[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Getfem-commits] [getfem-commits] branch devel-yves-octave updated: docu
|
From: |
Yves Renard |
|
Subject: |
[Getfem-commits] [getfem-commits] branch devel-yves-octave updated: documentation updates |
|
Date: |
Sun, 19 Apr 2020 02:45:43 -0400 |
This is an automated email from the git hooks/post-receive script.
renard pushed a commit to branch devel-yves-octave
in repository getfem.
The following commit(s) were added to refs/heads/devel-yves-octave by this push:
new 6e003fa documentation updates
6e003fa is described below
commit 6e003fa0474718dffbd1dbce03d94b88a738a044
Author: Yves Renard <address@hidden>
AuthorDate: Sun Apr 19 08:44:54 2020 +0200
documentation updates
---
.gitignore | 10 ++---
doc/sphinx/source/contents.rst | 2 +-
doc/sphinx/source/install/install_linux.rst | 2 +-
doc/sphinx/source/install/install_mac.rst | 14 ++++++-
.../code_samples/demo_laplacian.m | 0
.../code_samples/demo_step_by_step.m | 0
.../code_samples/demo_tripod.m | 0
.../source/{matlab => matlab_octave}/examples.rst | 41 ++++++++++-----------
.../{matlab => matlab_octave}/images/Makefile | 0
.../{matlab => matlab_octave}/images/hierarchy.fig | 0
.../images/tripodvonmiseswithmesh.png | Bin
.../source/{matlab => matlab_octave}/index.rst | 4 +-
.../source/{matlab => matlab_octave}/install.rst | 0
.../source/{matlab => matlab_octave}/intro.rst | 0
.../source/{matlab => matlab_octave}/mlabgf.rst | 12 +++---
.../source/{matlab => matlab_octave}/oocmd.rst | 12 +++---
.../{matlab => matlab_octave}/plotcmdref.rst | 0
.../source/{matlab => matlab_octave}/pre.rst | 0
doc/sphinx/source/project/libdesc_interface.rst | 14 +++----
doc/sphinx/source/userdoc/export.rst | 12 +++---
doc/sphinx/source/userdoc/gasm_high.rst | 10 ++---
interface/tests/matlab-octave/check_mesh_fem.m | 2 +-
interface/tests/matlab-octave/check_spmat.m | 2 +-
23 files changed, 73 insertions(+), 64 deletions(-)
diff --git a/.gitignore b/.gitignore
index 07ae121..d05871e 100644
--- a/.gitignore
+++ b/.gitignore
@@ -22,9 +22,9 @@ Makefile
/configure
/depcomp
/install-sh
-/interface/tests/matlab/check_all.sh.trs
-interface/tests/python/results/
-interface/tests/python/results1/
+/interface/tests/matlab-octave/check_all.sh.trs
+/interface/tests/python/results/
+/interface/tests/python/results1/
/ltmain.sh
/m4/libtool.m4
/m4/ltoptions.m4
@@ -173,11 +173,11 @@ interface/tests/python/results1/
/doc/sphinx/tools/
/doc/sphinx/source/scilab/cmdref*
/doc/sphinx/source/python/cmdref*
-/doc/sphinx/source/matlab/cmdref*
+/doc/sphinx/source/matlab_octave/cmdref*
/doc/sphinx/build/
/doc/sphinx/homepage/
/doc/sphinx/*.pdf
-/doc/sphinx/source/matlab/images/hierarchy.png
+/doc/sphinx/source/matlab_octave/images/hierarchy.png
/doc/sphinx/source/project/images/diagram.png
/doc/sphinx/source/project/images/getfemelemelem.png
/doc/sphinx/source/project/images/getfemtransgeo.png
diff --git a/doc/sphinx/source/contents.rst b/doc/sphinx/source/contents.rst
index 206f4ae..dc3df70 100644
--- a/doc/sphinx/source/contents.rst
+++ b/doc/sphinx/source/contents.rst
@@ -11,7 +11,7 @@
project/index.rst
gmm/index.rst
- matlab/index.rst
+ matlab_octave/index.rst
python/index.rst
scilab/index.rst
diff --git a/doc/sphinx/source/install/install_linux.rst
b/doc/sphinx/source/install/install_linux.rst
index 57b085d..528e569 100644
--- a/doc/sphinx/source/install/install_linux.rst
+++ b/doc/sphinx/source/install/install_linux.rst
@@ -181,7 +181,7 @@ You have first to install |octv| with the developpement
package
-The last step is to add the path to the toolbox in the matlab path:
+The last step is to add the path to the toolbox in the octave path:
* you can put ``addpath('toolbox_dir', '-begin')`` to your ``$HOME/.octaverc``
file
* you can simply use the ``addpath`` command in the octave command line.
