Merge branch 'JK/documentation' into 'develop'

   - Views have a copy-assignment operator which does a deep copy.

   - Views have all other methods present in the relevant container (data

     structure).

7. Smart pointers

   - TODO

TODO: formalize the concepts involving lambda functions (e.g. in `Reduction`)

## Programming principles

TNL follows common programming principles and design patterns to maintain a

comprehensible and efficient code base. We highlight some principles with

respect to the support for different compute architectures:

- CUDA kernels should not operate with needlessly extensive objects, e.g.

  objects which include smart pointers, because this wastes the device

  registers.

- CUDA kernels should not operate with "distributed" objects – they should

  operate only with the "local parts" of the distributed objects. MPI support is

  a higher layer than CUDA support and distributed objects generally contain

  attributes which should not be needed by CUDA kernels.

- Smart pointers should be cached if appropriate in order to avoid repeated

  memory allocations and copies.

  TNL provides a framework for the development of ODE or PDE solvers.

- \ref TNL::Images "Image processing".

  TNL provides structures for the representation of image data. Imports and

  exports from several file formats are provided using external libraries, such

  as [DCMTK](http://dicom.offis.de/dcmtk.php.en) for DICOM files,

  [libpng](http://www.libpng.org/pub/png/libpng.html) for PNG files, or

  [libjpeg](http://libjpeg.sourceforge.net/) for JPEG files.

  exports from several file formats such as DICOM, PNG, and JPEG are provided

  using external libraries (see below).

See also \ref comparison_with_other_libraries "Comparison with other libraries".

## Installation

TNL also provides several optional components:

<a name="optional-components"></a>

- TNL header files in the

  [src/TNL](https://mmg-gitlab.fjfi.cvut.cz/gitlab/tnl/tnl-dev/tree/develop/src/TNL)

  directory.

- Various pre-processing and post-processing tools in the

  [src/Tools](https://mmg-gitlab.fjfi.cvut.cz/gitlab/tnl/tnl-dev/tree/develop/src/Tools)

  directory.

- Python bindings and scripts in the

  [src/Python](https://mmg-gitlab.fjfi.cvut.cz/gitlab/tnl/tnl-dev/tree/develop/src/Python)

  directory.

- Examples of various numerical solvers in the

  [src/Examples](https://mmg-gitlab.fjfi.cvut.cz/gitlab/tnl/tnl-dev/tree/develop/src/Examples)

  directory.

- Benchmarks in the

  [src/Benchmarks](https://mmg-gitlab.fjfi.cvut.cz/gitlab/tnl/tnl-dev/tree/develop/src/Benchmarks)

  directory.

These components can be individually enabled or disabled and installed by a

convenient `install` script. See the [Installation](#installation) section for

details.

## Installation <a name="installation"></a>

You can either download the [stable version](http://tnl-project.org/download/)

or directly clone the git repository via HTTPS:

    git clone gitlab@mmg-gitlab.fjfi.cvut.cz:tnl/tnl-dev.git

Since TNL is a header-only library, no installation is necessary to actually use

the library. You can just extract the [src/TNL](

https://mmg-gitlab.fjfi.cvut.cz/gitlab/tnl/tnl-dev/tree/develop/src/TNL)

subdirectory, for example into `~/.local/include/`, and configure your compiler

to include headers from this path.

the library in your project. See the [Usage](#usage) section for details.

Optionally, you may want to compile and install various [tools and examples](

#optional-components).

You may also want to install some optional dependencies and/or compile and

install various tools and examples. See the following section for details.

### Supported compilers

### Dependencies <a name="dependencies"></a>

In order to use TNL, you need to install a compatible compiler, a parallel

computing platform, and (optionally) some libraries.

