack.repo.path.tag_index.keys())
available_tags = sorted(spack.repo.PATH.tag_index.keys())
if not available_tags:
tty.msg("No tagged packages")
return

View File

@@ -228,7 +228,7 @@ def create_reporter(args, specs_to_test, test_suite):
def test_list(args):
"""list installed packages with available tests"""
tagged = set(spack.repo.path.packages_with_tags(*args.tag)) if args.tag else set()
tagged = set(spack.repo.PATH.packages_with_tags(*args.tag)) if args.tag else set()
def has_test_and_tags(pkg_class):
tests = spack.install_test.test_functions(pkg_class)
@@ -237,7 +237,7 @@ def has_test_and_tags(pkg_class):
if args.list_all:
report_packages = [
pkg_class.name
for pkg_class in spack.repo.path.all_package_classes()
for pkg_class in spack.repo.PATH.all_package_classes()
if has_test_and_tags(pkg_class)
]

View File

@@ -209,12 +209,11 @@ def unit_test(parser, args, unknown_args):
# mock configuration used by unit tests
# Note: skip on windows here because for the moment,
# clingo is wholly unsupported from bootstrap
if sys.platform != "win32":
with spack.bootstrap.ensure_bootstrap_configuration():
spack.bootstrap.ensure_core_dependencies()
if pytest is None:
spack.bootstrap.ensure_environment_dependencies()
import pytest
with spack.bootstrap.ensure_bootstrap_configuration():
spack.bootstrap.ensure_core_dependencies()
if pytest is None:
spack.bootstrap.ensure_environment_dependencies()
import pytest
if args.pytest_help:
# make the pytest.main help output more accurate

View File

@@ -155,7 +155,7 @@ def url_list(args):
urls = set()
# Gather set of URLs from all packages
for pkg_cls in spack.repo.path.all_package_classes():
for pkg_cls in spack.repo.PATH.all_package_classes():
url = getattr(pkg_cls, "url", None)
urls = url_list_parsing(args, urls, url, pkg_cls)
@@ -192,7 +192,7 @@ def url_summary(args):
tty.msg("Generating a summary of URL parsing in Spack...")
# Loop through all packages
for pkg_cls in spack.repo.path.all_package_classes():
for pkg_cls in spack.repo.PATH.all_package_classes():
urls = set()
pkg = pkg_cls(spack.spec.Spec(pkg_cls.name))
@@ -336,7 +336,7 @@ def add(self, pkg_name, fetcher):
version_stats = UrlStats()
resource_stats = UrlStats()
for pkg_cls in spack.repo.path.all_package_classes():
for pkg_cls in spack.repo.PATH.all_package_classes():
npkgs += 1
for v in pkg_cls.versions:

View File

@@ -45,7 +45,7 @@ def setup_parser(subparser):
def versions(parser, args):
spec = spack.spec.Spec(args.package)
pkg_cls = spack.repo.path.get_pkg_class(spec.name)
pkg_cls = spack.repo.PATH.get_pkg_class(spec.name)
pkg = pkg_cls(spec)
safe_versions = pkg.versions

View File

@@ -135,7 +135,7 @@ def _init_compiler_config(*, scope):
def compiler_config_files():
config_files = list()
config = spack.config.config
config = spack.config.CONFIG
for scope in config.file_scopes:
name = scope.name
compiler_config = config.get("compilers", scope=name)
@@ -169,7 +169,7 @@ def remove_compiler_from_config(compiler_spec, scope=None):
"""
candidate_scopes = [scope]
if scope is None:
candidate_scopes = spack.config.config.scopes.keys()
candidate_scopes = spack.config.CONFIG.scopes.keys()
removal_happened = False
for current_scope in candidate_scopes:
@@ -523,7 +523,7 @@ def compiler_for_spec(compiler_spec, arch_spec):
@_auto_compiler_spec
def get_compiler_duplicates(compiler_spec, arch_spec):
config = spack.config.config
config = spack.config.CONFIG
scope_to_compilers = {}
for scope in config.scopes:

View File

@@ -28,6 +28,7 @@
import spack.abi
import spack.compilers
import spack.config
import spack.environment
import spack.error
import spack.platforms
@@ -37,7 +38,6 @@
import spack.tengine
import spack.util.path
import spack.variant as vt
from spack.config import config
from spack.package_prefs import PackagePrefs, is_spec_buildable, spec_externals
from spack.version import ClosedOpenRange, VersionList, ver
@@ -76,7 +76,7 @@ class Concretizer:
def __init__(self, abstract_spec=None):
if Concretizer.check_for_compiler_existence is None:
Concretizer.check_for_compiler_existence = not config.get(
Concretizer.check_for_compiler_existence = not spack.config.get(
"config:install_missing_compilers", False
)
self.abstract_spec = abstract_spec
@@ -113,7 +113,7 @@ def _valid_virtuals_and_externals(self, spec):
pref_key = lambda spec: 0 # no-op pref key
if spec.virtual:
candidates = spack.repo.path.providers_for(spec)
candidates = spack.repo.PATH.providers_for(spec)
if not candidates:
raise spack.error.UnsatisfiableProviderSpecError(candidates[0], spec)

View File

@@ -47,6 +47,8 @@
import spack.platforms
import spack.schema
import spack.schema.bootstrap
import spack.schema.cdash
import spack.schema.ci
import spack.schema.compilers
import spack.schema.concretizer
import spack.schema.config
@@ -64,7 +66,7 @@
from spack.util.cpus import cpus_available
#: Dict from section names -> schema for that section
section_schemas = {
SECTION_SCHEMAS = {
"compilers": spack.schema.compilers.schema,
"concretizer": spack.schema.concretizer.schema,
"mirrors": spack.schema.mirrors.schema,
@@ -80,16 +82,16 @@
# Same as above, but including keys for environments
# this allows us to unify config reading between configs and environments
all_schemas = copy.deepcopy(section_schemas)
all_schemas.update({spack.schema.env.TOP_LEVEL_KEY: spack.schema.env.schema})
_ALL_SCHEMAS = copy.deepcopy(SECTION_SCHEMAS)
_ALL_SCHEMAS.update({spack.schema.env.TOP_LEVEL_KEY: spack.schema.env.schema})
#: Path to the default configuration
configuration_defaults_path = ("defaults", os.path.join(spack.paths.etc_path, "defaults"))
CONFIGURATION_DEFAULTS_PATH = ("defaults", os.path.join(spack.paths.etc_path, "defaults"))
#: Hard-coded default values for some key configuration options.
#: This ensures that Spack will still work even if config.yaml in
#: the defaults scope is removed.
config_defaults = {
CONFIG_DEFAULTS = {
"config": {
"debug": False,
"connect_timeout": 10,
@@ -105,10 +107,10 @@
#: metavar to use for commands that accept scopes
#: this is shorter and more readable than listing all choices
scopes_metavar = "{defaults,system,site,user}[/PLATFORM] or env:ENVIRONMENT"
SCOPES_METAVAR = "{defaults,system,site,user}[/PLATFORM] or env:ENVIRONMENT"
#: Base name for the (internal) overrides scope.
overrides_base_name = "overrides-"
_OVERRIDES_BASE_NAME = "overrides-"
class ConfigScope:
@@ -134,7 +136,7 @@ def get_section_filename(self, section):
def get_section(self, section):
if section not in self.sections:
path = self.get_section_filename(section)
schema = section_schemas[section]
schema = SECTION_SCHEMAS[section]
data = read_config_file(path, schema)
self.sections[section] = data
return self.sections[section]
@@ -145,7 +147,7 @@ def _write_section(self, section):
# We copy data here to avoid adding defaults at write time
validate_data = copy.deepcopy(data)
validate(validate_data, section_schemas[section])
validate(validate_data, SECTION_SCHEMAS[section])
try:
mkdirp(self.path)
@@ -317,7 +319,7 @@ def __init__(self, name, data=None):
data = InternalConfigScope._process_dict_keyname_overrides(data)
for section in data:
dsec = data[section]
validate({section: dsec}, section_schemas[section])
validate({section: dsec}, SECTION_SCHEMAS[section])
self.sections[section] = _mark_internal(syaml.syaml_dict({section: dsec}), name)
def get_section_filename(self, section):
@@ -333,7 +335,7 @@ def _write_section(self, section):
"""This only validates, as the data is already in memory."""
data = self.get_section(section)
if data is not None:
validate(data, section_schemas[section])
validate(data, SECTION_SCHEMAS[section])
self.sections[section] = _mark_internal(data, self.name)
def __repr__(self):
@@ -430,7 +432,7 @@ def file_scopes(self) -> List[ConfigScope]:
return [
s
for s in self.scopes.values()
if (type(s) == ConfigScope or type(s) == SingleFileScope)
if (type(s) is ConfigScope or type(s) is SingleFileScope)
]
def highest_precedence_scope(self) -> ConfigScope:
@@ -711,11 +713,11 @@ def override(path_or_scope, value=None):
"""
if isinstance(path_or_scope, ConfigScope):
overrides = path_or_scope
config.push_scope(path_or_scope)
CONFIG.push_scope(path_or_scope)
else:
base_name = overrides_base_name
base_name = _OVERRIDES_BASE_NAME
# Ensure the new override gets a unique scope name
current_overrides = [s.name for s in config.matching_scopes(r"^{0}".format(base_name))]
current_overrides = [s.name for s in CONFIG.matching_scopes(r"^{0}".format(base_name))]
num_overrides = len(current_overrides)
while True:
scope_name = "{0}{1}".format(base_name, num_overrides)
@@ -725,19 +727,19 @@ def override(path_or_scope, value=None):
break
overrides = InternalConfigScope(scope_name)
config.push_scope(overrides)
config.set(path_or_scope, value, scope=scope_name)
CONFIG.push_scope(overrides)
CONFIG.set(path_or_scope, value, scope=scope_name)
try:
yield config
yield CONFIG
finally:
scope = config.remove_scope(overrides.name)
scope = CONFIG.remove_scope(overrides.name)
assert scope is overrides
#: configuration scopes added on the command line
#: set by ``spack.main.main()``.
command_line_scopes: List[str] = []
COMMAND_LINE_SCOPES: List[str] = []
def _add_platform_scope(cfg, scope_type, name, path):
@@ -781,14 +783,14 @@ def create():
cfg = Configuration()
# first do the builtin, hardcoded defaults
builtin = InternalConfigScope("_builtin", config_defaults)
builtin = InternalConfigScope("_builtin", CONFIG_DEFAULTS)
cfg.push_scope(builtin)
# Builtin paths to configuration files in Spack
configuration_paths = [
# Default configuration scope is the lowest-level scope. These are
# versioned with Spack and can be overridden by systems, sites or users
configuration_defaults_path
CONFIGURATION_DEFAULTS_PATH
]
disable_local_config = "SPACK_DISABLE_LOCAL_CONFIG" in os.environ
@@ -815,7 +817,7 @@ def create():
_add_platform_scope(cfg, ConfigScope, name, path)
# add command-line scopes
_add_command_line_scopes(cfg, command_line_scopes)
_add_command_line_scopes(cfg, COMMAND_LINE_SCOPES)
# we make a special scope for spack commands so that they can
# override configuration options.
@@ -825,7 +827,7 @@ def create():
#: This is the singleton configuration instance for Spack.
config: Union[Configuration, llnl.util.lang.Singleton] = llnl.util.lang.Singleton(create)
CONFIG: Union[Configuration, llnl.util.lang.Singleton] = llnl.util.lang.Singleton(create)
def add_from_file(filename, scope=None):
@@ -838,7 +840,7 @@ def add_from_file(filename, scope=None):
# update all sections from config dict
# We have to iterate on keys to keep overrides from the file
for section in data.keys():
if section in section_schemas.keys():
if section in SECTION_SCHEMAS.keys():
# Special handling for compiler scope difference
# Has to be handled after we choose a section
if scope is None:
@@ -849,7 +851,7 @@ def add_from_file(filename, scope=None):
new = merge_yaml(existing, value)
# We cannot call config.set directly (set is a type)
config.set(section, new, scope)
CONFIG.set(section, new, scope)
def add(fullpath, scope=None):
@@ -897,12 +899,12 @@ def add(fullpath, scope=None):
# merge value into existing
new = merge_yaml(existing, value)
config.set(path, new, scope)
CONFIG.set(path, new, scope)
def get(path, default=None, scope=None):
"""Module-level wrapper for ``Configuration.get()``."""
return config.get(path, default, scope)
return CONFIG.get(path, default, scope)
def set(path, value, scope=None):
@@ -910,26 +912,26 @@ def set(path, value, scope=None):
Accepts the path syntax described in ``get()``.
"""
return config.set(path, value, scope)
return CONFIG.set(path, value, scope)
def add_default_platform_scope(platform):
plat_name = os.path.join("defaults", platform)
plat_path = os.path.join(configuration_defaults_path[1], platform)
config.push_scope(ConfigScope(plat_name, plat_path))
plat_path = os.path.join(CONFIGURATION_DEFAULTS_PATH[1], platform)
CONFIG.push_scope(ConfigScope(plat_name, plat_path))
def scopes():
"""Convenience function to get list of configuration scopes."""
return config.scopes
return CONFIG.scopes
def _validate_section_name(section):
"""Exit if the section is not a valid section."""
if section not in section_schemas:
if section not in SECTION_SCHEMAS:
raise ConfigSectionError(
"Invalid config section: '%s'. Options are: %s"
% (section, " ".join(section_schemas.keys()))
% (section, " ".join(SECTION_SCHEMAS.keys()))
)
@@ -990,7 +992,7 @@ def read_config_file(filename, schema=None):
if data:
if not schema:
key = next(iter(data))
schema = all_schemas[key]
schema = _ALL_SCHEMAS[key]
validate(data, schema)
return data
@@ -1089,7 +1091,7 @@ def get_valid_type(path):
test_data = {component: test_data}
try:
validate(test_data, section_schemas[section])
validate(test_data, SECTION_SCHEMAS[section])
except (ConfigFormatError, AttributeError) as e:
jsonschema_error = e.validation_error
if jsonschema_error.validator == "type":
@@ -1278,9 +1280,9 @@ def default_modify_scope(section="config"):
If this is not 'compilers', a general (non-platform) scope is used.
"""
if section == "compilers":
return spack.config.config.highest_precedence_scope().name
return CONFIG.highest_precedence_scope().name
else:
return spack.config.config.highest_precedence_non_platform_scope().name
return CONFIG.highest_precedence_non_platform_scope().name
def default_list_scope():
@@ -1337,18 +1339,18 @@ def use_configuration(*scopes_or_paths):
Returns:
Configuration object associated with the scopes passed as arguments
"""
global config
global CONFIG
# Normalize input and construct a Configuration object
configuration = _config_from(scopes_or_paths)
config.clear_caches(), configuration.clear_caches()
CONFIG.clear_caches(), configuration.clear_caches()
saved_config, config = config, configuration
saved_config, CONFIG = CONFIG, configuration
try:
yield configuration
finally:
config = saved_config
CONFIG = saved_config
@llnl.util.lang.memoized

