This replaces the older pattern
:toggle (configuration-layer/package-usedp ..)
This implementation ensures that :disabled-for honors dependent packages, i.e.
if package a depends on package b, which is owned by layer c, and layer c is
disabled for layer d, then neither package a nor b will be configured for layer
d. Previously, this was only true for package a, but not b.
This commit also fixes:
- configuration-layer/describe-package now shows which post-init and pre-init
functions are disabled, if any
- Does not recreate all layer objects unconditionally when calling
configuration-layer/discover-layers. Previously, this led to all layers being
recreated after e.g. `SPC h SPC`, without any of the dotfile information.
Since this information is now necessary for
configuration-layer/describe-package, it’s important that we don’t clear the
indexed layers when invoking this function.
Problem:
Two which-key functions doc-strings, state that they are obsolete:
(which-key-declare-prefixes KEY-SEQUENCE REPLACEMENT &rest MORE)
This function is obsolete since 2016-10-05;
use ‘which-key-add-key-based-replacements’ instead.
(which-key-declare-prefixes-for-mode MODE KEY-SEQUENCE REPLACEMENT &rest MORE)
This function is obsolete since 2016-10-05;
use ‘which-key-add-major-mode-key-based-replacements’ instead.
Solution:
Replace the obsolete functions.
Searching with `SPC /` through the .emacs.d folder, didn't find any other
occurrences of these functions.
A recent change in the `.spacemacs.template` made it so that the
`dotspacemacs-configuration-layers` variable name can be found inside a
comment right before the expression that sets it's value. This makes the
lazy insertion fail to detect the proper place to add lazy loaded
layers.
This fix solves the immediate problem but maybe a better solution can
be found.
This is an attempt to tighten up the language of the docstrings for
initialization and configuration functions. I realize that's pretty
subjective, so please only use what makes sense. Below is a break-
down to avoid seeming _too_ arbitrary.
Rationale
---------
Headings: We know they're functions, so we don't need to say
'X function'. 'Layer configuration' is called that in the other
functions.
Instructions: Say what I should do.
Other: `dotspacemacs/init' -- 'very beginning' is the more common
English idiom.
If I sound strident, it's just 'cause I'm trying to be terse.
This is all opinion so YMMV. I hope it helps.
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.
This allows the user to, for instance, include the spacemacs tree as a
git submodule of their configuration repository without seeing
superfluous warnings.
Add two new functions: `spacemacs/transient-state-register-add-bindings`
and `spacemacs/transient-state-register-remove-bindings` to prevent layer
authors and end users from dealing with the underlying variables' subtleties.
At the moment a Quelpa recipe like `(recipe :fetcher local)` is being translated
to something like
`(recipe :fetcher file :path "my-layer/local/my-pkg/my-pkg.el")`. So we can
build simple single-file local packages.
This commit changes it to translate to the package directory instead of exact lisp
file, so we can build multi-file local packages. Thus, the above example will be
translated to `(recipe :fetcher file :path "my-layer/local/my-pkg")`.
Also, add the relevant info to LAYERS.org.
Some of the layers do not supply a README.org file which caused
invalid links in the layer list. To avoid this I have changed
spacemacs//generate-layers-from-path to only add links to layers
with a valid README.org file.
I have also removed an invalid check to exclude the non existing directory
"distribution" from the layers list. I think that originally the folder
"distributions" should have been excluded but this is not longer
a feasible action as there is at least one layer with a valid README.org
file in there today. So now we add links to all layers providing a
README.org file independent of their category.
In 2015 a file CONTRIBUTE.org was existing in the /doc folder.
In addition a CONTRIBUTING.md file was existing in the project root folder.
This was merged into a CONTRIBUTING.org file, which is still located in the project root.
However the documentation publishing system was still looking only in the doc folder.
In addition the external documentation is still referring to the old CONTRIBUTE.html which is not longer existing.
I have now included a new function called prior to publishing
which is copying the documents from the projects root into the doc folder.
As a side effect it renames the CONTRIBUTING.org to the old CONTRIBUTE.org
to support the external documentation.
Specifying parent modes (e.g. prog-mode) in :enabled-for-modes and
:disabled-for-modes keys in dotspacemacs-line-numbers catches derived
modes (e.g. c-mode) as well.
