* moved from NMesh to Mesh
* added more error checking.
* tested with 1660 lightwave models
* faster dictionary usage.
BPyImage - disallowed recursive searching of '/' or a windows drive letter.
update by the script author - may have killed some changes cambo did to the previous version but the layout of the script is too significantly different to merge them.
continual script stuff, minor stuff..
fix for bad return in group.
Added back group.objects.append/remove but print deprectaed warning. also some epydoc changes.
weightpaint_envelope_assign - can update active vgroup only
weightpaint_normalize - fixups
BKE_plugin_types - made the max length 32 ratehr then 16 so you can fill the text space in pupBlock
- Dont allow it to apply the matrix twice to a linked-dupe. (Same as Apply Loc/Size/Rot)
- Make sure that the source object is active, not just the first selected object.
- Use Mesh instead of NMesh.
- use mesh.transform(mat) rather then vert by vert vec*mat
Other scripts had pupBlock changes for better layout.
Click project, a nifty 3 click way to use a face to set a projection aurientation and then 2 more clicks for the U and V scales
with realtime updating as you move the mouse
should be faster then using view project all the time because you dont have to line the view up with the faces.
Changed...
me = NMesh.GetRaw(ob.data.name)
To
me = NMesh.GetRaw(ob.getData(name_only=1))
The line above converts the entire mesh to a python thick wrapper just to get its name, then only to that that same mesh again??
A lot of older scripts did this but theres no reason to do it.
in most (all?) cases..
me = ob.data - should be ok, but theres a subtle difference and I dont want to break anything.
xfig_export, xsi_export and videoscape_export need error checking too.
Minor changes to radiosity_export, updated to work with the new API (vertexColors MTFace)
obdatacopier - was getting the selection but not using it. made it ignore some read only properties.
BPyMesh - minor change: use face iterator rather then making a tuple()
curve2vecs(ob, WORLDSPACE= True):
'''
Takes a curve object and returns a list of vec lists (polylines)
one list per curve
This is usefull as a way to get a polyline per curve
so as not to have to deal with the spline types directly
'''
Added a new script - 'Dump All Images to Path'
This script copies all the images used by 1 blend to a spesified path and references the new images from Blender
Usefull for moving projects between computers or when you reference many images. naming collisions and multiple images using the same image path are delt with properly only creating new image names when needed.
Blender images will reference the newly copied files - So be mindfull when you save your blend after running the script.
Notes, images with the path "Untitled will be ignored"
Image path collisions are managed by enumerating the path names so images will never be overwritten at the target path.
an error in Scene.c - scn.Layers disallowd all layer bits to be set.
made image_billboard.py rotate all images to be verticle for more efficient packing, added the option not to pack resulting images into 1.
________________________
Render Billboard Script
This can texture a simple billboard mesh from any number of selected objects.
Renders objects in the selection to quad faces on the active mesh.
Usage
* Light your model or enable the shadless flag so it is visible
* Make a low poly mesh out of quads with 90d corners. (this will be you billboard mesh)
* Select the model and any lamps that light it
* Select the billboard mesh so that it is active
* Run this script, Adjust settings such as image size or oversampling.
* Select a place to save the PNG image.
* Once the script has finished running return to the 3d view by pressing Shift+F5
* To see the newly applied textures change the drawtype to 'Textured Solid'
uv_archimap still had python based line intersect
added plane2matrix function to BPyMathutils
added an optional arg to imageFromObjectsOrtho - camera_matrix
camera_matrix can be used to define a plane in 3d space where X and Y scale is used to set the width and height of the area to render.
BPyRender.imageFromObjectsOrtho() returns an image
Made all image bake scripts ask before overwriting a file. as well as displaying the newly created image once its rendered.
import_obj - comments, docstring additions, cleanup.
OBJ export: -renamed export_obj
faster edge exporting (use edges LOOSE flag rather then finding zero face user edges)
check for file before overwriting
use object iterators
http://mediawiki.blender.org/index.php/ScriptDev/Guidelines
* Recursive search was enabled by default, if your OBJ happened to be at root path, loading the OBJ could take ages. Made recursive searching optional.
* MAX's MTL exporter replaces spaces with underscores, image loading checks for this and loads the name with spaces if the one with underscores dosnt exist. not realy a problem with the importer but annoying to manually replace lots of images.
locations are: smart, middle, v1, and v2
where smart is an attempt to generate a loc that will collapse without loosing volume.
This can now collapse a subdivided cube properly.
