Installation
------------
Prerequisities
~~~~~~~~~~~~~~
The program requires the following general packages to run: `Numpy `_, `Matplotlib `_, `Scipy `_ and `Astropy `_. The following LIGO-related packages are also required for full functionality: `GWpy `_, `PyCBC `_, `Glue `_, `LAL `_, `LALburst `_ and `LALsimulation `_.
GWpy and PyCBC
~~~~~~~~~~~~~~
Both `GWpy `_ and `PyCBC `_ pacakges can easily be installed via the `pip `__ package manager as follows::
sudo pip install gwpy
sudo pip install pycbc
LALsuite repository
~~~~~~~~~~~~~~~~~~~
Github
''''''
The remaining of the LIGO-related packages are part of the `LALsuite `__ which can be installed straightforwardly from the `LALsuite Github repository `_ as follows:
1. Clone the repository on your computer::
git clone https://github.com/lscsoft/lalsuite
2. Configure and compile the entire software suite::
./00boot
./configure --enable-swig-python --prefix=/where/to/install
make -j install
.. tip:: In case the configure step doesn't work, e.g. ``./configure failed for lalframe`` error message, the installation of the following tools could fix it (shared by Xiang Zhang using Linux apt-get package manager):
1. Install the `FFTW library `_: ``sudo apt-get install fftw*``
2. Install the `GNU Scientific Library (GSL) `_: ``sudo apt-get install gsl*``
3. Download and install the `Common Data Frame Format Library `__
4. Install the `SWIG software development tool `_: ``sudo apt-get install swig``
5. Download and install the `metaio library `__
6. Install the `libXML2 XML C parser and toolkit `_: ``sudo apt-get install libxml2*``
7. Clone and install the `GLib application `_
8. Install the `HDF5 library `_: ``sudo apt-get install libhdf5*``
3. Pick up the environment by adding the following to your .profile::
source /where/to/install/etc/lal-user-env.sh
source /where/to/install/etc/lalapps-user-env.sh
4. Both glue and pylal packages can then be installed and brought into the environment::
cd glue/
python setup.py install --prefix=/where/to/install/
source /where/to/install/etc/glue-user-env.sh
cd pylal/
python setup.py install --prefix=/where/to/install/
source /where/to/install/etc/pylal-user-env.sh
Note that the sourcing has to take place in that order and should be done each time the user wants to obtain the environment.
MacPorts
''''''''
.. danger:: The LALsuite released code might be mangled or incomplete. Not recommend to install.
If a Mac user still decides to install the LALsuite via Macports, this can be done pretty easily. Detailed information can be found on `this page `__. The `MacPorts `_ package manager needs to be installed on your machine. The following commands should suffice to install the LALsuite package on your machine::
sudo port install lscsoft-deps
sudo port install glue
sudo port install lalapps
The first command will install all the dependencies needed for the LIGO software to be installed. The following 2 commands will install the actual packages.
apt-get
'''''''
.. danger:: The LALsuite released code might be mangled or incomplete. Not recommend to install.
Since the LIGO software is not a default package in the apt package manager system on Debian machine, additional steps will be needed. The following is extracted from the `LALSuite Installation page `_. The first step is to add the following links to the source list located at ``/etc/apt/sources.list``::
deb [arch=amd64] http://software.ligo.org/lscsoft/debian jessie contrib
deb-src [arch=amd64] http://software.ligo.org/lscsoft/debian jessie contrib
Note that the ``[arch=amd64]`` is needed to fix the architecture problem in case it tries to install i386 version on 64-bit Debian. Once the sources have been added, you must first install all the dependencies as follows::
apt-get install build-essential automake autoconf libtool devscripts
The LIGO software can finally be installed using the following command::
apt-get install lscsoft-all
GDAS installation
~~~~~~~~~~~~~~~~~
The GNOME Data Analysis Software can be installed very easily using the `pip `_ package manager as follows::
sudo pip install gdas
This method allows for easy uninstallation. You can also simply download the tarball from the `PyPI webpage `_, unpack it and then do::
sudo python setup.py install
If you wish to install a specific version of the software, you can either download the tarball version from `this page `__ or type the following::
sudo pip install gdas==0.4.0