Fix bug where an empty :disabled-for-modes and a non-empty :enabled-for-modes
enabled line numbers everywhere, instead of only in modes specified in
:enabled-for-modes. (see https://github.com/syl20bnr/spacemacs/issues/8482)
Add auxiliary function spacemacs/derived-mode-p, similar to derived-mode-p but
takes a major-mode as an argument instead of using the current major-mode.
In this case Spacemacs won't write custom variables to the
dotspacemacs/emacs-custom-settings variable.
Move setq and add-advice to spacemacs/initialize-custom-file which is now
called in spacemacs/init function just after dotspacemacs/user-init function.
New functions:
- configuration-layer/get-location-directory which return the location on disk
given a location
- spacemacs//get-theme-directory which returns the location on disk of the
theme
Add note in documentation to warn about the directory name when :location local
is used, the directory name is the package name not the theme name.
Also remove the docstring about SPC h SPC as it is already implied for the other
options in the dotfile.
Note that this commit modifies one condition, namely disabled mode condition.
It removes the first test:
(not (spacemacs/mplist-get dotspacemacs-line-numbers :enabled-for-modes))
as it is not necessary since we have a `or` in the enable-p condition. Disabled
modes will be checked only if the current major mode is not explicitly enabled
(as intented).
It is now possible to use package properties like :location in
dotspacemacs-themes.
Added hooks ran at the beginning and end of configuration-layer/sync:
- configuration-layer-pre-sync-hook
- configuration-layer-post-sync-hook
configuration-layer-pre-sync-hook is used to hook the new function
spacemacs//add-theme-packages-to-additional-packages. This new function updates
dotspacemacs--additional-theme-packages variables.
Update documentation to mention the new feature.
Add all theme packages defined in dotspacemacs-themes to the variable
dotspacemacs-additional-packages and delay the application of user theme at the
end of startup (only when user theme cannot be applied at the very beginning
of Emacs startup).
This has nice properties:
- we leverage the layer system to handle the theme packages installation and
cleanup.
- theme packages are automatically owned by the dotfile preventing them from
being garbage collected.
- the protected package mechanism is now obsolete since themes were the last
packages using it. This mechanism may be removed in a near future.
Thanks to TheBB for the initial idea.
Distribute spacemacs-theme with Spacemacs so we don't need to download the
package of the theme at startup. It was delaying the display of the home buffer.
Now Spacemacs fallback to spacemacs-dark theme if the user theme cannot be
applied. Spacemacs then tries to install and reapply the user theme. If
successful, at the subsequent startups the user theme is applied right away
instead of spacemacs-dark. If the installation failed then we display a warning
informing the user and suggesting some actions.
There is now no package left to be installed manually at the start of Spacemacs.
Since Emacs 25.1 I got screen flickering when packages are installed on
startup.
Shuffling the UI elements removal seems to fix it on my system using X.
It may have the opposite effect on other users setup.
Thanks to the power provided by Lisp, hack package-install to patch on the
fly org and org dependencies installations in order to install org-plus-contrib
instead.
It is now possible to add local elpa repositories to private variable
configuration-layer--elpa-archives for instance:
(defvar configuration-layer--elpa-archives
'(("spacelpa" . "~/.emacs.d/.cache/spacelpa/"))
"List of ELPA archives required by Spacemacs.")
New functions:
- configuration-layer//package-archive-absolute-pathp
- configuration-layer//package-archive-local-pathp
Fix computation of package installation lazyness when the variable
dotspacemacs-enable-lazy-installation is set to all.
Improve methods on cfgl-layer class to be able to return a list of packages with
or without their properties.
Update tests to reflect the changes.
Discard the warnings produced when packages are byte-compiled which fix the
popup of the *Compile-Log* window at the end of the installation.
If emacs has been launch with argument `--debug-init` we don't discard the
warnings.
Function configuration-layer/load-or-install-package now correctly add the
installed packages to indexed-package hash map.
Those packages are put under the special layer `system`.
This is the first stone toward Spacemacs own ELPA repository.
New function configuration-layer/create-spacelpa-repository which creates the
archive-contents file for the Spacelpa repository containing all the ELPA
packages supported by Spacemacs (and only them).
The layer files were loaded twice because the global variable controlling
the loading of layer files was not explicity set at the layer discovery phase.
Enabling a company backend for a specific mode was a tedious tasks with code
scattered at different locations, one for local variable definitions, one for
company hook function definitions and another where the backends were pushed to
the local variables (which was problematic, since we ended up pushing the same
backends over and over again with `SPC f e R`, pushes have been replaced by
add-to-list calls in the new macro).
