uri 0.12.2 → 1.1.1

data/lib/uri/common.rb

@@ -13,24 +13,54 @@ require_relative "rfc2396_parser"
 require_relative "rfc3986_parser"
 
 module URI
-  include RFC2396_REGEXP
+  # The default parser instance for RFC 2396.
+  RFC2396_PARSER = RFC2396_Parser.new
+  Ractor.make_shareable(RFC2396_PARSER) if defined?(Ractor)
 
-  REGEXP = RFC2396_REGEXP
-  Parser = RFC2396_Parser
+  # The default parser instance for RFC 3986.
   RFC3986_PARSER = RFC3986_Parser.new
   Ractor.make_shareable(RFC3986_PARSER) if defined?(Ractor)
 
-  # URI::Parser.new
-  DEFAULT_PARSER = Parser.new
-  DEFAULT_PARSER.pattern.each_pair do |sym, str|
-    unless REGEXP::PATTERN.const_defined?(sym)
-      REGEXP::PATTERN.const_set(sym, str)
+  # The default parser instance.
+  DEFAULT_PARSER = RFC3986_PARSER
+  Ractor.make_shareable(DEFAULT_PARSER) if defined?(Ractor)
+
+  # Set the default parser instance.
+  def self.parser=(parser = RFC3986_PARSER)
+    remove_const(:Parser) if defined?(::URI::Parser)
+    const_set("Parser", parser.class)
+
+    remove_const(:PARSER) if defined?(::URI::PARSER)
+    const_set("PARSER", parser)
+
+    remove_const(:REGEXP) if defined?(::URI::REGEXP)
+    remove_const(:PATTERN) if defined?(::URI::PATTERN)
+    if Parser == RFC2396_Parser
+      const_set("REGEXP", URI::RFC2396_REGEXP)
+      const_set("PATTERN", URI::RFC2396_REGEXP::PATTERN)
+    end
+
+    Parser.new.regexp.each_pair do |sym, str|
+      remove_const(sym) if const_defined?(sym, false)
+      const_set(sym, str)
     end
   end
-  DEFAULT_PARSER.regexp.each_pair do |sym, str|
-    const_set(sym, str)
+  self.parser = RFC3986_PARSER
+
+  def self.const_missing(const) # :nodoc:
+    if const == :REGEXP
+      warn "URI::REGEXP is obsolete. Use URI::RFC2396_REGEXP explicitly.", uplevel: 1 if $VERBOSE
+      URI::RFC2396_REGEXP
+    elsif value = RFC2396_PARSER.regexp[const]
+      warn "URI::#{const} is obsolete. Use URI::RFC2396_PARSER.regexp[#{const.inspect}] explicitly.", uplevel: 1 if $VERBOSE
+      value
+    elsif value = RFC2396_Parser.const_get(const)
+      warn "URI::#{const} is obsolete. Use URI::RFC2396_Parser::#{const} explicitly.", uplevel: 1 if $VERBOSE
+      value
+    else
+      super
+    end
   end
-  Ractor.make_shareable(DEFAULT_PARSER) if defined?(Ractor)
 
   module Util # :nodoc:
     def make_components_hash(klass, array_hash)
@@ -64,41 +94,101 @@ module URI
     module_function :make_components_hash
   end
 
-  module Schemes
+  module Schemes # :nodoc:
+    class << self
+      ReservedChars = ".+-"
+      EscapedChars = "\u01C0\u01C1\u01C2"
+      # Use Lo category chars as escaped chars for TruffleRuby, which
+      # does not allow Symbol categories as identifiers.
+
+      def escape(name)
+        unless name and name.ascii_only?
+          return nil
+        end
+        name.upcase.tr(ReservedChars, EscapedChars)
+      end
+
+      def unescape(name)
+        name.tr(EscapedChars, ReservedChars).encode(Encoding::US_ASCII).upcase
+      end
+
+      def find(name)
+        const_get(name, false) if name and const_defined?(name, false)
+      end
+
+      def register(name, klass)
+        unless scheme = escape(name)
+          raise ArgumentError, "invalid character as scheme - #{name}"
+        end
+        const_set(scheme, klass)
+      end
+
+      def list
+        constants.map { |name|
+          [unescape(name.to_s), const_get(name)]
+        }.to_h
+      end
+    end
   end
   private_constant :Schemes
 
