In 16219a31db we implemented a post-up
hook that checked for a problematic `/etc/zshenv` file. OSX/macOS has
changed the file responsible for executing `path_helper` and loading the
system paths multiple times, and this script change issued a warning to
users if `path_helper` might be called in such a way that their paths
could be unexpectedly reordered.
The script used colored output to highlight this problem for the user,
and relied on Bash's script path introspection (through `$BASH_SOURCE`)
to print the location of the `post-up` hook to assist the user in
troubleshooting and understanding where our warning came from.
This change removes the hook's reliance on GNU Bash in favor of `sh`. It
uses the POSIX-compliant `$0` to determine the hook location (this
method is less robust than Bash's `$BASH_SOURCE`, but is available in
all POSIX shells) and removes colored output in the warning to be
compatible with shells that don't support colored output.
Fix#517.
It seems like JavaScript/ES6 is becoming more common for us to run into
in the wild compared to CoffeeScript
with the advent of Babel / transpiling.
Eventually, we'll probably want to remove kchmck/vim-coffee-script.
Users working on CoffeeScript applications would need to move that
plugin into their personal dotfiles.
Some tmux users have reported that when running `rcup` that during the
post-up hook Vim would issue errors about input not coming from a
terminal and then the terminal would look and behave unexpectedly.
Running `reset` after the Vim plugin install step in `post-up` appears
to resolve the issue.
Fix#516.
The `UnPlug` command allows a user to remove a plugin from vim-plug's
list of plugins to install/activate. This command can be used to
selectively ignore shared plugins from `vimrc.bundles` without needing
to overwrite the entire file with a personal replacement.
This change is dependent on vim-plug continuing to use the `g:plugs`
global variable.
Why:
I occasionally will try to zoom in on a pane with the key combination of
C-s then z. However, instead of having the agility to remove my finger
from the ctrl key, I clumsily continue to have it depressed while
applying pressure to the z key. This results in the `suspend-client`
command being executed by tmux. I am then left not knowing what to do
and I can't reattach to the session the that terminal window. It's very
disconcerting and I wish it hadn't happened.
This PR:
Removes the binding for `suspend-client`.
I added this four years ago but haven't been using it.
Not common enough to be included in company-wide dotfiles?
Can move to personal dotfiles if desired.
This change prevents the `post-up` hook from breaking the terminal for
the rest of the session when running `rcup`. Since the session is
completely automated, we don't need input from the terminal once Vim
launches.
In #488 there was some confusion over where to clone the dotfiles
repository before running `rcm`. While the repository can be placed
anywhere, by default users should be placing it in `~/dotfiles` if they
are using the default installation instructions and an unmodified
`.rcrc`.
Close#488.
The use case for push-line-or-edit:
1. You are in the middle of typing a long command, perhaps something
involving `tar`.
2. You realize that you must first run another command, such as `man
tar`.
3. You invoke `push-line-or-edit`, which gives you a fresh prompt. You
type `man tar` and read as needed.
4. When the command (`man tar`) finishes, the long command you
half-typed is back, exactly as you left off.
Other use cases are realizing that you're in the wrong directory before
you press enter, changing your Ruby version before you press enter on
that `bundle` command, or in general being partway through something and
realizing that you're not quite ready to run it yet.
While here, turn off start/stop output control, which gives us back the
^Q keybinding.
There are, apparently, more things we need to configure. @BlakeWilliams shared
http://dance.computer.dance/posts/2015/04/using-ctags-on-modern-javascript.html
which outlines a bunch more things that don't work out of the box. For
example, ES2015 classes and their methods are not properly found, even
with this PR. There's more work to do here.
After using byebug on any Ruby file, the sessions history is written to
this file. It shouldn't be part of our repositories, so we ignore it.
See also: https://github.com/thoughtbot/suspenders/pull/710
Problem:
Currently the `for` in bash will break up lists by whitespace, so if a
filename has a space in it the `for` will break it up incorrectly.
AG will separate the files with a newline character, so this temporarily
overwrites the `IFS` to be used correctly for this use case.
[Docs](https://bash.cyberciti.biz/guide/$IFS)
* `noexpandtab` will ensure tabs are used instead of staces when tabbing
or auto-indenting.
* `softtabstop=2` isn't needed because we are already setting
`tabstop=2` in our `.vimrc`.
