7890yuiop

I feel like whenever I am browsing emacs content online there are still many topics for me to discover.

This will never not be the case.

My suggestions are:

• focus on addressing known pain points
• if something seems very popular, it might be worth trying to learn why and whether it could be beneficial to you as well
• if something else sounds super-interesting, take a look
• don't fret about not keeping abreast of everything -- it's almost certainly impractical

There is a thing which provides zsh completions. It's zsh. If that's what you're wanting, then maybe don't stop using it?

And in case the video doesn't say it:

Ediff has its own user manual: C-h i g (ediff)

Have you considered an org-capture template fitting your desired format?

I don't use use-package, but I've seen a lot of questions from users who do use it but don't understand how to use it, or what it's going to expand to, or what the things that it expands to actually do. My conclusion has been that for some users it introduces as many problems as it solves. I think those users would be better off if they learned how to manage their config without it first, and only considered use-package after understanding the more fundamental building blocks upon which it is built.

It's certainly not something you need to use, in any case. It's clearly an invaluable system to many users, but if you don't get along with it, don't use it.

Awesome; thank you.

The author of the emacs pgtk code says that no one who has X installed should use pgtk -- he's stated on several occasions that if you have X at all then you should use a supported X toolkit in Emacs for best results.

I've seen counter-arguments that pgtk is still beneficial if you happen to have a "high DPI display", but I believe that's the only argument I've ever seen for using pgtk under X.

I always build --with-x-toolkit=lucid myself, and can happily vouch for that one. I don't use Wayland, though.

Looks like a bug (whether documentation or code) as read-kbd-macro still claims to return a string if possible, but nowadays it forcibly returns a vector. Please M-x report-emacs-bug to get that clarified.

You could extract a string of characters from the vector:

(mapconcat (lambda (event)
             (and (characterp event)
                  (char-to-string event)))
           (read-kbd-macro "C-c"))

But if you look at the code for read-kbd-macro you'll see that it calls this:

(defun edmacro-parse-keys (string &optional _need-vector)
  (let ((result (kbd string)))
    (if (stringp result)
        (seq-into result 'vector)
      result)))

Hence the string value you wanted is coming from kbd:

(kbd "C-c") => "^C"

There are of course arguments you can pass to kbd which won't return a string, but that would always have been the case for your code, and presumably you're not attempting to use any of those.

https://www.youtube.com/watch?v=UtSu3ydxKJ4&t=20s

Is there a question here?

Are you asking about configure --prefix=... when building?

A couple of simple config suggestions:

(winner-mode 1) ;; Get familiar with `winner-undo'.
(global-set-key (kbd "C-x !") 'delete-other-windows-vertically)

I get a lot of mileage out of those two.

Also consider adding bindings for these, to tell the following command where to display its buffer:

• windmove-display-left
• windmove-display-right
• windmove-display-up
• windmove-display-down
• windmove-display-same-window

And the regular windmove commands are very convenient for moving between buffers.

The transpose-frame package available in MELPA is useful. Note that with only two windows, rotate-frame is "exchange the positions of the windows", and transpose-frame is "switch between vertical and horizontal split". I have those bound to C-c | and C-c \ respectively, and use them regularly.

You don't have to declare a new variable each time you want to remember an old value. You could, e.g., put a custom symbol property on the variable symbol with the old value (or list of old values) you wanted to store. Or maintain a single variable mapping all your variable symbols to their old values.