Merge CGI-0.3.5

5 files changed, 122 insertions, 39 deletions

diff --git a/lib/cgi.rb b/lib/cgi.rb
index af466bc673..dda5eb1712 100644
--- a/lib/cgi.rb
+++ b/lib/cgi.rb
@@ -162,7 +162,7 @@
 #   cgi.has_key?('field_name')
 #   cgi.include?('field_name')
 #
-# CAUTION! cgi['field_name'] returned an Array with the old
+# CAUTION! <code>cgi['field_name']</code> returned an Array with the old
 # cgi.rb(included in Ruby 1.6)
 #
 # === Get form values as hash
@@ -288,7 +288,7 @@
 #
 
 class CGI
-  VERSION = "0.3.2"
+  VERSION = "0.3.5"
 end
 
 require 'cgi/core'
diff --git a/lib/cgi/cgi.gemspec b/lib/cgi/cgi.gemspec
index 3ba62b93f6..381c55a5ca 100644
--- a/lib/cgi/cgi.gemspec
+++ b/lib/cgi/cgi.gemspec
@@ -22,10 +22,21 @@ Gem::Specification.new do |spec|
   spec.metadata["homepage_uri"] = spec.homepage
   spec.metadata["source_code_uri"] = spec.homepage
 
-  spec.files         = Dir.chdir(File.expand_path('..', __FILE__)) do
-    `git ls-files -z 2>/dev/null`.split("\x0").reject { |f| f.match(%r{\A(?:(?:test|spec|features)/|\.git)}) }
-  end
-  spec.extensions    = ["ext/cgi/escape/extconf.rb"]
   spec.executables   = []
+
+  spec.files = [
+    "LICENSE.txt",
+    "README.md",
+    *Dir["lib{.rb,/**/*.rb}", "bin/*"]  ]
+
   spec.require_paths = ["lib"]
+
+  if Gem::Platform === spec.platform and spec.platform =~ 'java' or RUBY_ENGINE == 'jruby'
+    spec.platform = 'java'
+    spec.require_paths << "ext/java/org/jruby/ext/cgi/escape/lib"
+    spec.files += Dir["ext/java/**/*.{rb}", "lib/cgi/escape.jar"]
+  else
+    spec.files += Dir["ext/cgi/**/*.{rb,c,h,sh}", "ext/cgi/escape/depend", "lib/cgi/escape.so"]
+    spec.extensions    = ["ext/cgi/escape/extconf.rb"]
+  end
 end
diff --git a/lib/cgi/cookie.rb b/lib/cgi/cookie.rb
index 6b0d89ca3b..1a9c1a82c1 100644
--- a/lib/cgi/cookie.rb
+++ b/lib/cgi/cookie.rb
@@ -40,6 +40,10 @@ class CGI
   class Cookie < Array
     @@accept_charset="UTF-8" unless defined?(@@accept_charset)
 
+    TOKEN_RE = %r"\A[[!-~]&&[^()<>@,;:\\\"/?=\[\]{}]]+\z"
+    PATH_VALUE_RE = %r"\A[[ -~]&&[^;]]*\z"
+    DOMAIN_VALUE_RE = %r"\A(?<label>(?!-)[-A-Za-z0-9]+(?<!-))(?:\.\g<label>)*\z"
+
     # Create a new CGI::Cookie object.
     #
     # :call-seq:
@@ -72,8 +76,8 @@ class CGI
       @domain = nil
       @expires = nil
       if name.kind_of?(String)
-        @name = name
-        @path = (%r|\A(.*/)| =~ ENV["SCRIPT_NAME"] ? $1 : "")
+        self.name = name
+        self.path = (%r|\A(.*/)| =~ ENV["SCRIPT_NAME"] ? $1 : "")
         @secure = false
         @httponly = false
         return super(value)
@@ -84,11 +88,11 @@ class CGI
         raise ArgumentError, "`name' required"
       end
 
-      @name = options["name"]
+      self.name = options["name"]
       value = Array(options["value"])
       # simple support for IE
-      @path = options["path"] || (%r|\A(.*/)| =~ ENV["SCRIPT_NAME"] ? $1 : "")
-      @domain = options["domain"]
+      self.path = options["path"] || (%r|\A(.*/)| =~ ENV["SCRIPT_NAME"] ? $1 : "")
+      self.domain = options["domain"]
       @expires = options["expires"]
       @secure = options["secure"] == true
       @httponly = options["httponly"] == true
@@ -97,11 +101,35 @@ class CGI
     end
 
