]>
https://gitweb.dealii.org/ - dealii.git/log
Timo Heister [Tue, 12 May 2015 18:39:13 +0000 (14:39 -0400)]
rename max tests
Wolfgang Bangerth [Tue, 12 May 2015 12:34:51 +0000 (07:34 -0500)]
Merge pull request #927 from tamiko/fix_test
Fix linear_operator_01 test
Matthias Maier [Tue, 12 May 2015 11:29:40 +0000 (13:29 +0200)]
Testsuite: Bugfix: Add missing interface in linear_operator_01
Martin Kronbichler [Tue, 12 May 2015 11:09:08 +0000 (13:09 +0200)]
Merge pull request #923 from bangerth/MatrixOut-for-more-matrix-types
Make MatrixOut work for Trilinos/PETSc matrices.
Matthias Maier [Mon, 11 May 2015 21:51:02 +0000 (23:51 +0200)]
Merge branch 'luca-heltai-bug-fix-linear-operator'
Closes #925
Matthias Maier [Mon, 11 May 2015 21:49:56 +0000 (23:49 +0200)]
Use operator= to set vectors to zero
Luca Heltai [Mon, 11 May 2015 16:51:40 +0000 (18:51 +0200)]
Fixed a bug in linear operator.
Matthias Maier [Mon, 11 May 2015 21:48:30 +0000 (23:48 +0200)]
Merge pull request #882 from tamiko/add_computation
Add a mechanism to store a partially applied computation
Matthias Maier [Mon, 11 May 2015 21:45:12 +0000 (23:45 +0200)]
Bugfix: Also allow to multiply a PackagedOperation with 0
Matthias Maier [Mon, 11 May 2015 21:33:18 +0000 (23:33 +0200)]
fix typos
Matthias Maier [Sun, 10 May 2015 11:40:59 +0000 (13:40 +0200)]
add a changes.h entry
Matthias Maier [Sun, 10 May 2015 12:19:27 +0000 (14:19 +0200)]
update documentation
Matthias Maier [Sun, 10 May 2015 11:35:30 +0000 (13:35 +0200)]
add a unit test for PackagedOperation
Matthias Maier [Sun, 10 May 2015 11:00:58 +0000 (13:00 +0200)]
add forgotten variant
Matthias Maier [Wed, 6 May 2015 13:34:47 +0000 (15:34 +0200)]
some takeouts
Those variants create LinearOperator or PackagedOperation by 'guessing'
information - remove them.
Matthias Maier [Tue, 5 May 2015 19:59:50 +0000 (21:59 +0200)]
Use PointerComparison::equal
Matthias Maier [Tue, 5 May 2015 16:57:38 +0000 (18:57 +0200)]
rename 'result' to 'apply'
Matthias Maier [Mon, 4 May 2015 15:26:09 +0000 (17:26 +0200)]
use intermediate storage if vmult is called with same dest and src
Matthias Maier [Sun, 3 May 2015 14:27:49 +0000 (16:27 +0200)]
Add missing variants
Matthias Maier [Sun, 3 May 2015 13:49:03 +0000 (15:49 +0200)]
rename and restructure
Matthias Maier [Fri, 1 May 2015 11:46:18 +0000 (13:46 +0200)]
make astyle happy
Matthias Maier [Fri, 1 May 2015 11:37:15 +0000 (13:37 +0200)]
Allow multiplication with matrices and creation by multiplication of matrices
Matthias Maier [Mon, 27 Apr 2015 10:02:29 +0000 (12:02 +0200)]
Restructure documentation
Matthias Maier [Sun, 26 Apr 2015 21:36:41 +0000 (23:36 +0200)]
[RFC] Add a mechanism to store a partially applied computation
Timo Heister [Mon, 11 May 2015 20:51:38 +0000 (16:51 -0400)]
Merge pull request #926 from bangerth/fix-indexing-error
Fix size of a temporary array.
Wolfgang Bangerth [Mon, 11 May 2015 20:46:00 +0000 (15:46 -0500)]
Fix size of a temporary array.
This was broken in the generic-FEValues patch the other day. It
manifested in an erroneous assertion in one of my students
programs while trying out what they had implemented over the
course of the semester.
Wolfgang Bangerth [Mon, 11 May 2015 13:40:26 +0000 (08:40 -0500)]
Also make MatrixOut work for PETSc matrices.
Wolfgang Bangerth [Mon, 11 May 2015 13:24:00 +0000 (08:24 -0500)]
Make MatrixOut also work for Trilinos matrices.