View File

@@ -5,8 +5,8 @@
"""Writers for different kind of recipes and related
convenience functions.
"""
import collections
import copy
from collections import namedtuple
from typing import Optional
import spack.environment as ev
@@ -159,13 +159,13 @@ def depfile(self):
@tengine.context_property
def run(self):
"""Information related to the run image."""
Run = collections.namedtuple("Run", ["image"])
Run = namedtuple("Run", ["image"])
return Run(image=self.final_image)
@tengine.context_property
def build(self):
"""Information related to the build image."""
Build = collections.namedtuple("Build", ["image"])
Build = namedtuple("Build", ["image"])
return Build(image=self.build_image)
@tengine.context_property
@@ -176,12 +176,13 @@ def strip(self):
@tengine.context_property
def paths(self):
"""Important paths in the image"""
Paths = collections.namedtuple("Paths", ["environment", "store", "hidden_view", "view"])
Paths = namedtuple("Paths", ["environment", "store", "view_parent", "view", "former_view"])
return Paths(
environment="/opt/spack-environment",
store="/opt/software",
hidden_view="/opt/._view",
view="/opt/view",
view_parent="/opt/views",
view="/opt/views/view",
former_view="/opt/view", # /opt/view -> /opt/views/view for backward compatibility
)
@tengine.context_property
@@ -257,7 +258,7 @@ def _package_info_from(self, package_list):
update, install, clean = commands_for(os_pkg_manager)
Packages = collections.namedtuple("Packages", ["update", "install", "list", "clean"])
Packages = namedtuple("Packages", ["update", "install", "list", "clean"])
return Packages(update=update, install=install, list=package_list, clean=clean)
def _os_pkg_manager(self):
@@ -273,7 +274,7 @@ def _os_pkg_manager(self):
@tengine.context_property
def extra_instructions(self):
Extras = collections.namedtuple("Extra", ["build", "final"])
Extras = namedtuple("Extra", ["build", "final"])
extras = self.container_config.get("extra_instructions", {})
build, final = extras.get("build", None), extras.get("final", None)
return Extras(build=build, final=final)
@@ -295,7 +296,7 @@ def bootstrap(self):
context = {"bootstrap": {"image": self.bootstrap_image, "spack_checkout": command}}
bootstrap_recipe = env.get_template(template_path).render(**context)
Bootstrap = collections.namedtuple("Bootstrap", ["image", "recipe"])
Bootstrap = namedtuple("Bootstrap", ["image", "recipe"])
return Bootstrap(image=self.bootstrap_image, recipe=bootstrap_recipe)
@tengine.context_property
@@ -303,7 +304,7 @@ def render_phase(self):
render_bootstrap = bool(self.bootstrap_image)
render_build = not (self.last_phase == "bootstrap")
render_final = self.last_phase in (None, "final")
Render = collections.namedtuple("Render", ["bootstrap", "build", "final"])
Render = namedtuple("Render", ["bootstrap", "build", "final"])
return Render(bootstrap=render_bootstrap, build=render_build, final=render_final)
def __call__(self):

View File

@@ -90,7 +90,7 @@ def spec_from_entry(entry):
name=entry["name"], version=entry["version"], compiler=compiler_str, arch=arch_str
)
pkg_cls = spack.repo.path.get_pkg_class(entry["name"])
pkg_cls = spack.repo.PATH.get_pkg_class(entry["name"])
if "parameters" in entry:
variant_strs = list()

