Some types of file opening don't load the powerline after startup. I think this is caused by over-zealous lazy loading. This only happens for certain file types, which leads me to suspect it is something to do with less-fancy major modes not trying to modify the modeline and thus not calling the hooks that lazy-load powerline.

It happens when loading files in any way and for certain types including: directories, .plist files, some other things I don't remember.

Reproducible Steps

• Start spacemacs
• Open a directory with emacsclient (or any other method)
• Modeline is blank

If this is due to lazy loading, this could potentially be fixed by not lazy-loading powerline. This makes sense since when would you ever start emacs and do a bunch of actions where you want none of them to have a powerline?

Error Messages

Note in the second message it mentions some flycheck-mode thing being void. This makes me suspect it is a lazy loading problem, and that this issue happens for any file type that flycheck does not support.

This is this error I get in messages after startup:

Eager macro-expansion failure: (error "Autoloading failed to define function defpowerline")
Error: (file-error "Cannot open load file" "no such file or directory" "powerline")
Error: (file-error "Cannot open load file" "no such file or directory" "tex-site")

And then this error after loading one of the aforementioned file types:

Error during redisplay: (eval (let* ((active (powerline-selected-window-active)) (line-face (if active (quote mode-line) (quote mode-line-inactive))) (face1 (if active (quote powerline-active1) (quote powerline-inactive1))) (face2 (if active (quote powerline-active2) (quote powerline-inactive2))) (state-face (if active (spacemacs/current-state-face) face2)) (flycheckp (and spacemacs-mode-line-flycheckp (symbolp flycheck-mode) (symbol-value flycheck-mode) (or flycheck-current-errors (eq (quote running) flycheck-last-status-change)))) (vc-face (if (or flycheckp spacemacs-mode-line-minor-modesp) face1 line-face)) (separator-left (intern (format "powerline-%s-%s" powerline-default-separator (car powerline-default-separator-dir)))) (separator-right (intern (format "powerline-%s-%s" powerline-default-separator (cdr powerline-default-separator-dir)))) (lhs (append (list (powerline-raw (spacemacs/window-number) state-face)) (if anzu--state (list (funcall separator-right state-face face1) (powerline-raw (anzu--update-mode-line) face1) (funcall separator-right face1 line-face)) (list (funcall separator-right state-face line-face))) (list (powerline-raw "%*" line-face (quote l)) (powerline-buffer-size line-face (quote l)) (powerline-buffer-id line-face (quote l)) (powerline-raw " " line-face) (funcall separator-left line-face face1) (powerline-major-mode face1 (quote l)) (powerline-raw " " face1) (funcall separator-right face1 line-face)) (if flycheckp (list (powerline-raw " " line-face) (powerline-raw (spacemacs//custom-flycheck-lighter error) (quote spacemacs-mode-line-error-face)) (powerline-raw (spacemacs//custom-flycheck-lighter warning) (quote spacemacs-mode-line-warning-face)) (powerline-raw (spacemacs//custom-flycheck-lighter info) (quote spacemacs-mode-line-info-face)))) (if (and flycheckp spacemacs-mode-line-minor-modesp) (list (funcall separator-left line-face face1) (powerline-raw "  " face1) (funcall separator-right face1 line-face))) (if spacemacs-mode-line-minor-modesp (list (powerline-minor-modes line-face (quote l)) (powerline-raw mode-line-process line-face (quote l)) (powerline-raw " " line-face))) (if (or flycheckp spacemacs-mode-line-minor-modesp) (list (funcall separator-left (if vc-face line-face face1) vc-face))) (list (powerline-vc vc-face) (powerline-raw " " vc-face) (funcall separator-right vc-face face2)))) (rhs (list (funcall separator-right face2 face1) (powerline-raw " " face1) (powerline-raw "%l:%2c" face1 (quote r)) (funcall separator-left face1 line-face) (powerline-raw " " line-face) (powerline-raw "%p" line-face (quote r)) (powerline-chamfer-left line-face face1) (let ((progress (format-mode-line "%p"))) (if (string-match "%" progress) (powerline-hud state-face face1)))))) (concat (powerline-render lhs) (powerline-fill face2 (powerline-width rhs)) (powerline-render rhs)))) signaled (void-variable flycheck-mode)

So typing "dotfiles" actually yields "dotiles". I presume it's something to do with the fd escape binding.

Say I want to type flycheck in the minibuffer, a bunch of fffffffffffffffffffff are printed in the minibuffer. If I type more slowly it is ok.