Matthias Maier [Mon, 11 May 2015 13:12:35 +0000 (15:12 +0200)]
Merge pull request #922 from bangerth/doc-update-28
Remove some comments that were not up to date any more.
Wolfgang Bangerth [Mon, 11 May 2015 12:54:56 +0000 (07:54 -0500)]
Remove some outdated comments.
Wolfgang Bangerth [Mon, 11 May 2015 12:25:17 +0000 (07:25 -0500)]
Merge pull request #921 from tamiko/remove_superfluous_typenames
remove superfluous 'typenames', fixes for #918
Matthias Maier [Mon, 11 May 2015 09:20:24 +0000 (11:20 +0200)]
remove superfluous 'typenames', remove tabs
Luca Heltai [Sun, 10 May 2015 17:48:10 +0000 (19:48 +0200)]
Merge pull request #920 from bangerth/fix-non-c++11-code
Fixes for #918
Wolfgang Bangerth [Sun, 10 May 2015 14:18:14 +0000 (09:18 -0500)]
Change code so that it is also correct wherever we use generic scalar types.
Wolfgang Bangerth [Sun, 10 May 2015 14:07:21 +0000 (09:07 -0500)]
Ensure that the new generic FEValues code also compiles in non-C++11 mode.
Matthias Maier [Sun, 10 May 2015 12:29:53 +0000 (14:29 +0200)]
Merge pull request #918 from bangerth/davydden-generic-scalar-types-for-FEValues
Introduce generic data types for values extracted from solution vectors
Matthias Maier [Sun, 10 May 2015 12:22:57 +0000 (14:22 +0200)]
Merge pull request #919 from bangerth/fix-dynamic-sparsity-pattern-iterators-3
Work around another issue with libc++.
Wolfgang Bangerth [Sat, 9 May 2015 20:48:55 +0000 (15:48 -0500)]
Work around another issue with libc++.
The same issue we had with operator== also exists with operator<,
namely that apparently the constructor of the one-past-the-end
object does not initialize the position within the row. This
fixes the problem in essentially the same way.
This should address the remaining failures referenced in #899.
Wolfgang Bangerth [Sat, 9 May 2015 03:47:11 +0000 (22:47 -0500)]
Fix a typo in a comment.
Wolfgang Bangerth [Sat, 9 May 2015 03:44:57 +0000 (22:44 -0500)]
Expand on certain FIXMEs.
Wolfgang Bangerth [Sat, 9 May 2015 03:36:20 +0000 (22:36 -0500)]
Minor cleanups.
Wolfgang Bangerth [Sat, 9 May 2015 03:26:02 +0000 (22:26 -0500)]
Add some more documentation.
Wolfgang Bangerth [Sat, 9 May 2015 03:10:12 +0000 (22:10 -0500)]
Also add a changelog entry.
Wolfgang Bangerth [Sun, 5 Apr 2015 18:32:44 +0000 (13:32 -0500)]
A few minor cleanups.
1/ In integrate_difference(), use the correct type.
2/ In point_gradient(), return an object of the correct type.
Denis Davydov [Fri, 3 Apr 2015 07:13:04 +0000 (09:13 +0200)]
indentation with astyle 2.04 instead of 2.05
Denis Davydov [Fri, 3 Apr 2015 06:50:25 +0000 (08:50 +0200)]
another comment
Denis Davydov [Fri, 3 Apr 2015 06:44:32 +0000 (08:44 +0200)]
Avoid a copy operation implemented by hand. This at least allows rvalue references in C++11 to work if the compiler supports it.