diff --git a/doc/sphinx/source/install/install_mac.rst
b/doc/sphinx/source/install/install_mac.rst
index c89605b..1cb9761 100644
--- a/doc/sphinx/source/install/install_mac.rst
+++ b/doc/sphinx/source/install/install_mac.rst
@@ -135,12 +135,24 @@ Configure Options
Note that there are other options to the configure script. A
``./configure --help`` will list them.
+Octave interface
+^^^^^^^^^^^^^^^^
+
+The compilation of the Octave interface is performed with the
``--enable-octave`` option of the ``configure`` script.
+
+First, you need ``octave`` and ``mkoct-file`` commands accessible from your
shell prompt, so that the correspoding packages have to be intalled.
+
+
+The last step is to add the path to the toolbox in the octave path:
+
+* you can put ``addpath('toolbox_dir', '-begin')`` to your ``$HOME/.octaverc``
file
+* you can simply use the ``addpath`` command in the octave command line.
Matlab interface
^^^^^^^^^^^^^^^^
-The compilation of the matlab interface (with the ``--enable-matlab`` option
of the ``configure`` script) may fail due to a bad configuration of the Matlab
compiler `mex`.
+The compilation of the Matlab interface (with the ``--enable-matlab`` option
of the ``configure`` script) may fail due to a bad configuration of the Matlab
compiler `mex`.
First, you need ``matlab`` and ``mex`` commands accessible from your shell
prompt. If not, add ``Applications/MATLAB_RXXXX.app/bin`` on your path (for
instance with ``export PATH=$PATH:Applications/MATLAB_RXXXX.app/bin`` if your
shell is ``bash`` and for ``XXXX`` your Matlab installed version.
Alternatively, you can make symbolic links to ``matlab`` and ``mex`` executable
in ``/usr/local/bin`` thanks to the command ``sudo ln -s
Applications/MATLAB_RXXXX.app/bin/matlab matlab`` and ``sudo [...]
diff --git a/doc/sphinx/source/matlab/code_samples/demo_laplacian.m
b/doc/sphinx/source/matlab_octave/code_samples/demo_laplacian.m
similarity index 100%
rename from doc/sphinx/source/matlab/code_samples/demo_laplacian.m
rename to doc/sphinx/source/matlab_octave/code_samples/demo_laplacian.m
diff --git a/doc/sphinx/source/matlab/code_samples/demo_step_by_step.m
b/doc/sphinx/source/matlab_octave/code_samples/demo_step_by_step.m
similarity index 100%
rename from doc/sphinx/source/matlab/code_samples/demo_step_by_step.m
rename to doc/sphinx/source/matlab_octave/code_samples/demo_step_by_step.m
diff --git a/doc/sphinx/source/matlab/code_samples/demo_tripod.m
b/doc/sphinx/source/matlab_octave/code_samples/demo_tripod.m
similarity index 100%
rename from doc/sphinx/source/matlab/code_samples/demo_tripod.m
rename to doc/sphinx/source/matlab_octave/code_samples/demo_tripod.m
diff --git a/doc/sphinx/source/matlab/examples.rst
b/doc/sphinx/source/matlab_octave/examples.rst
similarity index 93%
rename from doc/sphinx/source/matlab/examples.rst
rename to doc/sphinx/source/matlab_octave/examples.rst
index bc22ec0..f837e23 100644
--- a/doc/sphinx/source/matlab/examples.rst
+++ b/doc/sphinx/source/matlab_octave/examples.rst
@@ -18,7 +18,7 @@ This example shows the basic usage of getfem, on the
über-canonical problem abo
all others: solving the :envvar:`Laplacian`, :math:`-\Delta u = f` on a square,
with the Dirichlet condition :math:`u = g(x)` on the domain boundary. You can
find
the **m-file** of this example under the name **demo_step_by_step.m** in the
-directory ``interface/tests/matlab/`` of the |gf| distribution.