Make sources for helm to quickly access various part of Spacemacs (ie. an init function in a given layer)

I was using Vim for 6 years before I tried Evil, and had finally gotten in the hang of using Emacs with multiple require 'init-packages until I found this repo. I really like the way you set up extensions and packages, in addition to having a call to an init function on each package load, which is very reminiscent to how I hacked Vundle to load some autoload functions based on the package name loaded.

One question I had was whether you ever considered refactoring some of the package definitions into separate contribution layers for some advanced features and languages. I know you can add a package name to a variable to not load it, but I have about 40 packages in that variable, (mostly language packages I do not use). The idea being that you basically have a bare bones emacs environment to still have a decent experience and then maybe make a spacemacs-advanced contribution layer that loads everything else (it can even be set as default).

• Fixes from master
• Features from develop
• in any cases, PR onto develop
• rebase on top of upstream HEAD recommended before submitting

@trishume do you find this OK ?

The anzu-mode thing in the powerline appears in all windows, even ones that aren't being searched. I love the addition of anzu-mode and it is great, but this is kinda annoying.

It's visually distracting to have all my emacs windows and frames leap into motion with 10 different Anzu mode things changing all at once while I'm searching.

I might fix this out myself in a day or two once I get a better project-tree setup going. Currently the main downgrade from my previous editor (Sublime Text) is that I really don't like neotree. But after that I'll work on fixing up little things like this.

Right now everything is going into master (which I depend on with my work machine!) so stability is at risk. I propose moving to nvie's incredibly popular flow (which there's even tooling around).

So this repository should have a develop branch that is the very latest version. master is your latest stable version. When you want to add things you create feature/{FEATURE-NAME} and pull request it into develop. This is just touching on the subject though, I'd recommend reading that blog post I linked above.

Then we could implement semver so releases can be associated with a meaningful version. This allows users to stick at versions if they need to and only upgrade when desired. It should be more stable for day to day use and will allow you to write up release / upgrade notes when a new version is done.

For people following a certain branch, we need a function that will update spacemacs, this should both update the repo to the latest checkout, but also update all extension level submodules.

Perhaps also integrate this in startup, so in addition to adding packages, we also update submodules. The latest switch for nerd-commenter as an extension was missed in my setup when I just did a git pull

Issue to track progress toward a fill support of Emacs 24.4.

• OS X: Homebrew build with flag --with-x:
  • minibuffer takes more space (more vertical lines)
  • subword displays a lighter
  • mode-line colors from solarized are not accurate in the inactive buffers
  • deletion of orphan package seems broken

After updating my spacemacs to b50bda33ea2506acc374159c0df7e6cc7090a9ddh today, after the first launch it went through an orphan package deletion cycle where it says it deleted 11 packages, which it didn't list. I don't think there should have been that many orphaned packages, it might have been a bit overzealous.

Now on startup I get a warnings buffer opening up with the following error (yes it is repeated 3 times):

Warning (emacs): Unable to activate package `concurrent'.
Required package `deferred-0.3.1' is unavailable
Warning (emacs): Unable to activate package `concurrent'.
Required package `deferred-0.3.1' is unavailable
Warning (emacs): Unable to activate package `concurrent'.
Required package `deferred-0.3.1' is unavailable

I suspect the orphan deleter might have deleted the deferred package even though concurrent depends on it.

This bug causes neotree-rename-node and neotree-create-node fail for those who use helm as completion system.

Step to reproduce:

1. open a neotree window with <leader key> f t
2. in neotree window, M-x helm-mini. An error message should display:

Window too small for splitting

This is caused by line 1608 in spacemacs/packages.el,

(push '("^\*helm.+\*$" :regexp t  :position bottom) popwin:special-display-config)

Not sure what we use popwin with helm for. Is it possible to remove helm from popwin?

In Emacs 24.4 there are 2 functions to toggle the frame fullscreen:

• toggle-frame-fullscreen removes window manager decoration (no title bar, no taskbar on windows)
• toggle-frame-maximized respects window manager decoration (regular maximize)

On windows, it is not natural to hide the decorations.

2 possibilities for the behavior of dotspacemas-fullscreen-at-startup:

• dependending on the window system we dispatch to toggle-frame-fullscreen or toggle-frame-maximized (say the latter for Windows only)
• specific symbol value for the variable:
  • nil, with-decoration, without-decoration
  • nil, fullscreen, maximized

Need your feedback.

Getting this on startup under OS X, emacs 24.3:

Warning (initialization): An error occurred while loading `/Users/cc/.emacs.d/init.el':

error: Key sequence S c starts with non-prefix key S

You have evil-leader setting "S" to shell command, and then later using it as a prefix for another command.