     # Name of this cookie, as a +String+
-    attr_accessor :name
+    attr_reader :name
+    # Set name of this cookie
+    def name=(str)
+      if str and !TOKEN_RE.match?(str)
+        raise ArgumentError, "invalid name: #{str.dump}"
+      end
+      @name = str
+    end
+
     # Path for which this cookie applies, as a +String+
-    attr_accessor :path
+    attr_reader :path
+    # Set path for which this cookie applies
+    def path=(str)
+      if str and !PATH_VALUE_RE.match?(str)
+        raise ArgumentError, "invalid path: #{str.dump}"
+      end
+      @path = str
+    end
+
     # Domain for which this cookie applies, as a +String+
-    attr_accessor :domain
+    attr_reader :domain
+    # Set domain for which this cookie applies
+    def domain=(str)
+      if str and ((str = str.b).bytesize > 255 or !DOMAIN_VALUE_RE.match?(str))
+        raise ArgumentError, "invalid domain: #{str.dump}"
+      end
+      @domain = str
+    end
+
     # Time at which this cookie expires, as a +Time+
     attr_accessor :expires
     # True if this cookie is secure; false otherwise
diff --git a/lib/cgi/core.rb b/lib/cgi/core.rb
index bec76e0749..62e606837a 100644
--- a/lib/cgi/core.rb
+++ b/lib/cgi/core.rb
@@ -188,17 +188,28 @@ class CGI
   # Using #header with the HTML5 tag maker will create a <header> element.
   alias :header :http_header
 
+  def _no_crlf_check(str)
+    if str
+      str = str.to_s
+      raise "A HTTP status or header field must not include CR and LF" if str =~ /[\r\n]/
+      str
+    else
+      nil
+    end
+  end
+  private :_no_crlf_check
+
   def _header_for_string(content_type) #:nodoc:
     buf = ''.dup
     if nph?()
-      buf << "#{$CGI_ENV['SERVER_PROTOCOL'] || 'HTTP/1.0'} 200 OK#{EOL}"
+      buf << "#{_no_crlf_check($CGI_ENV['SERVER_PROTOCOL']) || 'HTTP/1.0'} 200 OK#{EOL}"
       buf << "Date: #{CGI.rfc1123_date(Time.now)}#{EOL}"
-      buf << "Server: #{$CGI_ENV['SERVER_SOFTWARE']}#{EOL}"
+      buf << "Server: #{_no_crlf_check($CGI_ENV['SERVER_SOFTWARE'])}#{EOL}"
       buf << "Connection: close#{EOL}"
     end
-    buf << "Content-Type: #{content_type}#{EOL}"
+    buf << "Content-Type: #{_no_crlf_check(content_type)}#{EOL}"
     if @output_cookies
-      @output_cookies.each {|cookie| buf << "Set-Cookie: #{cookie}#{EOL}" }
+      @output_cookies.each {|cookie| buf << "Set-Cookie: #{_no_crlf_check(cookie)}#{EOL}" }
     end
     return buf
   end # _header_for_string
@@ -213,9 +224,9 @@ class CGI
     ## NPH
     options.delete('nph') if defined?(MOD_RUBY)
     if options.delete('nph') || nph?()
-      protocol = $CGI_ENV['SERVER_PROTOCOL'] || 'HTTP/1.0'
+      protocol = _no_crlf_check($CGI_ENV['SERVER_PROTOCOL']) || 'HTTP/1.0'
       status = options.delete('status')
-      status = HTTP_STATUS[status] || status || '200 OK'
+      status = HTTP_STATUS[status] || _no_crlf_check(status) || '200 OK'
       buf << "#{protocol} #{status}#{EOL}"
       buf << "Date: #{CGI.rfc1123_date(Time.now)}#{EOL}"
       options['server'] ||= $CGI_ENV['SERVER_SOFTWARE'] || ''
@@ -223,38 +234,38 @@ class CGI
     end
     ## common headers
     status = options.delete('status')
-    buf << "Status: #{HTTP_STATUS[status] || status}#{EOL}" if status
+    buf << "Status: #{HTTP_STATUS[status] || _no_crlf_check(status)}#{EOL}" if status
     server = options.delete('server')
-    buf << "Server: #{server}#{EOL}" if server
+    buf << "Server: #{_no_crlf_check(server)}#{EOL}" if server
     connection = options.delete('connection')
-    buf << "Connection: #{connection}#{EOL}" if connection
+    buf << "Connection: #{_no_crlf_check(connection)}#{EOL}" if connection
     type = options.delete('type')
-    buf << "Content-Type: #{type}#{EOL}" #if type
+    buf << "Content-Type: #{_no_crlf_check(type)}#{EOL}" #if type
     length = options.delete('length')
-    buf << "Content-Length: #{length}#{EOL}" if length
+    buf << "Content-Length: #{_no_crlf_check(length)}#{EOL}" if length
     language = options.delete('language')
-    buf << "Content-Language: #{language}#{EOL}" if language
+    buf << "Content-Language: #{_no_crlf_check(language)}#{EOL}" if language
     expires = options.delete('expires')
     buf << "Expires: #{CGI.rfc1123_date(expires)}#{EOL}" if expires
     ## cookie
     if cookie = options.delete('cookie')
       case cookie
       when String, Cookie
-        buf << "Set-Cookie: #{cookie}#{EOL}"
+        buf << "Set-Cookie: #{_no_crlf_check(cookie)}#{EOL}"
       when Array
         arr = cookie
-        arr.each {|c| buf << "Set-Cookie: #{c}#{EOL}" }
+        arr.each {|c| buf << "Set-Cookie: #{_no_crlf_check(c)}#{EOL}" }
       when Hash
         hash = cookie
-        hash.each_value {|c| buf << "Set-Cookie: #{c}#{EOL}" }
+        hash.each_value {|c| buf << "Set-Cookie: #{_no_crlf_check(c)}#{EOL}" }
       end
     end
     if @output_cookies
-      @output_cookies.each {|c| buf << "Set-Cookie: #{c}#{EOL}" }
+      @output_cookies.each {|c| buf << "Set-Cookie: #{_no_crlf_check(c)}#{EOL}" }
     end
     ## other headers
     options.each do |key, value|
-      buf << "#{key}: #{value}#{EOL}"
+      buf << "#{_no_crlf_check(key)}: #{_no_crlf_check(value)}#{EOL}"
     end
     return buf
   end # _header_for_hash
diff --git a/lib/cgi/util.rb b/lib/cgi/util.rb
index 55e61bf984..5a5c77ac97 100644
--- a/lib/cgi/util.rb
+++ b/lib/cgi/util.rb
@@ -5,24 +5,57 @@ class CGI
   extend Util
 end
 module CGI::Util
-  @@accept_charset="UTF-8" unless defined?(@@accept_charset)
-  # URL-encode a string.
+  @@accept_charset = Encoding::UTF_8 unless defined?(@@accept_charset)
+
+  # URL-encode a string into application/x-www-form-urlencoded.
+  # Space characters (+" "+) are encoded with plus signs (+"+"+)
   #   url_encoded_string = CGI.escape("'Stop!' said Fred")
   #      # => "%27Stop%21%27+said+Fred"
   def escape(string)
     encoding = string.encoding
-    string.b.gsub(/([^ a-zA-Z0-9_.\-~]+)/) do |m|
+    buffer = string.b
+    buffer.gsub!(/([^ a-zA-Z0-9_.\-~]+)/) do |m|
       '%' + m.unpack('H2' * m.bytesize).join('%').upcase
-    end.tr(' ', '+').force_encoding(encoding)
+    end
+    buffer.tr!(' ', '+')
+    buffer.force_encoding(encoding)
   end
 
