uri-whatwg_parser 0.2.1 → 0.3.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: bc2c1e5428af4eaed582a417234dec63ac59554af14bbbbd1cc0c3017e8e32b9
-  data.tar.gz: 749b4ba051cb58a73f0ef8ebe64d6426a4d1f0803176641ff75af539757130c1
+  metadata.gz: 340c89cc9b0c80699547d7daa0a33c3c081250ebee1dd1178a36ee288a484a22
+  data.tar.gz: 961034f93393490c76ad015aea1524b27b5955ef38112892d50b2a1e15a3ac50
 SHA512:
-  metadata.gz: a28ffa266d8013c02ed0da9bfd285eaff113527772019a135ddd092310b942c01b6beab710336d37665f8ec3dcc18829edad14a0a97168ef1fc3c651148eb4f0
-  data.tar.gz: 1ba90031895fe24a39b1ae7ff127ea7af821cb39a0801b83234c40b63801c9ee6c886f1bb0e423bc9c6e9c817d31bea29661db7e22978c8b66891b9864a953f4
+  metadata.gz: 987525847dbe6f04d4a2a4c3c8e338d57ebd964d4312528bc16e5e515ba7460922c3fedf08a4151227dbd1fbca11be4af38db6db53e1d741992460fd89c86bd0
+  data.tar.gz: 12eae3fe44a8136aa66a55d1aa7e12b172980ae96470f890e34e721fae4c1f3f23155ff1a2e1db305ce2f239cdf239a7db43fe5feb153eb11f3f230c09fd77a4

data/CHANGELOG.md

@@ -1,3 +1,14 @@
+## 0.3.1
+
+* Fix a bug of parsing Windows drive letter
+* Correctly set a password when a username is nil
+* Fix setter methods behavior with `URI object created by `build` method
+
+## 0.3.0
+
+* Fix several incorrect parsing processes
+* Fix serialize URL
+
 ## 0.2.1
 
 * Improve the performance of `parse`

data/README.md

@@ -1,8 +1,7 @@
 # URI::WhatwgParser
-
 Ruby implementation of the [WHATWG URL Living Standard](https://url.spec.whatwg.org/).
 
-The latest revision that this package implements of the standard is [13 January 2026](https://url.spec.whatwg.org/commit-snapshots/b6b3251fe911ab33d68fb051efe0e4d39ae4145e/).
+The latest revision that this package implements of the standard is [14 April 2026](https://url.spec.whatwg.org/commit-snapshots/b11d73b8caefe90403afe19210db05acba897722/)
 
 ## Installation
 
@@ -27,10 +26,6 @@ After checking out the repo, run `bin/setup` to install dependencies. Then, run
 
 To install this gem onto your local machine, run `bundle exec rake install`. To release a new version, update the version number in `version.rb`, and then run `bundle exec rake release`, which will create a git tag for the version, push git commits and the created tag, and push the `.gem` file to [rubygems.org](https://rubygems.org).
 
-## TODO
-
-* Support validations
-
 ## Contributing
 
 Bug reports and pull requests are welcome on GitHub at https://github.com/y-yagi/uri-whatwg_parser.

data/lib/uri/whatwg_parser/generic.rb

@@ -3,16 +3,11 @@ require "uri/generic"
 module URI
   class WhatwgParser
     module Generic
-      def initialize(scheme,
-                     userinfo, host, port, registry,
-                     path, opaque,
-                     query,
-                     fragment,
-                     parser = DEFAULT_PARSER,
-                     arg_check = false)
-
-        return super unless URI::DEFAULT_PARSER.is_a?(URI::WhatwgParser)
-        return super if registry
+      def initialize(scheme, userinfo, host, port, registry, path, opaque, query, fragment, parser = DEFAULT_PARSER, arg_check = false)
+        @parsed_by_whatwg_parser = parser.is_a?(URI::WhatwgParser)
+        unless parser.is_a?(URI::WhatwgParser)
+          return super(scheme, userinfo, host, port, registry, path, opaque, query, fragment, URI::RFC3986_PARSER, arg_check)
+        end
 
         @scheme = nil
         @user = nil
@@ -20,6 +15,7 @@ module URI
         @host = nil
         @port = nil
         @path = nil
+        @raw_path = nil
         @query = nil
         @opaque = nil
         @fragment = nil
@@ -30,66 +26,69 @@ module URI
         self.set_port(port)
         self.set_userinfo(userinfo)
         self.set_path(path)
-        self.query = query
+        @query = query
         self.set_opaque(opaque)
-        self.fragment=(fragment)
+        @fragment = fragment
+        @raw_path = parser&.path
 
         self.set_path("") if !@path && !@opaque
-        DEFAULT_PARSER.parse(to_s) if arg_check
+        parser.parse(to_s) if arg_check
 
         @scheme&.freeze
         self.set_port(self.default_port) if self.default_port && !@port
       end
 
       def merge(oth)
-        URI::DEFAULT_PARSER.join(self.to_s, oth.to_s)
+        return super unless @parsed_by_whatwg_parser
+
+        parser.join(self.to_s, oth.to_s)
       end
       alias + merge
 
       def scheme=(v)
-        return super unless URI::DEFAULT_PARSER.is_a?(URI::WhatwgParser)
+        return super unless @parsed_by_whatwg_parser
         return if v.nil? || v.empty?
 
