2007-12-19
----------

    * Initial starting points, config parser and option parsers
    * Creation of directories for maildir usage
    * Initial creation of file per message (no collision detection or use of
      the state directory yet)

2007-12-20
----------

    * blockquote support
    * headings support
    * pre support
    * added documentation
    * change db key to be unique per feed + link url
    * add rudimentary entity support
    * paragraph wrapping
    * message id header
    * add both text/plain and text/html parts

2007-12-21
----------

    * fix typo for a particular entity
    * Add a prelimanary todo list
    * be slightly more forgiving on connection resets
    * if there's no date in the feed, use todays date/time
    * Only download feeds that have changed (or that don't give us enough data
      to workout if they've changed without downloading it all anyways)
    * updated posts are now "threaded" - adds a References header with the
      previous message-id in it, then adds the previous message id to the
      current message-id so that further updates can reference that properly
    * improve handling of unicode data
    * add unordered list support
    * tidy paragraph handling code to work better
    * improved entity handling

2007-12-22
----------

    * Update TODO list with further escaping needs
    * Add item url to html parts
    * Add item url to bottom of text only part
    * multiple posts with the same link but different guid support - still
      threaded, but don't keep delivering the same messages everytime until
      they leave the feed
    * Add redirect support

2007-12-24
----------

    * Add unittest for unordered list
    * make sure that the string that we use for plain text always ends in a new
      line character

2007-12-31
----------

    * Move some of the list handling above the paragraph handling so that it
      doesn't get confused (bless it!)
    * Make expected output match actual output, unittest now passes
    * Add unit test for some of the badly formed lists that we get after the
      feedparser "sanitizer" has a word with the HTML (currently fails)

2008-01-05
----------

    * serious reworking of the HTML2Text parser
    * fix README to have a more complete config example
    * stop text width from being hardcoded
    * add missing source files for unit tests
    * small fix to paragraph handling
    * add (first draft of) full test suite runner
    * add test for well formed paragraph handling
    * update UnorderedListTests to have better test naming scheme
    * add suite function to UnorderedListTests

2008-01-06
----------

    * Serious reworking of HTML2Text to handle nested lists reasonably
    * unittests improvements

2008-01-07
----------
    * Update list handling code to deal with nested lists better and badly
      formed html

2008-01-10
----------

    * Handle unicode data more effectively.
    * beginning of link support in HTML2Text

2008-01-12
----------

    * Change all entity refs in to unicode strings
    * Update <br> handling to be more effective
    * Ignore unknown tags and just pretend they're part of the flow
    * Add <img> support (very basic!)

2008-01-13
----------

    * Fix bad check on state directory
    * Begin fixes to list handling code
    * Small Unicode fix for img tags.

2008-03-01
----------

    * change images to ReST format
    * Normalise spaces where they should be.

2008-03-02
----------

    * fix ReST image support
    * add lots of entities