View File

@@ -21,10 +21,11 @@
import contextlib
import datetime
import os
import pathlib
import socket
import sys
import time
from typing import Dict, List, NamedTuple, Set, Type, Union
from typing import Any, Callable, Dict, Generator, List, NamedTuple, Set, Type, Union
try:
import uuid
@@ -141,22 +142,23 @@ class InstallStatuses:
def canonicalize(cls, query_arg):
if query_arg is True:
return [cls.INSTALLED]
elif query_arg is False:
if query_arg is False:
return [cls.MISSING]
elif query_arg is any:
if query_arg is any:
return [cls.INSTALLED, cls.DEPRECATED, cls.MISSING]
elif isinstance(query_arg, InstallStatus):
if isinstance(query_arg, InstallStatus):
return [query_arg]
else:
try: # Try block catches if it is not an iterable at all
if any(type(x) != InstallStatus for x in query_arg):
raise TypeError
except TypeError:
raise TypeError(
"installation query must be `any`, boolean, "
"InstallStatus, or iterable of InstallStatus"
)
return query_arg
try:
statuses = list(query_arg)
if all(isinstance(x, InstallStatus) for x in statuses):
return statuses
except TypeError:
pass
raise TypeError(
"installation query must be `any`, boolean, "
"InstallStatus, or iterable of InstallStatus"
)
class InstallRecord:
@@ -306,15 +308,16 @@ def __reduce__(self):
"""
#: Data class to configure locks in Database objects
#:
#: Args:
#: enable (bool): whether to enable locks or not.
#: database_timeout (int or None): timeout for the database lock
#: package_timeout (int or None): timeout for the package lock
class LockConfiguration(NamedTuple):
"""Data class to configure locks in Database objects
Args:
enable: whether to enable locks or not.
database_timeout: timeout for the database lock
package_timeout: timeout for the package lock
"""
enable: bool
database_timeout: Optional[int]
package_timeout: Optional[int]
@@ -348,13 +351,230 @@ def lock_configuration(configuration):
)
def prefix_lock_path(root_dir: Union[str, pathlib.Path]) -> pathlib.Path:
"""Returns the path of the prefix lock file, given the root directory.
Args:
root_dir: root directory containing the database directory
"""
return pathlib.Path(root_dir) / _DB_DIRNAME / "prefix_lock"
def failures_lock_path(root_dir: Union[str, pathlib.Path]) -> pathlib.Path:
"""Returns the path of the failures lock file, given the root directory.
Args:
root_dir: root directory containing the database directory
"""
return pathlib.Path(root_dir) / _DB_DIRNAME / "prefix_failures"
class SpecLocker:
"""Manages acquiring and releasing read or write locks on concrete specs."""
def __init__(self, lock_path: Union[str, pathlib.Path], default_timeout: Optional[float]):
self.lock_path = pathlib.Path(lock_path)
self.default_timeout = default_timeout
# Maps (spec.dag_hash(), spec.name) to the corresponding lock object
self.locks: Dict[Tuple[str, str], lk.Lock] = {}
def lock(self, spec: "spack.spec.Spec", timeout: Optional[float] = None) -> lk.Lock:
"""Returns a lock on a concrete spec.
The lock is a byte range lock on the nth byte of a file.
The lock file is ``self.lock_path``.
n is the sys.maxsize-bit prefix of the DAG hash. This makes likelihood of collision is
very low AND it gives us readers-writer lock semantics with just a single lockfile, so
no cleanup required.
"""
assert spec.concrete, "cannot lock a non-concrete spec"
timeout = timeout or self.default_timeout
key = self._lock_key(spec)
if key not in self.locks:
self.locks[key] = self.raw_lock(spec, timeout=timeout)
else:
self.locks[key].default_timeout = timeout
return self.locks[key]
def raw_lock(self, spec: "spack.spec.Spec", timeout: Optional[float] = None) -> lk.Lock:
"""Returns a raw lock for a Spec, but doesn't keep track of it."""
return lk.Lock(
str(self.lock_path),
start=spec.dag_hash_bit_prefix(bit_length(sys.maxsize)),
length=1,
default_timeout=timeout,
desc=spec.name,
)
def has_lock(self, spec: "spack.spec.Spec") -> bool:
"""Returns True if the spec is already managed by this spec locker"""
return self._lock_key(spec) in self.locks
def _lock_key(self, spec: "spack.spec.Spec") -> Tuple[str, str]:
return (spec.dag_hash(), spec.name)
@contextlib.contextmanager
def write_lock(self, spec: "spack.spec.Spec") -> Generator["SpecLocker", None, None]:
lock = self.lock(spec)
lock.acquire_write()
try:
yield self
except lk.LockError:
# This addresses the case where a nested lock attempt fails inside
# of this context manager
raise
except (Exception, KeyboardInterrupt):
lock.release_write()
raise
else:
lock.release_write()
def clear(self, spec: "spack.spec.Spec") -> Tuple[bool, Optional[lk.Lock]]:
key = self._lock_key(spec)
lock = self.locks.pop(key, None)
return bool(lock), lock
def clear_all(self, clear_fn: Optional[Callable[[lk.Lock], Any]] = None) -> None:
if clear_fn is not None:
for lock in self.locks.values():
clear_fn(lock)
self.locks.clear()
class FailureTracker:
"""Tracks installation failures.
Prefix failure marking takes the form of a byte range lock on the nth
byte of a file for coordinating between concurrent parallel build
processes and a persistent file, named with the full hash and
containing the spec, in a subdirectory of the database to enable
persistence across overlapping but separate related build processes.
The failure lock file lives alongside the install DB.
``n`` is the sys.maxsize-bit prefix of the associated DAG hash to make
the likelihood of collision very low with no cleanup required.
"""
def __init__(self, root_dir: Union[str, pathlib.Path], default_timeout: Optional[float]):
#: Ensure a persistent location for dealing with parallel installation
#: failures (e.g., across near-concurrent processes).
self.dir = pathlib.Path(root_dir) / _DB_DIRNAME / "failures"
self.dir.mkdir(parents=True, exist_ok=True)
self.locker = SpecLocker(failures_lock_path(root_dir), default_timeout=default_timeout)
def clear(self, spec: "spack.spec.Spec", force: bool = False) -> None:
"""Removes any persistent and cached failure tracking for the spec.
see `mark()`.
Args:
spec: the spec whose failure indicators are being removed
force: True if the failure information should be cleared when a failure lock
exists for the file, or False if the failure should not be cleared (e.g.,
it may be associated with a concurrent build)
"""
locked = self.lock_taken(spec)
if locked and not force:
tty.msg(f"Retaining failure marking for {spec.name} due to lock")
return
if locked:
tty.warn(f"Removing failure marking despite lock for {spec.name}")
succeeded, lock = self.locker.clear(spec)
if succeeded and lock is not None:
lock.release_write()
if self.persistent_mark(spec):
path = self._path(spec)
tty.debug(f"Removing failure marking for {spec.name}")
try:
path.unlink()
except OSError as err:
tty.warn(
f"Unable to remove failure marking for {spec.name} ({str(path)}): {str(err)}"
)
def clear_all(self) -> None:
"""Force remove install failure tracking files."""
tty.debug("Releasing prefix failure locks")
self.locker.clear_all(
clear_fn=lambda x: x.release_write() if x.is_write_locked() else True
)
tty.debug("Removing prefix failure tracking files")
try:
for fail_mark in os.listdir(str(self.dir)):
try:
(self.dir / fail_mark).unlink()
except OSError as exc:
tty.warn(f"Unable to remove failure marking file {fail_mark}: {str(exc)}")
except OSError as exc:
tty.warn(f"Unable to remove failure marking files: {str(exc)}")
def mark(self, spec: "spack.spec.Spec") -> lk.Lock:
"""Marks a spec as failing to install.
Args:
spec: spec that failed to install
"""
# Dump the spec to the failure file for (manual) debugging purposes
path = self._path(spec)
path.write_text(spec.to_json())
# Also ensure a failure lock is taken to prevent cleanup removal
# of failure status information during a concurrent parallel build.
if not self.locker.has_lock(spec):
try:
mark = self.locker.lock(spec)
mark.acquire_write()
except lk.LockTimeoutError:
# Unlikely that another process failed to install at the same
# time but log it anyway.
tty.debug(f"PID {os.getpid()} failed to mark install failure for {spec.name}")
tty.warn(f"Unable to mark {spec.name} as failed.")
return self.locker.lock(spec)
def has_failed(self, spec: "spack.spec.Spec") -> bool:
"""Return True if the spec is marked as failed."""
# The failure was detected in this process.
if self.locker.has_lock(spec):
return True
# The failure was detected by a concurrent process (e.g., an srun),
# which is expected to be holding a write lock if that is the case.
if self.lock_taken(spec):
return True
# Determine if the spec may have been marked as failed by a separate
# spack build process running concurrently.
return self.persistent_mark(spec)
def lock_taken(self, spec: "spack.spec.Spec") -> bool:
"""Return True if another process has a failure lock on the spec."""
check = self.locker.raw_lock(spec)
return check.is_write_locked()
def persistent_mark(self, spec: "spack.spec.Spec") -> bool:
"""Determine if the spec has a persistent failure marking."""
return self._path(spec).exists()
def _path(self, spec: "spack.spec.Spec") -> pathlib.Path:
"""Return the path to the spec's failure file, which may not exist."""
assert spec.concrete, "concrete spec required for failure path"
return self.dir / f"{spec.name}-{spec.dag_hash()}"
class Database:
#: Per-process lock objects for each install prefix
_prefix_locks: Dict[str, lk.Lock] = {}
#: Per-process failure (lock) objects for each install prefix
_prefix_failures: Dict[str, lk.Lock] = {}
#: Fields written for each install record
record_fields: Tuple[str, ...] = DEFAULT_INSTALL_RECORD_FIELDS
@@ -392,24 +612,10 @@ def __init__(
self._verifier_path = os.path.join(self.database_directory, "index_verifier")
self._lock_path = os.path.join(self.database_directory, "lock")
# This is for other classes to use to lock prefix directories.
self.prefix_lock_path = os.path.join(self.database_directory, "prefix_lock")
# Ensure a persistent location for dealing with parallel installation
# failures (e.g., across near-concurrent processes).
self._failure_dir = os.path.join(self.database_directory, "failures")
# Support special locks for handling parallel installation failures
# of a spec.
self.prefix_fail_path = os.path.join(self.database_directory, "prefix_failures")
# Create needed directories and files
if not is_upstream and not os.path.exists(self.database_directory):
fs.mkdirp(self.database_directory)
if not is_upstream and not os.path.exists(self._failure_dir):
fs.mkdirp(self._failure_dir)
self.is_upstream = is_upstream
self.last_seen_verifier = ""
# Failed write transactions (interrupted by exceptions) will alert
@@ -423,15 +629,7 @@ def __init__(
# initialize rest of state.
self.db_lock_timeout = lock_cfg.database_timeout
self.package_lock_timeout = lock_cfg.package_timeout
tty.debug("DATABASE LOCK TIMEOUT: {0}s".format(str(self.db_lock_timeout)))
timeout_format_str = (
"{0}s".format(str(self.package_lock_timeout))
if self.package_lock_timeout
else "No timeout"
)
tty.debug("PACKAGE LOCK TIMEOUT: {0}".format(str(timeout_format_str)))
self.lock: Union[ForbiddenLock, lk.Lock]
if self.is_upstream:
@@ -471,212 +669,6 @@ def read_transaction(self):
"""Get a read lock context manager for use in a `with` block."""
return self._read_transaction_impl(self.lock, acquire=self._read)
def _failed_spec_path(self, spec):
"""Return the path to the spec's failure file, which may not exist."""
if not spec.concrete:
raise ValueError("Concrete spec required for failure path for {0}".format(spec.name))
return os.path.join(self._failure_dir, "{0}-{1}".format(spec.name, spec.dag_hash()))
def clear_all_failures(self) -> None:
"""Force remove install failure tracking files."""
tty.debug("Releasing prefix failure locks")
for pkg_id in list(self._prefix_failures.keys()):
lock = self._prefix_failures.pop(pkg_id, None)
if lock:
lock.release_write()
# Remove all failure markings (aka files)
tty.debug("Removing prefix failure tracking files")
for fail_mark in os.listdir(self._failure_dir):
try:
os.remove(os.path.join(self._failure_dir, fail_mark))
except OSError as exc:
tty.warn(
"Unable to remove failure marking file {0}: {1}".format(fail_mark, str(exc))
)
def clear_failure(self, spec: "spack.spec.Spec", force: bool = False) -> None:
"""
Remove any persistent and cached failure tracking for the spec.
see `mark_failed()`.
Args:
spec: the spec whose failure indicators are being removed
force: True if the failure information should be cleared when a prefix failure
lock exists for the file, or False if the failure should not be cleared (e.g.,
it may be associated with a concurrent build)
"""
failure_locked = self.prefix_failure_locked(spec)
if failure_locked and not force:
tty.msg("Retaining failure marking for {0} due to lock".format(spec.name))
return
if failure_locked:
tty.warn("Removing failure marking despite lock for {0}".format(spec.name))
lock = self._prefix_failures.pop(spec.prefix, None)
if lock:
lock.release_write()
if self.prefix_failure_marked(spec):
try:
path = self._failed_spec_path(spec)
tty.debug("Removing failure marking for {0}".format(spec.name))
os.remove(path)
except OSError as err:
tty.warn(
"Unable to remove failure marking for {0} ({1}): {2}".format(
spec.name, path, str(err)
)
)
def mark_failed(self, spec: "spack.spec.Spec") -> lk.Lock:
"""
Mark a spec as failing to install.
Prefix failure marking takes the form of a byte range lock on the nth
byte of a file for coordinating between concurrent parallel build
processes and a persistent file, named with the full hash and
containing the spec, in a subdirectory of the database to enable
persistence across overlapping but separate related build processes.
The failure lock file, ``spack.store.STORE.db.prefix_failures``, lives
alongside the install DB. ``n`` is the sys.maxsize-bit prefix of the
associated DAG hash to make the likelihood of collision very low with
no cleanup required.
"""
# Dump the spec to the failure file for (manual) debugging purposes
path = self._failed_spec_path(spec)
with open(path, "w") as f:
spec.to_json(f)
# Also ensure a failure lock is taken to prevent cleanup removal
# of failure status information during a concurrent parallel build.
err = "Unable to mark {0.name} as failed."
prefix = spec.prefix
if prefix not in self._prefix_failures:
mark = lk.Lock(
self.prefix_fail_path,
start=spec.dag_hash_bit_prefix(bit_length(sys.maxsize)),
length=1,
default_timeout=self.package_lock_timeout,
desc=spec.name,
)
try:
mark.acquire_write()
except lk.LockTimeoutError:
# Unlikely that another process failed to install at the same
# time but log it anyway.
tty.debug(
"PID {0} failed to mark install failure for {1}".format(os.getpid(), spec.name)
)
tty.warn(err.format(spec))
# Whether we or another process marked it as a failure, track it
# as such locally.
self._prefix_failures[prefix] = mark
return self._prefix_failures[prefix]
def prefix_failed(self, spec: "spack.spec.Spec") -> bool:
"""Return True if the prefix (installation) is marked as failed."""
# The failure was detected in this process.
if spec.prefix in self._prefix_failures:
return True
# The failure was detected by a concurrent process (e.g., an srun),
# which is expected to be holding a write lock if that is the case.
if self.prefix_failure_locked(spec):
return True
# Determine if the spec may have been marked as failed by a separate
# spack build process running concurrently.
return self.prefix_failure_marked(spec)
def prefix_failure_locked(self, spec: "spack.spec.Spec") -> bool:
"""Return True if a process has a failure lock on the spec."""
check = lk.Lock(
self.prefix_fail_path,
start=spec.dag_hash_bit_prefix(bit_length(sys.maxsize)),
length=1,
default_timeout=self.package_lock_timeout,
desc=spec.name,
)
return check.is_write_locked()
def prefix_failure_marked(self, spec: "spack.spec.Spec") -> bool:
"""Determine if the spec has a persistent failure marking."""
return os.path.exists(self._failed_spec_path(spec))
def prefix_lock(self, spec: "spack.spec.Spec", timeout: Optional[float] = None) -> lk.Lock:
"""Get a lock on a particular spec's installation directory.
NOTE: The installation directory **does not** need to exist.
Prefix lock is a byte range lock on the nth byte of a file.
The lock file is ``spack.store.STORE.db.prefix_lock`` -- the DB
tells us what to call it and it lives alongside the install DB.
n is the sys.maxsize-bit prefix of the DAG hash. This makes
likelihood of collision is very low AND it gives us
readers-writer lock semantics with just a single lockfile, so no
cleanup required.
"""
timeout = timeout or self.package_lock_timeout
prefix = spec.prefix
if prefix not in self._prefix_locks:
self._prefix_locks[prefix] = lk.Lock(
self.prefix_lock_path,
start=spec.dag_hash_bit_prefix(bit_length(sys.maxsize)),
length=1,
default_timeout=timeout,
desc=spec.name,
)
elif timeout != self._prefix_locks[prefix].default_timeout:
self._prefix_locks[prefix].default_timeout = timeout
return self._prefix_locks[prefix]
@contextlib.contextmanager
def prefix_read_lock(self, spec):
prefix_lock = self.prefix_lock(spec)
prefix_lock.acquire_read()
try:
yield self
except lk.LockError:
# This addresses the case where a nested lock attempt fails inside
# of this context manager
raise
except (Exception, KeyboardInterrupt):
prefix_lock.release_read()
raise
else:
prefix_lock.release_read()
@contextlib.contextmanager
def prefix_write_lock(self, spec):
prefix_lock = self.prefix_lock(spec)
prefix_lock.acquire_write()
try:
yield self
except lk.LockError:
# This addresses the case where a nested lock attempt fails inside
# of this context manager
raise
except (Exception, KeyboardInterrupt):
prefix_lock.release_write()
raise
else:
prefix_lock.release_write()
def _write_to_file(self, stream):
"""Write out the database in JSON format to the stream passed
as argument.

View File

@@ -33,7 +33,7 @@ class OpenMpi(Package):
import functools
import os.path
import re
from typing import List, Optional, Set, Union
from typing import Any, Callable, List, Optional, Set, Tuple, Union
import llnl.util.lang
import llnl.util.tty.color
@@ -520,7 +520,8 @@ def _execute_conflicts(pkg):
# Save in a list the conflicts and the associated custom messages
when_spec_list = pkg.conflicts.setdefault(conflict_spec, [])
when_spec_list.append((when_spec, msg))
msg_with_name = f"{pkg.name}: {msg}" if msg is not None else msg
when_spec_list.append((when_spec, msg_with_name))
return _execute_conflicts
@@ -663,39 +664,35 @@ def _execute_patch(pkg_or_dep):
@directive("variants")
def variant(
name,
default=None,
description="",
values=None,
multi=None,
validator=None,
when=None,
sticky=False,
name: str,
default: Optional[Any] = None,
description: str = "",
values: Optional[Union[collections.abc.Sequence, Callable[[Any], bool]]] = None,
multi: Optional[bool] = None,
validator: Optional[Callable[[str, str, Tuple[Any, ...]], None]] = None,
when: Optional[Union[str, bool]] = None,
sticky: bool = False,
):
"""Define a variant for the package. Packager can specify a default
value as well as a text description.
"""Define a variant for the package.
Packager can specify a default value as well as a text description.
Args:
name (str): name of the variant
default (str or bool): default value for the variant, if not
specified otherwise the default will be False for a boolean
variant and 'nothing' for a multi-valued variant
description (str): description of the purpose of the variant
values (tuple or typing.Callable): either a tuple of strings containing the
allowed values, or a callable accepting one value and returning
True if it is valid
multi (bool): if False only one value per spec is allowed for
this variant
validator (typing.Callable): optional group validator to enforce additional
logic. It receives the package name, the variant name and a tuple
of values and should raise an instance of SpackError if the group
doesn't meet the additional constraints
when (spack.spec.Spec, bool): optional condition on which the
variant applies
sticky (bool): the variant should not be changed by the concretizer to
find a valid concrete spec.
name: Name of the variant
default: Default value for the variant, if not specified otherwise the default will be
False for a boolean variant and 'nothing' for a multi-valued variant
description: Description of the purpose of the variant
values: Either a tuple of strings containing the allowed values, or a callable accepting
one value and returning True if it is valid
multi: If False only one value per spec is allowed for this variant
validator: Optional group validator to enforce additional logic. It receives the package
name, the variant name and a tuple of values and should raise an instance of SpackError
if the group doesn't meet the additional constraints
when: Optional condition on which the variant applies
sticky: The variant should not be changed by the concretizer to find a valid concrete spec
Raises:
DirectiveError: if arguments passed to the directive are invalid
DirectiveError: If arguments passed to the directive are invalid
"""
def format_error(msg, pkg):
@@ -900,7 +897,8 @@ def _execute_requires(pkg):
# Save in a list the requirements and the associated custom messages
when_spec_list = pkg.requirements.setdefault(tuple(requirement_specs), [])
when_spec_list.append((when_spec, policy, msg))
msg_with_name = f"{pkg.name}: {msg}" if msg is not None else msg
when_spec_list.append((when_spec, policy, msg_with_name))
return _execute_requires

View File

@@ -11,6 +11,7 @@
import shutil
import sys
from contextlib import contextmanager
from pathlib import Path
import llnl.util.filesystem as fs
import llnl.util.tty as tty
@@ -104,7 +105,7 @@ def relative_path_for_spec(self, spec):
projection = spack.projections.get_projection(self.projections, spec)
path = spec.format(projection)
return path
return str(Path(path))
def write_spec(self, spec, path):
"""Write a spec out to a file."""