- __Supported compilers:__

  You need a compiler which supports the [C++14](

  https://en.wikipedia.org/wiki/C%2B%2B14) standard, for example [GCC](

  https://gcc.gnu.org/) 5.0 or later or [Clang](http://clang.llvm.org/) 3.4 or

later. For CUDA support, you also need [CUDA](

https://docs.nvidia.com/cuda/index.html) 9.0 or later.

  later.

- __Parallel computing platforms:__

  TNL can be used with one or more of the following platforms:

    - [OpenMP](https://en.wikipedia.org/wiki/OpenMP) -- for computations on

      shared-memory multiprocessor platforms.

    - [CUDA](https://docs.nvidia.com/cuda/index.html) 9.0 or later -- for

      computations on Nvidia GPUs.

    - [MPI](https://en.wikipedia.org/wiki/Message_Passing_Interface) -- TNL can

      use an MPI library such as [OpenMPI](https://www.open-mpi.org/) for

      distributed computing. For distributed CUDA computations, the library must

      be [CUDA-aware](

      https://developer.nvidia.com/blog/introduction-cuda-aware-mpi/).

- __Libraries:__

  Various libraries are needed to enable optional features or enhance the

  functionality of some TNL components. Make sure that all relevant packages are

  installed and use the appropriate flags when compiling your project.

  <table>

  <tr><th>Library</th>

      <th>Affected components</th>

      <th>Compiler flags</th>

      <th>Notes</th>

  </tr>

  <tr><td> [zlib](http://zlib.net/) </td>

      <td> \ref TNL::Meshes::Readers "XML-based mesh readers" and \ref TNL::Meshes::Writers "writers" </td>

      <td> `-DHAVE_ZLIB -lz` </td>

      <td> </td>

  </tr>

  <tr><td> [TinyXML2](https://github.com/leethomason/tinyxml2/) </td>

      <td> \ref TNL::Meshes::Readers "XML-based mesh readers" </td>

      <td> `-DHAVE_TINYXML2 -ltinyxml2` </td>

      <td> If TinyXML2 is not found as a system library, the `install` script

           will download, compile and install TinyXML2 along with TNL. </td>

  </tr>

  <tr><td> [Metis](http://glaros.dtc.umn.edu/gkhome/metis/metis/overview) </td>

      <td> `tnl-decompose-mesh` </td>

      <td> </td>

      <td> Only used for the compilation of the `tnl-decompose-mesh` tool. </td>

  </tr>

  <tr><td> [libpng](http://www.libpng.org/pub/png/libpng.html) </td>

      <td> \ref TNL::Images "Image processing" classes </td>

      <td> `-DHAVE_PNG_H -lpng` </td>

      <td> </td>

  </tr>

  <tr><td> [libjpeg](http://libjpeg.sourceforge.net/) </td>

      <td> \ref TNL::Images "Image processing" classes </td>

      <td> `-DHAVE_JPEG_H -ljpeg` </td>

      <td> </td>

  </tr>

  <tr><td> [DCMTK](http://dicom.offis.de/dcmtk.php.en) </td>

      <td> \ref TNL::Images "Image processing" classes </td>

      <td> `-DHAVE_DCMTK_H -ldcm...` </td>

      <td> </td>

  </tr>

  </table>

- __Other language toolchains/interpreters:__

    - Python – install an interpreter for using the Python scripts from TNL and

      the corresponding development package (depending on your operating system)

      for building the Python bindings.

### Optional components

    ./install

[CMake](https://cmake.org/) 3.12.2 or later is required for the compilation. The

script compiles and/or installs the following components into the `~/.local/`

directory:

[CMake](https://cmake.org/) 3.13 or later is required for the compilation.

- TNL header files from the

  [src/TNL](https://mmg-gitlab.fjfi.cvut.cz/gitlab/tnl/tnl-dev/tree/develop/src/TNL)

  directory.

- Various pre-processing and post-processing tools from the

  [src/Tools](https://mmg-gitlab.fjfi.cvut.cz/gitlab/tnl/tnl-dev/tree/develop/src/Tools)

  directory.

