| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
| |
Setting these options globally is probably what's been causing me to get
so confused by my indentation level changing to zero when editing other
files in the same Vim instance as a VimL file. This should correct it.
|
|
|
|
|
|
|
|
|
| |
It's misleading to label this target as installing plugins when what it
actually does at a directory level is install the Vim plugin submodules
into ~/.vim/bundle for loading by Pathogen.
This also allows scope for an `install-vim-plugins` target to actually
install into ~/.vim/bundle, if I do need that at some point.
|
|
|
|
|
|
|
|
|
| |
This method short-circuits the unwanted PHP expression-based indenting
configuration completely, rather than running it all and then undoing it
after the fact.
This involves creating a new direction ~/.vim/indent, and a Makefile
target install-vim-indent to copy everything into it.
|
|
|
|
|
|
|
|
|
| |
There's no particular reason to run these file detection rules after the
plugins have run, so we'll put them in a more expected directory.
I've created a new Makefile target to install this,
`install-vim-ftdetect`, which is included as a prerequisite of the
`install-vim` target.
|
|
|
|
|
|
|
|
|
|
| |
There's no particular need to set these options after netrw has loaded;
they can be set before it's loaded, and the plugin will still observe
them.
This empties the vim/after/plugin directory. It doesn't need to be
removed from the Makefile as there are no references to it; it was
installed by a glob cp(1).
|
|
|
|
|
|
|
| |
Commit feba460 set up the dist script helper for generating this, but
didn't actually run it, committing a version of the manual that included
the manual page header twice (with leading percent characters).
Re-running the dist script helper corrected this.
|
|
|
|
|
|
| |
See 2b13fc, where this document was first created to avoid an expensive
branch, and 1279419, where the branch for a multiple-file Vim
configuration (splitting my large .vimrc) was merged in.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
On an ancient Vim (6.1), the block of code checking the value of
COLORFGBG does not work at all, raising the following error output:
Error detected while processing function DetectBackground:
line 3:
E117: Unknown function: split
E15: Invalid expression: split($COLORFGBG, ';')
line 6:
E121: Undefined variable: l:colorfgbg
E15: Invalid expression: len(l:colorfgbg) ? l:colorfgbg[-1] : ''
line 9:
E121: Undefined variable: l:bg
E15: Invalid expression: l:bg == 'default' || l:bg == '7' || l:bg == '15'
line 11:
:else without :if: else
line 13:
:endif without :if: endif
I'm unlikely to need such an ancient Vim very often, so I've simply
added an error guard around the block.
|
| |
|
|\ |
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Now that this environment variable is kept and updated in tmux after
54553ae, we should be able to either configure terminals or explicitly
set it during startup if we want to use lighter terminals.
I'm much more comfortable with this than simply hardcoding it in the
configuration.
This doesn't solve the problem of carrying the environment variable over
an SSH session, however, but I'm not really sure there's a solution to
that besides configuring sshd(8) itself to accept these variables in
transit.
|
| |
| |
| |
| |
| |
| |
| | |
Stop stripping them from the session on startup, and update them in the
environment when new windows are reattached. This will allow me to use
light-background terminals in future if I need to and have Vim
accommodate it for syntax highlighting, at least in new windows.
|
|/
|
|
|
|
|
|
|
|
|
|
| |
This is in preparation for config in vim/config/syntax.vim that will do
a more comprehensive job of applying heuristics to figure out if the
background is light or dark and hence what colours should be loaded for
the appropriate scheme.
The test for the GUI or 256 colours is repeated in the colorscheme code
itself, but I think that's OK given that sahara.vim is distributed
separately and others probably wouldn't use the kind of guards
introduced in this commit.
|
| |
|
|
|
|
|
|
|
| |
While it would be ideal to keep the indentation-related configuration in
the config/indent.vim file, that approach ends up double-loading
filetype.vim from the core, so we'll merge them into a single call in
config/file.vim instead.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Per this suggestion from the `vim-pathogen` FAQ:
<https://github.com/tpope/vim-pathogen#faq>
>>Can I put pathogen.vim in a submodule like all my other plugins?
