Previous Section - SPARTA WWW Site - SPARTA Documentation - SPARTA Commands - Next Section

11. Python interface to SPARTA

This section describes how to build and use SPARTA via a Python interface.

The SPARTA distribution includes the file python/ which wraps the library interface to SPARTA. This file makes it possible to run SPARTA, invoke SPARTA commands or give it an input script, extract SPARTA results, and modify internal SPARTA variables, either from a Python script or interactively from a Python prompt. You can do the former in serial or parallel. Running Python interactively in parallel does not generally work, unless you have a package installed that extends your Python to enable multiple instances of Python to read what you type.

Python is a powerful scripting and programming language which can be used to wrap software like SPARTA and many other packages. It can be used to glue multiple pieces of software together, e.g. to run a coupled or multiscale model. See Section 4.7 of the manual and the examples/COUPLE directory of the distribution for more ideas about coupling SPARTA to other codes. See Section 2.4 about how to build SPARTA as a library, and Section 4.6 for a description of the library interface provided in src/library.cpp and src/library.h and how to extend it for your needs. As described below, that interface is what is exposed to Python. It is designed to be easy to add functions to. This can extend the Python inteface as well. See details below.

IMPORTANT NOTE: The examples/COUPLE dir has not been added to the distribution yet.

By using the Python interface, SPARTA can also be coupled with a GUI or other visualization tools that display graphs or animations in real time as SPARTA runs. Examples of such scripts are included in the python directory.

Two advantages of using Python are how concise the language is, and that it can be run interactively, enabling rapid development and debugging of programs. If you use it to mostly invoke costly operations within SPARTA, such as running a simulation for a reasonable number of timesteps, then the overhead cost of invoking SPARTA thru Python will be negligible.

Before using SPARTA from a Python script, you need to do two things. You need to build SPARTA as a dynamic shared library, so it can be loaded by Python. And you need to tell Python how to find the library and the Python wrapper file python/ Both these steps are discussed below. If you wish to run SPARTA in parallel from Python, you also need to extend your Python with MPI. This is also discussed below.

The Python wrapper for SPARTA uses the amazing and magical (to me) "ctypes" package in Python, which auto-generates the interface code needed between Python and a set of C interface routines for a library. Ctypes is part of standard Python for versions 2.5 and later. You can check which version of Python you have installed, by simply typing "python" at a shell prompt.

11.1 Building SPARTA as a shared library

Instructions on how to build SPARTA as a shared library are given in Section 2.4. A shared library is one that is dynamically loadable, which is what Python requires. On Linux this is a library file that ends in ".so", not ".a".

For make, from the src directory, type

make mode=shlib foo 

For CMake, from the build directory, tyoe

cmake -C /path/to/sparta/cmake/presets/foo.cmake -DBUILD_SHARED_LIBS=ON /path/to/sparta/cmake

where foo is the machine target name, such as icc or g++ or serial. This should create the file in the src directory, as well as a soft link, which is what the Python wrapper will load by default. Note that if you are building multiple machine versions of the shared library, the soft link is always set to the most recently built version.

If this fails, see Section 2.3 for more details, especially if your SPARTA build uses auxiliary libraries like MPI which may not be built as shared libraries on your system.

11.2 Installing the Python wrapper into Python

For Python to invoke SPARTA, there are 2 files it needs to know about: is the Python wrapper on the SPARTA library interface. is the shared SPARTA library that Python loads, as described above.

You can insure Python can find these files in one of two ways:

If you set the paths to these files as environment variables, you only have to do it once. For the csh or tcsh shells, add something like this to your ~/.cshrc file, one line for each of the two files:

setenv PYTHONPATH $PYTHONPATH:/home/sjplimp/sparta/python
setenv LD_LIBRARY_PATH $LD_LIBRARY_PATH:/home/sjplimp/sparta/src 

If you use the python/ script, you need to invoke it every time you rebuild SPARTA (as a shared library) or make changes to the python/ file.

You can invoke from the python directory as

% python [libdir] [pydir] 

The optional libdir is where to copy the SPARTA shared library to; the default is /usr/local/lib. The optional pydir is where to copy the file to; the default is the site-packages directory of the version of Python that is running the install script.

Note that libdir must be a location that is in your default LD_LIBRARY_PATH, like /usr/local/lib or /usr/lib. And pydir must be a location that Python looks in by default for imported modules, like its site-packages dir. If you want to copy these files to non-standard locations, such as within your own user space, you will need to set your PYTHONPATH and LD_LIBRARY_PATH environment variables accordingly, as above.

If the script does not allow you to copy files into system directories, prefix the python command with "sudo". If you do this, make sure that the Python that root runs is the same as the Python you run. E.g. you may need to do something like

% sudo /usr/local/bin/python [libdir] [pydir] 

You can also invoke from the make command in the src directory as

% make install-python 

In this mode you cannot append optional arguments. Again, you may need to prefix this with "sudo". In this mode you cannot control which Python is invoked by root.

