Merge URI-0.12.0

7 files changed, 64 insertions, 18 deletions

diff --git a/lib/uri.rb b/lib/uri.rb
index 394c156ac5..59a7c4ad28 100644
--- a/lib/uri.rb
+++ b/lib/uri.rb
@@ -101,3 +101,4 @@ require_relative 'uri/ldap'
 require_relative 'uri/ldaps'
 require_relative 'uri/mailto'
 require_relative 'uri/ws'
+require_relative 'uri/wss'
diff --git a/lib/uri/common.rb b/lib/uri/common.rb
index 26b179add2..ca38bec7ec 100644
--- a/lib/uri/common.rb
+++ b/lib/uri/common.rb
@@ -13,6 +13,8 @@ require_relative "rfc2396_parser"
 require_relative "rfc3986_parser"
 
 module URI
+  include RFC2396_REGEXP
+
   REGEXP = RFC2396_REGEXP
   Parser = RFC2396_Parser
   RFC3986_PARSER = RFC3986_Parser.new
@@ -62,14 +64,17 @@ module URI
     module_function :make_components_hash
   end
 
-  include REGEXP
-
   module Schemes
   end
   private_constant :Schemes
 
+  #
+  # 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).
+  #
   def self.register_scheme(scheme, klass)
-    Schemes.const_set(scheme, klass)
+    Schemes.const_set(scheme.to_s.upcase, klass)
   end
 
   # Returns a Hash of the defined schemes.
@@ -295,6 +300,7 @@ module URI
   256.times do |i|
     TBLENCWWWCOMP_[-i.chr] = -('%%%02X' % i)
   end
+  TBLENCURICOMP_ = TBLENCWWWCOMP_.dup.freeze
   TBLENCWWWCOMP_[' '] = '+'
   TBLENCWWWCOMP_.freeze
   TBLDECWWWCOMP_ = {} # :nodoc:
@@ -320,6 +326,33 @@ module URI
   #
   # See URI.decode_www_form_component, URI.encode_www_form.
   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.
+  #
+  # This decodes + to SP.
+  #
+  # See URI.encode_www_form_component, URI.decode_www_form.
+  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 +.
+  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.
+  def self.decode_uri_component(str, enc=Encoding::UTF_8)
+    _decode_uri_component(/%\h\h/, str, enc)
+  end
+
+  def self._encode_uri_component(regexp, table, str, enc)
     str = str.to_s.dup
     if str.encoding != Encoding::ASCII_8BIT
       if enc && enc != Encoding::ASCII_8BIT
@@ -328,19 +361,16 @@ module URI
       end
       str.force_encoding(Encoding::ASCII_8BIT)
     end
-    str.gsub!(/[^*\-.0-9A-Z_a-z]/, TBLENCWWWCOMP_)
+    str.gsub!(regexp, table)
     str.force_encoding(Encoding::US_ASCII)
   end
+  private_class_method :_encode_uri_component
 
-  # Decodes given +str+ of URL-encoded form data.
-  #
-  # This decodes + to SP.
-  #
-  # See URI.encode_www_form_component, URI.decode_www_form.
-  def self.decode_www_form_component(str, enc=Encoding::UTF_8)
+  def self._decode_uri_component(regexp, str, enc)
     raise ArgumentError, "invalid %-encoding (#{str})" if /%(?!\h\h)/.match?(str)
-    str.b.gsub(/\+|%\h\h/, TBLDECWWWCOMP_).force_encoding(enc)
+    str.b.gsub(regexp, TBLDECWWWCOMP_).force_encoding(enc)
   end
+  private_class_method :_decode_uri_component
 