Denis Davydov [Fri, 3 Apr 2015 06:29:28 +0000 (08:29 +0200)]
added comments and FIXME
Denis Davydov [Fri, 3 Apr 2015 06:20:57 +0000 (08:20 +0200)]
change to a straight initialization
Denis Davydov [Thu, 2 Apr 2015 12:48:53 +0000 (14:48 +0200)]
indentation
Denis Davydov [Thu, 2 Apr 2015 07:50:43 +0000 (09:50 +0200)]
update Allreduce operation to fit complex numbers in compute_mean_value
Denis Davydov [Thu, 2 Apr 2015 07:16:46 +0000 (09:16 +0200)]
Wolfgang's patch which fixes tests
Denis Davydov [Wed, 4 Feb 2015 14:20:16 +0000 (15:20 +0100)]
patched error_estimator
Denis Davydov [Mon, 2 Mar 2015 18:08:10 +0000 (19:08 +0100)]
fixed lines in fe_values.cc with += operator
Denis Davydov [Sun, 1 Mar 2015 17:14:10 +0000 (18:14 +0100)]
instantiate Vector<int>
Denis Davydov [Sun, 1 Mar 2015 16:26:25 +0000 (17:26 +0100)]
minor fix in do_function_symmetric_gradients
Denis Davydov [Sun, 1 Mar 2015 15:37:59 +0000 (16:37 +0100)]
temporary fix of vector_tools.templates.h
Denis Davydov [Wed, 4 Feb 2015 14:51:07 +0000 (15:51 +0100)]
fixed point_value_history
Denis Davydov [Wed, 4 Feb 2015 10:14:29 +0000 (11:14 +0100)]
patched fe_field_funciton.templates.h
Denis Davydov [Wed, 4 Feb 2015 08:17:56 +0000 (09:17 +0100)]
fixed KellyErrorEstimator in 1d
Denis Davydov [Tue, 3 Feb 2015 10:44:49 +0000 (11:44 +0100)]
fixing derivative approximation
Denis Davydov [Mon, 2 Feb 2015 22:27:17 +0000 (23:27 +0100)]
temporary fix to data_out_dof_data
Denis Davydov [Mon, 2 Feb 2015 08:28:11 +0000 (09:28 +0100)]
patch vector_selector
Denis Davydov [Sun, 1 Mar 2015 12:08:26 +0000 (13:08 +0100)]
fixed minor misprints in FEValues.cc
Denis Davydov [Sat, 28 Feb 2015 18:51:18 +0000 (19:51 +0100)]
fixed FEValuesBase instantiations
Denis Davydov [Sat, 28 Feb 2015 18:41:31 +0000 (19:41 +0100)]
made FEValuesBase::get_function_xxx to store results in containers of consistent type with the input vector
Denis Davydov [Sat, 28 Feb 2015 17:48:54 +0000 (18:48 +0100)]
fixing do_function_xxx used in the FEValuesBase class
Denis Davydov [Sat, 28 Feb 2015 17:04:29 +0000 (18:04 +0100)]
fixes in do_function_xxx
Denis Davydov [Sat, 28 Feb 2015 15:33:33 +0000 (16:33 +0100)]
fix FEValuesViews instantiations
Denis Davydov [Sat, 28 Feb 2015 15:19:01 +0000 (16:19 +0100)]
Convert FEValuesView::[Tensor|SymmetricTensor] to allow for generic types.
Denis Davydov [Sat, 28 Feb 2015 10:52:04 +0000 (11:52 +0100)]
Convert FEValuesView::Vector to allow for generic types.
Wolfgang Bangerth [Sun, 22 Feb 2015 22:14:18 +0000 (16:14 -0600)]
Convert FEValuesView::Scalar to allow for generic scalar types.
Wolfgang Bangerth [Fri, 8 May 2015 13:24:20 +0000 (08:24 -0500)]
Merge pull request #916 from mtezzele/vector_tools_boundary
Added instantiation for hp
Marco Tezzele [Fri, 8 May 2015 09:59:11 +0000 (11:59 +0200)]
Added instantiation for hp
Martin Kronbichler [Fri, 8 May 2015 06:41:35 +0000 (08:41 +0200)]
Merge pull request #915 from bangerth/doc-update-27
Doc update 27
Martin Kronbichler [Fri, 8 May 2015 06:38:18 +0000 (08:38 +0200)]
Merge pull request #914 from bangerth/doc-update-26
Extend the error message for ExcIndexRange.
Wolfgang Bangerth [Fri, 8 May 2015 02:21:45 +0000 (21:21 -0500)]
Do a better job explaining the output of count_dofs_per_component.
This, too, seems to unsettle students who are sent to read step-20 as an
example of a coupled, vector-valued problem, but who have never heard of
the Raviart-Thomas element and its intricacies.
Wolfgang Bangerth [Fri, 8 May 2015 02:20:21 +0000 (21:20 -0500)]
Discuss the reason why we even use such a complicated solver.
This seems to perpetually unsettle my students as they think that one
can only solve coupled systems using block solvers (step-20 is the
first coupled problem they encounter). But this is not true -- one
can just use a direct solver, for example.
Wolfgang Bangerth [Thu, 7 May 2015 22:57:04 +0000 (17:57 -0500)]
Extend the error message for ExcIndexRange.
