deal.II Newsletter #343

12 views
Skip to first unread message

Rene Gassmoeller

unread,
Jan 27, 2026, 6:00:45 AMJan 27
to dea...@googlegroups.com

Hello everyone!

This is deal.II newsletter #343.
It automatically reports recently merged features and discussions about the deal.II finite element library.

Below you find a list of recently proposed or merged features:

#19234: Mark some comparisons 'if constexpr' (proposed by kronbichler)
#19233: Fix: Ensure array consistency for auxiliary vectors (proposed by rzhang20)
#19232: Fix a typo in a comment (proposed by kronbichler)
#19231: Fix a compile warning (proposed by kronbichler)
#19230: Update links in CONTRIBUTING.md (proposed by kronbichler)
#19228: Some updates to step-7. (proposed by bangerth; merged)
#19227: Optimize SolverMinRes::solve by reducing redundant divisions and square roots (proposed by rzhang20)
#19226: Make a variable 'const'. (proposed by bangerth; merged)
#19225: Fix TrilinosWrappers::Vector::sadd() when the argument is a vector with ghost entries. (proposed by bangerth)
#19224: Some additions to the documentation of step-3. (proposed by bangerth)
#19223: Disable -Wmisleading-indentation. (proposed by bangerth)
#19222: Use 'writable', not 'writeable'. (proposed by bangerth; merged)
#19221: Project BC div-conf in 3D: Avoid temp vector of global problem size (proposed by kronbichler)
#19220: Remove a flippant comment. (proposed by bangerth)
#19219: Re-enable zeroing out vectors with ghost elements. (proposed by bangerth)
#19218: Refactor BlockSparseMatrixEZ to derive from BlockMatrixBase (proposed by FelixKimmerle)
#19216: Remove an unused variable. (proposed by drwells; merged)
#19215: p::f::T: sort a vector instead of using std::map. (proposed by drwells; merged)
#19214: `VectorTools::compute_mean_value()`: avoid allocation. (proposed by drwells; merged)
#19213: tests/performance/timing_step_3: add vmult() timing (proposed by tjhei)
#19212: `ReferenceCell`: use `MappingP1` in another place. (proposed by drwells; merged)
#19211: `ComputeDiagonalHelper`: avoid reallocating scratch arrays. (proposed by drwells; merged)
#19210: `MappingFEField`: `fix get_vertices()`. (proposed by drwells; merged)
#19209: DoFAccessor: use a bigger temporary array. (proposed by drwells; merged)
#19208: Make `DoFAccessor::get_interpolated_dof_values()` work with `ArrayView`. (proposed by drwells)
#19206: Add tutorial step-101 (proposed by lucia-agullo-marti)
#19204: Fix `refinement_06` (proposed by AndSte01; merged)
#19201: CI: maybe fix clang-module build (proposed by tjhei; merged)
#19200: More cleaning up of connectivity.h. (proposed by bangerth)
#19199: Choose better name for ROL interface class. (proposed by marcfehling; merged)
#19195: Implement isotropic refinement of Wedges. (proposed by AndSte01; merged)
#19193: Arkode - prepend namespaces inside the assertion macros (proposed by vovannikov; merged)
#19192: ReferenceCell: make things consistent with connectivity.h. (proposed by drwells; merged)
#19191: Extract implementations of the Arkode wrapper and some of its internal classes to template files (proposed by vovannikov; merged)
#19178: More changes to connectivity.h. (proposed by bangerth; merged)
#19161: Rework algorithms in `execute_refinement_isotropic(...)` to also work with cells that have multiple face types. (proposed by AndSte01; merged)
#19133: VTKWrappers utilities -- part V -- GridTools::parallel_to_serial_vertex_indices (proposed by luca-heltai; merged)

And this is a list of recently opened or closed discussions:

#19229 Refactor `maybe_update_jacobian_pushed_forward_grads` to use `dealii::ndarray` for consistency (opened)
#19217 Make `FEPointEvaluation` work with simplices (opened)
#19207 Optimize SolverMinRes::solve by reducing redundant divisions and square roots (opened)
#19205 Allow v=0 even for vectors that have ghost elements? (opened)
#19203 Regression tester regressed 4b2f91 (opened)
#19202 Disable -Wmisleading-indentation? (opened)
#19198 Wrap `std::inplace_vector` (opened)
#19197 Wedge reference cell has unfortunate documentation. (opened)
#19196 Regression tester regressed 6e0ef8 (opened)
#19194 Sundials - discontinue compatibility with versions <6 (opened)

A list of all major changes since the last release can be found at this website.

Thanks for being part of the community!

Let us know about questions, problems, bugs or just share your experience by writing to this mailing list, or by opening issues or pull requests on Github.
Additional information can be found at our official website.

Reply all
Reply to author
Forward
0 new messages