Re: Window splitting deactivates the mark?

[Richard M. Stallman]

I don't think window splitting does anything to deactivate the mark.
There is nothing in Fsplit_window that would do it.  Empirically, if I
set the mark (with Transient Mark mode) and then split the window, the
region remains active.

So this must be due, somehow, to whatever is done in tabbar.el.

Remember that deactivating the mark is normally done by setting
deactivate-mark to t.  When the command loop sees that, it deactivates
the mark.  It should be pretty easy to determine whether something
sets deactivate-mark to t.

split-window can run window-scroll-functions and it can run
window-configuration-change-hook.  If either of those does
editing in a buffer, it would set deactivate-mark to t.