View File

@@ -203,7 +203,7 @@ def activate(env, use_env_repo=False):
env.store_token = spack.store.reinitialize()
if use_env_repo:
spack.repo.path.put_first(env.repo)
spack.repo.PATH.put_first(env.repo)
tty.debug("Using environment '%s'" % env.name)
@@ -227,7 +227,7 @@ def deactivate():
# use _repo so we only remove if a repo was actually constructed
if _active_environment._repo:
spack.repo.path.remove(_active_environment._repo)
spack.repo.PATH.remove(_active_environment._repo)
tty.debug("Deactivated environment '%s'" % _active_environment.name)
@@ -1084,8 +1084,8 @@ def add(self, user_spec, list_name=user_speclist_name):
if list_name == user_speclist_name:
if spec.anonymous:
raise SpackEnvironmentError("cannot add anonymous specs to an environment")
elif not spack.repo.path.exists(spec.name) and not spec.abstract_hash:
virtuals = spack.repo.path.provider_index.providers.keys()
elif not spack.repo.PATH.exists(spec.name) and not spec.abstract_hash:
virtuals = spack.repo.PATH.provider_index.providers.keys()
if spec.name not in virtuals:
msg = "no such package: %s" % spec.name
raise SpackEnvironmentError(msg)
@@ -1262,7 +1262,7 @@ def develop(self, spec: Spec, path: str, clone: bool = False) -> bool:
# better if we can create the `source_path` directly into its final
# destination.
abspath = spack.util.path.canonicalize_path(path, default_wd=self.path)
pkg_cls = spack.repo.path.get_pkg_class(spec.name)
pkg_cls = spack.repo.PATH.get_pkg_class(spec.name)
# We construct a package class ourselves, rather than asking for
# Spec.package, since Spec only allows this when it is concrete
package = pkg_cls(spec)
@@ -1490,7 +1490,7 @@ def _concretize_separately(self, tests=False):
# for a write lock. We do this indirectly by retrieving the
# provider index, which should in turn trigger the update of
# all the indexes if there's any need for that.
_ = spack.repo.path.provider_index
_ = spack.repo.PATH.provider_index
# Ensure we have compilers in compilers.yaml to avoid that
# processes try to write the config file in parallel
@@ -1994,14 +1994,10 @@ def get_one_by_hash(self, dag_hash):
def all_matching_specs(self, *specs: spack.spec.Spec) -> List[Spec]:
"""Returns all concretized specs in the environment satisfying any of the input specs"""
# Look up abstract hashes ahead of time, to avoid O(n^2) traversal.
specs = [s.lookup_hash() for s in specs]
# Avoid double lookup by directly calling _satisfies.
return [
s
for s in traverse.traverse_nodes(self.concrete_roots(), key=traverse.by_dag_hash)
if any(s._satisfies(t) for t in specs)
if any(s.satisfies(t) for t in specs)
]
@spack.repo.autospec
@@ -2280,7 +2276,7 @@ def _add_to_environment_repository(self, spec_node: Spec) -> None:
repository = spack.repo.create_or_construct(repository_dir, spec_node.namespace)
pkg_dir = repository.dirname_for_package_name(spec_node.name)
fs.mkdirp(pkg_dir)
spack.repo.path.dump_provenance(spec_node, pkg_dir)
spack.repo.PATH.dump_provenance(spec_node, pkg_dir)
def manifest_uptodate_or_warn(self):
"""Emits a warning if the manifest file is not up-to-date."""
@@ -2448,13 +2444,13 @@ def make_repo_path(root):
def prepare_config_scope(env):
"""Add env's scope to the global configuration search path."""
for scope in env.config_scopes():
spack.config.config.push_scope(scope)
spack.config.CONFIG.push_scope(scope)
def deactivate_config_scope(env):
"""Remove any scopes from env from the global config path."""
for scope in env.config_scopes():
spack.config.config.remove_scope(scope.name)
spack.config.CONFIG.remove_scope(scope.name)
def manifest_file(env_name_or_dir):

View File

@@ -535,7 +535,7 @@ def edge_entry(self, edge):
def _static_edges(specs, deptype):
for spec in specs:
pkg_cls = spack.repo.path.get_pkg_class(spec.name)
pkg_cls = spack.repo.PATH.get_pkg_class(spec.name)
possible = pkg_cls.possible_dependencies(expand_virtuals=True, deptype=deptype)
for parent_name, dependencies in possible.items():

View File

@@ -49,7 +49,7 @@ def __call__(self, spec):
def _content_hash_override(spec):
pkg_cls = spack.repo.path.get_pkg_class(spec.name)
pkg_cls = spack.repo.PATH.get_pkg_class(spec.name)
pkg = pkg_cls(spec)
return pkg.content_hash()

View File

@@ -1147,12 +1147,12 @@ def write_test_result(self, spec, result):
def write_reproducibility_data(self):
for spec in self.specs:
repo_cache_path = self.stage.repo.join(spec.name)
spack.repo.path.dump_provenance(spec, repo_cache_path)
spack.repo.PATH.dump_provenance(spec, repo_cache_path)
for vspec in spec.package.virtuals_provided:
repo_cache_path = self.stage.repo.join(vspec.name)
if not os.path.exists(repo_cache_path):
try:
spack.repo.path.dump_provenance(vspec, repo_cache_path)
spack.repo.PATH.dump_provenance(vspec, repo_cache_path)
except spack.repo.UnknownPackageError:
pass # not all virtuals have package files

View File

@@ -519,13 +519,6 @@ def _try_install_from_binary_cache(
)
def clear_failures() -> None:
"""
Remove all failure tracking markers for the Spack instance.
"""
spack.store.STORE.db.clear_all_failures()
def combine_phase_logs(phase_log_files: List[str], log_path: str) -> None:
"""
Read set or list of logs and combine them into one file.
@@ -597,7 +590,7 @@ def dump_packages(spec: "spack.spec.Spec", path: str) -> None:
# Get the location of the package in the dest repo.
dest_pkg_dir = repo.dirname_for_package_name(node.name)
if node is spec:
spack.repo.path.dump_provenance(node, dest_pkg_dir)
spack.repo.PATH.dump_provenance(node, dest_pkg_dir)
elif source_pkg_dir:
fs.install_tree(source_pkg_dir, dest_pkg_dir)
@@ -1126,15 +1119,13 @@ class PackageInstaller:
instance.
"""
def __init__(self, installs: List[Tuple["spack.package_base.PackageBase", dict]] = []):
def __init__(self, installs: List[Tuple["spack.package_base.PackageBase", dict]] = []) -> None:
"""Initialize the installer.
Args:
installs (list): list of tuples, where each
tuple consists of a package (PackageBase) and its associated
install arguments (dict)
Return:
PackageInstaller: instance
"""
# List of build requests
self.build_requests = [BuildRequest(pkg, install_args) for pkg, install_args in installs]
@@ -1287,7 +1278,7 @@ def _check_deps_status(self, request: BuildRequest) -> None:
dep_id = package_id(dep_pkg)
# Check for failure since a prefix lock is not required
if spack.store.STORE.db.prefix_failed(dep):
if spack.store.STORE.failure_tracker.has_failed(dep):
action = "'spack install' the dependency"
msg = "{0} is marked as an install failure: {1}".format(dep_id, action)
raise InstallError(err.format(request.pkg_id, msg), pkg=dep_pkg)
@@ -1502,7 +1493,7 @@ def _ensure_locked(
if lock is None:
tty.debug(msg.format("Acquiring", desc, pkg_id, pretty_seconds(timeout or 0)))
op = "acquire"
lock = spack.store.STORE.db.prefix_lock(pkg.spec, timeout)
lock = spack.store.STORE.prefix_locker.lock(pkg.spec, timeout)
if timeout != lock.default_timeout:
tty.warn(
"Expected prefix lock timeout {0}, not {1}".format(
@@ -1627,12 +1618,12 @@ def _add_tasks(self, request: BuildRequest, all_deps):
# Clear any persistent failure markings _unless_ they are
# associated with another process in this parallel build
# of the spec.
spack.store.STORE.db.clear_failure(dep, force=False)
spack.store.STORE.failure_tracker.clear(dep, force=False)
install_package = request.install_args.get("install_package")
if install_package and request.pkg_id not in self.build_tasks:
# Be sure to clear any previous failure
spack.store.STORE.db.clear_failure(request.spec, force=True)
spack.store.STORE.failure_tracker.clear(request.spec, force=True)
# If not installing dependencies, then determine their
# installation status before proceeding
@@ -1888,7 +1879,7 @@ def _update_failed(
err = "" if exc is None else ": {0}".format(str(exc))
tty.debug("Flagging {0} as failed{1}".format(pkg_id, err))
if mark:
self.failed[pkg_id] = spack.store.STORE.db.mark_failed(task.pkg.spec)
self.failed[pkg_id] = spack.store.STORE.failure_tracker.mark(task.pkg.spec)
else:
self.failed[pkg_id] = None
task.status = STATUS_FAILED
@@ -2074,7 +2065,7 @@ def install(self) -> None:
# Flag a failed spec. Do not need an (install) prefix lock since
# assume using a separate (failed) prefix lock file.
if pkg_id in self.failed or spack.store.STORE.db.prefix_failed(spec):
if pkg_id in self.failed or spack.store.STORE.failure_tracker.has_failed(spec):
term_status.clear()
tty.warn("{0} failed to install".format(pkg_id))
self._update_failed(task)

View File

@@ -602,10 +602,10 @@ def setup_main_options(args):
key = syaml.syaml_str("repos")
key.override = True
spack.config.config.scopes["command_line"].sections["repos"] = syaml.syaml_dict(
spack.config.CONFIG.scopes["command_line"].sections["repos"] = syaml.syaml_dict(
[(key, [spack.paths.mock_packages_path])]
)
spack.repo.path = spack.repo.create(spack.config.config)
spack.repo.PATH = spack.repo.create(spack.config.CONFIG)
# If the user asked for it, don't check ssl certs.
if args.insecure:
@@ -930,7 +930,7 @@ def _main(argv=None):
# make spack.config aware of any command line configuration scopes
if args.config_scopes:
spack.config.command_line_scopes = args.config_scopes
spack.config.COMMAND_LINE_SCOPES = args.config_scopes
# ensure options on spack command come before everything
setup_main_options(args)

View File

@@ -442,7 +442,7 @@ def mirror_archive_paths(fetcher, per_package_ref, spec=None):
storage path of the resource associated with the specified ``fetcher``."""
ext = None
if spec:
pkg_cls = spack.repo.path.get_pkg_class(spec.name)
pkg_cls = spack.repo.PATH.get_pkg_class(spec.name)
versions = pkg_cls.versions.get(spec.version, {})
ext = versions.get("extension", None)
# If the spec does not explicitly specify an extension (the default case),
@@ -474,7 +474,7 @@ def get_all_versions(specs):
"""
version_specs = []
for spec in specs:
pkg_cls = spack.repo.path.get_pkg_class(spec.name)
pkg_cls = spack.repo.PATH.get_pkg_class(spec.name)
# Skip any package that has no known versions.
if not pkg_cls.versions:
tty.msg("No safe (checksummed) versions for package %s" % pkg_cls.name)

View File

@@ -833,7 +833,7 @@ def ensure_modules_are_enabled_or_warn():
return
# Check if we have custom TCL module sections
for scope in spack.config.config.file_scopes:
for scope in spack.config.CONFIG.file_scopes:
# Skip default configuration
if scope.name.startswith("default"):
continue

View File

@@ -143,7 +143,7 @@ def hierarchy_tokens(self):
# Check if all the tokens in the hierarchy are virtual specs.
# If not warn the user and raise an error.
not_virtual = [t for t in tokens if t != "compiler" and not spack.repo.path.is_virtual(t)]
not_virtual = [t for t in tokens if t != "compiler" and not spack.repo.PATH.is_virtual(t)]
if not_virtual:
msg = "Non-virtual specs in 'hierarchy' list for lmod: {0}\n"
msg += "Please check the 'modules.yaml' configuration files"

