problem:
The keyboard layouts main image, is displayed as a black rectangle, when the
readme.org file, it's opened in Spacemacs. This seems to be because there's an
issue with the transparency when a .png is saved with indexed colors.
When the transparency issue is fixed, by re-saving it as a non-indexed
(grayscale or RGB) image. Then there still is an issue, with themes that have a
dark background, i.e. spacemacs-dark. The black keys become almost invisible.
solution:
Add a 1px gray border, around the keys.
Looks like pass is trying to use variable which isn't available for some time,
at least there are no other packages using it.
`completing-read` should be universal for all completing needs I hope.
Old version worked only for method 'ssh' and did not work at all if
user was not specified (e.g. /ssh:hostname:/path). Since the symbol '@' was
required in a filename.
Commands and key bindings found via `alchemist-refcard` command.
I've tried to adhere to Spacemacs conventions but where I couldn't find relevant
conventions, I followed Alchemist's key bindings.
Also, delete alchemist-refcard - approved by @syl20bnr at
https://github.com/syl20bnr/spacemacs/pull/8746#issuecomment-301326850
If dotspacemacs-auto-generate-layout-names is non-nil, and the user
tries to open a layout in a position that doesn't yet have a layout,
then create a new layout with an automatically generated name.
Problem:
The current rainbow-mode screenshot, is from an older version. It shows white
text on light colors. This makes it hard to read the color codes. The old
screenshot hides the fact that issue has been solved.
Solution:
This updates the screenshot, to show it's current behaviour. Where the text
color is either white or black, depending on if the luminance is greater than
0.5 of 1.0.
org-mu4e was already handled in the mu4e layer which is the correct place for
this.
Added notmuch to mu4e layer, maybe not the right place for this but for now
it is OK.
* Make RET and TAB (C-m and C-i) perform the same actions as <return>
and <tab> so the maps work in the terminal.
* Catch <return> in workspaces transient state so that it exits without
executing a command (like already done for layouts).
* Add "workspace w/helm/ivy" to docstring, mirroring the layouts
docstring.
* layers/+lang/python/funcs.el (spacemacs/anaconda-view-forward-and-push):
Added helper function to go forward and hit RET
* layers/+lang/python/packages.el (python/init-anaconda-mode):
- use new evilification macro
- use helper function to go ahead and push button when hitting `RET`
- Add `C-j`, `C-k` for easy navigation in anaconda-view-mode when multiple
references are displayed. Doesn't conflict with normal `hjkl` navigation in
other views of `anaconda-view-mode`
Fix#7538Fix#5737
Problem:
Spacemacs rewrites the evil-mc-mode-line variable, to remove the evil-mc mode
line text, when there's only one cursor. The rewrite blocks the recent evil-mc
updates, that made the multi cursor, mode line text, more readable.
Solution:
Use the new evil-mc-one-cursor-show-mode-line-text variable, to only hide the
mode line text, when there's one cursor. This unblocks the current, and any
future updates, that the evil-mc package makes to the mode line text.
Currently the ocaml layer requires opam in order to work. However
opam is only really required for an easy installation of the
required ocaml binaries. The only place where opam is used
instead is to add the merlin files of the merlin installed
via opam instead of the merlin files from melp.
This commit introduces a fallback if opam is not installed but
ocamlmerlin is installed, since the melpa package of merlin
is already installed through packages.
SQL, by convention, uses upper-case keywords, although lower-case works just as
well. As humans, the separation between upper-case and lower-case helps scan and
parse the code much more quickly.
Clojure layer attempted to provide `C-j` and `C-k` keybindings
to the cider-repl-mode but there was a bug.
This fixes the bug and adds those keybindings to the documentation.
Mark up code that is mentioned in the documentation of some readmes.
In the case of “long options”, like for example `--with-emacs`, this is not just
cosmetic. On GitHub, Org files are apparently rendered in such a way that
strings like `--` in non-verbatim text (i.e. not verbatim-quoted nor
code-quoted) is transformed to `–` (EN DASH U+2013). So the string:
… --with-emacs option:
Will show up like this:
… –with-emacs option:
Also mark up nearby not-marked-up code mentions. But this pattern was what was
searched for, so this mostly changes the abovementioned kind of thing.
By convention, code markup (`~`) is reserved for keybindings in Org-based
documentation in Spacemacs. Verbatim markup (`=`) is reserved for code and
other code-like things. So change several readmes to reflect this convention.
Use verbatim markup for things like (non-exhaustive list):
- Emacs Lisp functions, modes, buffers, etc.
- Environment variables
- Directory paths
- Code in general
ADD:
- layers/+lang/go/config.el Added new variable `go-use-test-args` to allow
specifying additional arguments being passed to `go test.
CHANGE:
- layers/+lang/go/packages.el Updated `go-run-tests` to automatically concat
the new variable `go-use-test-args` to args passed to `go test`.
Most people won't take the time to write their own per-theme colors,
so having high-quality defaults matters more than anything else.
These defaults are carefully balanced statistical averages of all colors
suitable for both light and dark themes, and perfected to work for 90% of
all themes.
It's the BEST out-of-the-box experience we can offer without forcing
all users to write their own individual per-theme optimizations
for ALL of their themes...
It isn't enough to just update the variables! We must also refresh the "font
locking" (syntax highlighting) in all buffers that have rainbow-identifiers-mode
currently active, so that they instantly re-paint with their per-theme values.
Otherwise we get stuck with an ugly mishmash of old colors and a new theme.
This change loops through all buffers and marks matching ones for re-painting,
starting with the current buffer first so that the user sees quick results!
Added support for 9 new themes, and updated the defaults
for some old definitions to much more closely match the "spirit"
of each theme's individual intended brightness
and saturation level.
Now reverts lightness and saturation to the user's fallback defaults.
A suggested future improvement: Revert to whatever would be used
for the current theme, so that it is a true "reset". Such a change
would require decoupling the sat&light code in "colors//tweak-theme-colors".
Why?
- 'r' is free binding - not used anywhere for org related modes
- refiling is quite common activity if you use org heavily, so moving it for
easy access makes sense
