-
Franck Gauthier authoredFranck Gauthier authored
README 10.29 KiB
This file describes the installation instructions of SeSAM on all supported systems.
======================================================================
= =
= SeSAM installation notes for modern macOS systems =
= =
======================================================================
__________________
System requirement
__________________
MacOS release ≥ 10.13 (High Sierra)
__________________________________________
INSTALL FROM A PRE-COMPILED BINARY PACKAGE
__________________________________________
1. Download and install the last R release (>= 4.2.x) for macOS from this page:
https://cloud.r-project.org/bin/macosx/base/
2. Download the last binary SeSAM release for macOS from this page:
https://forgemia.inra.fr/gqe-acep/sesam/-/releases
3. Open a R 64-bits session and launch the following commands:
# Installation of R package doParallel
install.packages("doParallel", repos="https://cloud.r-project.org")
# Installation of SeSAM (depending on the release downloaded and the tar.gz archive location on your device)
install.packages("/path/to/SeSAM_<release_number>.tgz", type="binary", repos=NULL)
NB: To install SeSAM on older R releases, see the install from source instructions below
_______________________
INSTALL FROM THE SOURCE
_______________________
Building Prerequisite
=====================
1. Clang compilers with c++ support (clang++)
clang++ may already be installed on your macOS system.
Check the installed version using this command in a terminal window: clang --version
If you need to install or update the Clang compilers,
execute the command: xcode-select --install
and follow the instructions.
2. C++ Boost libraries version 1.56 or later
See section "Getting Boost and GMP libraries" below.
3. C++ GMP libraries
See section "Getting Boost and GMP libraries" below.
3. R version 3.3.3 or later
https://www.r-project.org/
Getting Boost and GMP libraries
===============================
NB: Skip this section if boost and gmp libraries are already installed on your macOS system.
1. Install homebrew (package manager for macOS). https://brew.sh/
# Execute the following commands in a terminal window :
/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"
brew update
2. Use homebrew to install Boost and GMP libraries
# Execute the following commands in a terminal window :
brew install boost
brew install gmp
Install SeSAM from source
=========================
We strongly recommend to have package doParallel installed before using SeSAM.
1. Install package doParallel:
# Open a R session and launch the following command:
install.packages("doParallel", repos="https://cloud.r-project.org")
2. Download the SeSAM source package from the section asset on the last release page:
https://forgemia.inra.fr/gqe-acep/sesam/-/releases
3. Install SeSAM:
# Open a R session where you downloaded the source package and launch the following command:
install.packages("sesam-x.y.tar.gz", type="source", repos=NULL)
# Where x.y corresponds to the release number
======================================================================
= =
= SeSAM installation notes for modern GNU/Linux systems =
= =
======================================================================
__________________
System requirement
__________________
Modern GNU/Linux distributions (tested on Debian 10 and 11, Ubuntu 20.04 and CentOS 7)
_______________________
INSTALL FROM THE SOURCE
_______________________
Building Prerequisite
=====================
1. GNU compiler collections with c++ support (g++) version 6 or later
Check the version installed with this command in a terminal window:
g++ --version
or
c++ --version
2. C++ Boost development libraries version 1.56 or later
See section "Getting Boost and GMP libraries" below.
3. C++ GMP development librairies
See section "Getting Boost and GMP libraries" below.
3. R version 3.3.3 or later
https://www.r-project.org/
Getting Boost and GMP libraries
===============================
We STRONGLY recommend to install these libraries via the package manager of your GNU/Linux distribution.
e.g.:
On Debian/Ubuntu/Mint plateforms (as root or sudoer):
apt-get install g++ libgmp-dev libboost-dev
on Redhat/Fedora/CentOS plateforms (as root or sudoer):
yum install gcc-c++ gmp-devel boost-devel
Install SeSAM from source
=========================
We strongly recommend to have package doParallel installed before using SeSAM.