All these steps are now put together at one place with the new macro
spacemacs|add-company-backends, check its docstring for more info on its
arguments.
This macro also allows to define arbitrary buffer local variables to tune
company for specific modes (similar to layer variables via a keyword :variables)
The code related to company backends management has been moved to the
auto-completion layer in the funcs.el file. A nice side effect of this move is
that it enforces correct encapsulation of company backends related code. We can
now easily detect if there is some configuration leakage when the
auto-completion layer is not used. But we loose macro expansion at file loading
time (not sue it is a big concern though).
The function spacemacs|enable-auto-complete was never used so it has been
deleted which led to the deletion of the now empty file core-auto-completion.el.
The example in LAYERS.org regarding auto-completion is now out of date and has
been deleted. An example to setup auto-completion is provided in the README.org
file of the auto-completion layer.
spacemacs/alternate-window was suggested to switch between the last selected
window, or the window-analog of SPC TAB (spacemacs/alternate-buffer). This is
part of a larger family of functions, such as spacemacs/jump-to-last-layout on
SPC l TAB and eyebrowse-last-window-config on SPC l w TAB.
Resolves#7845
Some directories were added twice.
Make it more explicit in the code which directories we look into and filter out
directories that don't exist. Emit a warning if an unknown directory is
provided.
Given the loading process of Spacemacs we have no choice but to set the
custom settings twice:
- once at the very beginning of startup
- once at the very end of loading
The first application of custom settings is to be sure that Emacs knows all the
defined settings before saving them to a file (otherwise we loose all the
settings that Emacs does not know of).
The second application is to override any settings set in dotfile functions
like `dotspacemacs/user-config`, users expect the custom settings to be the
effective ones.
This double loading issue is independent from the managment method used for
custom settings. Even with a separate custom-settings file explicitly loaded in
the dotfile we would have been forced to load this file twice to acheive the
expected result described here.
Note: Loading custom-settings twice is not ideal since they can have side
effects! Maybe an inhibit variable in Emacs can supress these side effects?
New function `dotspacmacs/emacs-custom-settings` wrapping Emacs
custom settings sexps.
`dotspacemacs/emacs-custom-settings` is called just after the user
configuration (`dotspacemacs/user-config`)
Customize cannot write its auto-generated sexps inside a function, to
accomplish this we trick Emacs by setting the custom file to a file
in `.cache` directory, the path to this file is defined by the variable
`spacemacs--custom-file`. At the startup of Emacs we read this file
to insert its content inside the function
`dotspacemacs/emacs-custom-settings` in the dotfile, this is done in the
function `spacemacs/write-custom-settings-to-dotfile`.
I don't think we need to write the custom settings to the dotfile when
exiting Emacs as well, since we do it at startup at the very beginning
(i.e. before actually loading the dotfile) we should be OK.
Fixes#5170
When upgrading packages, Spacemacs backs up the old versions of the
packages and then tells the user to restart Emacs using `SPC q r` so
that the new version will be auto-installed. But `SPC q r` doesn't
work if restart-emacs is one of the packages which is being upgraded,
and hence isn't currently installed.
enumeration lists of the form:
- lorem ipsum
* lorem ipsum
+ lorem ipsum
1. lorem ipsum
2) lorem ipsum
will be treated as separate paragraphes and not be squished by
fill-region anymore.
More flexible framed text generation:
- adjust frame width to content width
- define minimum and maximum width
- allow for inserting a caption at the bottom of the frame
Unless dotspacemacs-startup-buffer-respinsive is nil:
- framed notes are centered
- max width is adapted to the window width
Also add a note inviting the user to update his packages and dotfile on
every new release (fixes#7357).
With a nil value for dotspacemacs-switch-to-buffer-prefers-purpose,
switch-to-buffer prefers using the current window, same as vanilla
Emacs. With a non-nil value, switch-to-buffer prefers another window
with the same purpose as the buffer. This affects actions like
spacemacs/alternate-buffer, and opening buffers from Dired.
Cleaned up a warning that occurs when loading private layers
The layer discovery loop scans the private layer directory (
~/.spacemacs.d/layers) AND also ~/.spacemacs.d
The code then checks if a layer directory already exists
but the comparison fails due to trailing / in some paths.
The fix is to strip any trailing slash from both directories being
compared.
