Calculators for PDF, bond valence sum and other pair quantities

The diffpy.srreal package provides calculators for atomic pair distribution function (PDF), bond valence sums (BVS), atom overlaps for a hard-sphere model, bond distances and directions up to specified maximum distance. The atomic structure models are represented with internal classes as non-periodic, periodic or structures with space group symmetries. The package provides implicit adapters from diffpy.structure classes or from Crystal or Molecule objects from pyobjcryst. Adapters can be easily defined for any other structure representations in Python allowing their direct use with the calculators. Calculators support two evaluation models - BASIC, which performs a full pair-summation every time, and OPTIMIZED, which updates only pair contributions that have changed since the last evaluation. Calculations can be split among parallel jobs using Python multiprocessing package or any other library that provides parallel map function. PDF calculations can be done in two modes - either as a real-space summation of peak profiles (PDFCalculator) or as a reciprocal-space Debye summation and Fourier transform of the total scattering structure function (DebyePDFCalculator).

The diffpy.srreal package is a Python binding to the C library libdiffpy (https://github.com/diffpy/libdiffpy). Calculators are created as objects of a given calculator type and so multiple instances of the same calculator type can exist with different configurations. Calculators are composed of other objects that perform lower-level tasks, such as calculating peak profile or looking up atom scattering factors. These objects can be re-assigned at runtime allowing to easily customize the calculation procedure. New classes can be defined using object inheritance either in Python or in C and used with the existing calculators; as an example, this allows to calculate PDF with a user-defined profile function. A new calculator class can be also defined for any quantity that is obtained by iteration over atom pairs, by defining only the function that processes atom-pair contributions.

For more information about the diffpy.srreal library, see users manual at http://diffpy.github.io/diffpy.srreal.

The diffpy.srreal package requires Python 3.7, 3.6, 3.5 or 2.7, C compiler and the following software:

• setuptools - tools for installing Python packages
• NumPy - library for scientific computing with Python
• python-dev - header files for interfacing Python with C
• libboost-all-dev - Boost C libraries and development files
• libdiffpy - C library for PDF, bond valence sum and other pair quantity calculators https://github.com/diffpy/libdiffpy
• diffpy.structure - simple storage and manipulation of atomic structures https://github.com/diffpy/diffpy.structure
• scons - software construction tool (optional)

Optional software:

• periodictable - periodic table of elements in Python http://www.reflectometry.org/danse/elements.html
• pyobjcryst - Python bindings to ObjCryst , the Object Oriented Crystallographic library for C , https://github.com/diffpy/pyobjcryst.

We recommend to use Anaconda Python as it allows to install all software dependencies together with diffpy.srreal. For other Python distributions it is necessary to install the required software separately. As an example, on Ubuntu Linux some of the required software can be installed using

sudo apt-get install \
   python-setuptools python-numpy scons \
   build-essential python-dev libboost-all-dev

To install the remaining packages see the installation instructions at their respective web pages.

The preferred method is to use Anaconda Python and install from the "diffpy" channel of Anaconda packages

conda config --add channels diffpy
conda install diffpy.srreal

diffpy.srreal is also included in the "diffpy-cmi" collection of packages for structure analysis

conda install diffpy-cmi

If you prefer to install from sources, make sure all required software packages are in place and then run

python setup.py install

You may need to use sudo with system Python so the process is allowed to copy files to the system directories. If administrator (root) access is not available, see the output from python setup.py install --help for options to install to a user-writable location. The installation integrity can be verified by executing the included tests with

python -m diffpy.srreal.tests.run

An alternative way of installing diffpy.srreal is to use the SCons tool, which can speed up the process by compiling C files in several parallel jobs (-j4)

sudo scons -j4 install

See scons -h for decription of build targets and options.

diffpy.srreal is an open-source software developed as a part of the DiffPy-CMI complex modeling initiative at the Brookhaven National Laboratory. The diffpy.srreal sources are hosted at https://github.com/diffpy/diffpy.srreal.

Feel free to fork the project and contribute. To install diffpy.srreal in a development mode, where the sources are directly used by Python rather than copied to a system directory, use

python setup.py develop --user

To rebuild the C extension module and then optionally test the code integrity, use

scons -j4 build=debug develop [test]

When developing with Anaconda Python it is essential to specify header path, library path and runtime library path for the active Anaconda environment. This can be achieved by setting the CPATH, LIBRARY_PATH and LDFLAGS environment variables as follows:

# resolve the prefix directory P of the active Anaconda environment
P="$(conda info --json | grep default_prefix | cut -d\" -f4)"
export CPATH=$P/include
export LIBRARY_PATH=$P/lib
export LDFLAGS=-Wl,-rpath,$P/lib
# compile and re-install diffpy.srreal
scons -j4 build=debug develop

Note the Anaconda package for the required libdiffpy library is built with a C compiler provided by Anaconda. This may cause incompatibility with system C . In such case use Anaconda C to build diffpy.srreal.

For more information on diffpy.srreal please visit the project web-page

http://www.diffpy.org

or email Prof. Simon Billinge at [email protected].