Bug in compute wall/gran/local (Part -II)

There is another bug in "compute wall/gran/local". The number of contacts reported seems to change if you split the mesh into several .stl files instead of just one .stl file. I again attach a simple example which is exactly the same as the one attached before under "Bug in compute wall/gran/local". The mesh consists of three walls, bottom, left and right. There are two options, to use a single .stl file (mesh.stl) which contains all three walls. If you use that option, the reported number of contacts is 3, which is wrong since there should be 5 points of contact, one with bottom and two each with left and right wall. The other option is to split the mesh into three stl files, one for bottom, one for right wall and one for left wall, if you use that option, the number of contacts is correctly reported as 5, but the mesh id's are wrong, as discussed in the original post.

The commands for both options are in the Bug.in file, just comment/uncomment the lines as necessary.

Thanks,
Tapan