Symbol's function definition is void: spacemacs/cycle-spacemacs-theme

I haven't nailed down a cause, yet, just something I noticed. I'll let you know if I find anything out.

For now, installing it as a package has some issue regarding Emacs 24.4.

More info: redguardtoo/evil-nerd-commenter#37

Makefile or alike in order to easily fetch the dependencies required by some modes (ie. aspell for Flyspell)

• Flyspell --> aspell
• Pylookup --> run make
• Flycheck Python --> flake8
• Python test runner --> nose, nose-fixes

...

I'm not sure why yet, but I'm getting this error.

File mode specification error: (error "Autoloading failed to define function ac-js2-mode")

I'm planning on "fixing" this by removing the js2 completion and replacing it with tern.

Try to put this automatically generated configuration stuff in ~/.spacemacs:

(custom-set-variables
;; custom-set-variables was added by Custom.
;; If you edit it by hand, you could mess it up, so be careful.
;; Your init file should contain only one such instance.
;; If there is more than one, they won't work right.
)
(custom-set-faces
;; custom-set-faces was added by Custom.
;; If you edit it by hand, you could mess it up, so be careful.
;; Your init file should contain only one such instance.
;; If there is more than one, they won't work right.
)

OSX, emacs 24.3... After a fresh install of spacemacs with no .spacemacs I get this in the spacemacs buffer:

Loading.........................................................OK!
Found 23 orphan package(s) to delete...
--> deleting async... [1/23]Ready!
[126 packages loaded in 3.255s]

and this in the Warnings buffer:

Warning (initialization): An error occurred while loading `/Users/cc/.emacs.d/init.el':

Wrong type argument: arrayp, ([cl-struct-package-desc async (20141001 151) Asynchronous processing in Emacs nil nil nil /Users/cc/.emacs.d/elpa/async-20141001.151 nil nil])

To ensure normal operation, you should investigate and remove the cause of the error in your initialization file.  Start Emacs with the `--debug-init' option to view a complete error backtrace.

and here's the stack trace, apologies for the truncation:

Debugger entered--Lisp error: (wrong-type-argument arrayp ([cl-struct-package-desc async (20141001 $
aref(([cl-struct-package-desc async (20141001 151) "Asynchronous processing in Emacs" nil nil nil$
(package-version-join (aref (cdr (assq package package-alist)) 0))
contribsys/get-package-version(async)
(package-delete (symbol-name orphan) (contribsys/get-package-version orphan))
(while --dolist-tail-- (setq orphan (car --dolist-tail--)) (setq deleted-count (1+ deleted-count)$
(let ((--dolist-tail-- orphans) orphan) (while --dolist-tail-- (setq orphan (car --dolist-tail--)$
(progn (spacemacs/append-to-buffer "OK!\n") (spacemacs/append-to-buffer (format "Found %s orphan $
 (if orphans (progn (spacemacs/append-to-buffer "OK!\n") (spacemacs/append-to-buffer (format"Foun$
(let* ((dependencies (contribsys/get-package-dependencies)) (implicit-packages (contribsys/get-im$
contribsys/delete-orphan-packages()
(if (not (spacemacs/emacs-version-ok)) nil (defconst user-home-directory (expand-file-name (conca$ 
eval-buffer(#<buffer  *load*> nil "/Users/cc/.emacs.d/init.el" nil t)  ; Reading at buffer positi$      

Discussion started with issue #28.

we have 2 possibilities:

• we mode it to be usable (proper key bindings, remove mode-line, toggle on and off with automatic focus, make it play nice with window numbering and golden-ratio)
• we replace it

I lack knowledge for sidebar things since I don't use them (I use ido and helm projectile) so any proposal is welcome.

On both OSX Cocoa distributions of emacs the powerline dividers look terrible. They are missized and the edges also are wiggly and jagged. According to @syl20bnr this also happens on Windows.

Screenshot taken in Homebrew Cocoa Emacs

Screenshot taken in Railwaycat's Mac Port

It might just be my being a complete emacs beginner, but it isn't clear to me what to do after the git clone step of the installation. It's also possible that this is a regression and the package installation code just broke, but it seems more likely that I just don't understand how to install the package list.

I'm left with an emacs that looks like the first screenshot in the readme: no evil, no powerline, no fanciness of any kind. I have a feeling that the next step is to install all the required packages, but it isn't clear if there is a good way to do this.

Looking through the source I found the spacemacs-packages definition that looks like it provides the list of required packages to some code, but there doesn't seem to be any other references to this anywhere else in the project. Are new users supposed to map some kind of installation routine over this list in a lisp repl?