+  # Registers the given +klass+ as the class to be instantiated
+  # when parsing a \URI with the given +scheme+:
   #
-  # Register the given +klass+ to be instantiated when parsing URLs with the given +scheme+.
-  # Note that currently only schemes which after .upcase are valid constant names
-  # can be registered (no -/+/. allowed).
+  #   URI.register_scheme('MS_SEARCH', URI::Generic) # => URI::Generic
+  #   URI.scheme_list['MS_SEARCH']                   # => URI::Generic
   #
+  # Note that after calling String#upcase on +scheme+, it must be a valid
+  # constant name.
   def self.register_scheme(scheme, klass)
-    Schemes.const_set(scheme.to_s.upcase, klass)
+    Schemes.register(scheme, klass)
   end
 
-  # Returns a Hash of the defined schemes.
+  # Returns a hash of the defined schemes:
+  #
+  #   URI.scheme_list
+  #   # =>
+  #   {"MAILTO"=>URI::MailTo,
+  #    "LDAPS"=>URI::LDAPS,
+  #    "WS"=>URI::WS,
+  #    "HTTP"=>URI::HTTP,
+  #    "HTTPS"=>URI::HTTPS,
+  #    "LDAP"=>URI::LDAP,
+  #    "FILE"=>URI::File,
+  #    "FTP"=>URI::FTP}
+  #
+  # Related: URI.register_scheme.
   def self.scheme_list
-    Schemes.constants.map { |name|
-      [name.to_s.upcase, Schemes.const_get(name)]
-    }.to_h
+    Schemes.list
   end
 
+  # :stopdoc:
   INITIAL_SCHEMES = scheme_list
   private_constant :INITIAL_SCHEMES
   Ractor.make_shareable(INITIAL_SCHEMES) if defined?(Ractor)
+  # :startdoc:
 
+  # Returns a new object constructed from the given +scheme+, +arguments+,
+  # and +default+:
   #
-  # Construct a URI instance, using the scheme to detect the appropriate class
-  # from +URI.scheme_list+.
+  # - The new object is an instance of <tt>URI.scheme_list[scheme.upcase]</tt>.
+  # - The object is initialized by calling the class initializer
+  #   using +scheme+ and +arguments+.
+  #   See URI::Generic.new.
+  #
+  # Examples:
+  #
+  #   values = ['john.doe', 'www.example.com', '123', nil, '/forum/questions/', nil, 'tag=networking&order=newest', 'top']
+  #   URI.for('https', *values)
+  #   # => #<URI::HTTPS https://john.doe@www.example.com:123/forum/questions/?tag=networking&order=newest#top>
+  #   URI.for('foo', *values, default: URI::HTTP)
+  #   # => #<URI::HTTP foo://john.doe@www.example.com:123/forum/questions/?tag=networking&order=newest#top>
   #
   def self.for(scheme, *arguments, default: Generic)
-    const_name = scheme.to_s.upcase
+    const_name = Schemes.escape(scheme)
 
     uri_class = INITIAL_SCHEMES[const_name]
-    uri_class ||= if /\A[A-Z]\w*\z/.match?(const_name) && Schemes.const_defined?(const_name, false)
-      Schemes.const_get(const_name, false)
-    end
+    uri_class ||= Schemes.find(const_name)
     uri_class ||= default
 
     return uri_class.new(scheme, *arguments)
@@ -121,95 +211,49 @@ module URI
   #
   class BadURIError < Error; end
 
