Rib

by Lin Jen-Shin (godfat)

• github
• rubygems
• rdoc
• issues (feel free to ask for support)

Ruby-Interactive-ruBy -- Yet another interactive Ruby shell

Rib is based on the design of ripl and the work of ripl-rc, some of the features are also inspired by pry. The aim of Rib is to be fully featured and yet very easy to opt-out or opt-in other features. It shall be simple, lightweight and modular so that everyone could customize Rib.

• Tested with MRI (official CRuby) and JRuby.
• All gem dependencies are optional, but it's highly recommended to use Rib with bond for tab completion.

gem install rib

As IRB (reads ~/.rib/config.rb writes ~/.rib/history.rb)

rib

As Rails console

rib rails

You could also run in production and pass arguments normally as you'd do in rails console or ./script/console

rib rails production --sandbox --debugger

Note: You might need to add ruby-debug or ruby-debug19 to your Gemfile if you're passing --debugger and using bundler together.

For Rails Spring support, put this line in your ~/.spring.rb:

require 'rib/extra/spring'

As Rack console

rib rack

As a console for whichever the app in the current path it should be (for now, it's either Rails or Rack)

rib auto

If you're trying to use rib auto for a Rails app, you could also pass arguments as if you were using rib rails. rib auto is merely passing arguments.

rib auto production --sandbox --debugger

As a fully featured interactive Ruby shell (as ripl-rc)

rib all

As a fully featured app console (yes, some commands could be used together)

rib all auto # or `rib auto all`, the order doesn't really matter

You can customize Rib's behaviour by setting a config file located at $RIB_HOME/config.rb, or ./.rib/config.rb, or ~/.rib/config.rb, or ~/.config/rib/config.rb, searched by respected order. The default would be ~/.rib/config.rb. Since it's merely a Ruby script which would be loaded into memory before launching Rib shell session, You can put any customization or monkey patch there. Personally, I use all plugins provided by Rib.

My Personal ~/.config/rib/config

As you can see, putting require 'rib/all' into config file is exactly the same as running rib all without a config file. What rib all would do is merely require the file, and that file is also merely requiring all plugins, but without extra plugins, which you should enable them one by one. This is because most extra plugins are depending on other gems, or hard to work with other plugins, or having strong personal tastes, so you won't want to enable them all. Suppose you only want to use the core plugins and color plugin, you'll put this into your config file:

require 'rib/core'
require 'rib/more/color'

You can also write your plugins there. Here's another example:

require 'rib/core'
require 'pp'
Rib.config[:prompt] = '$ '

module RibPP
  Rib::Shell.send(:include, self)

  def format_result result
    result_prompt + result.pretty_inspect
  end
end

So that we override the original format_result to pretty_inspect the result. You can also build your own gem and then simply require it in your config file. To see a list of overridable API, please read api.rb

While it's convenient to just require 'rib/all', you might not want to use all the plugins. No worries, you don't have to list everything in order to not use something. For example, you might not get used to bottomup_backtrace and don't want to use it. You could put this in your config:

require 'rib/all'

Rib::BottomupBacktrace.disable

This could disable bottomup_backtrace so you get regular top-down backtrace with all other plugins. This is particularly useful whenever there's a bug in one of the plugins, and you might need to disable some plugins in order to debug. You could always enable it again with:

Rib::BottomupBacktrace.enable

You could do this any time, in the config, or in the shell session. No need to restart anything, because it takes effect immediately.

Rib home is used to store a config file and a history file, which is searched in this order:

• $RIB_HOME
• ./.rib
• ~/.rib
• ~/.config/rib

Rib would stop searching whenever the directory is found. If none could be found, the default would be:

• ~/.rib

So the default history file would be located at ~/.rib/history.rb.

Since ./.rib would be searched before ~/.rib, you could create project level config at the project directory, and the history would also be separated from each other, located at the respected ./.rib/history.rb.

To do this, you don't really have to create a project config. Creating an empty directory for Rib home at the project directory would also work.

You could set the project directory by using -p, --prefix command line option. So consider this:

cd ~/project
rib auto

Would work the same as:

cd /tmp
rib -p ~/project auto

And the project config and history would be located at ~/project/.rib.

To check for more command line options, run rib -h:

Usage: rib [ruby OPTIONS] [rib OPTIONS] [rib COMMANDS]
ruby options:
  -e, --eval LINE        Evaluate a LINE of code
  -d, --debug            Set debugging flags (set $DEBUG to true)
  -w, --warn             Turn warnings on (set $-w and $VERBOSE to true)
  -I, --include PATH     Specify $LOAD_PATH (may be used more than once)
  -r, --require LIBRARY  Require the library, before executing your script
rib options:
  -c, --config FILE      Load config from FILE
  -p, --prefix PATH      Prefix to locate the app. Default to .
  -n, --no-config        Suppress loading ~/.config/rib/config.rb
  -h, --help             Print this message
  -v, --version          Print the version
rib commands:
  all                    Load all recommended plugins
  auto                   Run as Rails or Rack console (auto-detect)
  min                    Run the minimum essence
  rack                   Run as Rack console
  rails                  Run as Rails console

require 'rib/core' # You get all of the followings:

• require 'rib/core/completion'

  Completion from bond.