View File

@@ -236,7 +236,7 @@ def install(self, prefix):
# Create a multimethod with this name if there is not one already
original_method = MultiMethodMeta._locals.get(method.__name__)
if not type(original_method) == SpecMultiMethod:
if not isinstance(original_method, SpecMultiMethod):
original_method = SpecMultiMethod(original_method)
if self.spec is not None:

View File

@@ -665,7 +665,7 @@ def __init__(self, spec):
self.win_rpath = fsys.WindowsSimulatedRPath(self)
if self.is_extension:
pkg_cls = spack.repo.path.get_pkg_class(self.extendee_spec.name)
pkg_cls = spack.repo.PATH.get_pkg_class(self.extendee_spec.name)
pkg_cls(self.extendee_spec)._check_extendable()
super().__init__()
@@ -728,11 +728,11 @@ def possible_dependencies(
continue
# expand virtuals if enabled, otherwise just stop at virtuals
if spack.repo.path.is_virtual(name):
if spack.repo.PATH.is_virtual(name):
if virtuals is not None:
virtuals.add(name)
if expand_virtuals:
providers = spack.repo.path.providers_for(name)
providers = spack.repo.PATH.providers_for(name)
dep_names = [spec.name for spec in providers]
else:
visited.setdefault(cls.name, set()).add(name)
@@ -756,7 +756,7 @@ def possible_dependencies(
continue
try:
dep_cls = spack.repo.path.get_pkg_class(dep_name)
dep_cls = spack.repo.PATH.get_pkg_class(dep_name)
except spack.repo.UnknownPackageError:
# log unknown packages
missing.setdefault(cls.name, set()).add(dep_name)
@@ -2209,7 +2209,7 @@ def uninstall_by_spec(spec, force=False, deprecator=None):
pkg = None
# Pre-uninstall hook runs first.
with spack.store.STORE.db.prefix_write_lock(spec):
with spack.store.STORE.prefix_locker.write_lock(spec):
if pkg is not None:
try:
spack.hooks.pre_uninstall(spec)
@@ -2459,8 +2459,8 @@ def possible_dependencies(*pkg_or_spec, **kwargs):
if not isinstance(pos, spack.spec.Spec):
pos = spack.spec.Spec(pos)
if spack.repo.path.is_virtual(pos.name):
packages.extend(p.package_class for p in spack.repo.path.providers_for(pos.name))
if spack.repo.PATH.is_virtual(pos.name):
packages.extend(p.package_class for p in spack.repo.PATH.providers_for(pos.name))
continue
else:
packages.append(pos.package_class)

View File

@@ -147,7 +147,7 @@ def preferred_variants(cls, pkg_name):
variants = " ".join(variants)
# Only return variants that are actually supported by the package
pkg_cls = spack.repo.path.get_pkg_class(pkg_name)
pkg_cls = spack.repo.PATH.get_pkg_class(pkg_name)
spec = spack.spec.Spec("%s %s" % (pkg_name, variants))
return dict(
(name, variant) for name, variant in spec.variants.items() if name in pkg_cls.variants
@@ -162,7 +162,7 @@ def spec_externals(spec):
from spack.util.module_cmd import path_from_modules # noqa: F401
def _package(maybe_abstract_spec):
pkg_cls = spack.repo.path.get_pkg_class(spec.name)
pkg_cls = spack.repo.PATH.get_pkg_class(spec.name)
return pkg_cls(maybe_abstract_spec)
allpkgs = spack.config.get("packages")
@@ -199,7 +199,7 @@ def is_spec_buildable(spec):
so_far = all_buildable # the default "so far"
def _package(s):
pkg_cls = spack.repo.path.get_pkg_class(s.name)
pkg_cls = spack.repo.PATH.get_pkg_class(s.name)
return pkg_cls(s)
# check whether any providers for this package override the default

View File

@@ -238,7 +238,7 @@ def to_dict(self):
def from_dict(dictionary, repository=None):
"""Create a patch from json dictionary."""
repository = repository or spack.repo.path
repository = repository or spack.repo.PATH
owner = dictionary.get("owner")
if "owner" not in dictionary:
raise ValueError("Invalid patch dictionary: %s" % dictionary)

View File

@@ -10,11 +10,12 @@
dependencies.
"""
import os
from pathlib import PurePath
import llnl.util.filesystem
#: This file lives in $prefix/lib/spack/spack/__file__
prefix = llnl.util.filesystem.ancestor(__file__, 4)
prefix = str(PurePath(llnl.util.filesystem.ancestor(__file__, 4)))
#: synonym for prefix
spack_root = prefix
@@ -88,7 +89,7 @@ def _get_user_cache_path():
return os.path.expanduser(os.getenv("SPACK_USER_CACHE_PATH") or "~%s.spack" % os.sep)
user_cache_path = _get_user_cache_path()
user_cache_path = str(PurePath(_get_user_cache_path()))
#: junit, cdash, etc. reports about builds
reports_path = os.path.join(user_cache_path, "reports")

View File

@@ -64,7 +64,7 @@ def use_platform(new_platform):
host = _PickleableCallable(new_platform)
# Clear configuration and compiler caches
spack.config.config.clear_caches()
spack.config.CONFIG.clear_caches()
spack.compilers._cache_config_files = []
yield new_platform
@@ -73,5 +73,5 @@ def use_platform(new_platform):
host = original_host_fn
# Clear configuration and compiler caches
spack.config.config.clear_caches()
spack.config.CONFIG.clear_caches()
spack.compilers._cache_config_files = []

View File

@@ -149,7 +149,7 @@ def compute_loader(self, fullname):
# If it's a module in some repo, or if it is the repo's
# namespace, let the repo handle it.
for repo in path.repos:
for repo in PATH.repos:
# We are using the namespace of the repo and the repo contains the package
if namespace == repo.full_namespace:
# With 2 nested conditionals we can call "repo.real_name" only once
@@ -163,7 +163,7 @@ def compute_loader(self, fullname):
# No repo provides the namespace, but it is a valid prefix of
# something in the RepoPath.
if path.by_namespace.is_prefix(fullname):
if PATH.by_namespace.is_prefix(fullname):
return SpackNamespaceLoader()
return None
@@ -184,9 +184,9 @@ def compute_loader(self, fullname):
def packages_path():
"""Get the test repo if it is active, otherwise the builtin repo."""
try:
return spack.repo.path.get_repo("builtin.mock").packages_path
return spack.repo.PATH.get_repo("builtin.mock").packages_path
except spack.repo.UnknownNamespaceError:
return spack.repo.path.get_repo("builtin").packages_path
return spack.repo.PATH.get_repo("builtin").packages_path
class GitExe:
@@ -282,7 +282,7 @@ def add_package_to_git_stage(packages):
git = GitExe()
for pkg_name in packages:
filename = spack.repo.path.filename_for_package_name(pkg_name)
filename = spack.repo.PATH.filename_for_package_name(pkg_name)
if not os.path.isfile(filename):
tty.die("No such package: %s. Path does not exist:" % pkg_name, filename)
@@ -647,7 +647,7 @@ class RepoPath:
"""
def __init__(self, *repos, **kwargs):
cache = kwargs.get("cache", spack.caches.misc_cache)
cache = kwargs.get("cache", spack.caches.MISC_CACHE)
self.repos = []
self.by_namespace = nm.NamespaceTrie()
@@ -966,7 +966,7 @@ def check(condition, msg):
# Indexes for this repository, computed lazily
self._repo_index = None
self._cache = cache or spack.caches.misc_cache
self._cache = cache or spack.caches.MISC_CACHE
def real_name(self, import_name):
"""Allow users to import Spack packages using Python identifiers.
@@ -1357,7 +1357,7 @@ def create_or_construct(path, namespace=None):
def _path(configuration=None):
"""Get the singleton RepoPath instance for Spack."""
configuration = configuration or spack.config.config
configuration = configuration or spack.config.CONFIG
return create(configuration=configuration)
@@ -1374,7 +1374,7 @@ def create(configuration):
#: Singleton repo path instance
path: Union[RepoPath, llnl.util.lang.Singleton] = llnl.util.lang.Singleton(_path)
PATH: Union[RepoPath, llnl.util.lang.Singleton] = llnl.util.lang.Singleton(_path)
# Add the finder to sys.meta_path
REPOS_FINDER = ReposFinder()
@@ -1383,7 +1383,7 @@ def create(configuration):
def all_package_names(include_virtuals=False):
"""Convenience wrapper around ``spack.repo.all_package_names()``."""
return path.all_package_names(include_virtuals)
return PATH.all_package_names(include_virtuals)
@contextlib.contextmanager
@@ -1398,21 +1398,21 @@ def use_repositories(*paths_and_repos, **kwargs):
Returns:
Corresponding RepoPath object
"""
global path
global PATH
# TODO (Python 2.7): remove this kwargs on deprecation of Python 2.7 support
override = kwargs.get("override", True)
paths = [getattr(x, "root", x) for x in paths_and_repos]
scope_name = "use-repo-{}".format(uuid.uuid4())
repos_key = "repos:" if override else "repos"
spack.config.config.push_scope(
spack.config.CONFIG.push_scope(
spack.config.InternalConfigScope(name=scope_name, data={repos_key: paths})
)
path, saved = create(configuration=spack.config.config), path
PATH, saved = create(configuration=spack.config.CONFIG), PATH
try:
yield path
yield PATH
finally:
spack.config.config.remove_scope(scope_name=scope_name)
path = saved
spack.config.CONFIG.remove_scope(scope_name=scope_name)
PATH = saved
class MockRepositoryBuilder:

View File

@@ -28,6 +28,12 @@
"unify": {
"oneOf": [{"type": "boolean"}, {"type": "string", "enum": ["when_possible"]}]
},
"duplicates": {
"type": "object",
"properties": {
"strategy": {"type": "string", "enum": ["none", "minimal", "full"]}
},
},
},
}
}

File diff suppressed because it is too large Load Diff

File diff suppressed because it is too large Load Diff

View File

@@ -0,0 +1,159 @@
# Copyright 2013-2023 Lawrence Livermore National Security, LLC and other
# Spack Project Developers. See the top-level COPYRIGHT file for details.
#
# SPDX-License-Identifier: (Apache-2.0 OR MIT)
import collections
from typing import List, Set, Tuple
import spack.dependency
import spack.package_base
PossibleDependencies = Set[str]
class Counter:
"""Computes the possible packages and the maximum number of duplicates
allowed for each of them.
Args:
specs: abstract specs to concretize
tests: if True, add test dependencies to the list of possible packages
"""
def __init__(self, specs: List["spack.spec.Spec"], tests: bool) -> None:
self.specs = specs
self.link_run_types: Tuple[str, ...] = ("link", "run", "test")
self.all_types: Tuple[str, ...] = spack.dependency.all_deptypes
if not tests:
self.link_run_types = ("link", "run")
self.all_types = ("link", "run", "build")
self._possible_dependencies: PossibleDependencies = set()
self._possible_virtuals: Set[str] = set(x.name for x in specs if x.virtual)
def possible_dependencies(self) -> PossibleDependencies:
"""Returns the list of possible dependencies"""
self.ensure_cache_values()
return self._possible_dependencies
def possible_virtuals(self) -> Set[str]:
"""Returns the list of possible virtuals"""
self.ensure_cache_values()
return self._possible_virtuals
def ensure_cache_values(self) -> None:
"""Ensure the cache values have been computed"""
if self._possible_dependencies:
return
self._compute_cache_values()
def possible_packages_facts(self, gen: "spack.solver.asp.PyclingoDriver", fn) -> None:
"""Emit facts associated with the possible packages"""
raise NotImplementedError("must be implemented by derived classes")
def _compute_cache_values(self):
raise NotImplementedError("must be implemented by derived classes")
class NoDuplicatesCounter(Counter):
def _compute_cache_values(self):
result = spack.package_base.possible_dependencies(
*self.specs, virtuals=self._possible_virtuals, deptype=self.all_types
)
self._possible_dependencies = set(result)
def possible_packages_facts(self, gen, fn):
gen.h2("Maximum number of nodes (packages)")
for package_name in sorted(self.possible_dependencies()):
gen.fact(fn.max_dupes(package_name, 1))
gen.newline()
gen.h2("Maximum number of nodes (virtual packages)")
for package_name in sorted(self.possible_virtuals()):
gen.fact(fn.max_dupes(package_name, 1))
gen.newline()
gen.h2("Possible package in link-run subDAG")
for name in sorted(self.possible_dependencies()):
gen.fact(fn.possible_in_link_run(name))
gen.newline()
class MinimalDuplicatesCounter(NoDuplicatesCounter):
def __init__(self, specs, tests):
super().__init__(specs, tests)
self._link_run: PossibleDependencies = set()
self._direct_build: PossibleDependencies = set()
self._total_build: PossibleDependencies = set()
self._link_run_virtuals: Set[str] = set()
def _compute_cache_values(self):
self._link_run = set(
spack.package_base.possible_dependencies(
*self.specs, virtuals=self._possible_virtuals, deptype=self.link_run_types
)
)
self._link_run_virtuals.update(self._possible_virtuals)
for x in self._link_run:
current = spack.repo.PATH.get_pkg_class(x).dependencies_of_type("build")
self._direct_build.update(current)
self._total_build = set(
spack.package_base.possible_dependencies(
*self._direct_build, virtuals=self._possible_virtuals, deptype=self.all_types
)
)
self._possible_dependencies = set(self._link_run) | set(self._total_build)
def possible_packages_facts(self, gen, fn):
build_tools = set(spack.repo.PATH.packages_with_tags("build-tools"))
gen.h2("Packages with at most a single node")
for package_name in sorted(self.possible_dependencies() - build_tools):
gen.fact(fn.max_dupes(package_name, 1))
gen.newline()
gen.h2("Packages with at multiple possible nodes (build-tools)")
for package_name in sorted(self.possible_dependencies() & build_tools):
gen.fact(fn.max_dupes(package_name, 2))
gen.fact(fn.multiple_unification_sets(package_name))
gen.newline()
gen.h2("Maximum number of nodes (virtual packages)")
for package_name in sorted(self.possible_virtuals()):
gen.fact(fn.max_dupes(package_name, 1))
gen.newline()
gen.h2("Possible package in link-run subDAG")
for name in sorted(self._link_run):
gen.fact(fn.possible_in_link_run(name))
gen.newline()
class FullDuplicatesCounter(MinimalDuplicatesCounter):
def possible_packages_facts(self, gen, fn):
build_tools = set(spack.repo.PATH.packages_with_tags("build-tools"))
counter = collections.Counter(
list(self._link_run) + list(self._total_build) + list(self._direct_build)
)
gen.h2("Maximum number of nodes")
for pkg, count in sorted(counter.items(), key=lambda x: (x[1], x[0])):
count = min(count, 2)
gen.fact(fn.max_dupes(pkg, count))
gen.newline()
gen.h2("Build unification sets ")
for name in sorted(self.possible_dependencies() & build_tools):
gen.fact(fn.multiple_unification_sets(name))
gen.newline()
gen.h2("Possible package in link-run subDAG")
for name in sorted(self._link_run):
gen.fact(fn.possible_in_link_run(name))
gen.newline()
counter = collections.Counter(
list(self._link_run_virtuals) + list(self._possible_virtuals)
)
gen.h2("Maximum number of virtual nodes")
for pkg, count in sorted(counter.items(), key=lambda x: (x[1], x[0])):
gen.fact(fn.max_dupes(pkg, count))
gen.newline()

