[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: A new user perspective about "Changes for emacs 28"

From: Jack Kamm
Subject: Re: A new user perspective about "Changes for emacs 28"
Date: Thu, 10 Sep 2020 00:23:17 -0700

> And try to consider how the visualization would look. I think it would 
> have to be like a tree (basically, like the one undo-tree paints). Then 
> it would probably be logical to have at least two different commands 
> which allow traversing the buffer states in both directions along the 
> branches of that tree. And presto, we got undo-tree again.

undo-tree is great. I used to think it was superior to native undo, but
had to disable it due to the corruption bugs.

Recently, I started using selective undo (i.e. undo in region). In my
opinion, it's a killer feature of native undo, and even more useful than
the undo-tree -- in particular, I think it's usually quicker to find an
old edit with selective-undo, than it is with undo-tree.

undo-tree can use selective undo, but in the past it was considered
buggy, and the common recommendation was to disable it [1]. I'm not sure
if that's still the case, but if you think about it, selective undo
violates the tree structure of the undo history, so naturally seems a
poor fit with undo-tree.

Using selective-undo, revert-buffer, undo-only, and undo-redo-only, it
becomes a lot easier to navigate emacs' undo. But these features are not
immediately obvious; I had to read the Info page to learn about
them. Whereas, undo-tree is quickly grokked due to the nice

[1] https://emacs.stackexchange.com/a/37399

reply via email to

[Prev in Thread] Current Thread [Next in Thread]