+directory ``interface/tests/matlab-octave/`` of the |gf| distribution.
The first step is to **create a mesh**. It is possible to create simple
structured meshes or unstructured meshes for simple geometries (see
``gf_mesh('generate', mesher_object mo, scalar h)``) or to rely on an external
mesher (see ``gf_mesh('import', string
FORMAT, string FILENAME))``). For this example, we
@@ -35,8 +35,7 @@ If you try to look at the value of ``m``, you'll notice that
it appears to be a
structure containing two integers. The first one is its identifier, the second
one
is its class-id, i.e. an identifier of its type. This small structure is just
an
"handle" or "descriptor" to the real object, which is stored in the |gf| memory
-and cannot be represented via |Mlab| data structures. Anyway, you can still
-inspect the |gf| objects via the command ``gf_workspace('stats')``.
+and cannot be represented via |octv| and |mlab| data structures. Anyway, you
can still inspect the |gf| objects via the command ``gf_workspace('stats')``.
Now we can try to have a **look at the mesh**, with its vertices numbering and
the
convexes numbering::
@@ -57,12 +56,12 @@ of FEM::
The first instruction builds a new |mlab_mf| object, the second argument
specifies
that this object will be used to interpolate scalar fields (since the unknown
-:math:`u` is a scalar field). The second instruction assigns the :math:`Q^2`
FEM
+:math:`u` is a scalar field). The second instruction assigns the :math:`Q_2`
FEM
to every convex (each basis function is a polynomial of degree 4, remember that
-:math:`P^k\rm I\hspace{-0.15em}Rightarrow` polynomials of degree :math:`k`,
while
-:math:`Q^k\rm I\hspace{-0.15em}Rightarrow` polynomials of degree :math:`2k`).
As :math:`Q^2` is a
+:math:`P_k` are polynomials of degree :math:`k`, while
+:math:`Q_k` are polynomials of degree :math:`2k`). As :math:`Q_2` is a
polynomial FEM, you can view the expression of its basis functions on the
-reference convex::
+reference element::
>> gf_fem_get(gf_fem('FEM_QK(2,2)'), 'poly_str');
ans =
@@ -76,10 +75,10 @@ reference convex::
'-4*x*y + 4*x^2*y + 8*x*y^2 - 8*x^2*y^2'
'x*y - 2*x^2*y - 2*x*y^2 + 4*x^2*y^2'
-It is also possible to make use of the "object oriented" features of |mlab|. As
+It is also possible to make use of the "object oriented" features of |octv|
and |mlab|. As
you may have noticed, when a class "foo" is provided by the |gfi|, it is build
with the function ``gf_foo``, and manipulated with the functions ``gf_foo_get``
-and ``gf_foo_set``. But (with matlab 6.x and better) you may also create the
+and ``gf_foo_set``. But you may also create the
object with the ``gfFoo`` constructor , and manipulated with the ``get(..)``
and
``set(..)`` methods. For example, the previous steps could have been::
@@ -212,7 +211,7 @@ with the instruction::
>> gf_model_get(md, 'solve');
The model now contains the solution (as well as other things, such as the
linear
-system which was solved). It is extracted, a display into a |mlab| figure::
+system which was solved). It is extracted, a display into a |octv| or |mlab|
figure::
>> % extracted solution
>> u = gf_model_get(md, 'variable', 'u');
@@ -223,7 +222,7 @@ system which was solved). It is extracted, a display into a
|mlab| figure::
Another Laplacian with exact solution
-------------------------------------
-This is the :file:`tests/matlab/demo_laplacian.m` example.
+This is the :file:`tests/matlab-octave/demo_laplacian.m` example.
.. literalinclude:: code_samples/demo_laplacian.m
@@ -233,7 +232,7 @@ Linear and non-linear elasticity
This example uses a mesh that was generated with `GiD`_. The object is meshed
with quadratic tetrahedrons. You can find the ``m-file`` of this example under
-the name :file:`demo_tripod.m` in the directory :file:`tests/matlab` of the
+the name :file:`demo_tripod.m` in the directory :file:`tests/matlab-octave` of
the
toolbox distribution.
