Taking a look at the built in windmove functions shows that Emacs has a
built in function for swapping windows called `window-swap-states`.
Using this ensures the window state is maintained.
Ref: #6023
Co-authored-by: luveti <luveti@users.noreply.github.com>
evil-mc's design is bizarre. Its variables and hooks are lazy loaded
rather than declared at top-level, some hooks aren't defined or
documented, it's a bit initializer-function drunk, and its minor modes
are intended to be perpetually active -- even when no cursors are active
(causing #6021). I undo all of that here.
Fix: #6021
Also limits the scope of org-display-inline-images to the current src
block. org-display-inline-images was previously used on the entire
subtree, which was slower than it needed to be (especially while
exporting/tangling org).
The C-c C-s binding mirrors the export and writable export bindings.
embark-export is used for type specific exports (such as buffers to an
ibuffer buffer), not for verbatim exports. This is most noticeable if
you want to export a variable list, which gets you to an apropos buffer.
if you just want a list of things to keep around,
embark-collect-snapshot is the tool for the job, but having it be only
on C-; S makes it harder to reach and less visible. C-c C-s is a nice
visible thing to have around.
BREAKING CHANGE: This commit changes the behaviour of the TAB and RET
keys in a consult-completing read multiple session, in order to make
them more intuitive. The behaviour is now:
- TAB: (unchanged) always select or deselect the current candidate, and
if the candidate is selected, move the index to the next one (this
allows for pressing TAB repeatedly to select multiple subsequent
candidates).
- RET: If no candidates have been selected, select the current candidate
and exit the completion session. If some have been selected, disregard
the current candidate and exit.
- S-TAB: (new) like TAB, but the keeps the input.
lsp-mode doesn't detect the correct indentation, since it treats the
derived tsx-mode as a web-mode. This change modifies the lsp-mode
internal alist to recognize tsx-mode correctly. Ideally lsp-mode would
have a better way to modify its known mode lists but I couldn't see it.
Ref: emacs-lsp/lsp-mode#3308
A mistake that snuck into 55c1cac while backporting some code from the
CLI rewrite. It was harmless and likely had no side effects, however.
Amend: 55c1cac43f
- Add @angular/core in detecting angular to set the web-angularjs minor
mode
- At the same time set the web-mode-engine to "angular" so that
`format-all` uses prettier to format ionic `*.page.html` files in
addition to `*.component.html` files.
When `format-all` sees the web-mode-engine is set to "angular", it
applies `prettier` as opposed to `html-tidy`, as `tidy` doesn't
recognise angular component tags and hence errors out. The current
web-mode only sets the engine to "angular" for "*.components.html".
However, ionic angular uses file names such as "*.page.html", setting
engine will apply prettier to these html fils.
set-{eval,repl}-handler! operate on major modes, not minor modes (this
should be corrected upstream, but I'll do so later). This doesn't apply
to set-lookup-handlers!, but I'm changing it too for consistency.
BREAKING CHANGE: Remove `company-erlang` which depends on
`ivy-erlang-complete`. `company-erlang` and `ivy-erlang-complte` have
not been maintained now. The lsp mode provides much better experience
than `company-erlang` and `ivy-erlang-complete`. The
`ivy-erlang-complete` package also has the defect that slow down the
erlang-mode buffer when the buffer is huge enough. It is not a good
choice to integrate ivy-erlang-complete nowadays even as a non-LSP
option.
code-review is built on magit-sections so it will have the same default
bindings for M-1, M-2 etc. To be consistent with the rest of doom, those
should be used for switching workspace and z1, z2 etc should be used for
toggling outlines in a magit-sections buffer.
This only affects users with :editor (evil +everywhere) enabled.
Amend: 2d3a68df49