- Apply `/` and `//` rules (double / is for private functions)
- Add missing `spacemacs/` prefixes
- Move functions used outside of spacemacs-base layer to
core/core-funcs.el
- Remove unused functions
Commit originally intented to only rename linum-update-window-scale-fix
to spacemacs/linum-update-window-scale-fix :-)
Currently j/k were only navigating between lines, but hitting RET
wouldn't work on a recent file and the like, when the cursor wasn't
directly over the filename. Changing j/k to widget navigation, ensures
that the cursor is always on something clickable. This change should
make the home buffer feel more at home for evil mode users.
Fixes some edge cases like SPC f e R performed after SPC h SPC which
could wrongly install or uninstall packages.
Side effects is contained using the variable
configuration-layer--package-properties-read-onlyp, if non nil then
properties value of a package cannot be overwritten.
Calling multiple times configuration-layer/make-package appends the
same layers to :owners, :pre-layers and :post-layers slot.
Use object-add-to-list instead of push.
Add some tests and mock some warning messages.
Rename configuration-layers/make-packages to
configuration-layers/make-packages-from-layers
Move all package initialization logic to configuration-layers/make-package
instead of having it split between make-packages and make-package.
Hook semantic is to be used with run-hooks API and run all hooks
sequentially, jump list semantic is different since the running
functions are not guaranteed to be executed so we prefer using regular
list API to manage jump-lists.
Previous change in 5d6e9ab789 dropped the
"/" marker to match at beginning of string. Since directory-files
matches a regex, use "\\`" to indicate match at beginning of string.
This commit defines:
- spacemacs-default-jump-handlers: a list of functions that can jump to
definition in ALL modes.
- spacemacs-jump-handlers-MODE: a list of functions that can jump to
definition in MODE.
- spacemacs-jump-handlers: a buffer-local list of functions that can
jump to definition. This is made up of the values of the two previous
variables whenever a given major mode is activated.
- spacemacs/jump-to-definition: a function that tries each function in
spacemacs-jump-handlers in order, and stops when one of them takes us
somewhere new.
- spacemacs|define-jump-handlers: a macro that
* defines spacemacs-jump-handlers-MODE, possibly filled with initial
functions
* defines a function that is added to the hook of the given MODE
* binds “SPC m g g” of that MODE to spacemacs/jump-to-definition
This is an attempt to harmonize all the different approaches to jumping.
Specifically,
- Existing intelligent jump packages that work for only a single mode
should go to the beginning of spacemacs-jump-handlers-MODE. E.g.
anaconda for python, ensime for scala, etc.
- Packages like gtags that work for several modes (but potentially not
all) and which is dumber than the intelligent jumpers should go the
the END of spacemacs-jump-handlers-MODE.
- Packages like dumb-jump that work for all modes should go to
spacemacs-default-jump-handlers.
In all cases the order of the jump handlers in each list should be from
most to least intelligent.
Fixes#6619
Rename dotspacemacs-download-packages to dotspacemacs-install-packages
to better reflect the changes in the previous commit.
Also change the value 'used to 'used-only (note that 'used is still
supported for backward compatibility).
See end of this message for important breaking changes.
Previous behavior was to configure any installed package which caused
a lot of bad side effects and could make spacemacs unusable. This
behavior made little sense and does not fit with spacemacs.
This commit fixes this behavior by separating installed packages from
configured packages. In short dostspacemacs-download-packages variable
now only affect package installation. Packages are now configured if and
only if they are effectively *used* (i.e. listed in variable
dotspacemacs-configuration-layers or dotspacemacs-additional-packages).
IMPORTANT CHANGE: functions `configuration-layer/declare-used-layer` and
`configuration-layer/declare-used-layers` have been removed. These
functions have been introduced in develop branch only so the impact
should be minimal.
Saving a pkg-match regex to a temporary variable lets us avoid recomputing it on each
iteration. I didn't benchmark the performance impact, but unless there's a downside to
using `let*` vs `let`, it seems like a no-brainer.
Passing pkg-match to `directory-files` reduces the returned files to those that match the
regex rolling our own reduction.
For emacs 24.5 packages will be installed in .emacs.d/elpa/24.5, for
emacs 24.4 packages will be installed in .emacs.d/elpa/24.4, etc.