.. literalinclude:: code_samples/demo_tripod.m
@@ -254,7 +253,7 @@ Avoiding the bricks framework
The model bricks are very convenient, as they hide most of the details of the
assembly of the final linear systems. However it is also possible to stay at a
lower level, and handle the assembly of linear systems, and their resolution,
-directly in |mlab|. For example, the demonstration :file:`demo_tripod_alt.m` is
+directly in |octv| or |mlab|. For example, the demonstration
:file:`demo_tripod_alt.m` is
very similar to the :file:`demo_tripod.m` except that the assembly is
explicit::
nbd=get(mfd, 'nbdof');
@@ -324,15 +323,13 @@ Other examples
order FEMs.
-Using Matlab Object-Oriented features
--------------------------------------
+Using Octave/Matlab Object-Oriented features
+--------------------------------------------
-The basic functions of the |gf| toolbox do not use any advanced |mlab| features
-(except that the handles to getfem objects are stored in a small |mlab|
-structure). But the toolbox comes with a set of |Mlab| objects, which
encapsulate
-the handles and make them look as real |mlab| objects. The aim is not to
provide
-extra-functionalities, but to have a better integration of the toolbox with
-|mlab|.
+The basic functions of the |gf| toolbox do not use any advanced |octv| or
|mlab| features
+(except that the handles to getfem objects are stored in a small structure).
But the toolbox comes with a set of objects, which encapsulate
+the handles and make them look as real |octv| / |mlab| objects. The aim is not
to provide
+extra-functionalities, but to have a better integration of the toolbox.
Here is an example of its use::
@@ -343,7 +340,7 @@ Here is an example of its use::
>> m2=gfMesh('cartesian',0:.1:1,0:.1:1)
gfMesh object ID=1 [17512 bytes], dim=2, nbpts=121, nbcvs=100
- % while \kw{m} is a simple structure, \kw{m2} has been flagged by |mlab|
+ % while \kw{m} is a simple structure, \kw{m2} has been flagged
% as an object of class gfMesh. Since the \texttt{display} method for
% these objects have been overloaded, the toolbox displays some
% information about the mesh instead of the content of the structure.
diff --git a/doc/sphinx/source/matlab/images/Makefile
b/doc/sphinx/source/matlab_octave/images/Makefile
similarity index 100%
rename from doc/sphinx/source/matlab/images/Makefile
rename to doc/sphinx/source/matlab_octave/images/Makefile
diff --git a/doc/sphinx/source/matlab/images/hierarchy.fig
b/doc/sphinx/source/matlab_octave/images/hierarchy.fig
similarity index 100%
rename from doc/sphinx/source/matlab/images/hierarchy.fig
rename to doc/sphinx/source/matlab_octave/images/hierarchy.fig
diff --git a/doc/sphinx/source/matlab/images/tripodvonmiseswithmesh.png
b/doc/sphinx/source/matlab_octave/images/tripodvonmiseswithmesh.png
similarity index 100%
rename from doc/sphinx/source/matlab/images/tripodvonmiseswithmesh.png
rename to doc/sphinx/source/matlab_octave/images/tripodvonmiseswithmesh.png
diff --git a/doc/sphinx/source/matlab/index.rst
b/doc/sphinx/source/matlab_octave/index.rst
similarity index 73%
rename from doc/sphinx/source/matlab/index.rst
rename to doc/sphinx/source/matlab_octave/index.rst
index f3a08c1..e2a4cc2 100644
--- a/doc/sphinx/source/matlab/index.rst
+++ b/doc/sphinx/source/matlab_octave/index.rst
@@ -4,8 +4,8 @@
.. _mlab:
-|mlab| Interface
-################
+|octv| and |mlab| Interfaces
+############################
.. toctree::
:maxdepth: 3
diff --git a/doc/sphinx/source/matlab/install.rst
b/doc/sphinx/source/matlab_octave/install.rst
similarity index 100%
rename from doc/sphinx/source/matlab/install.rst
rename to doc/sphinx/source/matlab_octave/install.rst
diff --git a/doc/sphinx/source/matlab/intro.rst
b/doc/sphinx/source/matlab_octave/intro.rst
similarity index 100%
rename from doc/sphinx/source/matlab/intro.rst
