Currently, when tab is pressed, spaces are inserted. It makes Python
REPL unable to recognize code indentation. We must use tab to indent.
Setting indent-tabs-mode is buffer local, so it does not affect its
value in other buffers.
Remove default auto-mode-alist mapping from *.h to c++-mode.
Introduce a variable and use that in the init: section of cc-mode to
conditionally update the auto-mode-alist.
Updated the documentation.
Conflicts:
contrib/!lang/c-c++/README.md
- enable ocamldoc in company completion
- enable merlin-document binding
- add binding for ML <-> MLI switch
- bind most merlin commands
user visible change:
printing type at point was moved to SPC mht, and SPC mhT asks for an expression
SPC mgg/mgG is now consistent with CONVENTIONS.md, and
SPC mgb can be used to go back(in addition to default C-c &)
When no Python executable is detected, Semantic causes an error that it cannot
retrieve include paths from the executable, causing an error that prevents
Python buffer to be switched and error is repeatedly reported whenever a Python
buffer is opened. This commit ignores the error from
semantic-python-get-system-include-path that is called when Semantic setup
itself for a Python buffer.
- Company should be enabled in Python REPL for displaying completion
candidates. Note that in Python REPL, we must use company-capf to
display the candidates from Python REPL, since in the REPL it uses
completion-at-point to display candidates from the REPL. Also, we
locally set company-minimum-prefix-length to 0, so that completion
starts immediately after the first character. We also need to set the
idle delay a bit higher (0.5) compared with Spacemacs default (0.2),
since setting it too low may cause company unable to retrieve completion
candidates.
- Anaconda should NOT be enabled in Python REPL for retrieving proper
completion candidates. This can be tested in Python REPL: when
anaconda-mode is enabled and without any prefix string before point, run
the command completion-at-point (bound to C-M-i by default), completion
candidates from Anaconda are displayed but NOT the ones from the real
REPL. So, when you import a module i.e. os and try to access the
functions in the module with "os." prefix, nothing inside the module is
displayed because Anaconda does not work in Python REPL. When
anaconda-mode is disabled, candidates in "os." are displayed properly
with company-capf. company-anaconda backend cannot be used in Python
REPL since it is buggy: it cannot retrieve any prefix before point and
narrow properly.
yas-global-mode is always enabled to ensure snippets are loaded
properly, but to use the snippets, a major mode must have yas-minor-mode
enabled. This commit ensure such condition happen.
By checking if the yas-minor-mode is enabled. If so, load yasnippet. We
must do so because some package might load yasnippet and make
yas-minor-mode definition available, thus prevent Spacemacs to load
yasnippet properly. By checking yas-global-mode variable, we can assure
that yas-global-mode is enabled properly.
Also exclude yasnippet from eshell.
- Only set Eshell by default on Windows; in other OSes we use `ansi-term`,
NOT `term`. The reason is that `ansi-term` can use Emacs key bindings
such as C-x b, C-x 1... even editing commands such as M-u/M-l. `term`
cannot and we have to use the default `C-c b` variants i.e. C-c b to
switch buffer instead of "C-x b". `ansi-term` is currently the best
terminal emulator in Emacs.
- Allow ansi-term/term to close its shell-pop window when a user type
"exit".
- With shell-pop, we can allow ansi-term/term to track current directory
of current buffer. It is a useful feature and it also makes ansi-term
consistent with eshell/shell.
- Swap "term" and "ansi-term" key bindings, since the old "ast" was used
for multi-term, which is a wrapper around ansi-term.
* `mtt` now reloads test ns before running all tests
* added `mtT` to reload test ns and run focused test
* added `mtr` to reload test ns and re-run failed tests for namespace
The test namespace reloading is a workflow optimization. The alternative
is a minimum of two commands (`SPC m e b` or `SPC m e f` and `SPC m t
t`) in order to re-run a single test. When driving development from
tests, this is onerous at best (and I couldn't think of a good reason
_not_ to reload the test ns).
- display a warning if agda has not been found
- add a micro-state to easily navigate between the goals
- use face-remapping-alist to redefine the faces
- add some missing key bindings
- add support for golden-ratio
- add quick cabal instruction to install agda
- sort key bindings alphabetically
This makes it possible to use the kbd tag. I'm not sure where should
this be in. Feel free to move it around.
