Undo-boundary in emacs

[Narendra Joshi]

Hi,

I have been trying to figure out where in the source of emacs this is happening:
https://emacs.stackexchange.com/questions/17578/why-does-modification-in-another-buffer-create-undo-boundary-in-current-buffer

I am trying to understand what were the design goals that caused such
things to happen?

Why does change in a buffer cause every other buffer to have a
undo-boundary? What would happen if this wasn't the case?

Best,
Narendra Joshi

[Phillip Lord]

This doesn't happen in Emacs-25, largely to address the use case you are
describing.

It was caused by this code in undo.c

if (current_buffer != last_undo_buffer)
Fundo_boundary ();
last_undo_buffer = current_buffer;

Most of the code handling automatic undo boundaries is now in simple.el.

Phil