-  # URL-decode a string with encoding(optional).
+  # URL-decode an application/x-www-form-urlencoded string with encoding(optional).
   #   string = CGI.unescape("%27Stop%21%27+said+Fred")
   #      # => "'Stop!' said Fred"
-  def unescape(string,encoding=@@accept_charset)
-    str=string.tr('+', ' ').b.gsub(/((?:%[0-9a-fA-F]{2})+)/) do |m|
+  def unescape(string, encoding = @@accept_charset)
+    str = string.tr('+', ' ')
+    str = str.b
+    str.gsub!(/((?:%[0-9a-fA-F]{2})+)/) do |m|
+      [m.delete('%')].pack('H*')
+    end
+    str.force_encoding(encoding)
+    str.valid_encoding? ? str : str.force_encoding(string.encoding)
+  end
+
+  # URL-encode a string following RFC 3986
+  # Space characters (+" "+) are encoded with (+"%20"+)
+  #   url_encoded_string = CGI.escape("'Stop!' said Fred")
+  #      # => "%27Stop%21%27%20said%20Fred"
+  def escapeURIComponent(string)
+    encoding = string.encoding
+    buffer = string.b
+    buffer.gsub!(/([^a-zA-Z0-9_.\-~]+)/) do |m|
+      '%' + m.unpack('H2' * m.bytesize).join('%').upcase
+    end
+    buffer.force_encoding(encoding)
+  end
+
+  # URL-decode a string following RFC 3986 with encoding(optional).
+  #   string = CGI.unescape("%27Stop%21%27+said%20Fred")
+  #      # => "'Stop!'+said Fred"
+  def unescapeURIComponent(string, encoding = @@accept_charset)
+    str = string.b
+    str.gsub!(/((?:%[0-9a-fA-F]{2})+)/) do |m|
       [m.delete('%')].pack('H*')
-    end.force_encoding(encoding)
+    end
+    str.force_encoding(encoding)
     str.valid_encoding? ? str : str.force_encoding(string.encoding)
   end