Have you made any progress on this? I have been working on a similar process to trim out xclip-ed blocks and xrefs. So far I found that i can get X-clip defenition data from the SpacialFilter object attached to the extended data of any given blockreference. I am still looking for a way to make the filter indexer, which is responsible for returning the cliped entities, to work for me.
I don't think anybody has done this in anything.net yet, so if you have, please help.
I'm working on something pretty similar (a clipbind command - erase non-visible objects in xrefs to keep bound file size down).. right now im getting the points via the spatialfilterdefinition, transforming the points from the definition.getpoints to real modelspace points using the following transforms:
So far I can draw plines in modelspace for each clipped block, which is a decent starting point. Where it gets tricky for me is that sometimes i have the same xref copied around, but each one is clipped differently. So i have to duplicate/clone the block definition multiple times so i can map the blockreferences to a distinct block definition, then modify the definition to erase the non-visible entities. Or something. This is my first .net project and its making my brain hurt.