header, directory, shadow status

[Benjamin Kosnik <bkoz at redhat dot com>]

Hey Y'all.

I think the directory shuffling that was a major part of last week's
episode is now officially completed. There are still some kinks to
work out, but for the most part things are back up and working.

Ok. 

Well, --enable-cshadow-header builds are completely not working.
Non-shadowed builds should work though. Can somebody else confirm this
please? 

As for the shadow header work: I can't spend any more time on it. I
had it working on 10-04, and I would think that it could survive a
directory rename, but apparently this is not the case. 

The ./include/c_std directory is supposed to re-create the ./shadow
directory. Currently, their contents are not identical: I'm thinking
of doing a check in to achieve equilibrium, and then revisiting
the situation when other, more pressing issues (solaris/bsd builds in
particular) are taken care of.

By then, perhaps some of the necessary compiler infrastructure will
also be in place. It's really kind of an exercise in frustration for
active work on this to continue while builtin types are placed in the
wrong namespace. Then again, perhaps I'm just venting.  

Thoughts?

-benjamin