   # Generates URL-encoded form data from given +enum+.
   #
diff --git a/lib/uri/file.rb b/lib/uri/file.rb
index 7671ad6470..4ff0bc097e 100644
--- a/lib/uri/file.rb
+++ b/lib/uri/file.rb
@@ -33,6 +33,9 @@ module URI
     # If an Array is used, the components must be passed in the
     # order <code>[host, path]</code>.
     #
+    # A path from e.g. the File class should be escaped before
+    # being passed.
+    #
     # Examples:
     #
     #     require 'uri'
@@ -44,6 +47,9 @@ module URI
     #       :path => '/ruby/src'})
     #     uri2.to_s  # => "file://host.example.com/ruby/src"
     #
+    #     uri3 = URI::File.build({:path => URI::escape('/path/my file.txt')})
+    #     uri3.to_s  # => "file:///path/my%20file.txt"
+    #
     def self.build(args)
       tmp = Util::make_components_hash(self, args)
       super(tmp)
diff --git a/lib/uri/generic.rb b/lib/uri/generic.rb
index cfa0de6b74..69698c4e2d 100644
--- a/lib/uri/generic.rb
+++ b/lib/uri/generic.rb
@@ -564,16 +564,26 @@ module URI
       end
     end
 
-    # Returns the user component.
+    # Returns the user component (without URI decoding).
     def user
       @user
     end
 
-    # Returns the password component.
+    # Returns the password component (without URI decoding).
     def password
       @password
     end
 
+    # Returns the user component after URI decoding.
+    def decoded_user
+      URI.decode_uri_component(@user) if @user
+    end
+
+    # Returns the password component after URI decoding.
+    def decoded_password
+      URI.decode_uri_component(@password) if @password
+    end
+
     #
     # Checks the host +v+ component for RFC2396 compliance
     # and against the URI::Parser Regexp for :HOST.
diff --git a/lib/uri/mailto.rb b/lib/uri/mailto.rb
index 87cb99656f..cb8024f301 100644
--- a/lib/uri/mailto.rb
+++ b/lib/uri/mailto.rb
@@ -15,7 +15,7 @@ module URI
   # RFC6068, the mailto URL scheme.
   #
   class MailTo < Generic
-    include REGEXP
+    include RFC2396_REGEXP
 
