Goal

Make an aperture datasource that can read files in iCal format, either from disk or from http/webcal. Transform each element into a aperture DataObject, some of the RDF descriptions can probably be stolen from the rdfical people (link below) This task can be very difficult, so keep in mind that we are more interested in something that works slightly than something that handles all bizarre timezone/reccurrence scenarios.

A good testcase would be to import a few simple non-reoccurring events from google calendar.

Steps

• Checkout aperture from the sourceforge CVS (always use latest cvs)
• create a new package in this code for ICAL
• read the tutorials:
  • ​http://aperture.sourceforge.net/tutorial/architecture.html
  • ​http://aperture.sourceforge.net/tutorial/crawlers.html
• copy/paste code from the filesystem crawler or any other crawler you find useful

Relevant resources

• Wikipedia: ​http://en.wikipedia.org/wiki/ICalendar
• iCal specs: ​http://tools.ietf.org/html/2445 (and syntax: ​http://en.wikipedia.org/wiki/RFC2445_Syntax_Reference)
• RDFIcal - ​http://www.w3.org/2002/12/cal/ - the html is old, but there is lots of new code (and schemas) - asking in SWIG for details.
• JICAL - ​http://jical.sourceforge.net/ has at least a partial java implementation of ical..
• aperture - ​http://aperture.sourceforge.net/ - the aperture project to create the crawler for