At some point, either Emacs, persp-mode, or I fixed what caused #319,
which swallows the *Warnings* popup when the new main workspace is
created. Until I can determine what fixed it (so I can determine whether
the whole hack should go), I'll simply add this guard.
Ref: #319
`centaur-tabs-buffer-update-groups` walks every buffer to determine the
group it belongs to. This function can be called execessively (upwards
of 1,000 times per second in certain conditions), as it is called on
every redisplay (it is indirectly attached to `tab-line-format`). This
rate limits its calls to about 10 per second.
Close: #7792
Ref: ema2159/centaur-tabs#222
Although it is rare for a user to not enable the :config default module,
enabling/disabling modules one-by-one is common practice for debugging
Doom Emacs. This PR fixes an error that occurs when you have :completion
vertico enabled without enabling :config default.
Although this error will not be triggered by most people, since
auth-source is loaded by a lot of packages, it can still happen if you
are debugging your configuration (e.g. enabling/disabling modules
one-by-one).
This ensures all these fuel keybinds on factor-mode-map (and the
fuel-based lookup handlers) are associated with the fuel-mode
package (in case the user disables the fuel package).
Fix: #7788
Co-authored-by: steve-ayerhart <steve-ayerhart@users.noreply.github.com>
Otherwise `+eval/open-repl-*` commands will emit "REPL handler X failed
to return a buffer" errors.
Fix: #7788
Co-authored-by: steve-ayerhart <steve-ayerhart@users.noreply.github.com>
- While seq.el is likely present, Doom does not guarantee it in
interactive sesions like it guarantees cl-lib's presence.
- There is no reason to ever use cl-first/cl-second.
The caching of rust-analyzer (RA) is too aggressive and leads to bad
results.
RA does not give a protocol-compliant way to extract 1 line information
from the hover response, so `lsp-mode` displays the first line of the
result, which currently is the byte alignment of the data
Fix: #6951Close: #7754
Co-authored-by: gagbo <gagbo@users.noreply.github.com>
Co-authored-by: scturtle <scturtle@users.noreply.github.com>
mu4e-clear-caches was removed in djcb/mu@f73aad2b41, so users on
1.12.2 or so will see void-function errors until djcb/mu@770a2396de is
propagated to distro package managers, which likely won't happen
soon (esp for LTS distros).
Fix: #7703
Ref: djcb/mu@f73aad2b41
Ref: djcb/mu@770a2396de
Co-authored-by: dltacube <dltacube@users.noreply.github.com>
A common challenge for beginners is distinguishing packages, modes, and
hooks, and since Doom's source code is meant to second as another layer
of documentation, I prefer to be explicit with hook symbols in
`add-hook!` calls (at least, for internal/module use).
Having the same rule in two separate package configs means users must be
aware of both to modify either, which is avoidable complexity. i.e.
(after! (:or julia-repl julia-snail)
(set-popup-rule! "^\\*julia" ...))
I backported `find-sibling-file` in 198fe82 and included with it some
rudimentary `find-sibling-rules` rules for CSS and its various
preprocessor languages. The CSS rule made *.((s[ac]|le)ss|styl) files
the siblings of *.css files, but not vice versa, so users couldn't jump
back to the source file with `find-sibling-file`; a second rule is
necessary.
These may be combinable, considering `find-sibling-file` deletes the
current buffer's filename from the list of candidates...
Ref: 198fe82b6d
I backported `find-sibling-file` in 198fe82 and included with it some
rudimentary `find-sibling-rules` rules for C++ and others. The C++ rule
made *.h(h|pp) files the siblings of *.c(c|pp) files, but not vice
versa, so users couldn't jump back to the source file with
`find-sibling-file`; a second rule is necessary.
These may be combinable, considering `find-sibling-file` deletes the
current buffer's filename from the list of candidates...
Fix: #7795
Ref: 198fe82b6d
Now that mu4e has its own notification feature since 1.10 (and as of
this commit version 1.12 is already out), we can just use that. Much
less code to maintain.
Closes: https://github.com/doomemacs/doomemacs/issues/6896
- Make `+fold/close-all` work for outline-minor-mode.
- Distinguish between `zo` and `zO` for outline-minor-mode alone, by
binding `zO` to a new function `+fold/open-rec`. `zO` will now trigger
`outline-show-subtree` in outline-minor-mode, while `zo` will only
trigger `outline-show-branches`, keeping the subheadings folded. This
matches the behavior in org-mode.
- Fix a bug in `+fold/open-all`, introduced in
7b9d00d661, that makes it only operate
on vimish-fold folds.
Note that outline-minor-mode is still not enabled by default anywhere.
These changes are only to prevent incorrect behavior for users who
decide to turn it on themselves.
This commit updates the smart tab functionality so that:
1. The only functionality checked is for the modules that are enabled.
2. The priority of the TAB behavior is tunable by the user.
This also updates the TAB behavior for the Corfu module to be
`indent-for-tab-command` instead of `completion-at-point` so that users
can use the TAB key to indent their code and navigating Org tables.
We also address #7372 by checking overriding-terminal-local-map, as that
is used by Embark.
At 0.1, Corfu pops up too aggressively and too often, even for fast
typists. This is especially disruptive with slow backends that might
block Emacs. I think 0.18-0.24 is the goldilocks zone, so I'll start
with 0.18, and maybe increase it later.
This removes the truncation of `package!` `:pin`s. This was originally
intended to make packages.el files easier to skim, but in hindsight it
didn't really. It served little other purpose but to make it harder for
folks to interact with the :pin string.
Due to upstream changes in ivy-rich, +workspace/switch-to throws a
`wrong-type-argument listp leaf` error. As I plan to phase out Ivy
support (and the Ivy module) in the long term, I'll simply remove
ivy (and ivy-rich) integration in the workspaces module, rather than
update it.
Fix: #7499Fix: #7173