I must have had half a dozen students this semester who were unable to
decipher what this error message really means:
An error occurred in line <1183> of file
<../include/deal.II/lac/vector.h> in function
Number &dealii::Vector<double>::operator()(const size_type) [number
= double]
The violated condition was:
i<vec_size
The name and call sequence of the exception was:
ExcIndexRangeType<size_type>(i,0,vec_size)
Additional Information:
Index 0 is not in [0,0[
This commit simply provides them with some more information.
Matthias Maier [Thu, 7 May 2015 17:07:01 +0000 (19:07 +0200)]
Merge pull request #913 from bangerth/cleanups-2
A couple of minor code cleanups
Matthias Maier [Thu, 7 May 2015 17:05:46 +0000 (19:05 +0200)]
Merge pull request #912 from bangerth/doc-update-25
Fix wrong HTML markup.
Matthias Maier [Thu, 7 May 2015 17:05:38 +0000 (19:05 +0200)]
Merge pull request #911 from bangerth/doc-update-24
Update documentation for a recent commit.
Wolfgang Bangerth [Thu, 7 May 2015 16:47:54 +0000 (11:47 -0500)]
Make a couple of tolerances relative to the size of the object.
Wolfgang Bangerth [Thu, 7 May 2015 16:47:33 +0000 (11:47 -0500)]
Remove a few #ifdefs by just moving a variable into the Assertion.
Wolfgang Bangerth [Thu, 7 May 2015 16:33:47 +0000 (11:33 -0500)]
Fix wrong HTML markup.
Wolfgang Bangerth [Thu, 7 May 2015 16:31:26 +0000 (11:31 -0500)]
Update documentation for a recent commit.
Wolfgang Bangerth [Thu, 7 May 2015 16:22:36 +0000 (11:22 -0500)]
Merge pull request #910 from fsonner/paramloopfix
Remove unused function from MultipleParameterLoop::UserClass
fsonner [Thu, 7 May 2015 14:08:35 +0000 (16:08 +0200)]
Remove unused function from param loop UserClass
Timo Heister [Thu, 7 May 2015 11:46:43 +0000 (07:46 -0400)]
Merge pull request #907 from tamiko/fix_condition
Bugfix: Fix misplaced brackets in condition
Martin Kronbichler [Thu, 7 May 2015 09:53:50 +0000 (11:53 +0200)]
Merge pull request #908 from tamiko/fix_warnings
silence a warning, part 2
Matthias Maier [Thu, 7 May 2015 08:20:32 +0000 (10:20 +0200)]
silence a warning, part 2
Martin Kronbichler [Thu, 7 May 2015 08:05:57 +0000 (10:05 +0200)]
Merge pull request #906 from tamiko/fix_warnings
silence a warning
Matthias Maier [Thu, 7 May 2015 08:03:57 +0000 (10:03 +0200)]
Bugfix: Fix misplaced brackets in condition
Matthias Maier [Thu, 7 May 2015 07:50:33 +0000 (09:50 +0200)]
silence a warning
Wolfgang Bangerth [Tue, 5 May 2015 21:01:37 +0000 (18:01 -0300)]
Merge pull request #883 from Rombur/serial_vector
[RFC] deprecate Trilinos and PETSc wrappers of serial vectors
Bruno Turcksin [Tue, 5 May 2015 18:57:33 +0000 (13:57 -0500)]
Don't throw an exception recommanding to use mpirun -np 1 in the tutorials.
Bruno Turcksin [Mon, 4 May 2015 15:09:08 +0000 (10:09 -0500)]
Fix double entry in changes.h
Bruno Turcksin [Mon, 4 May 2015 14:33:44 +0000 (09:33 -0500)]
Deprecate TrilinosWrappers constructors instead of the whole class to silence warnings.
Bruno Turcksin [Fri, 1 May 2015 17:43:32 +0000 (12:43 -0500)]
Deprecate the serial TrilinosWrappers and PETScWrappers classes instead of the constructor. Use complete_set_index in the tutorials.
Bruno Turcksin [Thu, 30 Apr 2015 16:14:44 +0000 (11:14 -0500)]
Throw an exception when one tries to use in parallel tutorials that are serial. Consistently use MPI_COMM_WORLD with reinit(IndesSet, MPI_Comm) in the tutorials.
In the beginning the Universe was created. This has made a lot of
people very angry and has been widely regarded as a bad move.
Douglas Adams
Typeset in
Trocchi and Trocchi Bold Sans Serif.