rename to doc/sphinx/source/matlab_octave/intro.rst
diff --git a/doc/sphinx/source/matlab/mlabgf.rst
b/doc/sphinx/source/matlab_octave/mlabgf.rst
similarity index 94%
rename from doc/sphinx/source/matlab/mlabgf.rst
rename to doc/sphinx/source/matlab_octave/mlabgf.rst
index c9d7858..322aa3f 100644
--- a/doc/sphinx/source/matlab/mlabgf.rst
+++ b/doc/sphinx/source/matlab_octave/mlabgf.rst
@@ -7,13 +7,13 @@
.. _mlab-mlabgf:
|gfm| organization
-=====================
+==================
The |gfm| toolbox is just a convenient interface to the |gf| library: you must
have a working |gf| installed on your computer. This toolbox provides a big
-:envvar:`mex-file` (c++ binary callable from |mlab|) and some additional
+:envvar:`mex-file` (c++ binary callable from |octv| or |mlab|) and some
additional
``m-files`` (documentation and extra-functionalities). All the functions of
|gfm|
-are prefixed by ``gf_`` (hence typing ``gf_`` at the |mlab| prompt and then
+are prefixed by ``gf_`` (hence typing ``gf_`` at the |octv| or |mlab| prompt
and then
pressing the ``<tab>`` key is a quick way to obtain the list of getfem
functions).
@@ -120,13 +120,13 @@ are not cleared when a::
>> clear all
-is issued at the |mlab| prompt, but instead the function::
+is issued at the |octv| or |mlab| prompt, but instead the function::
>> gf_workspace('clear all')
should be used. The various |gfm| object can be accessed via *handles* (or
-*descriptors*), which are just |mlab| structures containing 32-bits integer
-identifiers to the real objects. Hence the |mlab| command::
+*descriptors*), which are just |octv| / |mlab| structures containing 32-bits
integer
+identifiers to the real objects. Hence the |octv| or |mlab| command::
>> whos
diff --git a/doc/sphinx/source/matlab/oocmd.rst
b/doc/sphinx/source/matlab_octave/oocmd.rst
similarity index 87%
rename from doc/sphinx/source/matlab/oocmd.rst
rename to doc/sphinx/source/matlab_octave/oocmd.rst
index 13b8ee3..468382e 100644
--- a/doc/sphinx/source/matlab/oocmd.rst
+++ b/doc/sphinx/source/matlab_octave/oocmd.rst
@@ -9,16 +9,16 @@
|gfm| OO-commands
=================
-The toolbox comes with a set of |Mlab| objects `mathworks-oo`_, (look at the
+The toolbox comes with a set of |octv| / |mlab| objects `mathworks-oo`_, (look
at the
:file:`@gf*` sub-directories in the toolbox directory). These object are no
more
-than the getfem object handles, which are flagged by |mlab| as objects.
+than the getfem object handles, which are flagged by |octv| / |mlab| as
objects.
In order to use these objects, you have to call their constructors: ``gfMesh``,
``gfMeshFem``, ``gfGeoTrans``, ``gfFem``, ``gfInteg``. These constructor just
call the corresponding |gfm| function (i.e. ``gf_mesh``, ``gf_mesh_fem``,
...),
-and convert the structure returned by these function into a |mlab| object.
There
+and convert the structure returned by these function into a |octv| / |mlab|
object. There
is also a ``gfObject`` function which converts any getfem handle into the
-corresponding |mlab| object.
+corresponding |octv| / |mlab| object.
With such object, the most interesting feature is that you do not have to call
the "long" functions names ``gf_mesh_fem_get(obj,...)``,
@@ -76,7 +76,7 @@ As an example::
gfGeoTrans object ID= 0 dim=2, nbpts= 6 : GT_PK(2,2)
Although this interface seems more convenient, you must be aware that this
always
-induce a call to a mex-file, and additional |mlab| code::
+induce a call to a mex-file, and additional |octv| / |mlab| code::
>> tic; j=0; for i=1:1000, j=j+mf.nbdof; end; toc
elapsed_time =
@@ -88,7 +88,7 @@ induce a call to a mex-file, and additional |mlab| code::
elapsed_time =
0.0088
-Hence you should always try to store data in |mlab| arrays instead of
+Hence you should always try to store data in |octv| / |mlab| arrays instead of
repetitively calling the getfem functions.
