SportsML - Python implementation of the SportsML and SportsJS standards

SportsML is an open standard created by the International Press Telecommunications Council to share sports data. See http://www.sportsml.org/

This module is a part-implementation of the standard in Python. Currently it reads actions, entities (players, officials, teams) and some sports metadata objects from SportsML XML files and outputs Python objects or JSON.

Work in progress.

Currently built for Python 3 only - please let us know if you require Python 2 support.

Installation

Installing from PyPI (after we release it to PyPI...):

pip install sportsml

Usage

Example:

import sportsml

parser = sportsml.SportsMLParser("sportsml-file.xml")

print(parser.getSportsContent().to_json())

parser2 = sportsml.SportsMLParser("""
    <?xml version="1.0"?>
    <sports-content xmlns="http://iptc.org/std/nar/2006-10-01/">
      <sports-metadata date-time="2015-02-02T00:35:00-05:00" doc-id="xt.22956338-box" language="en-US" fixture-key="spfixt:event-stats" document-class="spct:event-summary">
        <catalogRef href="http://www.iptc.org/std/catalog/catalog.IPTC-Sports_1.xml"/>
      </sports-metadata>
      <sports-event>
        <event-metadata key="vendevent:l.nfl.com-2014-e.4481" temporal-unit-value="vendor:l.nfl.com-2014-e.4481" event-status="speventstatus:post-event" duration="PT3H36M" start-date-time="2015-02-01T18:30:00-05:00">
          <sports-content-codes/>
        </event-metadata>
      </sports-event>
    </sports-content>
""")

print(parser2.getSportsContent().to_json())

Testing

A very small unit test library is included.

Run it with:

python setup.py test

Tools

So far we have included one sample tool, a simple parser that converts an XML representation of SportsML to the JSON equivalent using the main SportsMLParser class.

Use it from the command line as follows:

$ tools/parser.py examples/xml/rugby-match-classic-generic-3.0.xml

We have also included an extremely simple shell script that runs the above tool over the included SportsML XML files (examples taken from the SportsML repository saved in examples/xml) and generates the corresponding SportsJS JSON files.

$ tools/convert-all-xml-to-json.sh

The converted files are available in the examples/json folder. They now all validate against the work-in-progress SportsJS JSON Schema.

Release notes

• 0.1 - First release, pinned to Python 3 only (use pip >9.0 to ensure pip's Python version requirement works properly)