File dolfin.changes of Package dolfin
-------------------------------------------------------------------
Fri Jan 14 17:44:34 UTC 2022 - Stefan Brüns <stefan.bruens@rwth-aachen.de>
- Do not claim directories provided by the filesystem package.
-------------------------------------------------------------------
Thu Oct 15 14:09:46 UTC 2020 - Stefan Brüns <stefan.bruens@rwth-aachen.de>
- Update to version 2019.1.0.post0, no changelog provided
- Fix build with Boost 1.74, add
* 0001-Add-missing-algorithm-include-for-std-min_element-co.patch
* 0001-Use-__BYTE_ORDER__-instead-of-removed-Boost-endian.h.patch
-------------------------------------------------------------------
Sat Feb 8 20:53:52 UTC 2020 - Stefan Brüns <stefan.bruens@rwth-aachen.de>
- Update to version 2019.1.0
* Support for shape calculus operations, see file:
python/test/unit/fem/test_coordinatederivative.py
* Support for PETSc and SLEPc 3.10.
* Documented Python elastodynamics demo.
* Various bug fixes, e.g. memory leaks, build issues.
* Enhancements and additions to pybind11-based Python interface.
- Update to version 2018.1.0
* Remove python2 support.
* Remove SWIG (now use pybind11 for wrapping C++)
* Rename mpi_comm_world() to MPI.comm_world.
* Removed from fenics import *, use from dolfin import *
* Rename ERROR, CRITICAL etc. to LogLevel.ERROR, LogLevel.CRITICAL.
- Update to version 2017.2.0
* Remove UnitQuadMesh and UnitHexMesh. Now use UnitSquareMesh and
UnitCubeMesh with cell type qualifiers.
* Remove MeshEditor::open without cell type. Now you must
explicitly specify CellType when opening a Mesh with MeshEditor.
* Rename Mesh::size_global to Mesh::num_entities_global.
* Remove Mesh::size. Use Mesh::num_entities instead.
* Improved mesh topology computation performance.
* Remove excessive calls to MPI init. It may now be necessary in
some cases to explicitly intialise MPI.
* Improvements to sparsity pattern computation.
* Addition of some interfaces using Eigen::Map/ref in addition to
dolfin::Array(View). ``dolfin::Array(View)``interfaces will be
removed in favour of Eigen interfaces in the next release.
* Update pkg-config (dolfin.pc) file.
* CMake modernisations, with more use of exported targets.
* Add experimental pybind11 generated Python interface. Will
replace the SWIG generated interface in the 2018.1 release.
* Remove redundant SLEPc interfaces. Fixes issue #908.
* Bump required Boost version to 1.55.
* Remove PETScUserPreconditioner (was unused and untested).
* Remove VTK plotting backend. Plotting is no longer available
from the C++ interface. Basic plotting is available using
matplotlib and x3dom backends via the plot() free function in
the Python interface. Users are advised to move to e.g.
Paraview for more demanding plotting needs.
* Updates for ufc::finite_element::evaluate_vertex_values
interface change.
* Add new methods XDMFFile::write_checkpoint,
XDMFFile::read_checkpoint to write and read (checkpoint)
functions.
* Implement marking vertex and edge mesh functions by
SubDomain::mark() using on_boundary flag.
* Fix quadratic scaling in PETSc matrix allocation with global
dofs; assembly with Real space now exhibits linear scaling in
number of dofs.
* Add assembly for quadrilateral and hexahedral meshes with CG
and DG elements.
* Updates for some demos and tests to show usage of quadrilateral
and hexahedral meshes.
* Deprecate CellSize (equivalent to 2*Circumradius) in favour of
new CellDiameter; add MinCellEdgeLength and MaxCellEdgeLength
* Deprecate subclassing of Expression in Python; new Python class
UserExpression introduced for user overloads
* Deprecate VertexFunction, EdgeFunction, FaceFunction,
FacetFunction, CellFunction; use MeshFunction instead
- Update to version 2017.1.0
* Refactor PETScLUSolver to use functionality from
PETScKrylovSolver. Simplify interface for solving transposed
systems. Fixes #815.
