KeepITcool,

I didnt see where your 'Jumbling Protest' came from,
since I'm only testing with a PURE invert (sans RngB)
and haven't tested on a 'Complement' style inversion.

Hardly a protest!
At most it's a trivial issue. Indeed for you it probably
won't arise as you are only going for the "Invert", the
title of this thread. However this code has bigger
possibilities - potentially will work well for subtracting
any intersecting ranges where either or both are
multiarea.

The basic principle has been discussed several times in
this NG (notably Dana DeLouis), but I think this is
working towards the best overall implementation, including
a lot of stuff I've done myself (not posted). Downside -
I'm going to have to re-work of my old code (:

So, I'm using avec un multi RngB - and getting un mélange
(with the Collection but not Array). I think using the
square will clear this up.

I'm currently having some problems with latest amendment
2707b and rSqu, Need to ensure all DV is collected from
the square range, then entirely deleted in the Square, do
stuff, then DV restored to the Square. Probably me missing
something obvious.


Following HAS happened:
due to editing or while testing I apparently
damaged/copied not removed
some dummy validation

Funny you should mention that. I've experienced similar
but ignored. Here's something else:

I have a recorded macro to replace identical DV to a
single area over everything before testing. But
occasionally it errors and I need to run the
line ".Delete" (DV) twice. I've got a feeling similar has
occurred in proper code without my knowing. I'm only
testing with small ranges - visible on the screen,
shouldn't need DoEvents.

If you can re-produce your "HAS", try adding a
second .Delete line before applying any DV.

Regards,
Peter