View File

@@ -0,0 +1,21 @@
% Copyright 2013-2023 Lawrence Livermore National Security, LLC and other
% Spack Project Developers. See the top-level COPYRIGHT file for details.
%
% SPDX-License-Identifier: (Apache-2.0 OR MIT)
%=============================================================================
% Avoid cycles in the DAG
%
% Some combinations of conditional dependencies can result in cycles;
% this ensures that we solve around them. Note that these rules are quite
% demanding on both grounding and solving, since they need to compute and
% consider all possible paths between pair of nodes.
%=============================================================================
#program no_cycle.
path(Parent, Child) :- depends_on(Parent, Child).
path(Parent, Descendant) :- path(Parent, A), depends_on(A, Descendant).
:- path(A, A).
#defined depends_on/2.

View File

@@ -0,0 +1,44 @@
% Copyright 2013-2023 Lawrence Livermore National Security, LLC and other
% Spack Project Developers. See the top-level COPYRIGHT file for details.
%
% SPDX-License-Identifier: (Apache-2.0 OR MIT)
%=============================================================================
% Heuristic to speed-up solves
%=============================================================================
%-----------------
% Domain heuristic
%-----------------
#heuristic attr("hash", node(0, Package), Hash) : literal(_, "root", Package). [45, init]
#heuristic attr("root", node(0, Package)) : literal(_, "root", Package). [45, true]
#heuristic attr("node", node(0, Package)) : literal(_, "root", Package). [45, true]
#heuristic attr("node", node(0, Package)) : literal(_, "node", Package). [45, true]
% Root node
#heuristic attr("version", node(0, Package), Version) : pkg_fact(Package, version_declared(Version, 0)), attr("root", node(0, Package)). [35, true]
#heuristic version_weight(node(0, Package), 0) : pkg_fact(Package, version_declared(Version, 0)), attr("root", node(0, Package)). [35, true]
#heuristic attr("variant_value", node(0, Package), Variant, Value) : variant_default_value(Package, Variant, Value), attr("root", node(0, Package)). [35, true]
#heuristic attr("node_target", node(0, Package), Target) : pkg_fact(Package, target_weight(Target, 0)), attr("root", node(0, Package)). [35, true]
#heuristic node_target_weight(node(0, Package), 0) : attr("root", node(0, Package)). [35, true]
#heuristic node_compiler(node(0, Package), CompilerID) : default_compiler_preference(ID, 0), compiler_id(ID), attr("root", node(0, Package)). [35, true]
% Providers
#heuristic attr("node", node(0, Package)) : default_provider_preference(Virtual, Package, 0), possible_in_link_run(Package). [30, true]
% node(ID, _)
#heuristic attr("version", node(ID, Package), Version) : pkg_fact(Package, version_declared(Version, ID)), attr("node", node(ID, Package)). [25-5*ID, true]
#heuristic version_weight(node(ID, Package), ID) : pkg_fact(Package, version_declared(Version, ID)), attr("node", node(ID, Package)). [25-5*ID, true]
#heuristic attr("variant_value", node(ID, Package), Variant, Value) : variant_default_value(Package, Variant, Value), attr("node", node(ID, Package)). [25-5*ID, true]
#heuristic attr("node_target", node(ID, Package), Target) : pkg_fact(Package, target_weight(Target, ID)), attr("node", node(ID, Package)). [25-5*ID, true]
#heuristic node_target_weight(node(ID, Package), ID) : attr("node", node(ID, Package)). [25-5*ID, true]
#heuristic node_compiler(node(ID, Package), CompilerID) : default_compiler_preference(ID, ID), compiler_id(ID), attr("node", node(ID, Package)). [25-5*ID, true]
% node(ID, _), split build dependencies
#heuristic attr("version", node(ID, Package), Version) : pkg_fact(Package, version_declared(Version, ID)), attr("node", node(ID, Package)), multiple_unification_sets(Package). [25, true]
#heuristic version_weight(node(ID, Package), ID) : pkg_fact(Package, version_declared(Version, ID)), attr("node", node(ID, Package)), multiple_unification_sets(Package). [25, true]
#heuristic attr("variant_value", node(ID, Package), Variant, Value) : variant_default_value(Package, Variant, Value), attr("node", node(ID, Package)), multiple_unification_sets(Package). [25, true]
#heuristic attr("node_target", node(ID, Package), Target) : pkg_fact(Package, target_weight(Target, ID)), attr("node", node(ID, Package)), multiple_unification_sets(Package). [25, true]
#heuristic node_target_weight(node(ID, Package), ID) : attr("node", node(ID, Package)), multiple_unification_sets(Package). [25, true]
#heuristic node_compiler(node(ID, Package), CompilerID) : default_compiler_preference(ID, ID), compiler_id(ID), attr("node", node(ID, Package)), multiple_unification_sets(Package). [25, true]

View File

@@ -3,9 +3,11 @@
%
% SPDX-License-Identifier: (Apache-2.0 OR MIT)
%=============================================================================
% OS compatibility rules for reusing solves.
% os_compatible(RecentOS, OlderOS)
% OlderOS binaries can be used on RecentOS
%=============================================================================
% macOS
os_compatible("monterey", "bigsur").

View File

@@ -0,0 +1,27 @@
% Copyright 2013-2023 Lawrence Livermore National Security, LLC and other
% Spack Project Developers. See the top-level COPYRIGHT file for details.
%
% SPDX-License-Identifier: (Apache-2.0 OR MIT)
%=============================================================================
% Minimize the number of literals that are not solved
%
% This minimization is used for the "when_possible" concretization mode,
% otherwise we assume that all literals must be solved.
%=============================================================================
% Give clingo the choice to solve an input spec or not
{ solve_literal(ID) } :- literal(ID).
literal_not_solved(ID) :- not solve_literal(ID), literal(ID).
% Make a problem with "zero literals solved" unsat. This is to trigger
% looking for solutions to the ASP problem with "errors", which results
% in better reporting for users. See #30669 for details.
1 { solve_literal(ID) : literal(ID) }.
opt_criterion(300, "number of input specs not concretized").
#minimize{ 0@300: #true }.
#minimize { 1@300,ID : literal_not_solved(ID) }.
#heuristic literal_solved(ID) : literal(ID). [1, sign]
#heuristic literal_solved(ID) : literal(ID). [50, init]