* Switch default Python version to Python 3. Use
-DDOLFIN_USE_PYTHON3=off to build with Python 2.
* Remove redundant solve_transpose functions (use solve with bool
argument instead)
* Remove OpenMpAsssmebler
* Remove MPI communicator as argument in GenericVector::init
functions (communicator should be passed via constructor)
* Remove Function::operator[+-*/] to prevent memory corruption
problems (does not affect Python interface)
* Fix XDMF3 output of time series. The default output method is
now to assume that all functions have different meshes, and
that the meshes change from time step to time step. Two
parameters control the output, one limits each function to only
one mesh for the whole time series, turn off the default on
parameter rewrite_function_mesh to enable this. You can also
make all functions share the same mesh and time series, which
currently is better supported in Paraview than the alternative,
turn on functions_share_mesh for this. These two parameters can
also be combined in case all functions share the same mesh at
all time steps. This creates minimal size files.
* Add PETScSNESSolver and PETScTAOSolver constructor accepting
both communicator and type
* Expression("f[0]*f[1]", f=obj) notation now supported for
non-scalar GenericFunction obj
* Expression("f", f=obj) notation now supports obj of
MeshFunction types (only cell based)
* Fix MPI deadlock in case of instant compilation failure
* Allow using Timer as context manager and add timed decorator
to measure timings of functions and methods
* Add NonlinearProblem::J_pc and support preconditioning matrix
in NewtonSolver, PETScSNESSolver and PETScTAOSolver
- Remove obsolete patches
* dolfin-1.1.0-suitesparse.patch
* python3-fixes.patch
* cmake-check-boostversion-163.patch
- Add 0001-Fix-linker-error-due-to-missing-lpthread.patch
- Removed unused build dependencies: VTK, Qt5, X11, Java
-------------------------------------------------------------------
Wed Mar 20 22:04:12 UTC 2019 - Stefan Brüns <stefan.bruens@rwth-aachen.de>
- Convert to _multibuild, sort BuildRequires
- Correctly specify PETSC_DIR/SLEPC_DIR using environment variables
- Use split Boost BuildRequires
- Cleanup BuildRequires, remove obsoleted packages
- Changed cmake-check-boostversion-163.patch to allow newer Boost versions
-------------------------------------------------------------------
Fri Apr 14 18:31:01 UTC 2017 - kkhere.geo@gmail.com
- build using vtk-compat_gl-devel because not all systems have brand
new graphics supporting OpenGL version 3.2
- more fixes for running scripts with python3
- require cmake >= 3.5 as reported by the CMakeLists.txt
-------------------------------------------------------------------
Wed Mar 8 15:11:46 UTC 2017 - kkhere.geo@gmail.com
- Update to upstream release 2016.2.0 [2016-11-30]
* Updates to XDMFFile interface, now fully supporting MeshFunction and
MeshValueCollection with multiple named datasets in one file (useful for
volume/boundary markers). Time series now only when a time is explicitly
specified for each step. Full support for ASCII/XML XDMF.
* Improved X3DOM support
* Improved detection of UFC
* Add CMake option `-DDOLFIN_USE_PYTHON3` to create a Python 3 build
* Require CMake version 3.5 or higher
* Add pylit to generate demo doc from rst
* More careful checks of Dirichlet BC function spaces
* Change definition of FunctionSpace::component()
* Adaptive solving now works for tensor-valued unknowns
* Improve logging of PETSc errors; details logged at level TRACE
release 2016.1.0 [2016-06-23]
* Remove support for 'uint'-valued MeshFunction (replaced by 'size_t')
* Major performance improvements and simplifications of the XDMF IO.
* Remove Zoltan graph partitioning interface
* Add new algorithm for computing mesh entiites. Typical speed-up of
two with gcc and four with clang. Reduced memory usage for meshes
with irregularly numbered cells.
