update.txt

This file contains notes for upgrades when special instructions are necessary.
In general, simply doing an

git pull

is all that's necessary. In the case that special instructions are needed, we
will attempt to put them here, with the most recent at the top.

------------------------------

2014-11-14:
 - Refines the _get_cookies() method that was added back in July by converting
   it to an attribute simply called cookies. This attribute should contain a
   dict, and a check is in place to ensure that it does. All scrapers have been
   updated to provide this value, and callers should be updated to use it.

   The purpose of this change is to move cookie normalization to the scraper,
   not the caller, simplifying use of Juriscraper.

2014-07-04:
 - Adds a new method on the AbstractSite object that should be implemented by
   all calling libraries. This method is used to pass cookies from Juriscraper
   to the caller. Unfortunately, Texas makes this necessary in order to
   download the PDFs off their site. See the bottom of AbstractSite for
   details, and see sample_caller for how this might be used upstream.

2014-07-13:
 - Adds a new method on the AbstractSite object that should be implemented by
   all calling libraries. This method is used to do post-download cleanup on
   PDFs and HTML files that Juriscraper unearths. See the bottom of
   AbstractSite for details.

2014-07-10:
 - Adds the PhantomJS dependency officially. Looks like there's no crawling
   Texas without it.

2013-06-19:
 - The dependency on the requests library has been updated to version 1.2.3 in
   both CourtListener and Juriscraper:

    sudo pip uninstall requests
    sudo pip install requests==1.2.3