MSVC compiler logic was using string parsing to extract version
from compiler spec, which was fragile. This broke in #37572, so has
been fixed and made more robust by using attribute access.
* CI: Expand E4S ROCm stack to include missing DaV packages
Ascent: Fixup for VTK-m with Kokkos backend
* DaV SDK: Removed duplicated openmp variant for ascent
* Drop visit and add conflict for Kokkos
* E4S: Drop ascent from CUDA builds
Ensure that requirements `packages:*:require:@x` and preferences `packages:*:version:[x]`
fail concretization when no version defined in the package satisfies `x`. This always holds
except for git versions -- they are defined on the fly.
In the past, Spack did not allow two different versions of the
same package within a DAG. That led to difficulties with packages
that still required Python 2 while other packages had already
switched to Python 3.
The libxcb and xcb-proto packages did not have Python 3 support
for a time. To get around this issue, Spack maintainers disabled
their dependency on an internal (i.e., Spack-provided) Python
(see #4145),forcing these packages to look for a system-provided
Python (see #7646).
This has worked for us all right, but with the arrival of our most
recent platform we seem to be missing the critical xcbgen Python
module on the system. Since most software has largely moved on to
Python 3 now, let's re-enable internal Spack dependencies for the
libxcb and xcb-proto packages.
Two bugs came in from #37438
1. `unify: when_possible` was broken, because of an incorrect assertion. abstract/concrete
spec pairs were compared against the results that were in the process of being computed,
rather than against the previous results.
2. `unify: true` had an ordering bug that could mix the association between abstract and
concrete specs
- [x] 1 is resolved by creating a lookup from old concrete specs to old abstract specs,
and we use that to associate the "new" concrete specs that happen to be the old
ones with their abstract specs (since those are stripped out for concretization
- [x] 2 is resolved by combining the new and old abstract as lists instead of combining
them as sets. This is important because `set() | set()` does not make any ordering
promises, even though set ordering is otherwise guaranteed in `python@3.7:`
* Upgrading kosh to 3.0.
* Accidentally regressed the package, changing back.
* Updating py-hdbscan versions for kosh.
* Fixing bug in patch.
* Adding 3.0.1
* Removing 3.0.
* Updating package deps for hdbscan to match requirements.txt.
* Version reqs for 3.0.*, need newer numpy and networkx
* spack style
* Reordering to match setup.py, adding "type" to python depends.
* trilinos@develop fixes
* Update var/spack/repos/builtin/packages/trilinos/package.py
Co-authored-by: Seth R. Johnson <johnsonsr@ornl.gov>
---------
Co-authored-by: Seth R. Johnson <johnsonsr@ornl.gov>