View File

@@ -985,16 +985,14 @@ def __iter__(self):
def __len__(self):
return len(self.edges)
def add(self, edge):
"""Adds a new edge to this object.
Args:
edge (DependencySpec): edge to be added
"""
def add(self, edge: DependencySpec):
key = edge.spec.name if self.store_by_child else edge.parent.name
current_list = self.edges.setdefault(key, [])
current_list.append(edge)
current_list.sort(key=_sort_by_dep_types)
if key in self.edges:
lst = self.edges[key]
lst.append(edge)
lst.sort(key=_sort_by_dep_types)
else:
self.edges[key] = [edge]
def __str__(self):
return "{deps: %s}" % ", ".join(str(d) for d in sorted(self.values()))
@@ -1299,7 +1297,7 @@ def __init__(self, spec, name, query_parameters):
original_spec = getattr(spec, "wrapped_obj", spec)
self.wrapped_obj = original_spec
self.token = original_spec, name, query_parameters
is_virtual = spack.repo.path.is_virtual(name)
is_virtual = spack.repo.PATH.is_virtual(name)
self.last_query = QueryState(
name=name, extra_parameters=query_parameters, isvirtual=is_virtual
)
@@ -1733,7 +1731,7 @@ def package(self):
self.name
)
if not self._package:
self._package = spack.repo.path.get(self)
self._package = spack.repo.PATH.get(self)
return self._package
@property
@@ -1741,11 +1739,11 @@ def package_class(self):
"""Internal package call gets only the class object for a package.
Use this to just get package metadata.
"""
return spack.repo.path.get_pkg_class(self.fullname)
return spack.repo.PATH.get_pkg_class(self.fullname)
@property
def virtual(self):
return spack.repo.path.is_virtual(self.name)
return spack.repo.PATH.is_virtual(self.name)
@property
def concrete(self):
@@ -1927,19 +1925,15 @@ def _lookup_hash(self):
store, or finally, binary caches."""
import spack.environment
matches = []
active_env = spack.environment.active_environment()
if active_env:
env_matches = active_env.get_by_hash(self.abstract_hash) or []
matches = [m for m in env_matches if m._satisfies(self)]
if not matches:
db_matches = spack.store.STORE.db.get_by_hash(self.abstract_hash) or []
matches = [m for m in db_matches if m._satisfies(self)]
if not matches:
query = spack.binary_distribution.BinaryCacheQuery(True)
remote_matches = query("/" + self.abstract_hash) or []
matches = [m for m in remote_matches if m._satisfies(self)]
# First env, then store, then binary cache
matches = (
(active_env.all_matching_specs(self) if active_env else [])
or spack.store.STORE.db.query(self, installed=any)
or spack.binary_distribution.BinaryCacheQuery(True)(self)
)
if not matches:
raise InvalidHashError(self, self.abstract_hash)
@@ -1960,19 +1954,17 @@ def lookup_hash(self):
spec = self.copy(deps=False)
# root spec is replaced
if spec.abstract_hash:
new = self._lookup_hash()
spec._dup(new)
spec._dup(self._lookup_hash())
return spec
# Get dependencies that need to be replaced
for node in self.traverse(root=False):
if node.abstract_hash:
new = node._lookup_hash()
spec._add_dependency(new, deptypes=(), virtuals=())
spec._add_dependency(node._lookup_hash(), deptypes=(), virtuals=())
# reattach nodes that were not otherwise satisfied by new dependencies
for node in self.traverse(root=False):
if not any(n._satisfies(node) for n in spec.traverse()):
if not any(n.satisfies(node) for n in spec.traverse()):
spec._add_dependency(node.copy(), deptypes=(), virtuals=())
return spec
@@ -1985,9 +1977,7 @@ def replace_hash(self):
if not any(node for node in self.traverse(order="post") if node.abstract_hash):
return
spec_by_hash = self.lookup_hash()
self._dup(spec_by_hash)
self._dup(self.lookup_hash())
def to_node_dict(self, hash=ht.dag_hash):
"""Create a dictionary representing the state of this Spec.
@@ -2272,7 +2262,7 @@ def override(init_spec, change_spec):
# TODO: this doesn't account for the case where the changed spec
# (and the user spec) have dependencies
new_spec = init_spec.copy()
package_cls = spack.repo.path.get_pkg_class(new_spec.name)
package_cls = spack.repo.PATH.get_pkg_class(new_spec.name)
if change_spec.versions and not change_spec.versions == vn.any_version:
new_spec.versions = change_spec.versions
for variant, value in change_spec.variants.items():
@@ -2546,7 +2536,7 @@ def validate_detection(self):
assert isinstance(self.extra_attributes, collections.abc.Mapping), msg
# Validate the spec calling a package specific method
pkg_cls = spack.repo.path.get_pkg_class(self.name)
pkg_cls = spack.repo.PATH.get_pkg_class(self.name)
validate_fn = getattr(pkg_cls, "validate_detected_spec", lambda x, y: None)
validate_fn(self, self.extra_attributes)
@@ -2645,7 +2635,7 @@ def _expand_virtual_packages(self, concretizer):
"""
# Make an index of stuff this spec already provides
self_index = spack.provider_index.ProviderIndex(
repository=spack.repo.path, specs=self.traverse(), restrict=True
repository=spack.repo.PATH, specs=self.traverse(), restrict=True
)
changed = False
done = False
@@ -2785,7 +2775,7 @@ def _old_concretize(self, tests=False, deprecation_warning=True):
visited_user_specs = set()
for dep in self.traverse():
visited_user_specs.add(dep.name)
pkg_cls = spack.repo.path.get_pkg_class(dep.name)
pkg_cls = spack.repo.PATH.get_pkg_class(dep.name)
visited_user_specs.update(x.name for x in pkg_cls(dep).provided)
extra = set(user_spec_deps.keys()).difference(visited_user_specs)
@@ -2868,7 +2858,7 @@ def inject_patches_variant(root):
# we can do it as late as possible to allow as much
# compatibility across repositories as possible.
if s.namespace is None:
s.namespace = spack.repo.path.repo_for_pkg(s.name).namespace
s.namespace = spack.repo.PATH.repo_for_pkg(s.name).namespace
if s.concrete:
continue
@@ -2926,7 +2916,7 @@ def ensure_external_path_if_external(external_spec):
# Get the path from the module the package can override the default
# (this is mostly needed for Cray)
pkg_cls = spack.repo.path.get_pkg_class(external_spec.name)
pkg_cls = spack.repo.PATH.get_pkg_class(external_spec.name)
package = pkg_cls(external_spec)
external_spec.external_path = getattr(
package, "external_prefix", md.path_from_modules(external_spec.external_modules)
@@ -2983,9 +2973,12 @@ def _new_concretize(self, tests=False):
providers = [spec.name for spec in answer.values() if spec.package.provides(name)]
name = providers[0]
assert name in answer
node = spack.solver.asp.SpecBuilder.make_node(pkg=name)
assert (
node in answer
), f"cannot find {name} in the list of specs {','.join([n.pkg for n in answer.keys()])}"
concretized = answer[name]
concretized = answer[node]
self._dup(concretized)
def concretize(self, tests=False):
@@ -3200,7 +3193,7 @@ def _find_provider(self, vdep, provider_index):
Raise an exception if there is a conflicting virtual
dependency already in this spec.
"""
assert spack.repo.path.is_virtual_safe(vdep.name), vdep
assert spack.repo.PATH.is_virtual_safe(vdep.name), vdep
# note that this defensively copies.
providers = provider_index.providers_for(vdep)
@@ -3266,7 +3259,7 @@ def _merge_dependency(self, dependency, visited, spec_deps, provider_index, test
# If it's a virtual dependency, try to find an existing
# provider in the spec, and merge that.
virtuals = ()
if spack.repo.path.is_virtual_safe(dep.name):
if spack.repo.PATH.is_virtual_safe(dep.name):
virtuals = (dep.name,)
visited.add(dep.name)
provider = self._find_provider(dep, provider_index)
@@ -3274,11 +3267,11 @@ def _merge_dependency(self, dependency, visited, spec_deps, provider_index, test
dep = provider
else:
index = spack.provider_index.ProviderIndex(
repository=spack.repo.path, specs=[dep], restrict=True
repository=spack.repo.PATH, specs=[dep], restrict=True
)
items = list(spec_deps.items())
for name, vspec in items:
if not spack.repo.path.is_virtual_safe(vspec.name):
if not spack.repo.PATH.is_virtual_safe(vspec.name):
continue
if index.providers_for(vspec):
@@ -3428,7 +3421,7 @@ def normalize(self, force=False, tests=False, user_spec_deps=None):
# Initialize index of virtual dependency providers if
# concretize didn't pass us one already
provider_index = spack.provider_index.ProviderIndex(
repository=spack.repo.path, specs=[s for s in all_spec_deps.values()], restrict=True
repository=spack.repo.PATH, specs=[s for s in all_spec_deps.values()], restrict=True
)
# traverse the package DAG and fill out dependencies according
@@ -3459,7 +3452,7 @@ def validate_or_raise(self):
for spec in self.traverse():
# raise an UnknownPackageError if the spec's package isn't real.
if (not spec.virtual) and spec.name:
spack.repo.path.get_pkg_class(spec.fullname)
spack.repo.PATH.get_pkg_class(spec.fullname)
# validate compiler in addition to the package name.
if spec.compiler:
@@ -3519,7 +3512,8 @@ def update_variant_validate(self, variant_name, values):
for value in values:
if self.variants.get(variant_name):
msg = (
"Cannot append a value to a single-valued " "variant with an already set value"
f"cannot append the new value '{value}' to the single-valued "
f"variant '{self.variants[variant_name]}'"
)
assert pkg_variant.multi, msg
self.variants[variant_name].append(value)
@@ -3527,7 +3521,7 @@ def update_variant_validate(self, variant_name, values):
variant = pkg_variant.make_variant(value)
self.variants[variant_name] = variant
pkg_cls = spack.repo.path.get_pkg_class(self.name)
pkg_cls = spack.repo.PATH.get_pkg_class(self.name)
pkg_variant.validate_or_raise(self.variants[variant_name], pkg_cls)
def constrain(self, other, deps=True):
@@ -3719,21 +3713,25 @@ def intersects(self, other: "Spec", deps: bool = True) -> bool:
"""
other = self._autospec(other)
lhs = self.lookup_hash() or self
rhs = other.lookup_hash() or other
return lhs._intersects(rhs, deps)
def _intersects(self, other: "Spec", deps: bool = True) -> bool:
if other.concrete and self.concrete:
return self.dag_hash() == other.dag_hash()
self_hash = self.dag_hash() if self.concrete else self.abstract_hash
other_hash = other.dag_hash() if other.concrete else other.abstract_hash
if (
self_hash
and other_hash
and not (self_hash.startswith(other_hash) or other_hash.startswith(self_hash))
):
return False
# If the names are different, we need to consider virtuals
if self.name != other.name and self.name and other.name:
if self.virtual and other.virtual:
# Two virtual specs intersect only if there are providers for both
lhs = spack.repo.path.providers_for(str(self))
rhs = spack.repo.path.providers_for(str(other))
lhs = spack.repo.PATH.providers_for(str(self))
rhs = spack.repo.PATH.providers_for(str(other))
intersection = [s for s in lhs if any(s.intersects(z) for z in rhs)]
return bool(intersection)
@@ -3742,7 +3740,7 @@ def _intersects(self, other: "Spec", deps: bool = True) -> bool:
virtual_spec, non_virtual_spec = (self, other) if self.virtual else (other, self)
try:
# Here we might get an abstract spec
pkg_cls = spack.repo.path.get_pkg_class(non_virtual_spec.fullname)
pkg_cls = spack.repo.PATH.get_pkg_class(non_virtual_spec.fullname)
pkg = pkg_cls(non_virtual_spec)
except spack.repo.UnknownEntityError:
# If we can't get package info on this spec, don't treat
@@ -3787,19 +3785,8 @@ def _intersects(self, other: "Spec", deps: bool = True) -> bool:
# If we need to descend into dependencies, do it, otherwise we're done.
if deps:
return self._intersects_dependencies(other)
else:
return True
def satisfies(self, other, deps=True):
"""
This checks constraints on common dependencies against each other.
"""
other = self._autospec(other)
lhs = self.lookup_hash() or self
rhs = other.lookup_hash() or other
return lhs._satisfies(rhs, deps=deps)
return True
def _intersects_dependencies(self, other):
if not other._dependencies or not self._dependencies:
@@ -3813,10 +3800,10 @@ def _intersects_dependencies(self, other):
# For virtual dependencies, we need to dig a little deeper.
self_index = spack.provider_index.ProviderIndex(
repository=spack.repo.path, specs=self.traverse(), restrict=True
repository=spack.repo.PATH, specs=self.traverse(), restrict=True
)
other_index = spack.provider_index.ProviderIndex(
repository=spack.repo.path, specs=other.traverse(), restrict=True
repository=spack.repo.PATH, specs=other.traverse(), restrict=True
)
# This handles cases where there are already providers for both vpkgs
@@ -3836,7 +3823,7 @@ def _intersects_dependencies(self, other):
return True
def _satisfies(self, other: "Spec", deps: bool = True) -> bool:
def satisfies(self, other: "Spec", deps: bool = True) -> bool:
"""Return True if all concrete specs matching self also match other, otherwise False.
Args:
@@ -3851,13 +3838,20 @@ def _satisfies(self, other: "Spec", deps: bool = True) -> bool:
# objects.
return self.concrete and self.dag_hash() == other.dag_hash()
# If the right-hand side has an abstract hash, make sure it's a prefix of the
# left-hand side's (abstract) hash.
if other.abstract_hash:
compare_hash = self.dag_hash() if self.concrete else self.abstract_hash
if not compare_hash or not compare_hash.startswith(other.abstract_hash):
return False
# If the names are different, we need to consider virtuals
if self.name != other.name and self.name and other.name:
# A concrete provider can satisfy a virtual dependency.
if not self.virtual and other.virtual:
try:
# Here we might get an abstract spec
pkg_cls = spack.repo.path.get_pkg_class(self.fullname)
pkg_cls = spack.repo.PATH.get_pkg_class(self.fullname)
pkg = pkg_cls(self)
except spack.repo.UnknownEntityError:
# If we can't get package info on this spec, don't treat
@@ -3939,8 +3933,8 @@ def patches(self):
# translate patch sha256sums to patch objects by consulting the index
if self._patches_assigned():
for sha256 in self.variants["patches"]._patches_in_order_of_appearance:
index = spack.repo.path.patch_index
pkg_cls = spack.repo.path.get_pkg_class(self.name)
index = spack.repo.PATH.patch_index
pkg_cls = spack.repo.PATH.get_pkg_class(self.name)
patch = index.patch_for_package(sha256, pkg_cls)
self._patches.append(patch)
@@ -4227,9 +4221,7 @@ def eq_node(self, other):
def _cmp_iter(self):
"""Lazily yield components of self for comparison."""
cmp_spec = self.lookup_hash() or self
for item in cmp_spec._cmp_node():
for item in self._cmp_node():
yield item
# This needs to be in _cmp_iter so that no specs with different process hashes
@@ -4240,10 +4232,10 @@ def _cmp_iter(self):
# TODO: they exist for speed. We should benchmark whether it's really worth
# TODO: having two types of hashing now that we use `json` instead of `yaml` for
# TODO: spec hashing.
yield cmp_spec.process_hash() if cmp_spec.concrete else None
yield self.process_hash() if self.concrete else None
def deps():
for dep in sorted(itertools.chain.from_iterable(cmp_spec._dependencies.values())):
for dep in sorted(itertools.chain.from_iterable(self._dependencies.values())):
yield dep.spec.name
yield tuple(sorted(dep.deptypes))
yield hash(dep.spec)
@@ -4315,6 +4307,19 @@ def format(self, format_string=default_format, **kwargs):
``\{`` and ``\}`` for literal braces, and ``\\`` for the
literal ``\`` character.
The ``?`` sigil may be used to conditionally add a
value. Conditional format values are used if constructing the
value would not throw any error, and are ignored if it would
throw an error. For example, ``{?^mpi.name}`` will print
``Spec["mpi"].name`` if such a node exists, and otherwise
prints nothing.
The ``?`` sigil may also be combined with a conditional
separator. This separator is prepended if anything is printed
for the conditional attribute. The syntax for this is
``?sep?attribute``,
e.g. ``{name}-{version}{?/?^mpi.name}{?-?^mpi.version}``.
Args:
format_string (str): string containing the format to be expanded
@@ -4338,6 +4343,15 @@ def write(s, c=None):
def write_attribute(spec, attribute, color):
attribute = attribute.lower()
conditional = False
conditional_sep = ""
matches_conditional_sep = re.match(r"^\?(.*)\?", attribute)
if matches_conditional_sep:
conditional = True
conditional_sep = matches_conditional_sep.group(1)
if attribute.startswith("?"):
conditional = True
sig = ""
if attribute.startswith(("@", "%", "/")):
# color sigils that are inside braces
@@ -4369,6 +4383,9 @@ def write_attribute(spec, attribute, color):
elif sig == " arch=" and attribute not in ("architecture", "arch"):
raise SpecFormatSigilError(sig, "the architecture", attribute)
# Now that we're done testing sig, combine it with conditional sep
sig = conditional_sep + sig
# find the morph function for our attribute
morph = transform.get(attribute, lambda s, x: x)
@@ -4398,7 +4415,12 @@ def write_attribute(spec, attribute, color):
else:
if isinstance(current, vt.VariantMap):
# subscript instead of getattr for variant names
current = current[part]
try:
current = current[part]
except KeyError:
if conditional:
return
raise
else:
# aliases
if part == "arch":
@@ -4414,6 +4436,8 @@ def write_attribute(spec, attribute, color):
try:
current = getattr(current, part)
except AttributeError:
if conditional:
return
parent = ".".join(parts[:idx])
m = "Attempted to format attribute %s." % attribute
m += "Spec %s has no attribute %s" % (parent, part)

