separate `launchctl unload` and comment that this command was
intended for unloading jobs by pathname rather than bundle ID.
Leave that undocumented for now as it is untested.
The `Hbc.outdated` boolean was in the way of my implementation of an
`outdated` scope for `brew cask outdated`, so I took the opportunity to:
* rename the flag internally to `cleanup_outdated`
* move the cleanup tests over to rspec and rework them to test in a
stubbed cleanroom environment
* refactor the implementation of cleanup to make it a bit more
testable - most importantly: inject the cache location and outdated
config dependencies instead of looking them up from constants
Note that there's no change to the user-facing interface, it's still
```
brew cask cleanup --outdated
```
Hbc is the namespace, Hbc::Cask is the object representing a Cask
One step on the path to reducing the surface area of the God-object
we've grown over the years. :)
and corresponding testing_env.rb.
* recast HOMEBREW_BREW_FILE as Hbc.homebrew_executable,
defined in Hbc::Locations
* recast HOMEBREW_REPOSITORY as Hbc.homebrew_repository
* recast HOMEBREW_PREFIX as Hbc.homebrew_prefix
* remove HOMEBREW_LIBRARY
* recast existing Hbc.tapspath as Hbc.homebrew_tapspath
to match new methods
fixes#8705
* convert existing Cask:: namespace to Hbc::
* move Homebrew-fork code under Hbc::
* move freestanding classes such as Tty and TopologicalHash under Hbc::
* recast HOMEBREW_CASK_ constants as HBC_
* modify our Homebrew Formula for backward compatibility
* devscripts and dev docs
* complete transition from MacOS#version to MacOS#release
throughout backend code
* recast related constants and comments as "release" instead
of "version", and "point release" instead of "full version"
* perform actual checksum calculation within download.rb
* use an accessor named sha256 instead of sums for checksum data
* migrate checksum-specific exceptions out of homebrew-fork
* accepts a Boolean value
* replaces the never-documented `postflight` methods `enable_accessibility_access` /
`disable_accessibility_access` (`remove_accessibility_access`)
* declarative: unlike the `postflight` approach, does not have to be
invoked in two places
* de-document obsolete form `caveats` / `assistive_devices`
* bump Travis seed to avoid false test failures
When invoking a `SystemCommand` with `:must_succeed => false`, the `SystemCommand::Result` class would mistake a `Process.Status` object for a `Fixnum`.
This commit fixes this by instantiating `Result` with the actual status code as a number.