• require 'rib/core/history'

  Remember history in a history file.

• require 'rib/core/strip_backtrace'

  Strip backtrace before Rib.

• require 'rib/core/readline'

  Readline support.

• require 'rib/core/multiline'

  You can interpret multiple lines.

• require 'rib/core/squeeze_history'

  Remove duplicated input from history.

• require 'rib/core/last_value'

  Save the last result in Rib.last_value and the last exception in Rib.last_exception.

require 'rib/more' # You get all of the followings:

• require 'rib/more/multiline_history_file'

  Not only readline could have multiline history, but also the history file.

• require 'rib/more/bottomup_backtrace'

  Show backtrace bottom-up instead of the regular top-down.

• require 'rib/more/color'

  Class based colorizing.

• require 'rib/more/multiline_history'

  Make readline aware of multiline history.

• require 'rib/more/beep'

  Print "\a" when the application was loaded and it's been too long. Configure the threshold via Rib.config[:beep_threshold].

• require 'rib/more/anchor'

  See As a debugging/interacting tool.

• require 'rib/more/caller'

  See Current call stack (backtrace, caller).

• require 'rib/more/edit'

  See In place editing.

There's no require 'rib/extra' for extra plugins because they might not be doing what you would expect or want, or having an external dependency, or having conflicted semantics.

• require 'rib/extra/autoindent' This plugin is depending on:

  1. readline_buffer
  2. readline plugin
  3. multiline plugin

  Which would autoindent your input.

• require 'rib/extra/hirb' This plugin is depending on:

  1. hirb

  Which would print the result with hirb.

• require 'rib/extra/paging' This plugin is depending on less and tput.

  Which would pass the result to less (or $PAGER if set) if the result string is longer than the screen.

• require 'rib/extra/spring' in your ~/.spring.rb for Rails Spring support.

Rib could be used as a kind of debugging tool which you can set break point in the source program.

require 'rib/config' # This would load your Rib config
require 'rib/more/anchor'
                     # If you enabled anchor in config, then needed not
Rib.anchor binding   # This would give you an interactive shell
                     # when your program has been executed here.
Rib.anchor 123       # You can also anchor on an object.

But this might be called in a loop, you might only want to enter the shell under certain circumstance, then you'll do:

require 'rib/debug'

Rib.enable_anchor do
  # Only `Rib.anchor` called in the block would launch a shell
end

Rib.anchor binding # No effect (no-op) outside the block

Anchor could also be nested. The level would be shown on the prompt, starting from 1.

Often time we would want to see current call stack whenever we're using Rib.anchor. We could do that by simply using caller but it's barely readable because it's just returning an array without any format and it also contains backtrace from Rib itself. You could use pretty formatting with Rib:

require 'rib/more/caller'

Rib.caller

It would use the same format for exception backtrace to show current call stack for you. Colors, bottom up order, etc, if you're also using the corresponding plugins.

Sometimes there are also too many stack frames which we don't care about. In this case, we could pass arguments to Rib.caller in order to filter against them. You could either pass:

• A String represents the name of the gem you don't care
• A Regexp which would be used to match against paths/methods you don't care

Examples:

require 'rib/more/caller'

Rib.caller 'activesupport', /rspec/

To remove backtrace from gem activesupport and paths or methods containing rspec as part of the name, like things for rspec or rspec-core and so on. Note that if a method name also contains rspec then it would also be filtered. Just keep that in mind when using regular expression.

Or if you don't care about any gems, only want to see application related calls, then try to match against %r{/gems/} because gems are often stored in a path containing /gems/:

Rib.caller %r{/gem/}

Happy debugging.

Whenever you called:

require 'rib/more/edit'

Rib.edit

Rib would open an editor according to $EDITOR (ENV['EDITOR']) for you. By default it would pick vim if no $EDITOR was set. After save and leave the editor, Rib would evaluate what you had input. This also works inside an anchor. To use it, require either rib/more/edit or rib/more or rib/all.

The essence is:

require 'rib'

All others are optional. The core plugins are lying in rib/core/*.rb, and more plugins are lying in rib/more/*.rb. You can read rib/app/rack.rb and bin/rib-rack as a Rib App reference implementation, because it's very simple, simpler than rib-rails.

• rest-more rib rest-core Run as interactive rest-core client
• rib-heroku rib heroku Run console on Heroku Cedar with your config

• Andrew Liu (@eggegg)
• ayaya (@ayamomiji)
• Lin Jen-Shin (@godfat)
• Mr. Big Cat (@miaout17)
• @alpaca-tc
• @bootleq
• @lulalala
• @MITSUBOSH
• @tka

Apache License 2.0 (Apache-2.0)

Copyright (c) 2011-2023, Lin Jen-Shin (godfat)

Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at

https://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.