These scripts were developed as part of an un-confrence session at the Analytics Camp held in Chapel Hill, NC on February 25, 2012, described on this page as http://wiki.analyticscamp.org/nc2011proposals.

The scripts demonstrate how to pull data from the Google Analytics API to generate an RSS feed of a website or blog's most popular pages, as well as creating a CSV file for further processing.

• During the presentation, the RSS feed was incorporated into a WordPress blog by using the RSS widget built into the theme I was using.
• Non-programmer analytics geeks also appreciated having their data in a spreadsheet so they could do their magic.
• Bear in mind, the data these tools collect are only as good as your Google Analytics implementation.

The presentation is available at Slideshare at: http://bitly.com/gaapi2mpp?git

This example code was provided in both Python and Perl.

• The Python example script is stand-alone and does NOT need the Perl example script nor its dependencies.
• The Perl example script is stand-alone and does NOT need the Python example script nor its dependencies.

Below are the installation instructions for each respectively:

The Python script requires the following libraries:

• python-googleanalytics
• PyRSS2Gen

You should be able to type in the following after checking-out the source:

Theoretically you should be able to type the following after checking out the source:

sudo python setup.py install

You can also just install using easy_install like so:

sudo easy_install python-googleanalytics sudo easy_install PyRSS2Gen

The Python script requires the following CPAN libraries:

• Net::Google::Analytics
• Net::Google::AuthSub
• XML::FeedPP

Installation is pretty straight forward:

perl -MCPAN -e shell install Net::Google::Analytics Net::Google::AuthSub XML::FeedPP

NOTE: You may encounter an error installing Net:Google::AuthSub in the form of a file permission error generated by ExtUtil::MakeMaker with regards to man page paths.

You'll want to see this page:

• http://search.cpan.org/~leont/Module-Build-0.40/lib/Module/Build.pm

It explains how you can modify your ~/.cpan/CPAN/MyConfig.pm to modify makepl_arg arg.

Here's what mine looked like:

'makepl_arg' => q[PREFIX=~/perlmods INSTALLMAN1DIR=~/perlmods/share/man/man1 INSTALLSITEMAN1DIR=~/perlmods/share/man/man1 INSTALLMAN3DIR=~/perlmods/share/man/man3 INSTALLS ITEMAN3DIR=~/perlmods/share/man/man3],

On a personal note, Dreamhost doesn't seem to currently want to incorporate this directive from the MyConfig.pm file.

Regardless of whether you're using the Perl or Python example scripts, you'll want to familiarize yourself with the arguments available on the Googl API Dimensions & Metrics Reference page.

Also note that regardless of whether you run the Python or Perl example script, you'll need to have access to a Google Analytics account, and have obtained a profile id (note, this is NOT the same thing as the ua-key).

The general syntax for both scritps is: [script name] [gmail/username] [password] [ga profile id]

As discussed in the demonstration, the script is made so it can run regularly using crontab. The online 'visual crontab utility' may provide assistance to thos individuals who can't always remember the cron syntax.

Below is the command-line syntax for running the Python example script:

python ga-api2mpp.py username\@gmail.com password 1234567

The script generates the follwing files:

• mpp-py.rss - a python generated RSS 2.0 compliant file of the top 20 most popular pages
• mpp-py.csv - a python generated tab-delimited file of all pages pulled from Google Analytics

Below is the command-line syntax for running the Perl example script:

perl ga-api2mpp.pl username\@gmail.com password 1234567

The script generates the follwing files:

• mpp-pl.rss - a perl generated RSS 2.0 compliant file of the top 20 most popular pages
• mpp-pl.csv - a perl generated tab-delimited file of all pages pulled from Google Analytics

Once the required libraries/modules are installed, I was able to run both script on the following platforms:

• Mac OS X 10.7
• Ubuntu 11.00
• CentOS 5
• Windows 7

Of course, your mileage may vary!