1. Install package doParallel:
# Open a R session and launch the following command:
install.packages("doParallel", repos="https://cloud.r-project.org")
2. Download the SeSAM source package from the section asset on the last release page:
https://forgemia.inra.fr/gqe-acep/sesam/-/releases
3. Install SeSAM:
# Open a R session where you downloaded the source package and launch the following command:
install.packages("sesam-x.y.tar.gz", type="source", repos=NULL)
# Where x.y corresponds to the release number
======================================================================
= =
= SeSAM installation notes for Windows 10 system =
= =
======================================================================
__________________
System requirement
__________________
MS Windows 10 64 bits version >= 1607 (Redstone 1 relesed on 2016-08-02)
SeSAM does not work on Windows 7 (NB: MS ended W7 support on 2020-01-14)
__________________________________________
INSTALL FROM A PRE-COMPILED BINARY PACKAGE
__________________________________________
1. We provide several pre-compiled binary packages build on R 4.1.x and R 4.2.x.
If needed, download and install the adequat R release (≥ 4.1.x) for Windows from this page:
https://cloud.r-project.org/bin/windows/base/
NB: make sure to install the 64-bits version
2. Download a SeSAM binary package for Windows 10, according to your installed R release, from this page:
https://forgemia.inra.fr/gqe-acep/sesam/-/releases
3. Open a R 64-bits session and launch the following commands:
# Installation of the R package doParallel
install.packages("doParallel", repos="https://cloud.r-project.org")
# Installation of SeSAM (depending on the release downloaded and the zip archive location on your device)
install.packages("/path/to/SeSAM_<release_number>.zip", type="binary", repos=NULL)
_______________________
INSTALL FROM THE SOURCE
_______________________
Building Prerequisite
=====================
1. R ≥ 4.1 (although the minimal required R version to run SeSAM is R 3.3.3, these instructions only apply to R 4.1 and later)
Download and install the last R release (≥ 4.1) for Windows from this page:
https://cloud.r-project.org/bin/windows/base/
NB: make sure to install the 64-bits version
2. Rtools 4.x
Download and install the 64-bits version using the apropriate link on this page :
https://cran.r-project.org/bin/windows/Rtools/
3. C++ Boost libraries version 1.56 or later
See section "Getting Boost and GMP libraries" below.
4. C++ GMP libraries
See section "Getting Boost and GMP libraries" below.
Getting Boost and GMP libraries
===============================
Open a Rtool 4.x terminal window (windows menu > Rtools 4.x > Rtools MinGW 64-bit or Rtools bash) and execute the following commands :
pacman -S mingw-w64-x86_64-boost
pacman -S mingw-w64-x86_64-gmp
Install SeSAM from source
=========================
We strongly recommend to have the R package doParallel installed before using SeSAM.
1. Install the package doParallel:
# Open a R session and launch the following command:
install.packages("doParallel", repos="https://cloud.r-project.org")
2. Download the SeSAM source package from the section asset on the last release page:
https://forgemia.inra.fr/gqe-acep/sesam/-/releases
3. Install SeSAM:
# Open a R session where you downloaded the source package and launch the following command:
install.packages("sesam-x.y.tar.gz", type="source", repos=NULL, INSTALL_opts="--no-multiarch")
# Where x.y corresponds to the release number
__________
ADDITIONAL
__________
Build a 64-bits binary package for Windows 10 and R 4.2
=======================================================
Open Rtool42 terminal (windows menu > Rtools42 > Rtools42 bash)
Change to the directory where you downloaded the SeSAM source archive (e.g. ~/Downloads):
cd ~/Downloads
Extract the source archive (exemple with a tar.gz archive)
tar xzf sesam-<release_number>.tar.gz
Change to the extracted directory
cd sesam-<release_number>
Open a R 4.2 (or later) 64-bit session (path to R executable depends on the chosen directory during R installation)
'C:\Program Files\R\R-4.2.x/bin/x64/R'
and launch the following commands (install R package devtools if needed):
install.packages("devtools", repos="https://cloud.r-project.org")
install.packages("doParallel", repos="https://cloud.r-project.org")
library(devtools)
build(binary=TRUE, args='--no-multiarch')
This will create a binary installation package in the parent directory: SeSAM_<x.x>.zip