Note that if you want Python to be able to load different versions of the SPARTA shared library (see this section below), you will need to manually copy files like into the appropriate system directory. This is not needed if you set the LD_LIBRARY_PATH environment variable as described above.

11.3 Extending Python with MPI to run in parallel

If you wish to run SPARTA in parallel from Python, you need to extend your Python with an interface to MPI. This also allows you to make MPI calls directly from Python in your script, if you desire.

There are several Python packages available that purport to wrap MPI as a library and allow MPI functions to be called from Python.

These include

All of these except pyMPI work by wrapping the MPI library and exposing (some portion of) its interface to your Python script. This means Python cannot be used interactively in parallel, since they do not address the issue of interactive input to multiple instances of Python running on different processors. The one exception is pyMPI, which alters the Python interpreter to address this issue, and (I believe) creates a new alternate executable (in place of "python" itself) as a result.

In principle any of these Python/MPI packages should work to invoke SPARTA in parallel and MPI calls themselves from a Python script which is itself running in parallel. However, when I downloaded and looked at a few of them, their documentation was incomplete and I had trouble with their installation. It's not clear if some of the packages are still being actively developed and supported.

The one I recommend, since I have successfully used it with SPARTA, is Pypar. Pypar requires the ubiquitous Numpy package be installed in your Python. After launching python, type

import numpy 

to see if it is installed. If not, here is how to install it (version 1.3.0b1 as of April 2009). Unpack the numpy tarball and from its top-level directory, type

python build
sudo python install 

The "sudo" is only needed if required to copy Numpy files into your Python distribution's site-packages directory.

To install Pypar (version pypar-2.1.4_94 as of Aug 2012), unpack it and from its "source" directory, type

python build
sudo python install 

Again, the "sudo" is only needed if required to copy Pypar files into your Python distribution's site-packages directory.

If you have successully installed Pypar, you should be able to run Python and type

import pypar 

without error. You should also be able to run python in parallel on a simple test script

% mpirun -np 4 python 

where contains the lines

import pypar
print "Proc %d out of %d procs" % (pypar.rank(),pypar.size()) 

and see one line of output for each processor you run on.

IMPORTANT NOTE: To use Pypar and SPARTA in parallel from Python, you must insure both are using the same version of MPI. If you only have one MPI installed on your system, this is not an issue, but it can be if you have multiple MPIs. Your SPARTA build is explicit about which MPI it is using, since you specify the details in your lo-level src/MAKE/ file. Pypar uses the "mpicc" command to find information about the MPI it uses to build against. And it tries to load "" from the LD_LIBRARY_PATH. This may or may not find the MPI library that SPARTA is using. If you have problems running both Pypar and SPARTA together, this is an issue you may need to address, e.g. by moving other MPI installations so that Pypar finds the right one.

11.4 Testing the Python-SPARTA interface

To test if SPARTA is callable from Python, launch Python interactively and type:

>>> from sparta import sparta
>>> spa = sparta() 

If you get no errors, you're ready to use SPARTA from Python. If the 2nd command fails, the most common error to see is

OSError: Could not load SPARTA dynamic library 

which means Python was unable to load the SPARTA shared library. This typically occurs if the system can't find the SPARTA shared library or one of the auxiliary shared libraries it depends on, or if something about the library is incompatible with your Python. The error message should give you an indication of what went wrong.

You can also test the load directly in Python as follows, without first importing from the file:

>>> from ctypes import CDLL
>>> CDLL("") 

If an error occurs, carefully go thru the steps in Section 2.4 and above about building a shared library and about insuring Python can find the necessary two files it needs.

Test SPARTA and Python in serial:

To run a SPARTA test in serial, type these lines into Python interactively from the bench directory:

>>> from sparta import sparta
>>> spa = sparta()
>>> spa.file("") 

Or put the same lines in the file and run it as

% python 

Either way, you should see the results of running the benchmark on a single processor appear on the screen, the same as if you had typed something like:

spa_g++ < 

You can also pass command-line switches, e.g. to set input script variables, through the Python interface.

Replacing the "spa = sparta()" line above with

spa = sparta("","-v","x","100","-v","y","100","-v","z","100") 

is the same as typing

spa_g++ -v x 100 -v y 100 -v z 100 < 

from the command line.

Test SPARTA and Python in parallel:

To run SPARTA in parallel, assuming you have installed the Pypar package as discussed above, create a file containing these lines:

import pypar
from sparta import sparta
spa = sparta()
print "Proc %d out of %d procs has" % (pypar.rank(),pypar.size()),lmp

You can then run it in parallel as:

% mpirun -np 4 python 

and you should see the same output as if you had typed

% mpirun -np 4 spa_g++ < in.lj 

Note that if you leave out the 3 lines from that specify Pypar commands you will instantiate and run SPARTA independently on each of the P processors specified in the mpirun command. In this case you should get 4 sets of output, each showing that a SPARTA run was made on a single processor, instead of one set of output showing that SPARTA ran on 4 processors. If the 1-processor outputs occur, it means that Pypar is not working correctly.