* Remove STLMatrix, STLVector, MUMPSLUSolver and PastixLUSolver
classes
* Remove PETScPreconditioner::set_near_nullspace and add
PETScMatrix::set_near_nullspace
* Build system updates for VTK 7.0
* Remove XDMF from File interface. XDMF is XML based, and has many
possibilities for file access, which are not accessible through the
limited File interface and "<<" ">>" operators. Instead of File, use
XDMFFile, and use XDMFFile.read() and XDMFFile.write() for
I/O. Demos and tests have been updated to show usage. XDMF now also
supports ASCII I/O in serial, useful for compatibility with users
who do not have the HDF5 library available.
* Require polynomial degree or finite element for Expressions in the
Python interface (fixes Issue #355,
https://bitbucket.org/fenics-project/dolfin/issues/355)
* Switch to Google Test framwork for C++ unit tests
* Fix bug when reading domain data from mesh file for a ghosted mesh
* Add interface for manipulating mesh geometry using (higher-order) FE
functions: free functions set_coordinates, get_coordinates,
create_mesh
* Fix bug when reading domain data from mesh file for a ghosted mesh.
* Remove reference versions of constructors for many classes that
store a pointer/reference to the object passed to the
constructor. This is an intrusive interface change for C++ users,
but necessary to improve code maintainabilty and to improve memory
safety. The Python interface is (virtually) unaffected.
* Remove class SubSpace. Using FunctionSpace::sub(...) instead
* Remove reference versions constructors of NonlinearVariationalSolver
* Remove setting of bounds from NonlinearVariationalSolver (was
already available through NonlinearVariationalProblem)
* Update Trilinos support to include Amesos2, and better support from
Python
* Rewrite interface of TensorLayout and SparsityPattern;
local-to-global maps now handled using new IndexMap class;
GenericSparsityPattern class removed
* Remove QT (was an optional dependency)
* PETScTAOSolver::solve() now returns a pair of number of iterations
(std::size_t) and whether iteration converged (bool)
* Better quality refinement in 2D in Plaza algorithm, by choosing
refinement pattern based on max/min edge ratio
* Removed refine_cell() method in CellTypes
* Enable marker refinement to work in parallel for 1D meshes too
* Add std::abort to Python exception hook to avoid parallel deadlocks
* Extend dof_to_vertex_map with unowned dofs, thus making
dof_to_vertex_map an inverse of vertex_to_dof_map
* Clean-up in PyDOLFIN function space design, issue #576
* Deprecate MixedFunctionSpace and EnrichedFunctionSpace in favour of
initialization by suitable UFL element
* Add experimental matplotlib-based plotting backend, see mplot demo
* Remove method argument of DirichletBC::get_boundary_values()
* Change return types of free functions adapt() to shared_ptr
- added python3-fixes.patch for compilation fixes with python3
- added cmake-check-boostversion-163.patch to fix finding boost-devel == 1.63
-------------------------------------------------------------------
Tue Oct 6 11:41:37 UTC 2015 - kkhere.geo@gmail.com
- removed unneeded patch
cmake-swig-output.patch (fixed upstream)
- fix build on Tumbleweed by disabling QT4
vtk on Tumbleweed uses QT5 so we can't build with QT4 too
-------------------------------------------------------------------
Thu Oct 1 09:51:22 UTC 2015 - kkhere.geo@gmail.com
- add build dependency on arpack-ng/parpack and trlan as SLEPc requires them
in SLEPcConfig.cmake
-------------------------------------------------------------------
Fri Sep 25 14:01:51 UTC 2015 - kkhere.geo@gmail.com
- remove deprecated Requires: python(abi) == %{py_ver}
- update version to 1.6.0 [2015-07-28]
* Remove redundant pressure boundary condition in Stokes demos
* Require Point in RectangleMesh and BoxMesh constructors
* Remove BinaryFile (TimeSeries now requires HDF5)
* Add (highly experimental) support for Tpetra matrices and
vectors from Trilinos, interfacing to Belos, Amesos2, IfPack2 and
Muelu.
* Enable (highly experimental) support for Quadrilateral and Hexahedral meshes,
including some I/O, but no assembly yet.