Avalaible object types are :envvar:`gfCvStruct`, :envvar:`gfGeoTrans`,
diff --git a/doc/sphinx/source/matlab/plotcmdref.rst
b/doc/sphinx/source/matlab_octave/plotcmdref.rst
similarity index 100%
rename from doc/sphinx/source/matlab/plotcmdref.rst
rename to doc/sphinx/source/matlab_octave/plotcmdref.rst
diff --git a/doc/sphinx/source/matlab/pre.rst
b/doc/sphinx/source/matlab_octave/pre.rst
similarity index 100%
rename from doc/sphinx/source/matlab/pre.rst
rename to doc/sphinx/source/matlab_octave/pre.rst
diff --git a/doc/sphinx/source/project/libdesc_interface.rst
b/doc/sphinx/source/project/libdesc_interface.rst
index a649048..aab1e41 100644
--- a/doc/sphinx/source/project/libdesc_interface.rst
+++ b/doc/sphinx/source/project/libdesc_interface.rst
@@ -118,7 +118,7 @@ Objects, methods and functions of the interface
The main concepts manipulated by the interface are a limited number of objects
(Fem, Mesh, MeshFem, Model ...), the associated methods and some functions
defined on these objects.
-A special effort has been done to facilitate the addition of new objects,
methods and functions to the interface without doing it separately for each
part supported script language (Python, Scilab, Matlab).
+A special effort has been done to facilitate the addition of new objects,
methods and functions to the interface without doing it separately for each
part supported script language (Python, Scilab, Octave, Matlab).
All the information needed to build the interface for the different objects,
methods and functions is contained in the files `interface/src/gf*.cc`. A
python script (`bin/extract_doc`) produces all the necessary files from the
information it takes there. In particular, it produces the python file
getfem.py, the matlab m-files for the different functions and objects
(including subdirectories) and it also produces the automatic documentations.
@@ -161,7 +161,7 @@ To make all the things work automatically, a certain number
of rules have to be
method/function with a special syntax and also gives a description of the
method/function which will be included in the documentations. The first
line of this commentary is important since it will be analyzed to produce
- the right interface for Python, Matlab and Scilab.
+ the right interface for Python, Octave, Matlab and Scilab.
The syntax for the description of the call of a method/function is the
following: After ``/*@`` a special keyword should be present. It is either
@@ -248,7 +248,7 @@ To make all the things work automatically, a certain number
of rules have to be
with the syntax ``INIT::OBJNAME('method-name', ...)``,
``GET::OBJNAME('method-name', ...)``, ``SET::OBJNAME('method-name', ...)``,
``FUNC::FUNCNAME('subcommand-name', ...)``. This will be replaced with
- the right syntax depending on the language (Matlab, Scilab or Python).
+ the right syntax depending on the language (Octave, Matlab, Scilab or
Python).
* Still in the documentations, parts for a specific language can be added by
``@MATLAB{specific part ...}``, ``@SCILAB{specific part ...}`` and
@@ -274,18 +274,18 @@ Adding a new function or object method to the getfem
interface
If one want to add a new function ``gf_mesh_get(m, "foobar", .)``, then the
main file to modify is :file:`gf_mesh_get.cc`. Remember to check every argument
-passed to the function in order to make sure that the user cannot crash
scilab, matlab or python when using that function. Use the macro defined in
:file:`gf_mesh_get.cc` to add your function.
+passed to the function in order to make sure that the user cannot crash
Scilab, Octave, Matlab or Python when using that function. Use the macro
defined in :file:`gf_mesh_get.cc` to add your function.
Do not forget to add documentation for that function: in
:file:`gf_mesh_get.cc`,
-this is the documentation that appears in the matlab/scilab/python help files
(that is when on
+this is the documentation that appears in the Octave/Matlab/Scilab/Python help
files (that is when on
type "``help gf_mesh_get``" at the matlab prompt), and in the getfem_python
autogenerated documentation.