-        parse_result = URI::DEFAULT_PARSER.split("#{v}:", url: self, state_override: :scheme_start_state)
+        parse_result = parser.split("#{v}:", url: self, state_override: :scheme_start_state)
         set_scheme(parse_result[0])
         set_port(parse_result[3])
       end
 
       def user=(v)
-        return super unless URI::DEFAULT_PARSER.is_a?(URI::WhatwgParser)
+        return super unless @parsed_by_whatwg_parser
         return v unless v
 
         if host.nil? || host.empty? || scheme == "file"
           raise InvalidURIError, "cannot set user when host is nil or file schme"
         end
-        set_user(URI::DEFAULT_PARSER.utf8_percent_encode_string(v, URI::WhatwgParser::USERINFO_PERCENT_ENCODE_SET))
+        set_user(parser.utf8_percent_encode_string(v, URI::WhatwgParser::USERINFO_PERCENT_ENCODE_SET))
       end
 
       def password=(v)
-        return super unless URI::DEFAULT_PARSER.is_a?(URI::WhatwgParser)
+        return super unless @parsed_by_whatwg_parser
         return v unless v
 
         if host.nil? || host.empty? || scheme == "file"
           raise InvalidURIError, "cannot set password when host is nil or file schme"
         end
-        set_password(URI::DEFAULT_PARSER.utf8_percent_encode_string(v, URI::WhatwgParser::USERINFO_PERCENT_ENCODE_SET))
+        set_password(parser.utf8_percent_encode_string(v, URI::WhatwgParser::USERINFO_PERCENT_ENCODE_SET))
       end
 
       def host=(v)
-        return super unless URI::DEFAULT_PARSER.is_a?(URI::WhatwgParser)
+        return super unless @parsed_by_whatwg_parser
         return if v.nil?
 
         if @opaque
-          raise InvalidURIError, "cannot set host with registry or opaque"
+          raise InvalidURIError, "cannot set host with opaque"
         end
 
-        parse_result = URI::DEFAULT_PARSER.split(v.to_s, url: self, state_override: :host_state)
+        parse_result = parser.split(v.to_s, url: self, state_override: :host_state)
         set_host(parse_result[2])
         set_port(parse_result[3])
       end
 
       def port=(v)
-        return super unless URI::DEFAULT_PARSER.is_a?(URI::WhatwgParser)
+        return super unless @parsed_by_whatwg_parser
         return if v.nil?
 
         if v.to_s.empty?
@@ -101,24 +100,55 @@ module URI
           raise InvalidURIError, "cannot set port when host is nil or scheme is file"
         end
 
-        parse_result = URI::DEFAULT_PARSER.split("#{v}:", url: self, state_override: :port_state)
+        parse_result = parser.split("#{v}:", url: self, state_override: :port_state)
         set_port(parse_result[3])
       end
 
       def path=(v)
-        return super unless URI::DEFAULT_PARSER.is_a?(URI::WhatwgParser)
+        return super unless @parsed_by_whatwg_parser
         return if v.nil?
 
         if @opaque
           raise InvalidURIError, "path conflicts with opaque"
         end
 
-        parse_result = URI::DEFAULT_PARSER.split(v.to_s, url: self, state_override: :path_start_state)
+        parse_result = parser.split(v.to_s, url: self, state_override: :path_start_state)
+        @raw_path = parser.path
         set_path(parse_result[5])
       end
 
+      def query=(v)
+        return super unless @parsed_by_whatwg_parser
+
+        if v.nil? || v.empty?
+          @query = nil
+          return
+        end
+
+        v = v.start_with?("?") ? v[1..-1] : v
+        @query = +""
+
+        parse_result = parser.split(v, url: self, state_override: :query_state)
+        @query = parse_result[7].to_s
+      end
+
+      def fragment=(v)
+        return super unless @parsed_by_whatwg_parser
+
+        if v.nil? || v.empty?
+          @fragment = nil
+          return
+        end
+
+        v = v.start_with?("#") ? v[1..-1] : v
+        @fragment = +""
+
+        parse_result = parser.split(v, url: self, state_override: :fragment_state)
+        @fragment = parse_result[8].to_s
+      end
+
       def userinfo=(userinfo)
-        return super unless URI::DEFAULT_PARSER.is_a?(URI::WhatwgParser)
+        return super unless @parsed_by_whatwg_parser
 
         user, password = split_userinfo(userinfo)
         self.user = user
@@ -126,17 +156,59 @@ module URI
       end
 
       def check_opaque(v)
-        return super unless URI::DEFAULT_PARSER.is_a?(URI::WhatwgParser)
+        return super unless @parsed_by_whatwg_parser
+
         return v unless v
 