We do most of our client work on Github,
and a large portion of our personal projects
are hosted there.
fugitive.vim currently supports
opening the current file in the current branch
in a browser at Github through the `Gbrowse` command.
That command is [slated to be removed to make fugitive
agnostic about coding sites](fugitive-docs).
The capability to quickly generate or view a Github URL
specific to the current branch
can be useful in PR reviews
or other discussions about code.
This change adds rhubarb.vim,
a plugin that implements the slated-to-be-removed
functionality of fugitive that works with Github.
It also provides the user with omni-completion (`<C-x><C-o>`)
to autocomplete Github issues, issue URLs, and collaborators.
[fugitive-docs]: 18d6d1ab82/doc/fugitive.txt
vim-rspec is an excellent Vim plugin for running RSpec tests but there
are many of us who run different kinds of tests on any given day.
vim-test is built to support a wide array of test frameworks and
strategies for executing those tests.
vim-test ships with support for several languages and frameworks in use
on thoughtbot projects today:
* Ruby (both RSpec and Minitest)
* Elixir (ExUnit and espec)
* Go (gotest)
* Python (nose, pytest, djangotest, djangonose)
* JavaScript (jasmine, mocha)
It also supports several execution strategies I have seen people use at
thoughtbot:
* Dispatch
* vim-tmux-runner
* Tslime
* Terminal.app/iTerm.app tabs
* neovim's native terminal
* `:!`
vim-test is extensible, both for additional languages/frameworks and for
execution contexts. See: https://github.com/janko-m/vim-test
The first time a `.py` file is run, or any time it's been modified, Python
creates a `.pyc` file containing a bytecode representation.
These bytecode files should always be ignored by Git, since keeping them
updated in a repo would be time consuming and would serve no purpose.
Add suggested Go configuration for Vim into the `ftplugin` folder and
the vim-go plugin. We may use this as a template for adding further
language-specific settings in the future for languages that aren't
universally used by the team.
* `fatih/vim-go`: The de-facto Golang Vim plugin.
* `softtabstop=2`: As a side effect of setting how "large" a tab is in
insert mode, Vim is able to delete tabs with the backspace key in
Insert mode.
* `compiler go`: Necessary for plugins using the `make` command, such as
vim-dispatch.
* `listchars=tab:\ \ `: Don't highlight tabs as "»·" as if they were
problem characters in go files. Also recommended for *sh files and any
others which which should also be using tabs.
* `go_fmt_command = "goimports"`: Tell vim-go to use the `goimports`
command rather than `gofmt`, which attempts to find dependencies and
adds them to your import statement.
When using ctrl-p without ag, the pattern given is not treated as a regular expression. Additionally, the value passed when first opening ctrl-p is not actually a valid PCRE regex, and produces an error when used on Windows. This modifies the options to treat the pattern as a string literal, not a regular expression, resolving the Windows problem, and better matching the behavior of ctrl-p without Ag.
Vim's documentation advises that buffer-specific configuration based on
a filetype should be performed via a file in `$HOME/.vim/ftplugin`:
http://vimdoc.sourceforge.net/htmldoc/usr_41.html#41.12
This change moves CSS/Sass-, Git commit message-, and Markdown-related
editor settings into `ftplugin/gitcommit.vim` and
`ftplugin/markdown.vim`, and rewrites those settings to avoid using
`autocmd`, letting Vim execute `setlocal` when a file of the filetype is
determined or set.
The existing implementation of the `replace` script does not work on
GNU/Linux due to differences in the implementation of `sed`. This change
reworks the process of the replacement to create temporary files,
apply the changes to the files, and then move them into place. It will
improve the script's portability for more OSes.
When using a join command (`gq`, `J`), vim defaults to adding two spaces
after '.', '?', and '!'.
Setting `nojoinspaces` causes vim to only insert a single space.
In cbdcbce01d we introduced a Tim
Pope-inspired Git template scheme that automatically ran `ctags`
asynchronously through a hook after certain source control operations.
This hook is bypassed if a user uses `<leader>ct` to reindex their tags
from within Vim.
This change binds `<leader>ct` to use a Vimscript function that will
identify if the `ctags` hook exists in the local repository and executes
it. The function falls back to existing functionality if the hook does
not exist. We are moving the function out of `vimrc` as a first step
toward breaking our Vim configuration up by topic area.