     # A Default port of nil for URI::MailTo.
     DEFAULT_PORT = nil
diff --git a/lib/uri/rfc3986_parser.rb b/lib/uri/rfc3986_parser.rb
index 3e07de4805..f3816d9ae5 100644
--- a/lib/uri/rfc3986_parser.rb
+++ b/lib/uri/rfc3986_parser.rb
@@ -2,9 +2,8 @@
 module URI
   class RFC3986_Parser # :nodoc:
     # URI defined in RFC3986
-    # this regexp is modified not to host is not empty string
-    RFC3986_URI = /\A(?<URI>(?<scheme>[A-Za-z][+\-.0-9A-Za-z]*):(?<hier-part>\/\/(?<authority>(?:(?<userinfo>(?:%\h\h|[!$&-.0-;=A-Z_a-z~])*)@)?(?<host>(?<IP-literal>\[(?:(?<IPv6address>(?:\h{1,4}:){6}(?<ls32>\h{1,4}:\h{1,4}|(?<IPv4address>(?<dec-octet>[1-9]\d|1\d{2}|2[0-4]\d|25[0-5]|\d)\.\g<dec-octet>\.\g<dec-octet>\.\g<dec-octet>))|::(?:\h{1,4}:){5}\g<ls32>|\h{1,4}?::(?:\h{1,4}:){4}\g<ls32>|(?:(?:\h{1,4}:)?\h{1,4})?::(?:\h{1,4}:){3}\g<ls32>|(?:(?:\h{1,4}:){,2}\h{1,4})?::(?:\h{1,4}:){2}\g<ls32>|(?:(?:\h{1,4}:){,3}\h{1,4})?::\h{1,4}:\g<ls32>|(?:(?:\h{1,4}:){,4}\h{1,4})?::\g<ls32>|(?:(?:\h{1,4}:){,5}\h{1,4})?::\h{1,4}|(?:(?:\h{1,4}:){,6}\h{1,4})?::)|(?<IPvFuture>v\h+\.[!$&-.0-;=A-Z_a-z~]+))\])|\g<IPv4address>|(?<reg-name>(?:%\h\h|[!$&-.0-9;=A-Z_a-z~])+))?(?::(?<port>\d*))?)(?<path-abempty>(?:\/(?<segment>(?:%\h\h|[!$&-.0-;=@-Z_a-z~])*))*)|(?<path-absolute>\/(?:(?<segment-nz>(?:%\h\h|[!$&-.0-;=@-Z_a-z~])+)(?:\/\g<segment>)*)?)|(?<path-rootless>\g<segment-nz>(?:\/\g<segment>)*)|(?<path-empty>))(?:\?(?<query>[^#]*))?(?:\#(?<fragment>(?:%\h\h|[!$&-.0-;=@-Z_a-z~\/?])*))?)\z/
-    RFC3986_relative_ref = /\A(?<relative-ref>(?<relative-part>\/\/(?<authority>(?:(?<userinfo>(?:%\h\h|[!$&-.0-;=A-Z_a-z~])*)@)?(?<host>(?<IP-literal>\[(?<IPv6address>(?:\h{1,4}:){6}(?<ls32>\h{1,4}:\h{1,4}|(?<IPv4address>(?<dec-octet>[1-9]\d|1\d{2}|2[0-4]\d|25[0-5]|\d)\.\g<dec-octet>\.\g<dec-octet>\.\g<dec-octet>))|::(?:\h{1,4}:){5}\g<ls32>|\h{1,4}?::(?:\h{1,4}:){4}\g<ls32>|(?:(?:\h{1,4}:){,1}\h{1,4})?::(?:\h{1,4}:){3}\g<ls32>|(?:(?:\h{1,4}:){,2}\h{1,4})?::(?:\h{1,4}:){2}\g<ls32>|(?:(?:\h{1,4}:){,3}\h{1,4})?::\h{1,4}:\g<ls32>|(?:(?:\h{1,4}:){,4}\h{1,4})?::\g<ls32>|(?:(?:\h{1,4}:){,5}\h{1,4})?::\h{1,4}|(?:(?:\h{1,4}:){,6}\h{1,4})?::)|(?<IPvFuture>v\h+\.[!$&-.0-;=A-Z_a-z~]+)\])|\g<IPv4address>|(?<reg-name>(?:%\h\h|[!$&-.0-9;=A-Z_a-z~])+))?(?::(?<port>\d*))?)(?<path-abempty>(?:\/(?<segment>(?:%\h\h|[!$&-.0-;=@-Z_a-z~])*))*)|(?<path-absolute>\/(?:(?<segment-nz>(?:%\h\h|[!$&-.0-;=@-Z_a-z~])+)(?:\/\g<segment>)*)?)|(?<path-noscheme>(?<segment-nz-nc>(?:%\h\h|[!$&-.0-9;=@-Z_a-z~])+)(?:\/\g<segment>)*)|(?<path-empty>))(?:\?(?<query>[^#]*))?(?:\#(?<fragment>(?:%\h\h|[!$&-.0-;=@-Z_a-z~\/?])*))?)\z/
