Installation, Testing & DevInfo

This page details installation instructions, how to run the tests, and some additional information if you want to develop new features for LASIF. If you encounter any issues, please don’t hesitate to contact us or open an issue on GitHub.

Installation

Requirements

LASIF is a Python based program and has a number of dependencies which are listed here. It might well work with other versions but only the versions listed here have been tested and are officially supported. It has been tested on Linux and Mac OS X but should also run just fine on Windows.

  • obspy >= 1.0.1 (www.obspy.org)
  • numpy >= 1.8
  • numexpr (newer versions are also oftentimes faster)
  • matplotlib >= 1.3
  • basemap >= 1.0.7
  • wfs_input_generator (website)
  • geographiclib
  • progressbar and colorama
  • mpi4py and joblib
  • For the webinterface, you additionally need: flask, flask-cache, and geojson.
  • The misfit GUI requires pyqt and pyqtgraph.
  • And for running the tests you need pytest, mock, nose, and flake8.

If you know what you are doing, just make sure these dependencies are available, otherwise please do yourself a favor and download the Anaconda Python distribution, a free package containing almost all dependencies. Download it, install it, and follow the upcoming instructions. It will install LASIF into a separate conda environment. This is very useful to separate the installation from the rest of your system. Additionally it does not require root privileges and thus can be installed almost everywhere.

# Create a new conda environment named "lasif".
$ conda create -n lasif python=2.7
# Activate that environment. You will have to do this every time you start LASIF.
# If you don't want to do that: put that line in your .bashrc/.bash_profile
$ source activate lasif
# Install most things via conda.
$ conda install -c obspy obspy basemap progressbar colorama joblib flask pyqt pyqtgraph pytest nose mock flake8 pip numexpr
# Install some missing things over pip.
$ pip install geographiclib flask-cache geojson
# Install the wfs_input_generator package.
$ pip install https://github.com/krischer/wfs_input_generator/archive/master.zip

Installing LASIF

The actual LASIF module can then be installed with

$ git clone https://github.com/krischer/LASIF.git
$ cd LASIF
$ pip install -v -e .

After the installation one should run the tests to ensure everything is installed correctly and works as intended on your machine.

Updating LASIF

To update LASIF, change into the LASIF directory and type

$ git pull
$ pip install -v -e .

Additionally you might have to update the wfs_input_generator:

$ pip install https://github.com/krischer/wfs_input_generator/archive/master.zi

Please note, that updating LASIF will not update your custom, user-defined functions within your projects (see Customizing LASIF). If you want to update those as well: delete them and execute any LASIF function - this will copy the latest versions of these files to your project directory. Make sure to save any changes you made to those functions!!

Testing

LASIF evolved into a fairly complex piece of code and extensive testing is required to assure that it works as expected.

Running the Tests

To run the tests, cd into the toplevel LASIF directory and execute:

$ py.test

This will recursively find and execute all tests below the current working directory. The output should look akin to the following:

===================================== test session starts =====================================
platform darwin -- Python 2.7.11, pytest-2.9.1, py-1.4.31, pluggy-0.3.1
rootdir: /Users/lion/workspace/code/LASIF, inifile: pytest.ini
collected 195 items

lasif/rotations.py .......
lasif/utils.py ..
lasif/window_selection.py .
...

===================== 189 passed, 4 skipped, 2 xfailed in 160.14 seconds ======================

No errors should occur. If you see nothing - make sure your MPI installation is correct (see above).

Assuming your machine has multiple cores, the test can also be sped up quite a bit by using pytest-xdist which can be installed via pip.

$ pip install pytest-xdist

It enables to distribute the tests across cores. To run on, for example, eight cores, use

$ py.test -n 8

Building the Documentation

sphinx is used to build the documentation so it needs to be installed. The theme is the standalone readthedocs theme. We will use the most up-to-date repository version here.

$ conda install sphinx sphinx_rtd_theme

To actually build the documentation (in this case in the HTML format), run

$ cd doc
$ make html

This might take a while if run for the first time. Subsequent runs are faster.

Developer Information

The following rules should be followed when developing for LASIF:

  • LASIF is written entirely in Python 2.7. Adding support for 3.x would not be a big issue if necessary.

  • Document the code.

  • Adhere to PEP8.

  • All contributed code must be contributed under the GPLv3.

  • Write tests where reasonable.

    • LASIF utilizes Travis CI for continuous integration testing. This means that every commit will be automatically tested and the responsible developer will receive an email in case her/his commit breaks LASIF.
    • The tests also verify the PEP8 conformance of the entire code base.

Terminology

In order to ease development, a consistent terminology should be used throughout the code base.

Assume a channel with a SEED identifier being equal to NET.STA.LOC.CHA, then the separate components should be called:

  • channel_id: NET.STA.LOC.CHA
  • station_id: NET.STA
  • network_code or network: NET
  • station_code or station: STA
  • location_code or location: LOC
  • channel_code or channel: CHA