For a user that uses several emacs versions with the same config, the
packages for version X will be in .emacs.d/elpa/X and the packages for
version Y will be in .emacs.d/elpa/Y. This is instead of using the same
.emacs.d/elpa and possibly having copmiled elisp packages the are
incompatible with one of the emacs versions in use.
Rollback directories are also separated by version:
.cache/.rollback/24.5, .cache/.rollback/24.4, etc.
Don't fallback to default init file when SPACMACSDIR is used and
$SPACEMACSDIR/init.el doesn't exist. The file will be created during
the startup process.
Misspellings:
"beginnign" changed to "beginning",
"Insert one of several lines" changed to "Insert one or several lines" in two functions,
"identation" changed to "indentation",
Missing comment copied from the function that inserts a line
in the opposite direction:
(defun spacemacs/insert-line-below-no-indent (count)
"Insert a new line below with no indentation."
and renamed the direction, resulting in:
"Insert a new line above with no indentation."
Duplicate comments removed, the comments inside the functions,
are better explanations of what the function does.
Functions reordered:
The functions:
(defun spacemacs/evil-insert-line-below (count)
and
(defun spacemacs/evil-insert-line-above (count)
were written in a illogical order,
the "above" function should be written before the "below" function,
with this change, the function order will match other functions
with "above" and "below" in their names,
for example these:
(defun spacemacs/insert-line-above-no-indent (count)
https://github.com/syl20bnr/spacemacs/blob/develop/layers/%2Bdistributions/spacemacs-base/funcs.el#L535
(defun spacemacs/insert-line-below-no-indent (count)
https://github.com/syl20bnr/spacemacs/blob/develop/layers/%2Bdistributions/spacemacs-base/funcs.el#L549
evil-commands.el
line 2205: (defun evil-open-above (count)
line 2219: (defun evil-open-below (count)
line 2310: (defun evil-copy-from-above (arg)
line 2326: (defun evil-copy-from-below (arg)
evil-common.el
line 1892: (defun evil-insert-newline-above ()
line 1901: (defun evil-insert-newline-below ()
Can force a check with the universal prefix argument, for instance:
SPC u SPC SPC spacemacs/check-for-new-version
will always check for a new version even if not on master branch.
Once per 24 hours by default.
Caveat: if a new version is detected on the current instance of Emacs
and Emacs is restarted before the end of the 24 hours window then
the new version lighter in the modeline disappear; i.e. we don't
remember the last check results.
Fixes#6692
TL;DR Should get 20~25% speed improvement on startup, should get a big
improvement when using ivy or helm SPC h SPC. Users with layers.el files
in their layers must use `configuration-layer/declare-used-layer`
instead of `configuration-layer/declare-layer`
The implementation of the layer system made heavy use of `object-assoc`
and `object-assoc-list` functions which are not efficient. This PR
mainly replaces those object lists with hash maps in order to index the
objects by their name and achieve an O(1) access time.
The old object lists `configuration-layer--layers` and
`configuration-layer--packages` have been each by two variables each:
- `configuration-layer--indexed-layers` which is a hash-map of all the
layer objects and `configuration-layer--used-layers` which is a list of
all _used_ layers symbols,
- symmetrically `configuration-layer--indexed-packages` which is a
hash-map of all the package objects and
`configuration-layer--used-packages` which is a list of all _used_
packages symbols.
The hash map `configuration-layer--layer-paths` is gone, now we create
directly layer objects when discovering the layers and set the :dir
property. Note that previously the layer paths were the parent directory
of the layer, now :dir is the layer path.
The function `configuration-layer//make-layer` is now similar to its
counterpart `configuration-layer//make-package` in the sense that it
takes an optional `obj` to be able to override its properties.
The functions `configuration-layer/declare-layer` and
`configuration-layer/declare-layers` now takes an optional parameter
`usedp` in order to declare used or not used layers. For convenience
new functions have been added: `configuration-layer/declare-used-layer`
and `configuration-layer/declare-used-layers`, users _must_ update all
occurrences of `configuration-layer/declare-layer` by
`configuration-layer/declare-used-layers` in their `layers.el` files.
`helm-spacemacs-help` and `ivy-spacemacs-help` are updated to match the
changes in `core-configuration-layer.el`.
Rename some variables to make them more explicit:
`configuration-layer-no-layer` -> `configuration-layer-exclude-all-layers`
`configuration-layer-distribution` -> `configuration-layer-force-distribution`
Choose to inherit from face lazy-highlight instead of region.