-  #
-  # == Synopsis
-  #
-  #   URI::split(uri)
-  #
-  # == Args
-  #
-  # +uri+::
-  #   String with URI.
-  #
-  # == Description
-  #
-  # Splits the string on following parts and returns array with result:
-  #
-  # * Scheme
-  # * Userinfo
-  # * Host
-  # * Port
-  # * Registry
-  # * Path
-  # * Opaque
-  # * Query
-  # * Fragment
-  #
-  # == Usage
-  #
-  #   require 'uri'
-  #
-  #   URI.split("http://www.ruby-lang.org/")
-  #   # => ["http", nil, "www.ruby-lang.org", nil, nil, "/", nil, nil, nil]
+  # Returns a 9-element array representing the parts of the \URI
+  # formed from the string +uri+;
+  # each array element is a string or +nil+:
+  #
+  #   names = %w[scheme userinfo host port registry path opaque query fragment]
+  #   values = URI.split('https://john.doe@www.example.com:123/forum/questions/?tag=networking&order=newest#top')
+  #   names.zip(values)
+  #   # =>
+  #   [["scheme", "https"],
+  #    ["userinfo", "john.doe"],
+  #    ["host", "www.example.com"],
+  #    ["port", "123"],
+  #    ["registry", nil],
+  #    ["path", "/forum/questions/"],
+  #    ["opaque", nil],
+  #    ["query", "tag=networking&order=newest"],
+  #    ["fragment", "top"]]
   #
   def self.split(uri)
-    RFC3986_PARSER.split(uri)
+    PARSER.split(uri)
   end
 
+  # Returns a new \URI object constructed from the given string +uri+:
   #
-  # == Synopsis
-  #
-  #   URI::parse(uri_str)
-  #
-  # == Args
+  #   URI.parse('https://john.doe@www.example.com:123/forum/questions/?tag=networking&order=newest#top')
+  #   # => #<URI::HTTPS https://john.doe@www.example.com:123/forum/questions/?tag=networking&order=newest#top>
+  #   URI.parse('http://john.doe@www.example.com:123/forum/questions/?tag=networking&order=newest#top')
+  #   # => #<URI::HTTP http://john.doe@www.example.com:123/forum/questions/?tag=networking&order=newest#top>
   #
-  # +uri_str+::
-  #   String with URI.
-  #
-  # == Description
-  #
-  # Creates one of the URI's subclasses instance from the string.
-  #
-  # == Raises
-  #
-  # URI::InvalidURIError::
-  #   Raised if URI given is not a correct one.
-  #
-  # == Usage
-  #
-  #   require 'uri'
-  #
-  #   uri = URI.parse("http://www.ruby-lang.org/")
-  #   # => #<URI::HTTP http://www.ruby-lang.org/>
-  #   uri.scheme
-  #   # => "http"
-  #   uri.host
-  #   # => "www.ruby-lang.org"
-  #
-  # It's recommended to first ::escape the provided +uri_str+ if there are any
-  # invalid URI characters.
+  # It's recommended to first URI::RFC2396_PARSER.escape string +uri+
+  # if it may contain invalid URI characters.
   #
   def self.parse(uri)
-    RFC3986_PARSER.parse(uri)
+    PARSER.parse(uri)
   end
 
+  # Merges the given URI strings +str+
+  # per {RFC 2396}[https://www.rfc-editor.org/rfc/rfc2396.html].
   #
-  # == Synopsis
-  #
-  #   URI::join(str[, str, ...])
-  #
-  # == Args
-  #
-  # +str+::
-  #   String(s) to work with, will be converted to RFC3986 URIs before merging.
-  #
-  # == Description
-  #
-  # Joins URIs.
+  # Each string in +str+ is converted to an
+  # {RFC3986 URI}[https://www.rfc-editor.org/rfc/rfc3986.html] before being merged.
   #
-  # == Usage
-  #
-  #   require 'uri'
+  # Examples:
   #
   #   URI.join("http://example.com/","main.rbx")
   #   # => #<URI::HTTP http://example.com/main.rbx>
@@ -227,7 +271,7 @@ module URI
   #   # => #<URI::HTTP http://example.com/foo/bar>
   #
   def self.join(*str)
-    RFC3986_PARSER.join(*str)
+    DEFAULT_PARSER.join(*str)
   end
 
