twiph

README

twiph
=====

``twiph`` draws fancy interactive graphs representation of Twitter connections (aka friends) given a list of names/surnames. It could be uselful to spot and analyze social connections whithin an arbitrary list of people.

Authentication
==============

You need to create your own app: https://apps.twitter.com/app/new

Then an ``Access Token`` must be created https://apps.twitter.com/app/<app-id>/keys

Examples
========

These are the output for the 100 younger italian parliamentarians grouped by political parties:

|image0|_ |image1|_

.. |image0| image:: http://www.eraclitux.com/img/twiph_100_younger.png
.. _image0: http://www.eraclitux.com/img/twiph_100_younger.png

and with avatars:

.. |image1| image:: http://www.eraclitux.com/img/twiph_100_younger_avatar.png
.. _image1: http://www.eraclitux.com/img/twiph_100_younger_avatar.png

Live demo
=========

http://www.eraclitux.com/twiph_demo/100_younger_camera_groups/index_groups.html

http://www.eraclitux.com/twiph_demo/100_younger_camera_groups/index_avatar.html

Usage
=====

Once retrieved auth credentils and created a ``cfg`` file (a sample is provided in ``conf/sample.cfg``)::

        export CFGP_FILE_PATH=./conf/mine.cfg;
        twiph -csv list.csv

Run tests
=========

Tests need Twitter api credentials. You can specify them by command line or in configuration file::

        export CFGP_FILE_PATH=./conf/test.cfg; go test

Notes
=====

Graphs with nodes >= ~100 start to be *very* heavy (cpu intensive) to display.

Credits
=======

The amazing ``D3.js`` is used to create graphs http://d3js.org/

Political data has been retrieved by ``Openpolis`` REST APIs http://api3.openpolis.it/

Disclaimer
==========

All trademarks, copyrights and other forms of intellectual property belong to their respective owners.

The author is not affiliated with any vendor cited above.

Documentation

Overview

twiph draws graph representation of Twitter connections of a given list of names. csv must have following format:

Name Surname (string, will be searched on Twitter), Group id in graph (int, differentiate colors), Group name in graph (string, showed in tooltips).