-        if @host || @port || @user || @path
-          raise InvalidURIError, "cannot set opaque with host, port, userinfo or path"
+        if @host || @port || @user
+          raise InvalidURIError, "cannot set opaque with host, port, or userinfo"
         end
 
         self.set_opaque(v)
-        DEFAULT_PARSER.parse(to_s)
+        # NOTE: WHATWG URL Living Standard doesn't define "opaque" setter. So parse a URL whole.
+        parser.parse(to_s)
         true
       end
+
+      def to_s
+        return super unless @parsed_by_whatwg_parser
+
+        str = "".dup
+        if @scheme
+          str << @scheme
+          str << ":"
+        end
+
+        if @host || %w[file postgres].include?(@scheme)
+          str << "//"
+        end
+        if self.userinfo
+          str << self.userinfo
+          str << "@"
+        end
+        if @host
+          str << @host
+        end
+        if @port && @port != self.default_port
+          str << ":"
+          str << @port.to_s
+        end
+        if @host.nil? && @opaque.nil? && @raw_path && @raw_path.length > 1 && @raw_path[0] == ""
+          str << "/."
+        end
+        str << @path if @path
+        str << @opaque if @opaque
+        if @query
+          str << "?"
+          str << @query
+        end
+
+        if @fragment
+          str << "#"
+          str << @fragment
+        end
+        str
+      end
     end
   end
 end

data/lib/uri/whatwg_parser/version.rb

@@ -2,6 +2,6 @@
 
 module URI
   class WhatwgParser
-    VERSION = "0.2.1"
+    VERSION = "0.3.1"
   end
 end

data/lib/uri/whatwg_parser.rb

@@ -25,7 +25,7 @@ module URI
 
     WINDOWS_DRIVE_LETTER = Regexp.new("\\A([a-zA-Z][:|])\\z")
     NORMALIZED_WINDOWS_DRIVE_LETTER = Regexp.new("\\A([a-zA-Z][:])\\z")
-    STARTS_WITH_WINDOWS_DRIVE_LETTER = Regexp.new("\\A([a-zA-Z][:|])(?:[/\\?#])?\\z")
+    FILE_OTHERWISE_CODE_POINTS = Set["/", "\\", "?", "#"]
 
     VALID_SIGNS_FOR_SCHEME = Set["+", "-", "."]
     DELIMITER_SIGNS = Set["/", "?", "#"]
@@ -36,6 +36,8 @@ module URI
     ASCII_ALPHA_UPPERCASE = Set.new(("A".."Z").to_a)
     ASCII_DIGIT = Set.new(("0".."9").to_a)
 
+    attr_reader :path
+
     def initialize
       reset
       @host_parser = HostParser.new
@@ -46,7 +48,7 @@ module URI
     end
 
     def parse(input, base: nil, url: nil, state_override: nil) # :nodoc:
-      URI.for(*self.split(input, base: base, url: url, state_override: state_override))
+      URI.for(*self.split(input, base: base, url: url, state_override: state_override), self)
     end
 
     def split(input, base: nil, url: nil, state_override: nil) # :nodoc:
@@ -54,8 +56,8 @@ module URI
       @base = nil
       if base != nil
         ary = split(base, base: nil)
-        @base = { scheme: ary[0], userinfo: ary[1], host: ary[2], port: ary[3], registry: ary[4], path: ary[5], opaque: ary[6], query: ary[7], fragment: ary[8]}
-        @base_paths = @paths
+        @base = { scheme: ary[0], userinfo: ary[1], host: ary[2], port: ary[3], query: ary[7], fragment: ary[8]}
+        @base_path = @path
         reset
       end
 