   #
@@ -254,9 +298,9 @@ module URI
   #   URI.extract("text here http://foo.example.org/bla and here mailto:test@example.com and here also.")
   #   # => ["http://foo.example.com/bla", "mailto:test@example.com"]
   #
-  def self.extract(str, schemes = nil, &block)
+  def self.extract(str, schemes = nil, &block) # :nodoc:
     warn "URI.extract is obsolete", uplevel: 1 if $VERBOSE
-    DEFAULT_PARSER.extract(str, schemes, &block)
+    PARSER.extract(str, schemes, &block)
   end
 
   #
@@ -291,16 +335,16 @@ module URI
   #     p $&
   #   end
   #
-  def self.regexp(schemes = nil)
+  def self.regexp(schemes = nil)# :nodoc:
     warn "URI.regexp is obsolete", uplevel: 1 if $VERBOSE
-    DEFAULT_PARSER.make_regexp(schemes)
+    PARSER.make_regexp(schemes)
   end
 
   TBLENCWWWCOMP_ = {} # :nodoc:
   256.times do |i|
     TBLENCWWWCOMP_[-i.chr] = -('%%%02X' % i)
   end
-  TBLENCURICOMP_ = TBLENCWWWCOMP_.dup.freeze
+  TBLENCURICOMP_ = TBLENCWWWCOMP_.dup.freeze # :nodoc:
   TBLENCWWWCOMP_[' '] = '+'
   TBLENCWWWCOMP_.freeze
   TBLDECWWWCOMP_ = {} # :nodoc:
@@ -314,44 +358,92 @@ module URI
   TBLDECWWWCOMP_['+'] = ' '
   TBLDECWWWCOMP_.freeze
 
-  # Encodes given +str+ to URL-encoded form data.
+  # Returns a URL-encoded string derived from the given string +str+.
+  #
+  # The returned string:
+  #
+  # - Preserves:
+  #
+  #   - Characters <tt>'*'</tt>, <tt>'.'</tt>, <tt>'-'</tt>, and <tt>'_'</tt>.
+  #   - Character in ranges <tt>'a'..'z'</tt>, <tt>'A'..'Z'</tt>,
+  #     and <tt>'0'..'9'</tt>.
+  #
+  #   Example:
+  #
+  #     URI.encode_www_form_component('*.-_azAZ09')
+  #     # => "*.-_azAZ09"
+  #
+  # - Converts:
+  #
+  #   - Character <tt>' '</tt> to character <tt>'+'</tt>.
+  #   - Any other character to "percent notation";
+  #     the percent notation for character <i>c</i> is <tt>'%%%X' % c.ord</tt>.
+  #
+  #   Example:
+  #
+  #     URI.encode_www_form_component('Here are some punctuation characters: ,;?:')
+  #     # => "Here+are+some+punctuation+characters%3A+%2C%3B%3F%3A"
   #
-  # This method doesn't convert *, -, ., 0-9, A-Z, _, a-z, but does convert SP
-  # (ASCII space) to + and converts others to %XX.
+  # Encoding:
   #
-  # If +enc+ is given, convert +str+ to the encoding before percent encoding.
+  # - If +str+ has encoding Encoding::ASCII_8BIT, argument +enc+ is ignored.
+  # - Otherwise +str+ is converted first to Encoding::UTF_8
+  #   (with suitable character replacements),
+  #   and then to encoding +enc+.
   #
-  # This is an implementation of
-  # https://www.w3.org/TR/2013/CR-html5-20130806/forms.html#url-encoded-form-data.
+  # In either case, the returned string has forced encoding Encoding::US_ASCII.
   #
-  # See URI.decode_www_form_component, URI.encode_www_form.
+  # Related: URI.encode_uri_component (encodes <tt>' '</tt> as <tt>'%20'</tt>).
   def self.encode_www_form_component(str, enc=nil)
     _encode_uri_component(/[^*\-.0-9A-Z_a-z]/, TBLENCWWWCOMP_, str, enc)
   end
 