- Python bindings and scripts from the

  [src/Python](https://mmg-gitlab.fjfi.cvut.cz/gitlab/tnl/tnl-dev/tree/develop/src/Python)

  directory.

- Examples of various numerical solvers from the

  [src/Examples](https://mmg-gitlab.fjfi.cvut.cz/gitlab/tnl/tnl-dev/tree/develop/src/Examples)

  directory.

- Benchmarks from the

  [src/Benchmarks](https://mmg-gitlab.fjfi.cvut.cz/gitlab/tnl/tnl-dev/tree/develop/src/Benchmarks)

  directory.

- Compiles and executes the unit tests from the

  [src/UnitTests](https://mmg-gitlab.fjfi.cvut.cz/gitlab/tnl/tnl-dev/tree/develop/src/UnitTests)

The script compiles and/or installs all [optional components](#optional-components)

into the `~/.local/` directory, and compiles and executes all unit tests from

the [src/UnitTests](

https://mmg-gitlab.fjfi.cvut.cz/gitlab/tnl/tnl-dev/tree/develop/src/UnitTests)

directory.

Individual components can be disabled and the installation prefix can be changed

by passing command-line arguments to the install script. Run `./install --help`

for details.

## Usage <a name="usage"></a>

TNL can be used with various build systems if you configure the compiler flags

as explained below. See also an [example project](

https://mmg-gitlab.fjfi.cvut.cz/gitlab/tnl/example-project) providing a simple

`Makefile`.

### C++ compiler flags

- Enable the C++14 standard: `-std=c++14`

- Configure the include path: `-I /path/to/include`

    - If you installed TNL with the install script, the include path is

      `<prefix>/include`, where `<prefix>` is the installation path (it is

      `~/.local` by default).

    - If you want to include from the git repository directly, you need to

      specify two include paths: `<git_repo>/src` and `<git_repo/src/3rdparty`,

      where `<git_repo>` is the path where you have cloned the TNL git

      repository.

    - Instead of using the `-I` flag, you can set the `CPATH` environment

      variable to a colon-delimited list of include paths. Note that this may

      affect the build systems of other projects as well. For example:

          export CPATH="$HOME/.local/include:$CPATH"

- Enable optimizations: `-O3 -DNDEBUG` (you can also add

  `-march=native -mtune=native` to enable CPU-specific optimizations).

- Of course, there are many other useful compiler flags. See, for example, our

  [CMakeLists.txt](https://mmg-gitlab.fjfi.cvut.cz/gitlab/tnl/tnl-dev/-/blob/develop/CMakeLists.txt)

  file for flags that we use when developing TNL (there are flags for e.g.

  hiding some useless compiler warnings).

### Compiler flags for parallel computing

To enable parallel computing platforms in TNL, additional compiler flags are

needed. They can be enabled by defining a corresponding C preprocessor macro

which has the form `HAVE_<PLATFORM>`, i.e.:

- `-D HAVE_OPENMP` enables OpenMP (also `-fopenmp` is usually needed to enable

  OpenMP support in the compiler)

- `-D HAVE_CUDA` enables CUDA (the compiler must actually support CUDA, use e.g.

  `nvcc` or `clang++`)

    - For `nvcc`, the following experimental flags are also required:

      `--expt-relaxed-constexpr --expt-extended-lambda`

- `-D HAVE_MPI` enables MPI (use a compiler wrapper such as `mpicxx` or link

  manually against the MPI libraries)

### Environment variables

If you installed some TNL tools or examples using the `install` script, we

recommend you to configure several environment variables for convenience. If you

used the default installation path `~/.local/`:

- `export PATH=$PATH:$HOME/.local/bin`

- If TinyXML2 was installed by the `install` script and not as a system package,

  also `export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$HOME/.local/lib`

These commands can be added to the initialization scripts of your favourite

shell, e.g. `.bash_profile`.

