I rebased the patch to 6766:e81d2bd66908 and attached as
working-dir-copies3.patch.txt (mostly trivial changes to the test output; kept
Alexis' name on it).  It passes the test suite on solaris, although the convert
tests were skipped because of missing sfw.

So it works for me, but I haven't tested it extensively.  Is there anything
that's known to be missing or incorrect?  I don't understand dirstate well
enough to make any judgement.  Would more testing help?

Mostly fixed, degrading to bug?

I'm attaching what I currently have.

While writing a test for this, I noticed that that patch is not enough, since
it doesn't update the dirstate, and this is turning out to be quite a bit
more tricky than I expected.

I've pulled this into main as it fixes another problem I ran into.

Patch looks fine.

The patch works for the simple test cases I have; thanks.

The problem resulted in a lot of tedium when trying to reorg many files in a repo,
and I have more rounds of this kind of thing planned.  So I'm going to bump the
priority; adjust if you disagree.

As mentioned in msg5349 in issue988, "hg update" doesn't take into account
uncommitted copies.  To reproduce:

----------------
hg init repo
cd repo

touch foo
hg ci -Am 'add foo'

echo >> foo
hg ci -m 'change foo'

hg up -C 0
hg mv foo bar
HGMERGE=false thg --debug -y update
--------------------------

A similar problem happens with hg merge --force.

I'm attaching a possible patch.