* Enable UMFPACK and CHOLMOD solvers with Eigen backend
* Add an MPI_Comm to logger, currently defaulted to MPI_COMM_WORLD
allowing better control over output in parallel
* Experimental output of quadratic geometry in XDMF files,
allows more exact visualisation of P2 Functions
* Remove GenericMatrix::compressed (see Issue #61)
* Deprecate and PETScKryloveSolver::set_nullspace() and add
PETScMatrix::set_nullspace()
* Remove uBLAS backend
* Remove UmfpackLUSolver and CholmodSolver
* Add EigenMatrix/Vector::data()
* Remove GenericMatrix/Vector::data() and GenericMatrix/Vector::data()
(to use backends that support data(), cast first to backend type, e.g.
A = A.as_backend_type()
* Remove cmake.local, replaced by fenics-install-component.sh
* Make interior facet integrals define - and + cells ordered by cell_domains
value.
* Remove deprecated arguments *_domains from assemble() and Form().
* Change measure definition notation from dx[mesh_function] to
dx(subdomain_data=mesh_function).
* Set locale to "C" before reading from file
* Change GenericDofMap::cell_dofs return type from const std::vector<..>&
to ArrayView<const ..>
* Add ArrayView class for views into arrays
* Change fall back linear algebra backend to Eigen
* Add Eigen linear algebra backend
* Remove deprecated GenericDofMap::geometric_dim function (fixes Issue #443)
* Add quadrature rules for multimesh/cut-cell integration up to order 6
* Implement MPI reductions and XML ouput of Table class
* list_timings() is now collective and returns MPI average across processes
* Add dump_timings_to_xml()
* Add enum TimingType { wall, user, system } for selecting wall-clock,
user and system time in timing routines
* Bump required SWIG version to 3.0.3
* Increase default maximum iterations in NewtonSolver to 50.
* Deprecate Python free function homogenize(bc) in favour of member
function DirichletBC::homogenize()
- refreshed dolfin-1.1.0-suitesparse.patch
-------------------------------------------------------------------
Thu Sep 24 23:29:24 UTC 2015 - kkhere.geo@gmail.com
- serial build isn't possible due to vtk-devel unconditionally requiring MPI
-------------------------------------------------------------------
Tue Feb 10 00:27:47 UTC 2015 - kkhere.geo@gmail.com
- Finally the correct patch after all
* added cmake-swig-output.patch for openSUSE > 13.2 (boo#874885)
* removed cmake-prevent-double-underscore.patch
-------------------------------------------------------------------
Mon Feb 9 16:17:41 UTC 2015 - kkhere.geo@gmail.com
- refresh cmake-prevent-double-underscore.patch with better logic
-------------------------------------------------------------------
Sat Feb 7 16:31:34 UTC 2015 - kkhere.geo@gmail.com
- add a patch to make sure the python modules don't have an extra _
in their names with cmake 3.1.1, which makes them fail to load
-------------------------------------------------------------------
Thu Feb 5 16:44:28 UTC 2015 - kkhere.geo@gmail.com
- Change dependency from python-ScientificPython to python-sympy
-------------------------------------------------------------------
Thu Jan 15 12:19:57 UTC 2015 - kkhere.geo@gmail.com
- update to version 1.5.0
* DG demos working is parallel
* Simplify re-use of LU factorisations
* CMake 3 compatibility
* Make underlying SLEPc object accessible
* Full support for linear algebra backends with 64-bit integers
* Add smoothed aggregation AMG elasticity demo
* Add support for slepc4py
* Some self-assignment fixes in mesh data structures
* Deprecated GenericDofMap::geometric_dimension()
* Experimental support for ghosted meshes (overlapping region in parallel)
* Significant memory reduction in dofmap storage
* Re-write dofmap construction with significant performance and scaling
improvements in parallel
* Switch to local (process-wise) indexing for dof indices
* Support local (process-wise) indexing in linear algerbra backends
* Added support for PETSc 3.5, require version >= 3.3
* Exposed DofMap::tabulate_local_to_global_dofs,
MeshEntity::sharing_processes in Python
* Added GenericDofmap::local_dimension("all"|"owned"|"unowned")
* Added access to SLEPc or slepc4py EPS object of SLEPcEigenSolver
(requires slepc4py version >= 3.5.1)
* LinearOperator can now be accessed using petsc4py
* Add interface (PETScTAOSolver) for the PETSc nonlinear (bound-constrained)
optimisation solver (TAO)
* Add GenericMatrix::nnz() function to return number of nonzero
entries in matrix (fixes #110)
* Add smoothed aggregation algerbraic multigrid demo for elasticity
* Add argument 'function' to project, to store the result into a preallocated
function
* Remove CGAL dependency and mesh generation, now provided by mshr
* Python 2.7 required
* Add experimental Python 3 support. Need swig version 3.0.3 or later
* Move to py.test, speed up unit tests and make tests more robust in parallel
* Repeated initialization of PETScMatrix is now an error
* MPI interface change: num_processes -> size, process_number -> rank
* Add optional argument project(..., function=f), to avoid superfluous
allocation
* Remove excessive printing of points during extrapolation
* Clean up DG demos by dropping restrictions of Constants: c('+') -> c
* Fix systemassembler warning when a and L both provide the same subdomain
data.
