things properly if they do. Simply stealing our target sprite's _bounds will
not result in the necessary calls to addDirtyRegion() if they change due to
calls to setLocation() on our (now unmanaged) target sprite.
git-svn-id: svn+ssh://src.earth.threerings.net/nenya/trunk@89 ed5b42cb-e716-0410-a449-f6a68f950b19
and efficiently dirty old and new bounds when our bounds change.
git-svn-id: svn+ssh://src.earth.threerings.net/nenya/trunk@88 ed5b42cb-e716-0410-a449-f6a68f950b19
- Ability to set a locale prefix in which to search for locale-specific images.
- Ability to clear the image cache (we want to do this when we switch locales, otherwise we end up with wrong-language images cached.)
git-svn-id: svn+ssh://src.earth.threerings.net/nenya/trunk@83 ed5b42cb-e716-0410-a449-f6a68f950b19
Note: This marks my official conversion to a 100 character line width. We've
been experimenting with it on MetaSOY for a while now and it reduces wrapping
by what feels to be a substantial amount. We also tried 120 which ended up
feeling too wide.
git-svn-id: svn+ssh://src.earth.threerings.net/nenya/trunk@82 ed5b42cb-e716-0410-a449-f6a68f950b19
force states to be reapplied for every batch when disabled.
git-svn-id: svn+ssh://src.earth.threerings.net/nenya/trunk@81 ed5b42cb-e716-0410-a449-f6a68f950b19
node name instead of the identity, to ensure consistent ordering.
git-svn-id: svn+ssh://src.earth.threerings.net/nenya/trunk@77 ed5b42cb-e716-0410-a449-f6a68f950b19
things we're importing from samskivert.
(Our style nowadays is to always explicitely specify imports.)
git-svn-id: svn+ssh://src.earth.threerings.net/nenya/trunk@72 ed5b42cb-e716-0410-a449-f6a68f950b19
of version safety (meaning we won't necessarily have to recompile all
the models when we add a new field) and should help avoid the frequent
binary changes to which Java serialization is prone.
git-svn-id: svn+ssh://src.earth.threerings.net/nenya/trunk@71 ed5b42cb-e716-0410-a449-f6a68f950b19
keep the world bound up-to-date; it won't change, and updateWorldVectors
has no effect.
git-svn-id: svn+ssh://src.earth.threerings.net/nenya/trunk@56 ed5b42cb-e716-0410-a449-f6a68f950b19
textures, etc., that share the same mesh and animations.
git-svn-id: svn+ssh://src.earth.threerings.net/nenya/trunk@49 ed5b42cb-e716-0410-a449-f6a68f950b19
bounds after temporarily setting the transform to identity. I think
there are still issues with JME's bounding volume transformations, but
they seem to err on the side of larger bounding volumes.
git-svn-id: svn+ssh://src.earth.threerings.net/nenya/trunk@45 ed5b42cb-e716-0410-a449-f6a68f950b19
applet. We don't need to do the same restore from back buffer stuff we do for a
ManagedJFrame.
git-svn-id: svn+ssh://src.earth.threerings.net/nenya/trunk@43 ed5b42cb-e716-0410-a449-f6a68f950b19
bounds weren't being recorded correctly (hence they were interpreted as
out of view when they weren't, and the models failed to appear).
git-svn-id: svn+ssh://src.earth.threerings.net/nenya/trunk@42 ed5b42cb-e716-0410-a449-f6a68f950b19
points on either the x or y axis are equal and will interpolate the z
values.
git-svn-id: svn+ssh://src.earth.threerings.net/nenya/trunk@38 ed5b42cb-e716-0410-a449-f6a68f950b19
vertex) to compute the depth used for sorting. There are still annoying
artifacts, but I've already spent too much time exploring all the ways
to sort these things.
git-svn-id: svn+ssh://src.earth.threerings.net/nenya/trunk@31 ed5b42cb-e716-0410-a449-f6a68f950b19