Adds initial basic support for HCP based configuration in vagrant-go.
The initalization process has been updated to remove Vagrantfile parsing
from the client, moving it to the runner using init jobs for the basis
and the project (if there is one). Detection is done on the file based
on extension for Ruby based parsing or HCP based parsing.
Current HCP parsing is extremely simple and currently just a base to
build off. Config components will be able to implement an `Init`
function to handle receiving configuration data from a non-native source
file. This will be extended to include a default approach for injecting
defined data in the future.
Some cleanup was done in the state around validations. Some logging
adjustments were applied on the Ruby side for better behavior
consistency.
VirtualBox provider now caches locale detection to prevent multiple
checks every time the driver is initialized.
This also removes the "Running (command)..." output since users will
expect to be able to redirect the ssh-config output to a file without
that noise.
If we really like the "Running" thing we can try and swing around and
get fancy by detecting whether or not we have a terminal or a file as
output and doing conditional stuff, but this seemed like the simplest
way forward for now.
Found this while running `./vagrant box` with no args to get the help
output. It turns out you can have an empty RunResult but also a nil
error. I took the occasion to unwind the conditional tree a bit which
hopefully makes it a bit easier to read.
In legacy Vagrant, any exception raised that's a subclass of
Vagrant::Errors::VagrantError is considered user-facing and so causes
the error message to be printed to the console and the process to use
exit code 1. Anything outside of that causes the process to use exit
code 255. (See `bin/vagrant` for the code.)
Here we mirror that behavior by treating errors that have a
LocalizedMessage as user-facing and those without as unexpected. This
allows the basic virtualbox component to pass in vagrant-spec!
* `vagrant ssh` has a --no-tty flag colliding with the one defined in
`bin/vagrant` - in fact none of the flags in `bin/vagrant` are
processed in `serve` mode, so remove the code that captures them from
the CommandInfo OptionParser dance
* `vagrant ssh` has a `--plain` flag colliding with the one defined in
`internal/cli/base.go` - this flag was inherited from Waypoint, so we
can just rename it to line up with the (inversely defined) `--color`
flag used in legacy vagrant
* Makes all help output showup on stdout like in rubyland
* Moves verbosity flags to capital Vs to get out of the way of version
* Passes version down to mitchellh/cli so it handles -v and --version
flags
* Tweaks version output to remove the leading `v` which is not output
in legacy vagrant
Having the modules stored in `./vendor` causes issues with `go mod`.
Follow waypoint's convention to store in `./thirdparty` and grab
the same Makefile modifications to alert when the submodules need
to be initialized. Update generators to use new path.
There are still some things left to address like how to handle
a target (or targets) and if that is something we should even
be doing. It may be best to just let the command deal with
target loading. There are also some considerations to make
around remote source.