Ideally a theme should not set lazy-highlight to the same face as
region.
Also move some function to funcs.el and remove some empty lines.
Providing a min-version allows to fetch the elpa version of a built-in
package. For intance to install python mode from GNU elpa, add to
the dotspacemacs-additional-packages:
(python :location elpa :min-version "0.25.2")
This feels weird to set as default a variable using a third-party
service like Dropbox. Was OK when Spacemacs was less popular, now this
variable has nothing to do in this project.
- make it clear that setting dotspacemacs-default-font has no effect in
terminal
- make it clear that it's user responsibility to install Source Code Pro
font
Random banner choice was broken by the presence of the `img` directory.
Even if `random` was set instead of `random*` in .spacemacs file the
cate banner would still show.
This commit fixes this bug while simplifying the existing code by adding
a regexp filter to only choose "*.txt" files.
Previous behavior, if a micro-state binding had :doc not nil, was an error
Wrong type argument: stringp, 1
Detailed reproduction:
cat > /tmp/test.el <<EOF
(spacemacs|define-micro-state test
:doc "[a] a [b] b [q] quit"
:bindings
("a" nil :doc "a")
("b" nil :doc "b")
("q" nil :doc "quit" :exit t))
(spacemacs/test-micro-state)
EOF
emacs --load /tmp/test.el
Type a, b, or q and see the error in the echo buffer.
With this new variable, user can load spacemacs anywhere, e.g.
"~/.emacs.d/spacemacs/". Only user's cache directory is still hard-coded
as "~/.emacs.d/.cache/". If user want to use spacemacs this way, drop
one line as the below in "~/.emacs.d/init.el":
(setq spacemacs-start-directory "~/.emacs.d/spacemacs/")
(load-file (concat spacemacs-start-directory "init.el"))
Helm-flx, which is included as a core package, requires a minimum Emacs
version of 24.4. As it stands attempting to install Spacemacs on Emacs
24.3 or below will break on helm-flx.
Old implementation excluded package that were not selected, this
implementation does not excluded them, it simply ignore it in the
layer where they are not selected. This reimplementation comes from
a refactor of the way packages.el files are loaded, instead of loading
these files at the moment of resolving the list of used packages, they
are now loaded when making the layer objects. A neat consequence is that
side effects is better confined and the configuration-layer/get-packages
is now pure (at least a lot more pure than before).
In the `cfgl-layer` class the slot `:user-packages` has been renamed
to `:selected-packages` which defaults to `'all` meaning that all
the packages in `:packages` are selected. `:selected-packages` value
is given by the new function `configuration-layer//select-packages`.
Effectively selected packages are given by a new method for `cfgl-layer`
class called `cfgl-layer-get-packages`.
Tests have been updated to reflect the changes.
Also documentation on configuration layer declaration in the dotfile
section of DOCUMENTATION.org has been greatly improved (I hope) and
reflect the last feature added to the
`dotspacemacs-configuration-layers` list.
I had a duplicated layer, but the message that's printed broke spacemacs starting up because it was a format string but wasn't in a `format` form. Simple fix. Spacemacs works again! Yay!
This issue was introduced in c0851ddcb3, according to a quick `git bisect`. :)
The car of :owners is the real owner.
Now we can easily gather the list of all owners of a package.
Also fix unit tests about missing spacemacs-insecure variable.
Revert "Parse CLI args after dotspacemacs/init"
This reverts commit 0a35502de4.
We cannot delay the parsing of CLI params since we need their values
early during the loading process (i.e. debug info).
Replace the previous commit with a new variable `spacemacs-insecure`
which takes the value of --insecure boolean parameter.
* layers/+spacemacs/spacemacs-editing/packages.el (spacemacs-editing/init-origami):
add origami package and initialize it
* core/core-dotspacemacs.el (dotspacemacs-override-evil-folding): new
variable to allow choosing between different code folding methods.
Currently supported `evil' and `origami'
- Use completing-read when called interactively
- Check for string locations in addition to built-in etc.
- Don’t double-princ on/off for toggles
- Bind to SPC hdP
This new variable replace dotspacemacs-delete-orphan-packages
It defines the behaviour of Spacemacs when downloading packages.
Possible values are `used', `used-but-keep-unused' and `all'.
- `used' will download only explicitly used packages and remove any
unused packages as well as their dependencies.