-  # Decodes given +str+ of URL-encoded form data.
+  # Returns a string decoded from the given \URL-encoded string +str+.
   #
-  # This decodes + to SP.
+  # The given string is first encoded as Encoding::ASCII-8BIT (using String#b),
+  # then decoded (as below), and finally force-encoded to the given encoding +enc+.
   #
-  # See URI.encode_www_form_component, URI.decode_www_form.
+  # The returned string:
+  #
+  # - Preserves:
+  #
+  #   - Characters <tt>'*'</tt>, <tt>'.'</tt>, <tt>'-'</tt>, and <tt>'_'</tt>.
+  #   - Character in ranges <tt>'a'..'z'</tt>, <tt>'A'..'Z'</tt>,
+  #     and <tt>'0'..'9'</tt>.
+  #
+  #   Example:
+  #
+  #     URI.decode_www_form_component('*.-_azAZ09')
+  #     # => "*.-_azAZ09"
+  #
+  # - Converts:
+  #
+  #   - Character <tt>'+'</tt> to character <tt>' '</tt>.
+  #   - Each "percent notation" to an ASCII character.
+  #
+  #   Example:
+  #
+  #     URI.decode_www_form_component('Here+are+some+punctuation+characters%3A+%2C%3B%3F%3A')
+  #     # => "Here are some punctuation characters: ,;?:"
+  #
+  # Related: URI.decode_uri_component (preserves <tt>'+'</tt>).
   def self.decode_www_form_component(str, enc=Encoding::UTF_8)
     _decode_uri_component(/\+|%\h\h/, str, enc)
   end
 
-  # Encodes +str+ using URL encoding
-  #
-  # This encodes SP to %20 instead of +.
+  # Like URI.encode_www_form_component, except that <tt>' '</tt> (space)
+  # is encoded as <tt>'%20'</tt> (instead of <tt>'+'</tt>).
   def self.encode_uri_component(str, enc=nil)
     _encode_uri_component(/[^*\-.0-9A-Z_a-z]/, TBLENCURICOMP_, str, enc)
   end
 
-  # Decodes given +str+ of URL-encoded data.
-  #
-  # This does not decode + to SP.
+  # Like URI.decode_www_form_component, except that <tt>'+'</tt> is preserved.
   def self.decode_uri_component(str, enc=Encoding::UTF_8)
     _decode_uri_component(/%\h\h/, str, enc)
   end
 
+  # Returns a string derived from the given string +str+ with
+  # URI-encoded characters matching +regexp+ according to +table+.
   def self._encode_uri_component(regexp, table, str, enc)
     str = str.to_s.dup
     if str.encoding != Encoding::ASCII_8BIT
@@ -366,39 +458,112 @@ module URI
   end
   private_class_method :_encode_uri_component
 
+  # Returns a string decoding characters matching +regexp+ from the
+  # given \URL-encoded string +str+.
   def self._decode_uri_component(regexp, str, enc)
     raise ArgumentError, "invalid %-encoding (#{str})" if /%(?!\h\h)/.match?(str)
     str.b.gsub(regexp, TBLDECWWWCOMP_).force_encoding(enc)
   end
   private_class_method :_decode_uri_component
 
