.. |
lib |
test |
.gitignore |
.travis.yml |
CHANGES.md |
Gemfile |
LICENSE |
README.md |
Rakefile |
slop.gemspec |
Slop is a simple option parser with an easy to remember syntax and friendly API. API Documentation is available here.
opts = Slop.parse do banner 'Usage: foo.rb [options]' on 'name=', 'Your name' on 'p', 'password', 'An optional password', argument: :optional on 'v', 'verbose', 'Enable verbose mode' end # if ARGV is `--name Lee -v` opts.verbose? #=> true opts.password? #=> false opts[:name] #=> 'lee' opts.to_hash #=> {:name=>"Lee", :password=>nil, :verbose=>true}
gem install slop
Slop attempts to build a good looking help string to print to your users. You
can see this by calling opts.help
or simply puts opts
.
All of these options can be sent to Slop.new
or Slop.parse
in Hash form.
strict
- Enable strict mode. When processing unknown options, Slop will
raise an InvalidOptionError
. default: false.help
- Automatically add the --help
option. default: false.banner
- Set the help banner text. default: nil.ignore_case
- When enabled, -A
will look for the -a
option if -A
does not exist. default: false.autocreate
- Autocreate options on the fly. default: false.arguments
- Force all options to expect arguments. default: false.optional_arguments
- Force all options to accept optional arguments.
default: false.multiple_switches
- When disabled, Slop will parse -abc
as the option a
with the argument bc
rather than 3 separate options. default: true.longest_flag
- The longest string flag, used to aid configuring help
text. default: 0.opts = Slop.parse do on :list=, as: Array end # ruby run.rb --list one,two opts[:list] #=> ["one", "two"] # ruby run.rb --list one,two --list three opts[:list] #=> ["one", "two", "three"]
You can also specify a delimiter and limit.
opts = Slop.parse do on :list=, as: Array, delimiter: ':', limit: 2 end # ruby run.rb --list one:two:three opts[:list] #=> ["one", "two:three"]
opts = Slop.parse do on :range=, as: Range end # ruby run.rb --range 1..10 opts[:range] #=> 1..10 # ruby run.rb --range 1...10 opts[:range] #=> 1...10 # ruby run.rb --range 1-10 opts[:range] #=> 1..10 # ruby run.rb --range 1,10 opts[:range] #=> 1..10
Slop has an 'autocreate' feature. This feature is intended to create options on the fly, without having to specify them yourself. In some case, using this code could be all you need in your application:
# ruby run.rb --foo bar --baz --name lee opts = Slop.parse(autocreate: true) opts.to_hash #=> {:foo=>"bar", :baz=>true, :name=>"lee"} opts.fetch_option(:name).expects_argument? #=> true
Slop supports git style sub-commands, like so:
opts = Slop.parse do on '-v', 'Print the version' do puts "Version 1.0" end command 'add' do on :v, :verbose, 'Enable verbose mode' on :name=, 'Your name' run do |opts, args| puts "You ran 'add' with options #{opts.to_hash} and args: #{args.inspect}" end end end # ruby run.rb -v #=> Version 1.0 # ruby run.rb add -v foo --name Lee #=> You ran 'add' with options {:verbose=>true,:name=>"Lee"} and args ["foo"] opts.to_hash(true) # Pass true to tell Slop to merge sub-command option values. # => { :v => nil, :add => { :v => true, :name => "Lee" } }
I'm not, honestly! I love OptionParser. I really do, it's a fantastic library. So why did I build Slop? Well, I find myself using OptionParser to simply gather a bunch of key/value options, usually you would do something like this:
require 'optparse' things = {} opt = OptionParser.new do |opt| opt.on('-n', '--name NAME', 'Your name') do |name| things[:name] = name end opt.on('-a', '--age AGE', 'Your age') do |age| things[:age] = age.to_i end # you get the point end opt.parse things #=> { :name => 'lee', :age => 105 }
Which is all great and stuff, but it can lead to some repetition. The same thing in Slop:
require 'slop' opts = Slop.parse do on :n, :name=, 'Your name' on :a, :age=, 'Your age', as: Integer end opts.to_hash #=> { :name => 'lee', :age => 105 }