- `used-but-keep-unused' will download only the used packages but won't
delete them if they become unused.
- `all' will download all the packages regardless if they are used or
not and packages won't be deleted by Spacemacs.
Default value is `used`.
The new layer file `layers.el` is used to declared additional layers.
It is like the sibling of packages.el except that for now it does not
take a list (can do this in a futur commit).
The new order for file loading is the following:
layers.el > packages.el > funcs.el > config.el > keybindings.el
Since packages.el relies on some undefined stuff encapsulated
in init functions, it is not meant to be byte compiled. OTOH funcs.el
(where lies most of the computation added by a package config) should
be compilable.
Since we load packages.el very early it is not possible to use
`configuration-layer/package-usedp` in funcs.el.
This commit also fixes the tests.
Add argument `action` as first positional argument, the second
positonal argument is for tests.
Add actions `doc` and `config`, doc works on org files and config works
on configuration file like packages.el.
The special action `all` performs all actions.
Move test files to `test` sub-folder.
Choose between `md5sum` and `md5` (on OS X) executables.
Add a new formatting function `move-packages-to-config` to move
package lists to config.el file.
Change the semantic of the files for a better control over the
configuration (like being able to use package-usedp function to
guard package configuration and function definition).
This is a big change!
The list of packages of the layer must now be placed in config.el
packages.el only contains pre-init, init and post-init functions
Now both packages.el and funcs.el can use
configuration-layer/package-usedp which is much more powerful than
just configuration-layer/layer-usedp.
Also merging into one file the layer variable and the list of packages
make sense (in this case config.el).
It also means that we can quickly load all the packages declared in
all layer for helm and ivy sources by just loading the config.el file.
Expect some regressions introduced by this change, like the helm
and ivy sources, they will be fixed in other commits.
This commit introduces two new options to add-toggle:
:prefix, a symbol that is bound to the raw prefix argument (as
in `(interactive "P") forms).
:on-message, an expression overriding the default 'on' toggle
message (useful to document a toggle's argument).
These new options are applied to long-lines' toggle, so we can choose
how many lines to toggle it on via a raw prefix argument.
Visiting the dump file before writing to it is slow, because it triggers
all the regular actions that happen when a user opens a file, e.g.
enable major mode and minor modes. We don't need all that,
with-temp-file is what we really want.
spacemacs/report-issue opens a buffer prepopulated with a template and some
system information. The buffer is markdown-mode augmented with a keybinding to
send the contents to github.
Or if that doesn't work, the text remains in the buffer to be manually copied
and pasted. This makes the error reporting more robust: right after the user has
experienced one bug is the worst time to manifest another one.
This commit also rewrites spacemacs//describe-last-keys-string because I
couldn't figure out how else to prevent view-lossage from burying the new
REPORT_SPACEMACS_ISSUE buffer. Suggestions welcome.
--no-layer: desactive all the layers except the distribution layer
--distribution: allow the change temporarily the distribution
example
Start emacs with spacemacs-base and no layers:
emacs --no-layer --distribution spacemacs-base
Add test mode: {repo-root}/core/tools/doc-fmt/run.bash test
applies formatting only to {repo-root}/core/tools/doc-fmt/test.org
Use "gsed" if present.
Add new filters, refactor code.
test.org now represents most of the .org file formatting errors.
If emacs is launched with -q option and spacemacs files are loaded via
--eval option, then spacemacs fails due to depending `after-init-time'
variable being set properly which is done only when emacs is launched
without -q. This patch is to remove this dependence so that spacemacs
launched with -q can still work. This is done by introducing
`spacemacs-initialized' variable which is set properly regardless of
which command line options are present or absent.
New package :step 'bootstrap', this step happens before 'pre' packages.
A new layer names 'spacemacs-bootstrap' gather all the ':step bootstrap'
packages. This layer is special and is always the first element of
the variable 'configuration-layer--layers' which assure that all
bootstrap packages are configured first.
This new layer leverages the configuration layer system, removes the
clutter of package installations in the function 'spacemacs/init' and
isolate the bootstrap packages in one place.
The https link-type opens the local copies of the Spacemacs documentation files with
the spacemacs/view-org-file function. It supports GitHub style heading links
For example, the link:
https://github.com/syl20bnr/spacemacs/blob/develop/layers/org/README.org#links
Will be handled similary to as if it was:
file:~/.emacs.d/layers/org/README.org::*links
Also the `space-doc' mode will be applied.