+    RFC3986_URI = /\A(?<URI>(?<scheme>[A-Za-z][+\-.0-9A-Za-z]*):(?<hier-part>\/\/(?<authority>(?:(?<userinfo>(?:%\h\h|[!$&-.0-;=A-Z_a-z~])*)@)?(?<host>(?<IP-literal>\[(?:(?<IPv6address>(?:\h{1,4}:){6}(?<ls32>\h{1,4}:\h{1,4}|(?<IPv4address>(?<dec-octet>[1-9]\d|1\d{2}|2[0-4]\d|25[0-5]|\d)\.\g<dec-octet>\.\g<dec-octet>\.\g<dec-octet>))|::(?:\h{1,4}:){5}\g<ls32>|\h{1,4}?::(?:\h{1,4}:){4}\g<ls32>|(?:(?:\h{1,4}:)?\h{1,4})?::(?:\h{1,4}:){3}\g<ls32>|(?:(?:\h{1,4}:){,2}\h{1,4})?::(?:\h{1,4}:){2}\g<ls32>|(?:(?:\h{1,4}:){,3}\h{1,4})?::\h{1,4}:\g<ls32>|(?:(?:\h{1,4}:){,4}\h{1,4})?::\g<ls32>|(?:(?:\h{1,4}:){,5}\h{1,4})?::\h{1,4}|(?:(?:\h{1,4}:){,6}\h{1,4})?::)|(?<IPvFuture>v\h+\.[!$&-.0-;=A-Z_a-z~]+))\])|\g<IPv4address>|(?<reg-name>(?:%\h\h|[!$&-.0-9;=A-Z_a-z~])*))(?::(?<port>\d*))?)(?<path-abempty>(?:\/(?<segment>(?:%\h\h|[!$&-.0-;=@-Z_a-z~])*))*)|(?<path-absolute>\/(?:(?<segment-nz>(?:%\h\h|[!$&-.0-;=@-Z_a-z~])+)(?:\/\g<segment>)*)?)|(?<path-rootless>\g<segment-nz>(?:\/\g<segment>)*)|(?<path-empty>))(?:\?(?<query>[^#]*))?(?:\#(?<fragment>(?:%\h\h|[!$&-.0-;=@-Z_a-z~\/?])*))?)\z/
+    RFC3986_relative_ref = /\A(?<relative-ref>(?<relative-part>\/\/(?<authority>(?:(?<userinfo>(?:%\h\h|[!$&-.0-;=A-Z_a-z~])*)@)?(?<host>(?<IP-literal>\[(?:(?<IPv6address>(?:\h{1,4}:){6}(?<ls32>\h{1,4}:\h{1,4}|(?<IPv4address>(?<dec-octet>[1-9]\d|1\d{2}|2[0-4]\d|25[0-5]|\d)\.\g<dec-octet>\.\g<dec-octet>\.\g<dec-octet>))|::(?:\h{1,4}:){5}\g<ls32>|\h{1,4}?::(?:\h{1,4}:){4}\g<ls32>|(?:(?:\h{1,4}:){,1}\h{1,4})?::(?:\h{1,4}:){3}\g<ls32>|(?:(?:\h{1,4}:){,2}\h{1,4})?::(?:\h{1,4}:){2}\g<ls32>|(?:(?:\h{1,4}:){,3}\h{1,4})?::\h{1,4}:\g<ls32>|(?:(?:\h{1,4}:){,4}\h{1,4})?::\g<ls32>|(?:(?:\h{1,4}:){,5}\h{1,4})?::\h{1,4}|(?:(?:\h{1,4}:){,6}\h{1,4})?::)|(?<IPvFuture>v\h+\.[!$&-.0-;=A-Z_a-z~]+))\])|\g<IPv4address>|(?<reg-name>(?:%\h\h|[!$&-.0-9;=A-Z_a-z~])+))?(?::(?<port>\d*))?)(?<path-abempty>(?:\/(?<segment>(?:%\h\h|[!$&-.0-;=@-Z_a-z~])*))*)|(?<path-absolute>\/(?:(?<segment-nz>(?:%\h\h|[!$&-.0-;=@-Z_a-z~])+)(?:\/\g<segment>)*)?)|(?<path-noscheme>(?<segment-nz-nc>(?:%\h\h|[!$&-.0-9;=@-Z_a-z~])+)(?:\/\g<segment>)*)|(?<path-empty>))(?:\?(?<query>[^#]*))?(?:\#(?<fragment>(?:%\h\h|[!$&-.0-;=@-Z_a-z~\/?])*))?)\z/
     attr_reader :regexp
 
     def initialize
diff --git a/lib/uri/version.rb b/lib/uri/version.rb
index 82188e25ad..a9643ef8bc 100644
--- a/lib/uri/version.rb
+++ b/lib/uri/version.rb
@@ -1,6 +1,6 @@
 module URI
   # :stopdoc:
-  VERSION_CODE = '001100'.freeze
+  VERSION_CODE = '001200'.freeze
   VERSION = VERSION_CODE.scan(/../).collect{|n| n.to_i}.join('.').freeze
   # :startdoc:
 end