Renamed endless to spacemacs
Added fontification to @@html: stuff (could be changed)
This is because the list of snippet dirs had our
~/.emacs.d/private/snippets at the end so it doesn't guess that.
Something must have changed upstream. Also it's better to have our own
custom snippets to be first so as to override the default ones if we
pleased.
Fix activation of auto-complete
Sort packages
Improvement with hook usage
Move opam init function in funcs.el and call it only in tuareg config
Fix errors if opam is not installed, warn if it has not been found
Small edition of the README
Rename additional REPL functions with spacemacs prefix
This is a nice feature taken from Prelude. This commit advises yank and
yank-pop and evil-paste-after to indent after pasting, unless a prefix
arugment is specified or the major mode is one of
spacemacs-indent-sensitive-modes.
Conflicts:
contrib/better-defaults/README.md
It is not needed in this case because a post-init does not mark
the package as used (so if there is not init function for company
it won't be installed and the post-init function won't be evaluated)
It is reported in this issue: https://github.com/company-mode/company-mode/issues/180
that company has distorted completion popup with fci-mode turned on.
Company maintainer proposed a fix for this issue that is added in this
commit.
Having fci-mode to visually indicate the limit of commit message region
is nice. With it, we know exactly our text does not cross the limit (i.e
72 columns). Sometimes, text can cross the boundary if we add something
but use C-n or j to move to next line.
The `progn` in the eval-after-load was not quoted, so it was being
evaluated unconditionally. This led to the haskell-indentation guides
being unconditionally enabled.
Additionally, only enable the indentation guides when
haskell-indentation-mode is on.
Resolves#1136.
- Replace `SPC h h` and `SPC h H` for `SPC h i` and `SPC h I`
- Replace `SPC r` prefix for `SPC x`
- Replace `SPC x SPC` for `SPC x r`
- Sort key bindings alphabetically
Originally `load-ess-on-demand` used `use-package` for the purposes of
actually loading `ess-site` and related elisp for dealing with R. But
since `use-package` doesn't return true or false when a package is
actually successfully loaded anymore, and the semantics of how it will
work aren't precisely clear, it makes sense to use `require` here
instead.
(defun load-ess-on-demand ()
(interactive)
(-all? '---truthy? (list
(require 'ess-site)
(require 'ess-R-object-popup)
(require 'ess-R-data-view))))
All the normal hooks setup by `use-package` will work as normal. I
actually don't see a reason we should use `use-package` instead of
`require` in this specific instance. Since `use-package` often defers
loading packages, it's arguably clearer to use `require` in this
particular instance.
Currently by default company knows when to flip the popup. However, when
the above option is set to t, it causes a strange behaviour: when the
tooltip is flipped, press down (i.e. M-n or arrow key) go up and press
up (i.e. M-p or arrow key) go down.
With current configuration, .h file is only associated with c++-mode
only when cc-mode is loaded. So, when a .h file is opened the first
time, it is opened and is applied with c-mode. Only the 2nd time then
c++-mode is properly used.
- RET is not used anymore
- TAB complete the common prefix and cycle between candidates
- To complete the selection use C-l or `jk` pressed quickly
- Yasnippet expand uses hippie-expand and M-/ or C-p
As per http://www.emacswiki.org/emacs/IswitchBuffers :
> Note: IswitchB was part of GnuEmacs, but it has been removed starting
> with Emacs 24.4. The Emacs ‘NEWS’ (‘C-h N’) says to use IcompleteMode
> instead, but for those wanting more familiar functionality,
> InteractivelyDoThings is a closer fit.
Add open-line command to "SPC j" group: open-line breaks the current
line into two lines without moving the cursor down. It's different from
`o` in Vim.
Use C-o to enter evil-execute-in-normal-state.
In i3, whenever you switch to a new workspace, it starts with a blank
frame with default wallpaper. It would be useful when we switch
workspace, the window is reset to a default buffer so it's more obvious
for first time users and is closer to i3 or Vim's tabs. Since eyebrowse
supports this feature already, we simply set it to switch to
spacemacs/home function.
Also, eyebrowse-get becomes eyebrowse--get to express that it is now
private function. So this commit fixes it appropriately.
- ggtags eldoc works only if we enable ggtags-mode.
- For that reason, move the code that enable eldoc into its own
function.
- Also enable gtags for some other modes that have no layer.