Refactored GH style anchor search.
`space-doc-mode` - Buffer local minor mode for Spacemacs documentation files. The mode hides org meta tags.
Enabled when viewing documentation via `SPC` `h` `SPC`
Add new html preprocessors to the org-html-publish-to-html pipeline:
- `spacemacs//toc-org-unhrefify-toc`
- remove the `toc-org` modifications for the TOC
- `spacemacs//org-heading-annotate-custom-id`
- annotate org headings with the indexes that GitHub uses for linking.
`org-html-publish-to-html` will use them instead of the default
`#orgheadline{N}`. This way the GitHub links and the
http://spacemacs.org/ links will be compatible."
- `spacemacs//reroot-links`
- find the links that start with
https://github.com/syl20bnr/spacemacs/blob/ and end with
.org{#an-optional-heading-link} (i.e the links between the local org
files). Change their root to http://spacemacs.org/ so the links will
point at files located on the site.
*For the "file to file" links to work properly the expor
Several things have been fixed:
- correctly set the :lazy-install flag on all packages of a lazy
installed layer
- correctly unset the :lazy-install flag when installing the packages
- sort the packages before installation and configuration
Fixes#5610
Make cfgl-layer-owned-packages return a list of package objects instead
of just package names.
Use this function to correctly excluded non owned package when flagging
package for lazy installation.
Fixes#5608
- only consider distant packages (i.e. requiring an installation)
- remove lazy install flag of a layer if at least one owned package is
already installed
Now Spacemacs will ask for layer installation when opening a file with
a known file type.
The auto-mode-alist entries are added by the file auto-layer.el in
the layers directory.
Easy insert of forms for lazy initialization can be done with the
interactive function configuration-layer//insert-lazy-install-form.
Change default value of dotspacemacs-enable-lazy-installation to t.
Motivation
While disabling Evil in holy-mode makes its implementation shorter and
sounds elegant on the paper, in practice it puts a big burden on the
configuration parts which need to know if Evil is enable or not. This is
a bad separation of concerns and the bunch of fixes that we were forced
to do in the past weeks shows this issue. Those fixes were about
removing the knowledge of the activation of Evil by implementing new
dispatching functions to be used by layers, this is cumbersome and makes
Spacemacs layer configuration more subtle which is not good. There was
additional bad consequences of the removal of Evil state like the
impossibility to use Evil lisp state or iedit states, or we would have
been forced to implement a temporary activation of Evil which is
awkward.
Instead I reintroduce Evil as the central piece of Spacemacs design thus
Evil is now re-enabled in holy-mode. It provides the abstraction we need
to isolate editing styles and be able to grow the Spacemacs
configuration coverage sanely. Layers don't need to check whether the
holy mode is active or not and they don't need to know if Evil is
available (it is always available). We also don't need to write
additional dispatching functions, this is the job of Evil, and I think
it provides everything for this. Ideally configuration layer should be
implemented with only Evil in mind and the holy-mode (and hybrid-mode)
should magically make it work for Emacs style users, for instance we can
freely use `evil-insert-state` anywhere in the code without any guard.
Evil is now even more part of Spacemacs, we can really say that
Spacemacs is Emacs+Evil which is now an indivisible pair. Spacemacs
needed this stable API to continue on the right track.
While these changes should be rather transparent to the user, I'm sorry
for this experimental period, I failed to see all the implications of
such a change, I was just excited about the possibility to make Evil
optional. The reality is that Spacemacs has to embrace it and keep its
strong position on being Emacs+Evil at the core.
Implementation
- insert, motion and normal states are forced to emacs state using an
advice on `evil-insert-state`, `evil-motion-state` and
`evil-normal-state` respectively. These functions can be used freely in
the layer configuration.
- A new general hook `spacemacs-editing-style-hook` allow to hook any
code that need to be configured based on the editing style. Functions
hooked to this hook takes the current style as parameter, this
basically generalize the hook used to setup hjkl navigation bindings.
- ESC has been removed from the emacs state map.
- Revert unneeded changes
- Revert "evil: enter insert-state only from normal-state"
commit bdd702dfbe.
- Revert "avoid being evil in deft with emacs editing style"
commit f3a16f49ed.
Additional changes
All editing style packages have been moved to a layer called
`spacemacs-editing-styles`
Notes
I did not have time to attack hybrid mode, I should be able to do it
later.
