It only uses the face's :foreground, which should be enough, but it
prevents the minibuffer from being resized when eldoc displays
breadcrumbs in org-mode.
+ Changed org-export-directory to +org-export-dir (conform to naming
convention). It turns out org-export-directory never existed in org.
+ Make org-export-backends addition (for ox-pandoc) idempotent.
+ Fix redundant forward slash in org-publish-timestamp-directory.
+ Resolve export directory later, giving the user a larger window to
change +org-export-dir.
evil-org changes the behavior of o/O to create new headlines, plain list
items or table rows. I disable its new behavior in plain lists only.
This was done because:
1. It isn't uncommon to want o/O's default behavior in plain list bodies
of text. Unlike tables, where a new line in the middle of a table
doesn't make much sense.
2. M-RET/S-M-RET exists.
At some point, org changed how soon it popped up this window, so
+popup-shrink-to-fit would run before there was any content in the
buffer, causing it to take up way too much space.
elisp-mode is loaded at startup, so the usual methods won't work.
Instead, we tie a transient advice to the emacs-lisp-mode
function, *however*, this function is commonly called by various
packages to parse elisp code! So we have to make sure the emacs lisp
module only initializes the first time it is used interactively.
+ It's too much trouble supporting the evil-org-set-key-theme workflow.
Perhaps I'll make it complain when you do.
+ Don't add +org|setup-ui to doom-load-theme-hook, it's unnecsesary.
+ Use faces in org-priority-faces rather than colors.
Caused when evil-org-set-key-theme is called too early (somehow).
Also makes evil-org-key-theme customizable, so it doesn't override a
user's changes to it.
Reported by @majorgreys
`set-repl-handler!` helps with opening a repl when a particular mode is
active in a buffer. We want to be able to open a `sly-mrepl` whenever we
are in a lisp buffer, so we should have the repl handler look for
`'lisp-mode` to define opening a sly repl.
I assume the lookup handlers should be making sure we are in a lisp
buffer (similar reasoning to the repl-handler).
This prevents the unnecessary eager-loading of many autodefs (and
evil-collection-elisp-mode), since the elisp-mode package is always
available at startup.
+ Enables web-mode's autopairing functionality
+ Prevents duplicate >'s after web-mode autopairing.
+ Leave longer-than-3-character pairs to smartparens (there was only
<!-- -->)
+ Remove <?p -> <?php | ?> autopairer. Leave that to snippets (web-mode
doesn't support prefixes longer than 3 letters, and <?php ?> is too
much of a performance drain to have smartparens support it, so I added
a "php" snippet that expands to <?php | ?>).
+ Tags now auto-close when you finish typing the opening
tag. (web-mode-auto-close-style = 2)
After some profiling, it turns out map-put and map-delete are 5-7x
slower (more on Emacs 25) than delq, setf/alist-get and add-to-list for
small lists (under 250 items), which is exactly how I've been using
them.
The only caveat is alist-get's signature is different on Emacs 25, thus
a polyfill is necessary in core-lib.
str => clojure.string is already defined in `cljr-magic-require-namespaces'
re-frame + reagent is subjective (for personal configuration)
fighweel aspect is now handled upstream by giving a prompt
see: `cider-cljs-repl-types'