I feel like there's one step missing from the installation section of the readme. If I figure out what it is I'll submit a PR to add it to the Readme, but until then I'll just leave this issue here.

@syl20bnr

I work around this issue by patching evil-nerd-commenter.el to the previous commit redguardtoo/evil-nerd-commenter@a88146b.

More info here: redguardtoo/evil-nerd-commenter#37

Take care of the absence of aspell.

After upgrading to v0.2.0 I'm a problem with projectile where projectile won't find the files in some repos. I think the pattern might be repos that I've visited while my emacs was in a borked state, so it might have something to do with caching.

On some repos when I do either a find in project or a helm find, the only files that show up are the ones I've opened in emacs. When I search for any other file, even the exact names of files I know exist, nothing comes up.

This only seems to happen on repos I've visited in emacs within the last little while, i'm not sure what the exact time period is. If I open a project I haven't touched in a long time projectile works fine.

I'll keep looking into this, I think it has something to do with the caching.

If we are concerned about ergonomics, what do you all think about adjusting some of the keybindings to less intensive keys?

I know with Vim when I got in that mind set, I remapped ; to : to save on pressing the shift. What about changing helm-M-x to SPC ; in the same vein? I have a hard time pressing all 3 keys.

In general, I think it would be good to remap bindings to 3 or less keys. The yank ring for example is typically tied to the y key. Why not SPC s y instead of the current SPC k i l?

I know that I can change everything on my side, but I am asking on a crowd level.

Also, has anyone seen what @tpope has done with unimpaired-vim, where he maps most movement specific keybindings to [ or ] + key? This extends the bracket key usage in Vim to multiple different aspects. For example next or previous error is ]l or [l respectively (referencing the listwin). Would a change away from using SPC for everything be frowned on, as I was thinking of trying to port over some of those features.

I will bootstrap this first and post updates in this issue.

This task depends on #110 because we need to retain backward compatibility with the current layers format.
The idea is to have only one list per configuration layer to declare all the packages and extensions.
This list will have convenient properties like:

• :type with possible values melpa, melpa-stable, el-get or quelpa or even submodule (although we definitely don't want them).
• :step with values pre or post to control the loading time.
• :disabled t to mark a package as not activated by default (useful to be able to incorporate contrib layers with only one package in spacemacs layer).

See: bbatsov/projectile#493

After talking with @trishume on Gitter, we will make the .spacemacs 100% optional which should be pretty straightforward.

Was thinking about this, but how about making some spacemacs framework specific snippets to help to standardize creation of the package and init and config functions for the contribution layers?

Should be trivial to fix

I found the built in comment system slightly lacking, evil-nerd-commenter seems far more powerful and mimics Vim functionality. It seems to handle multi line comments such as this far better too:

/**
 * Something.
 *
 * @param {String} name Blah
 */

It will complete the asterisks for you when using <M-j>.

There is defintely something in spacemacs with projectile.

We have this issue: bbatsov/projectile#486 (comment)
And @Wolfy87 reported that projectile-bookmarks does not work as well.

I have my personal overlay within my dotfiles respository, I keep that in my home directory because it's used by multiple applications. I have tried using ln -s ~/dotfiles/emacs/Wolfy87 ~/.emacs.d/contrib/Wolfy87 but it doesn't pick any of my configuration up.

If I copy the directory in it works fine though. I'm running Arch Linux with Emacs 24.4. I'll see if I can work it out too, I'd really like to help maintain this project as I think I'm going to be using it a lot.

When using projectile grep (SPC p g) it creates a grep buffer that includes a 10 line find grep pipeline, which means you can't see any results unless you scroll down. That brings me to the second problem: the buffer isn't focused by default, so you have to switch to it before you can scroll down.

Ideally the find command should be hidden and the buffer should be selected by default. I'll try to fix this myself if this is something that you never use.

