Define a new custom layout named "@Notmuch" with binding "n" for notmuch
buffers. The custom layout's name can be customized using the new
notmuch-spacemacs-layout-name variable. The custom layout's binding can
be customized using the new notmuch-spacemacs-layout-binding variable.
A fix for #10091. The default behavour for which-key is for
'which-key-idle-delay' to affect both the initial trigger and subsequent
actions. By setting 'which-key-idle-secondary-delay' to something non-nil,
that delay acts over the subsequent actions instead.
The which-key wiki recommends setting it to a 'non-zero value', as zero could
cause issues, so instead set it to 0.01.
- Make the notmuch layer more idiomatic.
- Move org related stuff to a use-package hook for org.
- Simplify configuration of helm and ivy
- Sort package config
- Move sections in README to avoid repetition of Evilified State sections
- Add docstrings to functions and use // convention for private function names
abo-abo/swiper@4f57b5a9 removed the counsel--git-dir variable. Code
that used that variable is now expected to use (ivy-state-directory
ivy-last) instead. Update the code to reflect this change.
Signed-off-by: Chirantan Ekbote <chirantan@chromium.org>
When a layer doesn't have a package, its name was added to the candidate list as
a symbol instead of a string, breaking actions such as
layer-action-open-packages which expect a string.
When ivy is used, the default org-set-tags will only complete for the
first tag, and any additional tags must be typed manually. counsel
provides a replacement function to fix this.
Unfortunately, we can't just alias org-set-tags because it is used in
other places, so we must change C-c C-c using a hook.
See https://github.com/syl20bnr/spacemacs/issues/6918
Like is has been done for evil-cleverparens and lisp dialects.
It means that the mode-line lighter is now a plain squared G and that the
toggle key bindings is now under `SPC , T g` instead of the global `SPC t G`.
- Move rtags key bindings under `SPC m g` prefix
- Document rtags support and key bindings in README.org
- attempt to implement gtags fallback using
https://github.com/Andersbakken/rtags#fall-back-to-other-taggers
- make c-c++ layer dependend on gtags layer (gtags layer is automatically
installed when the c-c++ layer is used)
- sort packages config in packages.el
- use a dolist instead of a function to define rtags key bindings
Resolve#10041
The spacemacs/clang-format-buffer function doesn't exist. It should be replaced
with clang-format-buffer. This error was introduced in 01ddc98cef.
Problem
-------
If the symbol is not resolved due to a missing import there is no automatic way
in the Python layer to import it.
Solution
--------
Use 'importmagic' package.
Pressing `SPC m r f` at point that requires an import will suggest a list of
possible options.
Delete layer evil-cleverparens and move the package to spacemacs-evil layer.
The feature is called "Safe structurral editing" for lisp dialects. Support for
it is added via pre-init functions in each of the concerned layer and proper
documentation is added to their README.org files.
This also removes the recently added package evil-smartparens. The goal is
to choose the best package for evil safe structural editing. For now we use
evil-cleverparens as we supported it first, if evil-smartparens is shown to be
a better package we will be able to switch to it.
Convention for major mode specific toggles is to put them under SPC m T prefix.
Used `n` instead of `>` because there is no unicode squared symbol with `>`.
Also use the new generated function to register the hooks.
This is more dwim in the sense that the normal deleting commands now behave more
like paredit, i.e. region deletion is adjusted to always keep the pairs
balanced.
This change makes the c-c++ layer configure company-c-headers to search for and
list completion alternatives according to the priority list found in 'man gcc'.
If supported, the default system include paths are also fetched from gcc's
configuration instead of being hard coded to "/usr/include" and
"/usr/local/include". This also remedies the problem where the C++ standard
library headers does not show up as completion alternatives without manual
addition of include paths (see #4493, #8655).
Solve part of #7628. Additionally, #7629 is no longer needed.
Add support for formatting lines, functions, and buffers with clang-format. Each
function produces a clear message on success.
The README was updated to reflect the usage of the new key bindings. References
to binding clang-format to the tab key were removed.
- Change nameless prefix to > instead of default :
- Change toggle to SPC m >
- Set nameless separator to nil in order to make it work with any type or separator
- Add diminish unicode ⧁
- Define some default global aliases for Spacemacs source code
- Rename layer variable emacs-lisp-nameless-mode to emacs-lisp-hide-namespace-prefix
- Make variable nameless-current-name safe as a local variable for string values
- Set nameless prefix for core-configuration-layer.el
Fix issue #9981: rcirc /reconnect command reverts to default port.
rcirc has included a built-in /reconnect command since Emacs 25.1, which
makes the rcirc-reconnect package shipped with the Spacemacs rcirc layer
superfluous.
Modify the rcirc layer's initialization not to load rcirc-reconnect, and
delete rcirc-reconnect from the layer's local packages.
The `helm-split-window-in-side-p` variable has been renamed to:
`helm-split-window-inside-p` here: 8b5df59a14
This caused the following variable `helm-always-two-windows` to become enabled.
This patch adds a notmuch[1] layer to Spacemacs!
notmuch is a mail indexing system written in C. It provides an emacs
interface, and this patch adds Spacemacs support for this interface.
Thanks to @FrancescElies for starting this work, and working with me to
get it finished. We have been using the layer for awhile now and are
happy with the initial release!
Closes#2163
[1] https://notmuchmail.org/
Tested-by: Francesc Elies <elies@posteo.net>
Tested-by: William Casarin <jb55@jb55.com>
Signed-off-by: William Casarin <jb55@jb55.com>