View File

@@ -197,7 +197,9 @@ def _expand_matrix_constraints(matrix_config):
for combo in itertools.product(*expanded_rows):
# Construct a combined spec to test against excludes
flat_combo = [constraint for constraint_list in combo for constraint in constraint_list]
flat_combo = [Spec(x) for x in flat_combo]
# Resolve abstract hashes so we can exclude by their concrete properties
flat_combo = [Spec(x).lookup_hash() for x in flat_combo]
test_spec = flat_combo[0].copy()
for constraint in flat_combo[1:]:

View File

@@ -326,7 +326,7 @@ def __init__(
self.keep = keep
# File lock for the stage directory. We use one file for all
# stage locks. See spack.database.Database.prefix_lock for
# stage locks. See spack.database.Database.prefix_locker.lock for
# details on this approach.
self._lock = None
if lock:
@@ -484,7 +484,7 @@ def fetch(self, mirror_only=False, err_msg=None):
if self.default_fetcher.cachable:
for rel_path in reversed(list(self.mirror_paths)):
cache_fetcher = spack.caches.fetch_cache.fetcher(
cache_fetcher = spack.caches.FETCH_CACHE.fetcher(
rel_path, digest, expand=expand, extension=extension
)
fetchers.insert(0, cache_fetcher)
@@ -577,7 +577,7 @@ def check(self):
self.fetcher.check()
def cache_local(self):
spack.caches.fetch_cache.store(self.fetcher, self.mirror_paths.storage_path)
spack.caches.FETCH_CACHE.store(self.fetcher, self.mirror_paths.storage_path)
def cache_mirror(self, mirror, stats):
"""Perform a fetch if the resource is not already cached

View File

@@ -25,13 +25,14 @@
from typing import Any, Callable, Dict, Generator, List, Optional, Union
import llnl.util.lang
import llnl.util.tty as tty
from llnl.util import tty
import spack.config
import spack.database
import spack.directory_layout
import spack.error
import spack.paths
import spack.spec
import spack.util.path
#: default installation root, relative to the Spack install path
@@ -134,18 +135,21 @@ def parse_install_tree(config_dict):
class Store:
"""A store is a path full of installed Spack packages.
Stores consist of packages installed according to a
``DirectoryLayout``, along with an index, or _database_ of their
contents. The directory layout controls what paths look like and how
Spack ensures that each unique spec gets its own unique directory (or
not, though we don't recommend that). The database is a single file
that caches metadata for the entire Spack installation. It prevents
us from having to spider the install tree to figure out what's there.
Stores consist of packages installed according to a ``DirectoryLayout``, along with a database
of their contents.
The directory layout controls what paths look like and how Spack ensures that each unique spec
gets its own unique directory (or not, though we don't recommend that).
The database is a single file that caches metadata for the entire Spack installation. It
prevents us from having to spider the install tree to figure out what's there.
The store is also able to lock installation prefixes, and to mark installation failures.
Args:
root: path to the root of the install tree
unpadded_root: path to the root of the install tree without padding.
The sbang script has to be installed here to work with padded roots
unpadded_root: path to the root of the install tree without padding. The sbang script has
to be installed here to work with padded roots
projections: expression according to guidelines that describes how to construct a path to
a package prefix in this store
hash_length: length of the hashes used in the directory layout. Spec hash suffixes will be
@@ -170,6 +174,19 @@ def __init__(
self.upstreams = upstreams
self.lock_cfg = lock_cfg
self.db = spack.database.Database(root, upstream_dbs=upstreams, lock_cfg=lock_cfg)
timeout_format_str = (
f"{str(lock_cfg.package_timeout)}s" if lock_cfg.package_timeout else "No timeout"
)
tty.debug("PACKAGE LOCK TIMEOUT: {0}".format(str(timeout_format_str)))
self.prefix_locker = spack.database.SpecLocker(
spack.database.prefix_lock_path(root), default_timeout=lock_cfg.package_timeout
)
self.failure_tracker = spack.database.FailureTracker(
self.root, default_timeout=lock_cfg.package_timeout
)
self.layout = spack.directory_layout.DirectoryLayout(
root, projections=projections, hash_length=hash_length
)
@@ -195,7 +212,7 @@ def create(configuration: ConfigurationType) -> Store:
Args:
configuration: configuration to create a store.
"""
configuration = configuration or spack.config.config
configuration = configuration or spack.config.CONFIG
config_dict = configuration.get("config")
root, unpadded_root, projections = parse_install_tree(config_dict)
hash_length = configuration.get("config:install_hash_length")
@@ -217,7 +234,7 @@ def create(configuration: ConfigurationType) -> Store:
def _create_global() -> Store:
result = create(configuration=spack.config.config)
result = create(configuration=spack.config.CONFIG)
return result
@@ -355,10 +372,10 @@ def use_store(
# Swap the store with the one just constructed and return it
ensure_singleton_created()
spack.config.config.push_scope(
spack.config.CONFIG.push_scope(
spack.config.InternalConfigScope(name=scope_name, data={"config": {"install_tree": data}})
)
temporary_store = create(configuration=spack.config.config)
temporary_store = create(configuration=spack.config.CONFIG)
original_store, STORE = STORE, temporary_store
try:
@@ -366,7 +383,7 @@ def use_store(
finally:
# Restore the original store
STORE = original_store
spack.config.config.remove_scope(scope_name=scope_name)
spack.config.CONFIG.remove_scope(scope_name=scope_name)
class MatchError(spack.error.SpackError):

View File

@@ -94,15 +94,15 @@ class TestState:
def __init__(self):
if _SERIALIZE:
self.config = spack.config.config
self.config = spack.config.CONFIG
self.platform = spack.platforms.host
self.test_patches = store_patches()
self.store = spack.store.STORE
def restore(self):
if _SERIALIZE:
spack.config.config = self.config
spack.repo.path = spack.repo.create(self.config)
spack.config.CONFIG = self.config
spack.repo.PATH = spack.repo.create(self.config)
spack.platforms.host = self.platform
spack.store.STORE = self.store
self.test_patches.restore()

View File

@@ -32,10 +32,10 @@ def packages_with_tags(tags, installed, skip_empty):
"""
tag_pkgs = collections.defaultdict(lambda: list)
spec_names = _get_installed_package_names() if installed else []
keys = spack.repo.path.tag_index if tags is None else tags
keys = spack.repo.PATH.tag_index if tags is None else tags
for tag in keys:
packages = [
name for name in spack.repo.path.tag_index[tag] if not installed or name in spec_names
name for name in spack.repo.PATH.tag_index[tag] if not installed or name in spec_names
]
if packages or not skip_empty:
tag_pkgs[tag] = packages

View File

@@ -199,15 +199,11 @@ def test_satisfy_strict_constraint_when_not_concrete(architecture_tuple, constra
],
)
@pytest.mark.usefixtures("mock_packages", "config")
@pytest.mark.only_clingo("Fixing the parser broke this test for the original concretizer.")
def test_concretize_target_ranges(root_target_range, dep_target_range, result, monkeypatch):
# Monkeypatch so that all concretization is done as if the machine is core2
monkeypatch.setattr(spack.platforms.test.Test, "default", "core2")
# use foobar=bar to make the problem simpler for the old concretizer
# the new concretizer should not need that help
if spack.config.get("config:concretizer") == "original":
pytest.skip("Fixing the parser broke this test for the original concretizer.")
spec_str = "a %%gcc@10 foobar=bar target=%s ^b target=%s" % (
root_target_range,
dep_target_range,

View File

@@ -37,7 +37,7 @@
from spack.paths import test_path
from spack.spec import Spec
pytestmark = pytest.mark.skipif(sys.platform == "win32", reason="does not run on windows")
pytestmark = pytest.mark.not_on_windows("does not run on windows")
mirror_cmd = spack.main.SpackCommand("mirror")
install_cmd = spack.main.SpackCommand("install")
@@ -51,7 +51,7 @@
def cache_directory(tmpdir):
fetch_cache_dir = tmpdir.ensure("fetch_cache", dir=True)
fsc = spack.fetch_strategy.FsCache(str(fetch_cache_dir))
spack.config.caches, old_cache_path = fsc, spack.caches.fetch_cache
spack.config.caches, old_cache_path = fsc, spack.caches.FETCH_CACHE
yield spack.config.caches
@@ -115,8 +115,8 @@ def default_config(tmpdir, config_directory, monkeypatch, install_mockery_mutabl
]
)
spack.config.config, old_config = cfg, spack.config.config
spack.config.config.set("repos", [spack.paths.mock_packages_path])
spack.config.CONFIG, old_config = cfg, spack.config.CONFIG
spack.config.CONFIG.set("repos", [spack.paths.mock_packages_path])
njobs = spack.config.get("config:build_jobs")
if not njobs:
spack.config.set("config:build_jobs", 4, scope="user")
@@ -138,9 +138,9 @@ def default_config(tmpdir, config_directory, monkeypatch, install_mockery_mutabl
if not timeout:
spack.config.set("config:connect_timeout", 10, scope="user")
yield spack.config.config
yield spack.config.CONFIG
spack.config.config = old_config
spack.config.CONFIG = old_config
mutable_dir.remove()

View File

@@ -26,11 +26,11 @@ def test_store_is_restored_correctly_after_bootstrap(mutable_config, tmpdir):
user_path = str(tmpdir.join("store"))
with spack.store.use_store(user_path):
assert spack.store.STORE.root == user_path
assert spack.config.config.get("config:install_tree:root") == user_path
assert spack.config.CONFIG.get("config:install_tree:root") == user_path
with spack.bootstrap.ensure_bootstrap_configuration():
assert spack.store.STORE.root == spack.bootstrap.config.store_path()
assert spack.store.STORE.root == user_path
assert spack.config.config.get("config:install_tree:root") == user_path
assert spack.config.CONFIG.get("config:install_tree:root") == user_path
@pytest.mark.regression("38963")
@@ -40,11 +40,11 @@ def test_store_padding_length_is_zero_during_bootstrapping(mutable_config, tmpdi
"""
user_path = str(tmpdir.join("store"))
with spack.store.use_store(user_path, extra_data={"padded_length": 512}):
assert spack.config.config.get("config:install_tree:padded_length") == 512
assert spack.config.CONFIG.get("config:install_tree:padded_length") == 512
with spack.bootstrap.ensure_bootstrap_configuration():
assert spack.store.STORE.root == spack.bootstrap.config.store_path()
assert spack.config.config.get("config:install_tree:padded_length") == 0
assert spack.config.config.get("config:install_tree:padded_length") == 512
assert spack.config.CONFIG.get("config:install_tree:padded_length") == 0
assert spack.config.CONFIG.get("config:install_tree:padded_length") == 512
@pytest.mark.regression("38963")
@@ -54,15 +54,15 @@ def test_install_tree_customization_is_respected(mutable_config, tmp_path):
"""
spack.store.reinitialize()
store_dir = tmp_path / "store"
spack.config.config.set("config:install_tree:root", str(store_dir))
spack.config.CONFIG.set("config:install_tree:root", str(store_dir))
with spack.bootstrap.ensure_bootstrap_configuration():
assert spack.store.STORE.root == spack.bootstrap.config.store_path()
assert (
spack.config.config.get("config:install_tree:root")
spack.config.CONFIG.get("config:install_tree:root")
== spack.bootstrap.config.store_path()
)
assert spack.config.config.get("config:install_tree:padded_length") == 0
assert spack.config.config.get("config:install_tree:root") == str(store_dir)
assert spack.config.CONFIG.get("config:install_tree:padded_length") == 0
assert spack.config.CONFIG.get("config:install_tree:root") == str(store_dir)
assert spack.store.STORE.root == str(store_dir)
@@ -185,12 +185,12 @@ def test_bootstrap_custom_store_in_environment(mutable_config, tmpdir):
def test_nested_use_of_context_manager(mutable_config):
"""Test nested use of the context manager"""
user_config = spack.config.config
user_config = spack.config.CONFIG
with spack.bootstrap.ensure_bootstrap_configuration():
assert spack.config.config != user_config
assert spack.config.CONFIG != user_config
with spack.bootstrap.ensure_bootstrap_configuration():
assert spack.config.config != user_config
assert spack.config.config == user_config
assert spack.config.CONFIG != user_config
assert spack.config.CONFIG == user_config
@pytest.mark.parametrize("expected_missing", [False, True])

View File

@@ -5,7 +5,6 @@
import os
import os.path
import sys
import pytest
@@ -16,7 +15,7 @@
install = spack.main.SpackCommand("install")
pytestmark = pytest.mark.skipif(sys.platform == "win32", reason="does not run on windows")
pytestmark = pytest.mark.not_on_windows("does not run on windows")
def test_build_tarball_overwrite(install_mockery, mock_fetch, monkeypatch, tmpdir):

Some files were not shown because too many files have changed in this diff Show More