LGEO/LDBoxer folder structure

[Michael Horvath]

Currently, all LGEO parts (except Anton Raves and Darats STL parts) are stored in "lgeo/lg". So there's no need to worry about sub-parts or primitives located inside sub-directories. However, when one is utilizing the LDBoxer library, boxed parts are stored in the "ldraw/parts/b", "ldraw/parts/s" and "ldraw/parts/s/b" directories. Here are some examples from one of my models:

Code:

1 0 -170 -224 -90 0 0 1 0 1 0 -1 0 0 b\a2339.dat
1 0 170 -224 -90 0 0 -1 0 1 0 1 0 0 b\a2339.dat
1 0 -190 -128 100 0 0 -1 0 1 0 1 0 0 b\a3004.dat
1 0 -190 -104 100 0 0 -1 0 1 0 1 0 0 b\a3004.dat
1 0 -190 -80 100 0 0 -1 0 1 0 1 0 0 b\a3004.dat
1 0 -190 -56 100 0 0 -1 0 1 0 1 0 0 b\a3004.dat
1 0 190 -128 100 0 0 -1 0 1 0 1 0 0 b\a3004.dat
1 0 190 -104 100 0 0 -1 0 1 0 1 0 0 b\a3004.dat
1 0 190 -80 100 0 0 -1 0 1 0 1 0 0 b\a3004.dat
1 0 190 -56 100 0 0 -1 0 1 0 1 0 0 b\a3004.dat
1 0 190 -200 100 0 0 -1 0 1 0 1 0 0 b\a3004.dat
1 0 -190 -200 100 0 0 -1 0 1 0 1 0 0 b\a3004.dat
1 0 -190 -224 100 0 0 -1 0 1 0 1 0 0 b\a3010.dat
1 0 190 -224 100 0 0 -1 0 1 0 1 0 0 b\a3010.dat

Each of these boxed parts is located in the "ldraw/parts/b" directory.

What would be the best way to deal with these sub-directories within files like "lg_elements.lst" and "LGEO.xml"? I'm not sure what the best solution is, especially given the number of programs that rely on "lg_elements.lst" and/or "LGEO.xml" to behave in a certain way, and would likely need to be reprogrammed as a result. Does anyone have any suggestions? Thanks.