We removed `db:test:prepare` from the `migrate` alias in #260 because
Rails 4.1 discouraged users from running the deprecated (at the time)
Rake task. Rails brought back `db:test:load` in rails/rails#17739 due to
user complaints about being unable to force a test database
synchronization.
Without dropping and recreating the test database from scratch, the
current migration strategy of solely using `db:migrate` and
`db:rollback` will never bring changes into test from a migration that
was run, rolled back, modified, and then re-run.
Running `db:test:load` or `db:test:prepare` on each migrate operation
has a small performance penalty, but the task is only being run when you
have a reason to want to check or force a synchronization of the
database. Knowing for sure that your test and development databases are
at the same point in their evolution is worthwhile.
Running `db:test:prepare` or `db:test:load` after running our existing
migrate alias added just under 1.0 seconds on average to the migrate
operation on an application with a reasonably sized DB schema.
diff --git a/aliases b/aliases index 6a0f602..102caca 100644 ---
a/aliases +++ b/aliases @@ -9,7 +9,7 @@ alias v="$VISUAL" alias
b="bundle"
# Rails
-alias migrate="rake db:migrate db:rollback && rake db:migrate"
+alias migrate="rake db:migrate db:rollback && rake db:migrate db:test:prepare"
alias s="rspec"
# Pretty print the path
This solution was heavily influenced by @keith's original contribution
in #356, which I think may have worked on some but not all setups due to
personal configuration differences.
The solution in #356 added the `_git_delete_branch` function in the
`zsh/completions` folder, which resulted in the tab-completion working
against `git`, but not against our `g` function. With our new loading
order where functions are loaded first and completions are loaded last,
we can load the `_git_delete_branch` function before any completions are
defined on `git`, `g`, or `hub`, allowing the `_git_delete_branch` shim
to be used by all three of those git-invoking commands/functions.
Close#355.
The `$PS1` variable does not need to be exported in order for `zsh` to
use its contents to set prompt options. `export`ing it has the negative
side-effect of polluting prompts in shells launched from `zsh`. The
`export`ing of `$PS1` is the cause of #270, where using `sudo su` to
become another user (launching a non-`zsh` shell) resulted in a
corrupted prompt.
Removing the `export` will allow `sh`/`bash`/`dash` and other shells to
be launched from `zsh` without inheriting its prompt settings.
Fix#270.
In f7c73f7c1b we started to deduplicate
`$PATH` using `typeset -U`, but that did not prevent duplicate `$PATH`
entries when processes were launched that inherited the environment from
an existing shell.
Using `export -U` keeps the `$PATH` deduplicated even when tmux launches
a new shell.
Fix#443.
Why:
* Remove dupes from PATH
How:
$PATH is a scalar composed of contents of $path, so it uses typeset -U
("U" as in Unique) to ensure there are no dupes. It preserves the first
occurrence in the array, since it would be searched first before others
when matching a command in the directories in the PATH anyway.
This allows people to incorporate the thoughtbot dotfiles into their own
dotfiles in a more fine-grained manner.
I left some things in zshrc that we should eventually handle more
precisely:
- Load `.zsh/functions/*`. This could instead be replaced with: `mv
.zsh/functions/* .zsh/configs`.
- Load `.aliases`. This could instead be replaced with: `mv .aliases
.zsh/configs/aliases.zsh`.
- Load `.zshrc.local`. This file can realistically go away entirely,
with people adding their own files to `.zsh/configs`.
A further refactoring, which I have done locally, is to introduce a
`~/.sh/configs` directory, in which people can put POSIX-specific
configuration that can be shared between GNU Bash, zsh, ksh, etc:
aliases, functions, paths, prompts, and so on. But one step at a time.
Other changes:
* Move aliases setup to occur after loading other config, as some of our
aliases depend on environment variables having been set, so alias
loading must come last after we've sourced `zsh/configs`.
* Move autocompletion for `g` function from the function definition to
to `zsh/completions/_g`
* Move `PATH` setup to `zsh/configs/post` to ensure it happens after
other configuration that might alter the `PATH`
Unfortunately, the convention for most JavaScript linters is to name
their configuration `.${TOOL}rc`.
In spite of this name, the configuration files are `JSON`. Forcing vim
to recognize them as JSON would enable syntax highlighting and linting
plugins (like syntastic) to recognize invalidly structured JSON.
