CHANGES.txt

version 0.4 (28th May, 2012)
-----------------------------

- All JavaScript animations are now stopped before layout bug detection
  to reduce the possibility of false positives. Additionally the default
  text detector AnimationAwareTextDetector will now hide all images on
  the analyzed web page when taking screenshots, which includes all
  animated GIF images.
- Fixed issue 8: The detector DetectInvalidImageUrls can now handle data URLs.
- Fixed issue 7: The default text detector AnimationAwareTextDetector
  uses less memory now, is faster and even more robust.
  Further details: Less screenshots are cached now, if animation is found on
  the web page. Furthermore all cached screenshots are now stored in SoftReferences
  and can therefore be garbage collected, if free heap space becomes rare.
- Added new detector DetectElementsWithInvisibleFocus, which detects focusable
  elements like links, input fields or buttons, which do not change their
  visual appearance when they get the focus.
- Updated version of Selenium to 2.21.0


version 0.3 (3rd April, 2012)
-----------------------------

- The TextDetector implementations ignore the new HTML 5 <video> element now too
- Fixed issue 5: Added method setHttpClient(...) to the DetectInvalidImageUrls
  detector, this enables you to configure the HttpClient, which is used for
  downloading CSS resources and checking image URLs.
  Now you can for example add proxy configuration like in this sample code:
    HttpClient myHttpClient = ...;
    FightingLayoutBugs flb = new FightingLayoutBugs();
    flb.configure(DetectInvalidImageUrls.class).setHttpClient(myHttpClient);
- Updated involved third party libraries, e.g. Selenium 2.20.0 and jQuery 1.7.2