* Require mesh instead of cell argument to FacetArea, FacetNormal,
CellSize, CellVolume, SpatialCoordinate, Circumradius,
MinFacetEdgeLength, MaxFacetEdgeLength
* Remove argument reset_sparsity to assemble()
* Simplify assemble() and Form() signature: remove arguments mesh,
coefficients, function_spaces, common_cell. These are now
all found by inspecting the UFL form
* Speed up assembly of forms with multiple integrals depending on
different functions, e.g. f*dx(1) + g*dx(2).
* Handle accessing of GenericVectors using numpy arrays in python layer
instead of in hard-to-maintain C++ layer
* Add support for mpi groups in jit-compilation
* Make access to HDFAttributes more dict like
* Add 1st and 2nd order Rush Larsen schemes for the PointIntegralSolver
* Add vertex assembler for PointIntegrals
* Add support for assembly of custom_integral
* Add support for multimesh assembly, function spaces, dofmaps and functions
* Fix to Cell-Point collision detection to prevent Points inside the mesh
from falling between Cells due to rounding errors
* Enable reordering of cells and vertices in parallel via SCOTCH and the
Giibs-Poole-Stockmeyer algorithm
* Efficiency improvements in dof assignment in parallel, working on HPC up to
24000 cores
* Introduction of PlazaRefinement methods based on refinement of the Mesh
skeleton, giving better quality refinement in 3D in parallel
* Basic support for 'ghost cells' allowing integration over interior facets in
parallel
-------------------------------------------------------------------
Tue Sep 2 14:49:41 UTC 2014 - kkhere.geo@gmail.com
- update to version 1.4.0
* Feature: Add set_diagonal (with GenericVector) to GenericMatrix
* Fix many bugs associated with cell orientations on manifolds
* Force all global dofs to be ordered last and to be on the last process
in parallel
* Speed up dof reordering of mixed space including global dofs by removing
the latter from graph reordering
* Force all dofs on a shared facet to be owned by the same process
* Add FEniCS ('fenics') Python module, identical with DOLFIN Python module
* Add function Form::set_some_coefficients()
* Remove Boost.MPI dependency
* Change GenericMatrix::compresss to return a new matrix (7be3a29)
* Add function GenericTensor::empty()
* Deprecate resizing of linear algebra via the GenericFoo interfaces
(fixes #213)
* Deprecate MPI::process_number() in favour of MPI::rank(MPI_Comm)
* Use PETSc built-in reference counting to manage lifetime of wrapped PETSc
objects
* Remove random access function from MeshEntityIterator (fixes #178)
* Add support for VTK 6 (fixes #149)
* Use MPI communicator in interfaces. Permits the creation of distributed and
local objects, e.g. Meshes.
