Notes:
* This is a somewhat reworked version of what is currently in bf-translation's trunk/po/tools, not yet fully functionnal (well, 95% is ;) ) nor fully tested. ultimately, it will replace it (being "svn-linked" in bf-translation).
* Added feature: more complete/strict tests (yet some work to be done here).
* Added spell checking (huge spellcheck commit incomming...).
* Trying to get rid of xgettext itself (should e.g. allow us to use #defines as contexts, among other things...). But currently captures less strings, work needed here too.
Please note this includes libfribidi.dll, as it is hard to find it for windows (unixes should have no problems here).
Issue was caused by linking to grease pencil from direct_link* function
which lead to NULL GP data because it's being read a way later.
Link to GP data in lib_link* instead.
One important thing to keep in mind when using this feature is that you'll need to flip your textures vertically (both the GIMP and Photoshop DDS tools I've seen have support for this on export). This is a quirk in using a texture format originally made for DirectX/DirectDraw, and flipping the compressed data is a real headache. Another quick fix for this issue is to change the Y value for the Size in the Mapping panel in the Texture properties to -1 (default is 1).
The transform operators in nodes will now use the unselected nodes to generate snapping points. Unlike object snapping, node snapping works for the x/y axes separately and snaps node borders to same borders of unselected nodes. The sensitive area for node borders extends over the whole view2D range, to enable simple alignment of nodes in both x and y direction.
For snap points in the node editor an additional enum value is stored to indicate the type of node border (left/right/top/bottom). This works as a constraint on possible node alignments: only same border types align with each other.
Enabled use_gridfill for edgesubdivide called by loopcut. This will break edgeslide in this specific case (intersecting faceloop), but imho makes more sense this way than the other. Very easy to revert anyway, and this should only affect this specific cornercase.
The tool works OK, except it was messing vertices' order of polys, often giving ugly results! Now only using sorted list of vertices indices to find matching polys.
The bug is related to 31581 and the main cause is the small offset that
BM_loop_interp_from_face introduces before calculating barycentric
weights. Solved by only calculating displacement layer.