include Makefile.inc

SUBDIRS = 

HEADERS =

SOURCES = tnl-make-example.cpp

CUDA_SOURCES = tnl-make-example-cuda.cu

TARGETS = tnl-make-example 

CUDA_TARGETS = tnl-make-example-cuda

FILES = Makefile \

        Makefile.inc \

        $(CUDA_SOURCES) \

        $(SOURCES) \

        $(HEADERS)

SUBDIRSCLEAN=$(addsuffix clean,$(SUBDIRS))

all: bin subdirs $(TARGETS) $(CUDA_TARGETS)

.PHONY:	subdirs $(SUBDIRS)

subdirs:    $(SUBDIRS)

$(SUBDIRS):

	$(MAKE) -C $@	

bin:

	mkdir -p bin

install: all

	mkdir -p $(INSTALL_DIR)/bin

	cp bin/* $(INSTALL_DIR)/bin

.PHONY:	clean

clean:	$(SUBDIRSCLEAN) clean_curdir

clean_curdir:

	rm -f *.o

%clean:	%

	$(MAKE) -C $< clean

dist: clean

	tar zcvf $(PROJECT_NAME)-src.tgz $(SUBDIRS) $(FILES)

$(TARGETS): % : %.o

	$(CXX) $(LDFLAGS) -o $@ $< $(LDLIBS)

$(CUDA_TARGETS): % : %.cu.o

	$(CUDA_CXX) $(CUDA_LDFLAGS) -o $@ $< $(CUDA_LDLIBS)

$(SOURCES:%.cpp=%.o): %.o: %.cpp

	$(CXX) $(CPPFLAGS) $(CXXFLAGS) -c -o $@ $<

$(CUDA_SOURCES:%.cu=%.cu.o): %.cu.o : %.cu

	$(CUDA_CXX) $(CUDA_CPPFLAGS) $(CUDA_CXXFLAGS) -c -o $@ $<

 No newline at end of file

# Replace the following with your project name

PROJECT_NAME = tnl-make-example

# Replace the following with your TNL installation path

TNL_HEADERS = ${HOME}/.local/include

INSTALL_DIR = ${HOME}/.local

WITH_CUDA = yes

WITH_OPENMP = yes

WITH_DEBUG = no

# If TNL is installed on your system, CUDA arch can be detected automatically using

# a tool 'tnl-cuda-arch'. This is done by default if CUDA_ARCH is set to 'auto'. 

# Otherwise, if you set it manually by telling the CUDA architecture number

# i.e 50, 60 etc.

CUDA_ARCH = auto

# Set-up compilers

CXX = g++

CUDA_CXX = nvcc

# Set-up CXX_FLAGS

CXXFLAGS = -pthread -Wall -Wno-unused-local-typedefs -Wno-unused-variable -Wno-unknown-pragmas -std=c++14 -I$(TNL_HEADERS)

ifeq ( $(WITH_DEBUG), yes )

	CXXFLAGS += -O0 -g

else

   CXXFLAGS += -DNDEBUG -O3 -funroll-loops

endif

# Set-up CUDA_CXXFLAGS

CUDA_CXXFLAGS = -Wno-deprecated-gpu-targets --expt-relaxed-constexpr --expt-extended-lambda -Xcudafe --std c++14 -I$(TNL_HEADERS) 

ifeq ( $(WITH_CUDA), yes )

   CUDA_CXXFLAGS += -DHAVE_CUDA

   ifeq ( $(CUDA_ARCH), auto )

      CUDA_CXXFLAGS += `tnl-cuda-arch`

   else

      CUDA_CXXFLAGS += -gencode arch=compute_$(CUDA_ARCH),code=sm_$(CUDA_ARCH)

   endif

endif

# Set-up CPPFLAGS

CPPFLAGS = -MD -MP

# Set-up LDFLAGS

LDFLAGS += -lm

ifeq ( $(WITH_OPENMP), yes )

   LDFLAGS += -lgomp

endif