>
>Sure, stick it under `~/.vim/bundle`, and prepend the following to
>your vimrc:
>
> runtime bundle/vim-pathogen/autoload/pathogen.vim
This method avoids using symbolic links, which is desirable in general,
and also removes the need for the `install-vim-pathogen` dependency of
the `install-vim-plugin` target, since this is now done in Vim
configuration.
This also takes away another of the steps required for setting up the
Vim configuration on Windows.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Commit 70fcb35 incorrectly built the dotfiles(7) manual without the
header required by the Pandoc converter. Rebuilt it properly by making a
script in a new directory "dist" which is to be run by the maintainer
whenever its source file README.md is updated. This should probably be
automated on my end with Git hooks.
The reason we don't include the Pandoc recipe for making this manual as
a target in the Makefile is to do with the heavy dependency of Pandoc,
for which packages are not available on some desirable operating
systems, as arranged in a8ab2cf.
Also added the new install-man target as one of the default subtargets
of `install`.
|
| |
|
|
|
|
| |
Mostly to reflect its split into subfiles in ~/.vim/config.
|
|\
| |
| |
| |
| | |
This development branch split .vimrc into many logically-grouped
subfiles in ~/.vim/config/*.vim.
|
| |
| |
| |
| |
| |
| |
| | |
This is an awkward filename and very unlikely to ever have anything but
this one setting in it, but I can't think of any logical other place to
put it. number.vim applies to line numbering, which is a distinct
concept.
|
| | |
|
| |
| |
| |
| |
| |
| | |
This is an awkward filename and very unlikely to ever have anything but
this one setting in it, but I can't think of any logical other place to
put it.
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
| |
| |
| | |
Just the 'shortmess' setting for now.
|
| |
| |
| |
| |
| |
| | |
Per 6ca11a5, I've confirmed I do still need this, otherwise the default
colorschemes (not sahara.vim) assume a bright background and show very
dark colours.
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Some refactoring is done here, because as noted in 5caa13c, my custom
colorscheme is implemented as a plugin to be loaded by Pathogen, and
hence isn't available into after it's done its work.
I've removed the :set background line for now until I'm sure it's
needed, because at the moment I'm not sure.
|
| |
| |
| |
| |
| |
| | |
Only one setting at the moment, but there's enough completion stuff even
just in core Vim that I'm barely using, so this could be expanded upon
later on.
|
| | |
|
| | |
|
| | |
|
| |
| |
| |
| |
| | |
Not completely sure this grouping is meaningful; I may refactor it a bit
more later on.
|
| |
| |
| |
| |
| |
| | |
"Matching" here refers to using % as a motion to the matching character
or closing statement for a block, as enabled by Vim and enhanced by the
optional matchit.vim included with the distribution.
|
| | |
|
| | |
|
| |
| |
| |
| |
| | |
A little bit iffy on the grouping here, but it's still better than
having it all lumped in the one file.
|
| |
| |
| |
| |
| | |
This file is rather short; it may turn out to make more sense to put
these settings elsewhere a bit later.
|
| | |
|
| |
| |
| |
| | |
The StripTrailingWhitespace() function should perhaps be its own plugin.
|
| |
| |
| |
| |
| | |
By "list" here I am referring to options for Vim's 'list' display
setting, showing control characters visually.
|
| |
| |
| |
| |
| |
| | |
Also add a note to IDEAS.md for later to consider packaging this as a
proepr plugin, even if it doesn't actually leave the dotfiles repository
just yet.
|
| |
| |
| |
| |
| | |
The ToggleFormatFlag function might actually be better implemented as
some sort of plugin.
|
| |
| |
| |
| |
| |
| | |
I'm not quite so sure about this one. The ToggleBreak() function might
actually be better in a plugin on its own. The rest of it makes sense
though.
|
| |
| |
| |
| | |
Not the operating system; Vim editor windows.
|
| | |
|