These instructions are for compiling from source code on a Linux system.
If you wish to run from another operating system it is recommended to use
the VirtualBox image, as described in the Crosslink manual (docs/crosslink_manual.pdf).
You can also use the VirtualBox image on Linux if you do not wish to compile
the program from the source code.

Summary of Install Process
==========================
This quick summary of the install process assumes Ubuntu Linux is being used,
and was tested on version 14.04. If you use another distribution use the
appropriate package manager and package names for your system.

sudo apt-get install git gcc
git clone https://github.com/eastmallingresearch/crosslink
cd crosslink/src
./make.sh

sudo apt-get install libsdl2-2.0-0 libsdl2-dev libsdl2-image-2.0.0 libsdl2-image-dev
./viewer_make.sh

export CROSSLINK_PATH=~/crosslink
export PATH=${CROSSLINK_PATH}/bin:${CROSSLINK_PATH}/scripts:${PATH}

sudo apt-get install python-scipy python-numpy python-matplotlib

Details
=======

To install Crosslink on a Linux computer begin by downloading the latest
release of the source code from github at https://github.com/eastmallingresearch/crosslink/releases
(choose either Source code (zip) or Source code (tar.gz)) and extract the
files somewhere. You may wish to move the folder to a more convenient location
such as ~/crosslink. To compile the main Crosslink programs you will need
gcc installed. Open a terminal. Type the following to see if you have it
already installed:

gcc ‑‑version

If not, on Ubuntu or Debian based systems gcc can usually be installed using:

sudo apt-get update
sudo apt-get install gcc

If you do not have the ability to install packages on your system ask your
system administrator to help you. If you use a different Linux distribution
use the appropriate package management tool for your system. Now compile
the main Crosslink programs using (assuming the crosslink source code is
now in a directory called ~/crosslink):

cd ~/crosslink/src
./make.sh

If no error messages appear the compilation was probably successful, and
the programs crosslink_group, crosslink_map and crosslink_pos should have
appear in the bin directory. Test you can run crosslink_group as follows:

../bin/crosslink_group ‑‑help

You should see a help message starting “Crosslink Copyright (C) 2016 NIAB EMR...”.
To compile the data visualisation tools you will need SDL2 installed, including
the development files. On Ubuntu or Debian based system you should be able
to install these using:

sudo apt-get install libsdl2-2.0-0 libsdl2-dev
sudo apt-get install libsdl2-image-2.0.0 libsdl2-image-dev

Now compile the data visualisation programs (crosslink_viewer and crosslink_graphical)
using:

./viewer_make.sh

Check you can run crosslink_viewer using:

../bin/crosslink_viewer ‑‑help

You should see a similar help message appear. Next, to allow running the
programs from any directory on your system add the path to Crosslink’s programs
to your .bashrc file. Here I will assume that the editor nano will be used
(install first using sudo apt-get install nano if required). Launch the
nano editor using:

nano ~/.bashrc

Add the following lines to the bottom of the file (again assuming Crosslink
files were installed to ~/crosslink):

export CROSSLINK_PATH=~/crosslink
export PATH=${CROSSLINK_PATH}/bin:${CROSSLINK_PATH}/scripts:${PATH}

Make sure not to omit the final ${PATH} from the second line! Press CTRL-O
then ENTER to save and CTRL-X to close the editor. Then close the terminal
and open a new one (or log out and then back in again if you are logged in
remotely). You should now be able to run Crosslink’s programs from any directory:

cd ~
mkdir test_crosslink
cd test_crosslink
crosslink_group ‑‑help

You should see the same help message as before. Now try running one of the
example pipelines:

${CROSSLINK_PATH}/example_pipelines/simple_simulated.sh

The pipeline should run and produce new output files in your current directory,
including one called final000.loc. Test you can view this file now using
crosslink_viewer:

crosslink_viewer ‑‑inp=final000.loc

which should display a coloured plot similar to the one shown for the virtual
machine installation section above. Some of the python helper scripts require
one or more of the following python modules: scipy, numpy, matplotlib. Install
them using:

sudo apt-get install python-scipy python-numpy python-matplotlib

Or, if you have pip available you may wish to try installing them only in
your user account rather than system-wide, using:

pip install ‑‑user numpy scipy matplotlib