-IMPORTANT. Note that the array indices start at 0 in Python and 1 in Matlab
and Scilab. A specific function::
+IMPORTANT. Note that the array indices start at 0 in Python and 1 in Octave,
Matlab and Scilab. A specific function::
config::base_index()
-whose value is 0 in python and 1 in Matlab and Scilab has to be used to
exchange indices and array of indices. Take care not to make the correction
twice. Some Array of indices are automatically shifted.
+whose value is 0 in python and 1 in Octave, Matlab and Scilab has to be used
to exchange indices and array of indices. Take care not to make the correction
twice. Some Array of indices are automatically shifted.
Adding a new object to the getfem interface
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
diff --git a/doc/sphinx/source/userdoc/export.rst
b/doc/sphinx/source/userdoc/export.rst
index b67bf42..945b656 100644
--- a/doc/sphinx/source/userdoc/export.rst
+++ b/doc/sphinx/source/userdoc/export.rst
@@ -11,8 +11,8 @@ Export and view a solution
There are essentially four ways to view the result of getfem computations:
-* Matlab, with the matlab-interface.
-* The open-source Mayavi or any other VTK files viewer.
+* Scilab, Octave or Matlab, with the interface.
+* The open-source Paraview or Mayavi or any other VTK files viewer.
* The open-source OpenDX program.
* The open-source Gmsh program.
@@ -21,9 +21,9 @@ The objects that can be exported are, |m|, |mf| objects, and
|smsl|.
Saving mesh and mesh_fem objects for the Matlab interface
---------------------------------------------------------
-If you have installed the Matlab interface, you can simply use
+If you have installed the Scilab, Octave or Matlab interface, you can simply
use
``mesh_fem::write_to_file`` and save the solution as a plain text file, and
then,
-load them into Matlab. For example, supposing you have a solution ``U`` on a
|mf|
+load them with the interface. For example, supposing you have a solution ``U``
on a |mf|
``mf``,::
std::fstream f("solution.U",std::ios::out);
@@ -33,7 +33,7 @@ load them into Matlab. For example, supposing you have a
solution ``U`` on a |mf
// when the 2nd arg is true, the mesh is saved with the |mf|
mf.write_to_file("solution.mf", true);
-and then, under matlab:
+and then, under Scilab, Octave or Matlab:
.. code-block:: matlab
@@ -156,7 +156,7 @@ right choice, but for P2, P3, non linear transformation
etc, it is better to ref
each convex of the original mesh during the slicing operation. This allows an
accurate representation of any finite element field onto a very simple
structure
(linear segment/triangles/tetrahedrons with P1 discontinuous data on them)
which is
-what most visualization programs (gmsh, mayavi, opendx, matlab, etc.) expect.
+what most visualization programs (gmsh, mayavi, opendx, scilab, octave,
matlab, etc.) expect.
Example of use (cut the boundary of a mesh ``m`` with a half-space, and save
the
result into a |smsl|)::
diff --git a/doc/sphinx/source/userdoc/gasm_high.rst
b/doc/sphinx/source/userdoc/gasm_high.rst
index 391f7d7..8d4e9ba 100644
--- a/doc/sphinx/source/userdoc/gasm_high.rst
+++ b/doc/sphinx/source/userdoc/gasm_high.rst
@@ -219,7 +219,7 @@ Note also that the method::
allows to cancel all furnished expressions and allows to re-use the same
workspace for another assembly.
-It is also possible to call the generic assembly from the Python/Scilab/Matlab
interface. See ``gf_asm`` command of the interface for more details.
+It is also possible to call the generic assembly from the
Python/Scilab/Octave/Matlab interface. See ``gf_asm`` command of the interface
for more details.
C++ assembly examples
---------------------
@@ -359,7 +359,7 @@ with ``D`` the flexion modulus and ``nu`` the Poisson ratio.
Script languages call of the assembly
-------------------------------------
-For the use with Python, Scilab or Matlab interfaces, see the respective
documentation, in particular the ``gf_asm`` command and the ``model`` object.