* Reduce memory usage and increase speed of mesh topology computation
- Removed backport patch, vtk-6 support is now in released version
* dolfin-1.3-vtk-6-enable.patch
-------------------------------------------------------------------
Sun Jan 26 21:33:01 UTC 2014 - kkhere.geo@gmail.com
- backport vtk-6 support from upstream, otherwise no visualisation:
* dolfin-1.3-vtk-6-enable.patch
- corrected installation path due to failing compilation of demo files
-------------------------------------------------------------------
Mon Jan 13 08:03:19 UTC 2014 - kkhere.geo@gmail.com
- update to version 1.3.0
changelog available only in upstream git repository
-------------------------------------------------------------------
Tue Jun 25 12:21:29 UTC 2013 - kkhere.geo@gmail.com
- Activate features by using optional modules
CGAL for geometry manipulation
HDF5 format support in parallel
SLEPc support for eigenvalues
- Cleanup spec file
-------------------------------------------------------------------
Mon Jun 24 09:31:36 UTC 2013 - kkhere.geo@gmail.com
- update to version 1.2.0
* Fixes bug where child/parent hierarchy in Python were destroyed
* Add utility script dolfin-get-demos
* MeshFunctions in python now support iterable protocol
* Expose ufc::dofmap::tabulate_entity_dofs to GenericDofMap interface
* Expose ufc::dofmap::num_entity_dofs to GenericDofMap interface
* Allow setting of row dof coordinates in preconditioners (only works with PETSc backed for now)
* Expose more PETSc/ML parameters
* Improve speed to tabulating coordinates in some DofMap functions
* Feature: Add support for passing a Constant as default value to a CompiledExpression
* Fix bug in dimension check for 1-D ALE
* Remove some redundant graph code
* Improvements in speed of parallel dual grapj builder
* Fix bug in XMDF output for cell-based Functions
* Fixes for latest version of clang compiler
* LocalSolver class added to efficiently solve cell-wise problems
* New implementation of periodic boundary conditions. Now incorporated into the dofmap
* Optional arguments to assemblers removed
* SymmetricAssembler removed
* Domains for assemblers can now only be attached to forms
* SubMesh can now be constructed without a CellFunction argument, if the MeshDomain contains marked celldomains.
* MeshDomains are propagated to a SubMesh during construction
* Simplify generation of a MeshFunction from MeshDomains: No need to call mesh_function with mesh
* Rename dolfin-config.cmake to DOLFINConfig.cmake
* Use CMake to configure JIT compilation of extension modules
* Feature: Add vertex_to_dof_map to DofMap, which map vertex indices to dolfin dofs
* Feature: Add support for solving on m dimensional meshes embedded in n >= m dimensions
-------------------------------------------------------------------
Fri Mar 1 21:25:11 UTC 2013 - scorot@free.fr
- add missing petsc-devel in build requirements
- remove unneeded patch dolfin-1.1.0-cmake-find-scotch.patch
- disable hdf5 support on serial version due to build failure
- add explicit configure flags to disable support for libs
we don't want or which are not available
-------------------------------------------------------------------
Wed Feb 27 21:43:10 UTC 2013 - scorot@free.fr
- add parallel openmpi packages
- add patches in order to fix thirdparty libs detection
* dolfin-1.1.0-cmake-find-hdf5.patch
* dolfin-1.1.0-cmake-find-scotch.patch
-------------------------------------------------------------------
Mon Jan 28 20:40:24 UTC 2013 - scorot@free.fr
- update to version 1.1.0
* Changelog to long see ChangeLog file for a full list
- update patch dolfin-1.1.0-suitesparse.patch
- remove patch dolfin-1.0.0-petsc-3.3.patch not needed anymore
-------------------------------------------------------------------
Sun Oct 14 21:17:09 UTC 2012 - scorot@free.fr
- recompile pyc files of python modules to avoid presence of
rpm build root
- remove remaining duplicates files
-------------------------------------------------------------------
Mon Oct 1 22:08:46 UTC 2012 - scorot@free.fr
- Buildrequires openmpi-devel
- use fdupes for doc duplicated files
-------------------------------------------------------------------
Tue Sep 25 21:09:55 UTC 2012 - scorot@free.fr
- fix build with recent petsc thanks to patch0
- fix suitesparse detection and linking thanks to patch1
- build sphinx documentation
-------------------------------------------------------------------
Sat Sep 22 21:57:40 UTC 2012 - scorot@free.fr
- first package
