rack-sanitizer 2.0.3 → 2.1.0

data/bin/rubocop

@@ -0,0 +1,27 @@
+#!/usr/bin/env ruby
+# frozen_string_literal: true
+
+#
+# This file was generated by Bundler.
+#
+# The application 'rubocop' is installed as part of a gem, and
+# this file is here to facilitate running it.
+#
+
+ENV["BUNDLE_GEMFILE"] ||= File.expand_path("../Gemfile", __dir__)
+
+bundle_binstub = File.expand_path("bundle", __dir__)
+
+if File.file?(bundle_binstub)
+  if File.read(bundle_binstub, 300).include?("This file was generated by Bundler")
+    load(bundle_binstub)
+  else
+    abort("Your `bin/bundle` was not generated by Bundler, so this binstub cannot run.
+Replace `bin/bundle` by running `bundle binstubs bundler --force`, then run this command again.")
+  end
+end
+
+require "rubygems"
+require "bundler/setup"
+
+load Gem.bin_path("rubocop", "rubocop")

data/dev.yml

@@ -0,0 +1,17 @@
+name: rack-sanitizer
+
+type: ruby
+nix: true
+
+up:
+  - ruby
+  - bundler
+
+commands:
+  test:
+    syntax:
+      argument: file
+      optional: args...
+    run: bundle exec rake spec
+  style:
+    run: bin/rubocop

data/gemfiles/rack_2.2.gemfile

@@ -0,0 +1,9 @@
+# frozen_string_literal: true
+
+source "https://rubygems.org"
+
+gemspec path: ".."
+gem "rubocop-shopify"
+gem "rake"
+gem "ostruct"
+gem "rack", "~> 2.2.0"

data/gemfiles/rack_3.0.gemfile

@@ -0,0 +1,9 @@
+# frozen_string_literal: true
+
+source "https://rubygems.org"
+
+gemspec path: ".."
+gem "rubocop-shopify"
+gem "rake"
+gem "ostruct"
+gem "rack", "~> 3.0.0"

data/gemfiles/rack_3.1.gemfile

@@ -0,0 +1,9 @@
+# frozen_string_literal: true
+
+source "https://rubygems.org"
+
+gemspec path: ".."
+gem "rubocop-shopify"
+gem "rake"
+gem "ostruct"
+gem "rack", "~> 3.1.0"

data/gemfiles/rack_3.2.gemfile

@@ -0,0 +1,9 @@
+# frozen_string_literal: true
+
+source "https://rubygems.org"
+
+gemspec path: ".."
+gem "rubocop-shopify"
+gem "rake"
+gem "ostruct"
+gem "rack", "~> 3.2.0"

data/lib/rack/sanitizer.rb

@@ -7,9 +7,11 @@ module Rack
   class Sanitizer
     # options[:sanitizable_content_types] Array
     # options[:additional_content_types] Array
-    def initialize(app, options={})
+    def initialize(app, options = {})
       @app = app
-      @strategy = build_strategy(options)
+      strategy = options.fetch(:strategy) { :replace }
+      @strategy = build_strategy(strategy)
+      @exception_strategy = strategy == :exception
       @sanitizable_content_types = options[:sanitizable_content_types]
       @sanitizable_content_types ||= SANITIZABLE_CONTENT_TYPES + (options[:additional_content_types] || [])
     end
@@ -17,25 +19,28 @@ module Rack
     def call(env)
       env = sanitize(env)
       begin
+        if @exception_strategy && env['rack.input'].is_a?(SanitizedRackInput)
+          env['rack.input'].ensure_sanitized
+        end
         @app.call(env)
       rescue SanitizedRackInput::FailedToReadBody
-        return [400, { "Content-Type" => "text/plain" }, ["Bad Request"]]
+        [400, { "content-type" => "text/plain" }, ["Bad Request"]]
       end
     end
 