Also note that once you import the PyPar module, Pypar initializes MPI for you, and you can use MPI calls directly in your Python script, as described in the Pypar documentation. The last line of your Python script should be pypar.finalize(), to insure MPI is shut down correctly.

Running Python scripts:

Note that any Python script (not just for SPARTA) can be invoked in one of several ways:

% python foo.script
% python -i foo.script
% foo.script 

The last command requires that the first line of the script be something like this:

#!/usr/local/bin/python -i 

where the path points to where you have Python installed, and requires that you have made the script file executable:

% chmod +x foo.script 

Without the "-i" flag, Python will exit when the script finishes. With the "-i" flag, you will be left in the Python interpreter when the script finishes, so you can type subsequent commands. As mentioned above, you can only run Python interactively when running Python on a single processor, not in parallel.

11.5 Using SPARTA from Python

The Python interface to SPARTA consists of a Python "sparta" module, the source code for which is in python/, which creates a "sparta" object, with a set of methods that can be invoked on that object. The sample Python code below assumes you have first imported the "sparta" module in your Python script, as follows:

from sparta import sparta 

These are the methods defined by the sparta module. If you look at the file src/library.cpp you will see that they correspond one-to-one with calls you can make to the SPARTA library from a C++ or C or Fortran program.

spa = sparta()           # create a SPARTA object using the default library
spa = sparta("g++")      # create a SPARTA object using the library
spa = sparta("",list)    # ditto, with command-line args, e.g. list = ["-echo","screen"]
spa = sparta("g++",list) 
spa.close()              # destroy a SPARTA object 
spa.file(file)           # run an entire input script, file = "in.lj"
spa.command(cmd)         # invoke a single SPARTA command, cmd = "run 100" 
fnum = spa.extract_global(name,type) # extract a global quantity
                                     # name = "dt", "fnum", etc
				     # type = 0 = int
				     #        1 = double 
temp = spa.extract_compute(id,style,type) # extract value(s) from a compute
                                          # id = ID of compute
					  # style = 0 = global data
					  #	    1 = per particle data
					  #	    2 = per grid cell data
					  #	    3 = per surf element data
					  # type = 0 = scalar
					  #	   1 = vector
					  #        2 = array 
var = spa.extract_variable(name,flag)  # extract value(s) from a variable
	                               # name = name of variable
				       # flag = 0 = equal-style variable
				       #        1 = particle-style variable 

IMPORTANT NOTE: Currently, the creation of a SPARTA object from within does not take an MPI communicator as an argument. There should be a way to do this, so that the SPARTA instance runs on a subset of processors if desired, but I don't know how to do it from Pypar. So for now, it runs with MPI_COMM_WORLD, which is all the processors. If someone figures out how to do this with one or more of the Python wrappers for MPI, like Pypar, please let us know and we will amend these doc pages.

Note that you can create multiple SPARTA objects in your Python script, and coordinate and run multiple simulations, e.g.

from sparta import sparta
spa1 = sparta()
spa2 = sparta()

The file() and command() methods allow an input script or single commands to be invoked.

The extract_global(), extract_compute(), and extract_variable() methods return values or pointers to data structures internal to SPARTA.

For extract_global() see the src/library.cpp file for the list of valid names. New names can easily be added. A double or integer is returned. You need to specify the appropriate data type via the type argument.

For extract_compute(), the global, per particle, per grid cell, or per surface element results calulated by the compute can be accessed. What is returned depends on whether the compute calculates a scalar or vector or array. For a scalar, a single double value is returned. If the compute or fix calculates a vector or array, a pointer to the internal SPARTA data is returned, which you can use via normal Python subscripting. See Section 6.4 of the manual for a discussion of global, per particle, per grid, and per surf data, and of scalar, vector, and array data types. See the doc pages for individual computes for a description of what they calculate and store.

For extract_variable(), an equal-style or particle-style variable is evaluated and its result returned.

For equal-style variables a single double value is returned and the group argument is ignored. For particle-style variables, a vector of doubles is returned, one value per particle, which you can use via normal Python subscripting.

As noted above, these Python class methods correspond one-to-one with the functions in the SPARTA library interface in src/library.cpp and library.h. This means you can extend the Python wrapper via the following steps:

11.6 Example Python scripts that use SPARTA

There are demonstration Python scripts included in the python/examples directory of the SPARTA distribution, to illustrate what is possible when Python wraps SPARTA.

See the python/README file for more details.