django-flux 0.3.0

locally fetch, store, and present social media flux

Here’s the situation. You’re on a website and you think to yourself, “whoa, this site is gnarly! I’d really like to stay up to date with this.” You hesitate. Will you be inundated with irrelevant content on Twitter, etc.?

Although this information is readily available and often a click or two away, these additional clicks create an unnecessary barrier to entry. The goal of flux is to make it possible to quickly assess the utility of following a particular feed of content and whether the flux of relevant content will be useful for you. Currently, flux supports:

• Twitter via python-twitter

• RSS via feedparser

• Facebook via fbconsole

• LinkedIn via oauth2

Quick start

1. Install flux with pip:

   [shell]$ pip install django-flux

2. Add flux to the INSTALLED_APPS in settings.py of your django project:

   INSTALLED_APPS += ('flux', )

3. Run syncdb to create the necessary tables:

   [shell]$ python manage.py syncdb

4. Make sure the admin is enabled on your site and add accounts to monitor by visiting the admin page of your site (e.g., http://localhost:8000/admin/flux/account/add)

5. Run the update_flux management command:

   [shell]$ python manage.py update_flux

6. Use the flux_timeseries template tag on Account instances (account below) in your templates:

   <link rel="stylesheet" href="{{STATIC_URL}}flux/css/timeseries.css" />
   
   {% load flux %}
   {% flux_timeseries account %}

   and you should see something like this:

   

7. Customize the styling and layout by altering the CSS, and content accordingly or by taking advantage of any of the other ways of displaying the flux information:

Labels on mouseover with bars

Optionally include labels for the bars with d3.js by including the following in your templates:

<link rel="stylesheet" href="{{STATIC_URL}}flux/css/timeseries.css" />
<link rel="stylesheet" href="{{STATIC_URL}}flux/css/bar_mouseover_labels.css" />

{% load flux %}
{% flux_timeseries account %}

<script src="//d3js.org/d3.v2.js"></script>
<script src="//ajax.googleapis.com/ajax/libs/jquery/1.8.3/jquery.min.js"></script>
<script src="{{STATIC_URL}}flux/js/bar_mouseover_labels.js"></script>

and you should see something like this:

Sparklines

Optionally include sparklines with d3.js by including the following in your templates:

<link rel="stylesheet" href="{{STATIC_URL}}flux/css/timeseries.css" />
<link rel="stylesheet" href="{{STATIC_URL}}flux/css/sparkline.css" />

{% load flux %}
{% flux_timeseries account %}

<script src="//d3js.org/d3.v2.js"></script>
<script src="//ajax.googleapis.com/ajax/libs/jquery/1.8.3/jquery.min.js"></script>
<script src="{{STATIC_URL}}flux/js/sparkline.js"></script>

and you should see something like this:

Labels on mouseover of sparkline

Optionally include labels for each sparkline with d3.js by including the following in your templates:

<link rel="stylesheet" href="{{STATIC_URL}}flux/css/timeseries.css" />
<link rel="stylesheet" href="{{STATIC_URL}}flux/css/sparkline.css" />

{% load flux %}
{% flux_timeseries account %}

<script src="//d3js.org/d3.v2.js"></script>
<script src="//ajax.googleapis.com/ajax/libs/jquery/1.8.3/jquery.min.js"></script>
<script src="{{STATIC_URL}}flux/js/sparkline.js"></script>
<script src="{{STATIC_URL}}flux/js/sparkline_mouseover_labels.js"></script>

and you should see something like this:

Production usage

To have the Accounts monitored be continuously updated, add the following line to your crontab on your production server:

0 0 * * * python /path/to/manage.py update_flux

Account configuration

Account.type=”twitter”

Account.name is the Twitter username (e.g., for http://twitter.com/DsAtweet, Account.name=”DsAtweet”).

No additional information is needed to access Twitter Accounts and Account.other is ignored.

Account.type=”rss”

Account.name is the full URL of the RSS feed you want to track (e.g., for http://datascopeanalytics.com/rss/, Account.name=”http://datascopeanalytics.com/rss/”).

No additional information is needed to access Twitter Accounts and Account.other is ignored.

Account.type=”facebook”

Account.name is the name of the Facebook page that you want to track (e.g., for http://facebook.com/datascopeanalytics, Account.name=”datascopeanalytics”)

The Account.other JSON must also include several attributes in order to authenticate to the Facebook API using fbconsole with something like:

{
    "app_id":"123456789012345",                          // [0-9]+
    "client_secret": "1234567890abcdef1234567890abcdef", // [0-9a-f]+
    "scope": ["read_stream"],
    "email":"[e-mail]",
    "password": "this.is.your.facebook.password"
}

Account.type=”linkedin”

Account.name is the name of the LinkedIn company page that you want to track (e.g., for http://linkedin.com/company/datascope-analytics-llc, Account.name=”datascope-analytics-llc”)

The Account.other JSON must also include several attributes in order to authenticate to the LinkedIn API with something like:

{
    "api_key": "1234567890ab",                      // [0-9a-f]+
    "api_secret": "1234567890ABCDEF",               // [0-9a-zA-Z]+
    "token":"12345678-90ab-cdef-1234-567890abcdef", // [0-9a-f\-]+
    "secret":"12345678-90ab-cdef-1234-567890abcdef" // [0-9a-f\-]+
}

Contribute!

1. Clone the code from github

2. Setup the virtualenv by following the instructions in example_project/virtualenv_requirements.txt

3. Edit, test, and share your code. See the issues page for inspiration and to coordinate with the community.