-  # Generates URL-encoded form data from given +enum+.
+  # Returns a URL-encoded string derived from the given
+  # {Enumerable}[https://docs.ruby-lang.org/en/master/Enumerable.html#module-Enumerable-label-Enumerable+in+Ruby+Classes]
+  # +enum+.
+  #
+  # The result is suitable for use as form data
+  # for an \HTTP request whose <tt>Content-Type</tt> is
+  # <tt>'application/x-www-form-urlencoded'</tt>.
+  #
+  # The returned string consists of the elements of +enum+,
+  # each converted to one or more URL-encoded strings,
+  # and all joined with character <tt>'&'</tt>.
+  #
+  # Simple examples:
+  #
+  #   URI.encode_www_form([['foo', 0], ['bar', 1], ['baz', 2]])
+  #   # => "foo=0&bar=1&baz=2"
+  #   URI.encode_www_form({foo: 0, bar: 1, baz: 2})
+  #   # => "foo=0&bar=1&baz=2"
   #
-  # This generates application/x-www-form-urlencoded data defined in HTML5
-  # from given an Enumerable object.
+  # The returned string is formed using method URI.encode_www_form_component,
+  # which converts certain characters:
   #
-  # This internally uses URI.encode_www_form_component(str).
+  #   URI.encode_www_form('f#o': '/', 'b-r': '$', 'b z': '@')
+  #   # => "f%23o=%2F&b-r=%24&b+z=%40"
   #
-  # This method doesn't convert the encoding of given items, so convert them
-  # before calling this method if you want to send data as other than original
-  # encoding or mixed encoding data. (Strings which are encoded in an HTML5
-  # ASCII incompatible encoding are converted to UTF-8.)
+  # When +enum+ is Array-like, each element +ele+ is converted to a field:
   #
-  # This method doesn't handle files.  When you send a file, use
-  # multipart/form-data.
+  # - If +ele+ is an array of two or more elements,
+  #   the field is formed from its first two elements
+  #   (and any additional elements are ignored):
   #
-  # This refers https://url.spec.whatwg.org/#concept-urlencoded-serializer
+  #     name = URI.encode_www_form_component(ele[0], enc)
+  #     value = URI.encode_www_form_component(ele[1], enc)
+  #     "#{name}=#{value}"
   #
-  #    URI.encode_www_form([["q", "ruby"], ["lang", "en"]])
-  #    #=> "q=ruby&lang=en"
-  #    URI.encode_www_form("q" => "ruby", "lang" => "en")
-  #    #=> "q=ruby&lang=en"
-  #    URI.encode_www_form("q" => ["ruby", "perl"], "lang" => "en")
-  #    #=> "q=ruby&q=perl&lang=en"
-  #    URI.encode_www_form([["q", "ruby"], ["q", "perl"], ["lang", "en"]])
-  #    #=> "q=ruby&q=perl&lang=en"
+  #   Examples:
+  #
+  #     URI.encode_www_form([%w[foo bar], %w[baz bat bah]])
+  #     # => "foo=bar&baz=bat"
+  #     URI.encode_www_form([['foo', 0], ['bar', :baz, 'bat']])
+  #     # => "foo=0&bar=baz"
+  #
+  # - If +ele+ is an array of one element,
+  #   the field is formed from <tt>ele[0]</tt>:
+  #
+  #     URI.encode_www_form_component(ele[0])
+  #
+  #   Example:
+  #
+  #     URI.encode_www_form([['foo'], [:bar], [0]])
+  #     # => "foo&bar&0"
+  #
+  # - Otherwise the field is formed from +ele+:
+  #
+  #     URI.encode_www_form_component(ele)
+  #
+  #   Example:
+  #
+  #     URI.encode_www_form(['foo', :bar, 0])
+  #     # => "foo&bar&0"
+  #
+  # The elements of an Array-like +enum+ may be mixture:
+  #
+  #   URI.encode_www_form([['foo', 0], ['bar', 1, 2], ['baz'], :bat])
+  #   # => "foo=0&bar=1&baz&bat"
+  #
+  # When +enum+ is Hash-like,
+  # each +key+/+value+ pair is converted to one or more fields:
+  #
+  # - If +value+ is
+  #   {Array-convertible}[https://docs.ruby-lang.org/en/master/implicit_conversion_rdoc.html#label-Array-Convertible+Objects],
+  #   each element +ele+ in +value+ is paired with +key+ to form a field:
+  #
+  #     name = URI.encode_www_form_component(key, enc)
+  #     value = URI.encode_www_form_component(ele, enc)
+  #     "#{name}=#{value}"
+  #
+  #   Example:
+  #
+  #     URI.encode_www_form({foo: [:bar, 1], baz: [:bat, :bam, 2]})
+  #     # => "foo=bar&foo=1&baz=bat&baz=bam&baz=2"
+  #
+  # - Otherwise, +key+ and +value+ are paired to form a field:
+  #
+  #     name = URI.encode_www_form_component(key, enc)
+  #     value = URI.encode_www_form_component(value, enc)
+  #     "#{name}=#{value}"
+  #
+  #   Example:
+  #
+  #     URI.encode_www_form({foo: 0, bar: 1, baz: 2})
+  #     # => "foo=0&bar=1&baz=2"
+  #
+  # The elements of a Hash-like +enum+ may be mixture:
+  #
+  #   URI.encode_www_form({foo: [0, 1], bar: 2})
+  #   # => "foo=0&foo=1&bar=2"
   #
-  # See URI.encode_www_form_component, URI.decode_www_form.
   def self.encode_www_form(enum, enc=nil)
     enum.map do |k,v|
       if v.nil?
@@ -419,22 +584,39 @@ module URI
     end.join('&')
   end
 