Here's the cruft that is printed, except the one long line wraps into around 20 lines:

-*- mode: grep; default-directory: "~/Box/Dev/Projects/MacRanger/" -*-
Grep started at Fri Oct 24 20:52:11

find . -type d \( -path \*/.eunit -o -path \*/.git -o -path \*/.hg -o -path \*/.fslckout -o -path \*/.bzr -o -path \*/_darcs -o -path \*/.tox -o -path \*/.svn -o -path \*/build -o -path \*/SCCS -o -path \*/RCS -o -path \*/CVS -o -path \*/MCVS -o -path \*/_MTN -o -path \*/\{arch\} \) -prune -o \! -type d \( -name TAGS -o -name .\#\* -o -name \*.beam -o -name \*.vee -o -name \*.jam -o -name \*.o -o -name \*\~ -o -name \*.bin -o -name \*.lbin -o -name \*.so -o -name \*.a -o -name \*.ln -o -name \*.blg -o -name \*.bbl -o -name \*.elc -o -name \*.lof -o -name \*.glo -o -name \*.idx -o -name \*.lot -o -name \*.fmt -o -name \*.tfm -o -name \*.class -o -name \*.fas -o -name \*.lib -o -name \*.mem -o -name \*.x86f -o -name \*.sparcf -o -name \*.dfsl -o -name \*.pfsl -o -name \*.d64fsl -o -name \*.p64fsl -o -name \*.lx64fsl -o -name \*.lx32fsl -o -name \*.dx64fsl -o -name \*.dx32fsl -o -name \*.fx64fsl -o -name \*.fx32fsl -o -name \*.sx64fsl -o -name \*.sx32fsl -o -name \*.wx64fsl -o -name \*.wx32fsl -o -name \*.fasl -o -name \*.ufsl -o -name \*.fsl -o -name \*.dxl -o -name \*.lo -o -name \*.la -o -name \*.gmo -o -name \*.mo -o -name \*.toc -o -name \*.aux -o -name \*.cp -o -name \*.fn -o -name \*.ky -o -name \*.pg -o -name \*.tp -o -name \*.vr -o -name \*.cps -o -name \*.fns -o -name \*.kys -o -name \*.pgs -o -name \*.tps -o -name \*.vrs -o -name \*.pyc -o -name \*.pyo \) -prune -o  -type f \( -iname \* -o -iname .\* \) -exec grep -i -nH -e resize {} +

Currently evil-normal-state is the default in spacemacs-mode, which prevent a user from using q to call quit-window function in spacemacs-mode-map to quit that buffer. Will you consider using evil-motion-state in spacemacs-mode buffer by default?

I apologize for what is probably an unhelpfully broad issue, but I installed spacemacs on Windows (by moving my .emacs.d and replacing it with this repo) and basically nothing seems to work as "advertised" in the README. Some of the obvious problems:

• evil-mode doesn't seem to be on and I can't get to normal mode. ESC doesn't take me to normal mode. Neither does fd.
• Adding (evil-mode t) to the dotspacemacs/config function in my ~/.spacemacs does not cause evil-mode to work when I start Emacs either.
• After I turn on evil-mode with M-x evil-mode, none of the bindings in the README work. fd still does not leave insert mode. SPC does not work as a leader, which means none of the other bindings work either. They don't work even if I replace SPC with the evil-leader default of \.
• C-j etc. don't work when I bring up helm-M-x.
• Possibly related, but I'm not sure: Spacemacs says "Found 16 new package(s) to install" every time I start emacs up, which I've done several times today. Surely there aren't actually 16 new packages every time?
• powerline doesn't seem to work. I get a normal emacs modeline.

But there is evidence that parts of spacemacs do work:

• The color scheme is solarized, as expected.
• I get the cute little Ⓗ in the modeline.
• The expected packages do seem to be installed and operational. I can bring up Helm and turn evil-mode on without having to install those packages myself.

But still, none of the fancy things and helpful bindings in the README seem to work at all. Help! What am I missing here?

I'm running emacs 24.4 in Terminal.app on OS X 10.9 (Mavericks). I've run into a series of issues that seem to be related to calling functions not available in terminal mode, starting with this appearing in the Warnings buffer:

Symbol's function definition is void: tool-bar-mode

That's from line 52 of spacemacs/config.el. I changed it to the following which seems to resolve the issue:

(if (functionp 'tool-bar-mode) (tool-bar-mode 0))

Same error on line 56 with scroll-bar-mode, similar fix applied.

That gets me as far as fetching and installing packages, but then this appears:

Symbol's function definition is void: define-fringe-bitmap

That one seemed a bit more involved so I did not attempt a fix.

There's at least one marmalade only package listed in packages.el: python

since the removal of marmalade, install fails.

Produces a strange error:

package "rainbow-mode-" not available

Note the trailing dash.
Does not occur on Mac OS X (in XQuartz).

For now rainbow-mode is disabled.

Disabled EDTS for now until this dependency is pushed to MELPA.

Spend some time making the mode-line responsive to graphical UI and terminals.

First step to improve the documentation is to have a straight to the point readme + wiki pages.

The fd quest continues, surprisingly enough I never noticed that f does not work while entering an ex command.