     DEFAULT_STRATEGIES = {
       replace: lambda do |input|
-        input.
-          force_encoding(Encoding::ASCII_8BIT).
-          encode!(Encoding::UTF_8,
-                  invalid: :replace,
-                  undef:   :replace)
+        input
+          .force_encoding(Encoding::ASCII_8BIT)
+          .encode!(Encoding::UTF_8,
+            invalid: :replace,
+            undef:   :replace)
         input
       end,
       exception: lambda do |input|
-        input.
-          force_encoding(Encoding::ASCII_8BIT).
-          encode!(Encoding::UTF_8)
+        input
+          .force_encoding(Encoding::ASCII_8BIT)
+          .encode!(Encoding::UTF_8)
         input
       end
     }.freeze
@@ -73,8 +78,6 @@ module Rack
             env[key] = sanitize_uri_encoded_string(value)
           end
         elsif key.to_s.start_with?("HTTP_")
-          # Just sanitize the headers and leave them in UTF-8. There is
-          # no reason to have UTF-8 in headers, but if it's valid, let it be.
           if value.frozen?
             env[key] = sanitize_string(value.dup).freeze
           else
@@ -86,9 +89,7 @@ module Rack
 
     private
 
-    def build_strategy(options)
-      strategy = options.fetch(:strategy) { :replace }
-
+    def build_strategy(strategy)
       return strategy unless DEFAULT_STRATEGIES.key?(strategy)
 
       DEFAULT_STRATEGIES[strategy]
@@ -128,7 +129,7 @@ module Rack
       return unless env['HTTP_COOKIE']
 
       env['HTTP_COOKIE'] = env['HTTP_COOKIE']
-        .split(/[;,] */n)
+        .split(/; */n)
         .map { |cookie| sanitize_uri_encoded_string(cookie) }
         .join('; ')
     end
@@ -157,8 +158,8 @@ module Rack
 
       def decode_string(input)
         unescape_unreserved(
-          sanitize_string(input).
-            force_encoding(Encoding::ASCII_8BIT))
+          sanitize_string(input)
+            .force_encoding(Encoding::ASCII_8BIT))
       end
 
       # RFC3986, 2.2 states that the characters from 'reserved' group must be
@@ -196,16 +197,18 @@ module Rack
 
       def sanitize_string(input)
         if input.is_a? String
-          input = input.force_encoding(Encoding::UTF_8)
+          # Handle chilled strings
+          # Rack values aren't supposed to be frozen, but it's common in test suites.
+          input = +input
 
-          if input.valid_encoding?
-            input
-          else
-            @strategy.call(input)
-          end
-        else
-          input
+          input.force_encoding(Encoding::UTF_8)
+
+          input = @strategy.call(input) unless input.valid_encoding?
+          # Rack requires non-ASCII strings to use ASCII-8BIT encoding.
+          input = input.b if input.is_a?(String) && !input.ascii_only?
         end
+
+        input
       end
     end
 
@@ -236,6 +239,10 @@ module Rack
         sanitized_io.each(&block)
       end
 
+      def ensure_sanitized
+        sanitized_io
+      end
+
       def rewind
         sanitized_io.rewind
       end

data/rack-sanitizer.gemspec

@@ -1,27 +1,22 @@
-# -*- encoding: utf-8 -*-
+# frozen_string_literal: true
 
 Gem::Specification.new do |gem|
   gem.name          = "rack-sanitizer"
-  gem.version       = '2.0.3'
-  gem.authors       = ["Jean Boussier", "whitequark"]
+  gem.version       = '2.1.0'
+  gem.authors       = ["Shopify - Ruby and Rails Infratructure", "Jean Boussier", "whitequark"]
   gem.license       = "MIT"
-  gem.email         = ["jean.boussier@gmail.org"]
-  gem.description   = %{Rack::Sanitizer is a Rack middleware which cleans up } <<
+  gem.email         = ["ruby-and-rails-infrastructure@shopify.com"]
+  gem.description   = %{Rack::Sanitizer is a Rack middleware which cleans up } +
                       %{invalid UTF8 characters in request URI and headers.}
   gem.summary       = "It is a mordernized and optimized fork of rack-utf8_sanitizer"
   gem.homepage      = "http://github.com/Shopify/rack-sanitizer"
 
   gem.files         = `git ls-files`.split($/)
-  gem.test_files    = gem.files.grep(%r{^(test|spec|features)/})
   gem.require_paths = ["lib"]
 
   gem.metadata["allowed_push_host"] = "https://rubygems.org/"
 
   gem.required_ruby_version = '>= 2.5'
 
-  gem.add_dependency             "rack", '>= 1.0', '< 4.0'
-
-  gem.add_development_dependency "bacon"
-  gem.add_development_dependency "bacon-colored_output"
-  gem.add_development_dependency "rake"
+  gem.add_dependency "rack", '>= 1.0', '< 4.0'
 end