Closed Jake-Moss closed 3 years ago
The issue was most likely introduced by 6dcafc4 at this line.
I honestly don't understand why this is happening on your box. Are you not using evil?
If you look into the code that introduced the error:
(when org-ol-tree-packages--evil-p
(evil-define-key '(normal) org-ol-tree-mode-map
"h" #'org-ol-tree-navigation--collapse-current
"l" #'org-ol-tree-navigation--expand-current
(kbd "<left>") #'org-ol-tree-navigation--collapse-current
(kbd "<right>") #'org-ol-tree-navigation--expand-current
))
You can see that I only use the evil-define-key
if org-ol-tree-packages--evil-p
is not nil
. And here is how I populate org-ol-tree-packages--evil-p
:
(defconst org-ol-tree-packages--evil-p (require 'evil-core nil 'noerror)
"Constant indicating if package evil is installed.")
As far as I understood from the lisp documentation, when calling require
with 'noerror
, if Emacs can't load the requested feature, it returns nil
. So, in theory, the use of evil-define-key
should only happen if evil-core
is loaded (which is the feature that defines evil-define-key
).
I'll try to use fboundp
on the check to see if this gets any better on the next patch.
This is working now. I do use evil so I have no idea what was happening either
After updating the package using
doom sync
org-ol-tree/display-sections raises(invalid-function evil-define-key)
.Unfortunately I couldn't find the specific commit that introduced. A backtrace is bellow