In the TitledBorder tests I believe we had a slight mistake. We were
treating the font height to be equal to ascent+descent, which is not
necessarily the case. In our case we really want to check for ascent
+descent except for one case (BELOW_BOTTOM) where we have to include the
line distance. The problem here is that line distance is really 0 for
many fonts, so this is a corner case that is hard to test. But I'm quite
certain that the tests are now correct and they PASS now both with Sun
and Classpath.

2006-11-28  Roman Kennke <kennke@aicas.com>

	* gnu/testlet/javax/swing/border/TitledBorder/getBorderInsets.java:
	Fixed test for differentiation between font ascent+descent and
	height.

/Roman