bug#23906: 25.0.95; Undo boundary after process output is not consistent

[Stefan Monnier]

> Yep, but Stefan did say how to implement this! We could do the same
> thing as viper, which works, but has the issues he pointed out. Off
> hand, I do not see an easy solution.

I think Viper's approach isn't fundamentally wrong.  The use of an
invalid element is a problem, but that shouldn't be hard to fix.

> At the moment, I am interested in fixing the regressions that I've
> caused by my changes to the undo system which I've nearly done -- you
> have a workaround at least, and I'll put something easier onto master.

I think what you consider as a regression is not worth fixing.
It just made an earlier problem slightly more visible.  Let's focus on
fixing the more general problem.

I'll see when I find time to devote to undo-begin/start-transaction.


        Stefan