Getting started with Anaconda

A minimal set of data processing tools could include

• TIMEleSS Viewer : to look at diffraction images and median calculations
• ImageD11: script for peak search, experimental parameters refinement, peak assignment, can do grain indexing, and many others
• GrainSpotter: grain indexing
• Specific TIMEleSS tools: a collection of scripts from the TIMEleSS projects, for both pre- and post-processing

Those are themselves built against a suite of libraries and could be hard to get working.

Anaconda is a wrapper the install programming environments that runs on Windows, macOS, and Linux. Conda should be able to quickly install, run, and update packages and their dependencies. It also easily creates, saves, loads, and switches between environments on your local computer.

In recent years, however, Anaconda has become very bloated. You need to pay for services, and it is challenging to use. The main author of this wiki hence does not recommend Anaconda anymore and recommends installing a dedicated python environement.

If you have Anaconda working and would like to use it. The instructions are below.

You first step is hence to install Anaconda on your mac.

Once Anaconda is installed,

• Open the MacOS Terminal and type: python
• You should see something like Python 3.6.3 | Anaconda Inc. |
• At the Python REPL (the Python »> prompt), exit out of the Python REPL using the command exit().

The following packages should be installed from anaconda to avoid compatibility issues

• numpy: conda install -c anaconda numpy
• scipy: conda install -c anaconda scipy
• matplotlib: conda install -c anaconda matplotlib
• pyqt: conda install -c anaconda pyqt
• Pmw: conda install -c conda-forge pmw
• OpenPG: conda install pyopenGL

Then there are two packages wich ended up conflicting when trying to install using conda. Resorting to pip instead:

• fabio: python -m pip install fabio
• pyopengltk: python -m pip install pyopengltk

You also need to install silx, either with the conda command or pip, whatever works.

Python packages can be installed from source, downloading the source code from the original website, or through pip. Source download will always get you the latest version, with the latest developments. Installation with pip is much easier, but may not be the most up-to-date.

If you are super comfortable with python, anaconda, and such, jump to the next section and install everything from the source files at github. If not, stay here and do part of the work with pip.

Some of the Fable-3XRD packages are easy to install

• ImageD11, simply type the following command in your anaconda terminal python -m pip install ImageD11
• TIMEleSS-tools: python -m pip install timeless-tools
• silx: python -m pip install silx (not a Fable-3XRD but could become useful in the future)

GrainSpotter, also, is available automatically on Anaconda if you are using linux (GrainSpotter is a C program, hence the difference procedure)

• In a terminal: conda install -c fable grainspotter
• This does not seem to work with a mac, see below (installing from source)

Those few lines will install the packages, but also thousands of various libraries that they depend on.

Other packages are required but need to be installed from the Fable-3DXRD source at github:

• polyxsim (used by some of the TIMEleSS-tools)
• TIMEleSS Viewer

For both, procedure is identical

• go to the Fable-3DXRD source at github,
• enter the corresponding repository (polyxsim or fabian)
• at the top right of the page, click on the green Code button, and download a zip archive,
• on your machine, unzip the archive,
• from the terminal, move into the folder that was created,
• type python setup.py build,
• type python setup.py install

GrainSpotter is a C program which, on a mac, is not readily available in anaconda

• Download the GrainSpotter source code from https://sourceforge.net/p/fable/code/HEAD/tree/GrainSpotter/trunk/,
• Move into the folder with GrainSpotter and type make,
• Copy GrainSpotter.0.90 somewhere where you can access it easily. This is the executable.

It should work out of the box.

By now, you should be done and have a minimal working installation on your system.