-  # Decodes URL-encoded form data from given +str+.
+  # Returns name/value pairs derived from the given string +str+,
+  # which must be an ASCII string.
+  #
+  # The method may be used to decode the body of Net::HTTPResponse object +res+
+  # for which <tt>res['Content-Type']</tt> is <tt>'application/x-www-form-urlencoded'</tt>.
+  #
+  # The returned data is an array of 2-element subarrays;
+  # each subarray is a name/value pair (both are strings).
+  # Each returned string has encoding +enc+,
+  # and has had invalid characters removed via
+  # {String#scrub}[https://docs.ruby-lang.org/en/master/String.html#method-i-scrub].
   #
-  # This decodes application/x-www-form-urlencoded data
-  # and returns an array of key-value arrays.
+  # A simple example:
   #
-  # This refers http://url.spec.whatwg.org/#concept-urlencoded-parser,
-  # so this supports only &-separator, and doesn't support ;-separator.
+  #   URI.decode_www_form('foo=0&bar=1&baz')
+  #   # => [["foo", "0"], ["bar", "1"], ["baz", ""]]
   #
-  #    ary = URI.decode_www_form("a=1&a=2&b=3")
-  #    ary                   #=> [['a', '1'], ['a', '2'], ['b', '3']]
-  #    ary.assoc('a').last   #=> '1'
-  #    ary.assoc('b').last   #=> '3'
-  #    ary.rassoc('a').last  #=> '2'
-  #    Hash[ary]             #=> {"a"=>"2", "b"=>"3"}
+  # The returned strings have certain conversions,
+  # similar to those performed in URI.decode_www_form_component:
+  #
+  #   URI.decode_www_form('f%23o=%2F&b-r=%24&b+z=%40')
+  #   # => [["f#o", "/"], ["b-r", "$"], ["b z", "@"]]
+  #
+  # The given string may contain consecutive separators:
+  #
+  #   URI.decode_www_form('foo=0&&bar=1&&baz=2')
+  #   # => [["foo", "0"], ["", ""], ["bar", "1"], ["", ""], ["baz", "2"]]
+  #
+  # A different separator may be specified:
+  #
+  #   URI.decode_www_form('foo=0--bar=1--baz', separator: '--')
+  #   # => [["foo", "0"], ["bar", "1"], ["baz", ""]]
   #
-  # See URI.decode_www_form_component, URI.encode_www_form.
   def self.decode_www_form(str, enc=Encoding::UTF_8, separator: '&', use__charset_: false, isindex: false)
     raise ArgumentError, "the input of #{self.name}.#{__method__} must be ASCII only string" unless str.ascii_only?
     ary = []
@@ -713,7 +895,18 @@ end # module URI
 module Kernel
 
   #
-  # Returns +uri+ converted to an URI object.
+  # Returns a \URI object derived from the given +uri+,
+  # which may be a \URI string or an existing \URI object:
+  #
+  #   require 'uri'
+  #   # Returns a new URI.
+  #   uri = URI('http://github.com/ruby/ruby')
+  #   # => #<URI::HTTP http://github.com/ruby/ruby>
+  #   # Returns the given URI.
+  #   URI(uri)
+  #   # => #<URI::HTTP http://github.com/ruby/ruby>
+  #
+  # You must require 'uri' to use this method.
   #
   def URI(uri)
     if uri.is_a?(URI::Generic)