+For the use with Python, Scilab, Octave or Matlab interfaces, see the
respective documentation, in particular the ``gf_asm`` command and the
``model`` object.
The tensors
@@ -765,7 +765,7 @@ For instance, the assembly expression to prescribe the
equality of a variable ``
(Interpolate(u,my_transformation)-u)*lambda
-(see :file:`demo\_periodic\_laplacian.m` in :file:`interface/tests/matlab`
directory).
+(see :file:`demo\_periodic\_laplacian.m` in
:file:`interface/tests/matlab-octave` directory).
In some situations, the interpolation of a point may fail if the transformed
point is outside the target mesh. Both in order to treat this case and to allow
the transformation to differentiate some other cases (see
:ref:`ud-model-contact-friction_raytrace_inter_trans` for the differentiation
between rigid bodies and deformable ones in the
Raytracing_interpolate_transformation) the transformation returns an integer
identifier to the weak form language. A value 0 of this identifier means t [...]
@@ -837,7 +837,7 @@ Note that the region representing the set of all internal
faces of a mesh can be
where ``mr`` is an optional mesh region. If ``mr`` is specified only the face
internal with respect to this region are returned. An important aspect is that
each face is represented only once and is arbitrarily chosen between the two
neighbor elements.
-See for instance :file:`interface/tests/python/demo_laplacian_DG.py` or
:file:`interface/tests/matlab/demo_laplacian_DG.m` for an example of use.
+See for instance :file:`interface/tests/python/demo_laplacian_DG.py` or
:file:`interface/tests/matlab-octave/demo_laplacian_DG.m` for an example of use.
Compared to other interpolate transformations, this transformation is more
optimized and benefits from finite element and geometric transformation
pre-computations.
@@ -872,7 +872,7 @@ Once a secondary domain has been declared, it can be
specified that a GWFL expre
add_linear_twodomain_term(model, mim, expr, region, secondary_domain)
add_nonlinear_twodomain_term(model, mim, expr, region, secondary_domain)
-For the utilisation with the Python/Scilab/Matlab interface, see the
documentation on ``gf_asm`` command and the ``model`` object.
+For the utilisation with the Python/Scilab/Octave/Matlab interface, see the
documentation on ``gf_asm`` command and the ``model`` object.
Inside an expression of GWFL, one can refer to the unit normal vector to a
boundary, to the current position or to the value of a variable thanks to the
expressions::
diff --git a/interface/tests/matlab-octave/check_mesh_fem.m
b/interface/tests/matlab-octave/check_mesh_fem.m
index 9147969..9b4355a 100644
--- a/interface/tests/matlab-octave/check_mesh_fem.m
+++ b/interface/tests/matlab-octave/check_mesh_fem.m
@@ -254,7 +254,7 @@ function check_mesh_fem(iverbose,idebug)
gf_mesh_set(m,'boundary',51,oo);
o=gf_mesh_get(m,'boundary',51);
- gfassert('size(o)==size(oo) && sum(sum(o))==sum(sum(oo))');
+ gfassert('size(o)==size(oo) & sum(sum(o))==sum(sum(oo))');
o=gf_mesh_get(mf2,'boundary',1);
gfassert('isempty(o)');
diff --git a/interface/tests/matlab-octave/check_spmat.m
b/interface/tests/matlab-octave/check_spmat.m
index b7179b2..1f26c77 100644
--- a/interface/tests/matlab-octave/check_spmat.m
+++ b/interface/tests/matlab-octave/check_spmat.m
@@ -40,7 +40,7 @@ function check_spmat(iverbose,idebug)
B = gf_spmat('empty', 11111);
C = gf_spmat('copy', A);
- if (!is_octave) % The import of complex sparse matrices is not working in
Octave for the moment
+ if (~is_octave) % The import of complex sparse matrices is not working in
Octave for the moment
C = sprand(50,50,.1); C(2,2)=1+2i; I = 1:40; J = [6 7 8 3 10];
D = gf_spmat('copy', C, I, J);
DD = gf_spmat_get(D,'full');
| [Prev in Thread] |
Current Thread |
[Next in Thread] |
- [Getfem-commits] [getfem-commits] branch devel-yves-octave updated: documentation updates,
Yves Renard <=