@@ -94,9 +96,19 @@ module URI
         @pos += 1
       end
 
-      userinfo = [@username, @password].compact.reject(&:empty?).join(":")
-      path = "/#{@paths.join("/")}" if @paths && !@paths.empty?
-      [@parse_result[:scheme], userinfo, @parse_result[:host], @parse_result[:port], @parse_result[:registry], path, @parse_result[:opaque], @parse_result[:query], @parse_result[:fragment]]
+      if @password
+        userinfo = "#{@username}:#{@password}"
+      else
+        userinfo = @username
+      end
+      if @path
+        if @path.is_a?(Array)
+          path = "/#{@path.join("/")}"
+        else
+          opaque = @path
+        end
+      end
+      [@parse_result[:scheme], userinfo, @parse_result[:host], @parse_result[:port], nil, path, opaque, @parse_result[:query], @parse_result[:fragment]]
     end
 
     def join(*uris)
@@ -143,10 +155,10 @@ module URI
       @at_sign_seen = nil
       @password_token_seen = nil
       @inside_brackets = nil
-      @paths = nil
+      @path = nil
       @username = nil
       @password = nil
-      @parse_result = { scheme: nil, host: nil, port: nil, registry: nil, path: nil, opaque: nil, query: nil, fragment: nil }
+      @parse_result = {}
       @state_override = nil
       @state = :scheme_start_state
       @special_url = nil
@@ -207,7 +219,7 @@ module URI
           @state = :path_or_authority_state
           @pos += 1
         else
-          @parse_result[:opaque] = +""
+          @path = +""
           @state = :opaque_path_state
         end
       elsif @state_override.nil?
@@ -220,14 +232,14 @@ module URI
     end
 
     def no_scheme_state(c)
-      raise ParseError, "scheme is missing" if @base.nil? || (!@base[:opaque].nil? && c != "#")
+      raise ParseError, "scheme is missing" if @base.nil? || (has_opaque_path?(@base_path) && c != "#")
 
-      if !@base[:opaque].nil? && c == "#"
+      if has_opaque_path?(@base_path) && c == "#"
         @parse_result[:scheme] = @base[:scheme]
         @special_url = special_url?(@base[:scheme])
-        @paths = @base_paths
+        @path = @base_path
         @parse_result[:query] = @base[:query]
-        @parse_result[:fragment] = nil
+        @parse_result[:fragment] = +""
         @state = :fragment_state
       elsif @base[:scheme] != "file"
         @state = :relative_state
@@ -268,14 +280,14 @@ module URI
         @username, @password = @base[:userinfo].split(":") if @base[:userinfo]
         @parse_result[:host] = @base[:host]
         @parse_result[:port] = @base[:port]
-        @paths = @base_paths
+        @path = @base_path
         @parse_result[:query] = @base[:query]
 
         if c == "?"
-          @parse_result[:query] = nil
+          @parse_result[:query] = +""
           @state = :query_state
         elsif c == "#"
-          @parse_result[:fragment] = nil
+          @parse_result[:fragment] = +""
           @state = :fragment_state
         elsif !c.nil?
           @parse_result[:query] = nil
@@ -386,7 +398,7 @@ module URI
         @buffer << c
       elsif c.nil? || DELIMITER_SIGNS.include?(c) || (@special_url && c == "\\") || @state_override
         unless @buffer.empty?
-          port = Integer(@buffer, 10)
+          port = @buffer.to_i
           raise ParseError, "port is invalid value" if port < 0 || port > 65535
           if SPECIAL_SCHEME[@parse_result[:scheme]] == port
             @parse_result[:port] = nil
@@ -401,7 +413,6 @@ module URI
           end
         end
 
-        raise ParseError, "port is invalid value" if @state_override
         @state = :path_start_state
         @pos -= 1
       else
@@ -413,24 +424,24 @@ module URI
       @parse_result[:scheme] = "file"
       @special_url = true
       @parse_result[:host] = nil
-
       if c == "/" || c == "\\"
         @state = :file_slash_state
       elsif !@base.nil? && @base[:scheme] == "file"
         @parse_result[:host] = @base[:host]
+        @path = @base_path
         @parse_result[:query] = @base[:query]
         if c == "?"
-          @parse_result[:query] = nil
+          @parse_result[:query] = +""
           @state = :query_state
         elsif c == "#"
-          @parse_result[:fragment] = nil
+          @parse_result[:fragment] = +""
           @state = :fragment_state
         elsif !c.nil?
           @parse_result[:query] = nil
           if !starts_with_windows_drive_letter?(rest)
             shorten_url_path
           else
-            @paths = nil
+            @path = nil
           end
           @state = :path_state
           @pos -= 1
@@ -447,11 +458,9 @@ module URI
       else
         if !@base.nil? && @base[:scheme] == "file"
           @parse_result[:host] = @base[:host]
-          if !starts_with_windows_drive_letter?(rest) && @base_paths && normalized_windows_drive_letter?(@base_paths[0])
-            if @paths.nil?
-              @paths ||= []
-              @paths[0] = @base_paths[0]
-            end
+          if !starts_with_windows_drive_letter?(rest) && @base_path && normalized_windows_drive_letter?(@base_path[0])
+            @path = [] if @path.nil?
+            @path << @base_path[0]
           end
         end
         @state = :path_state
@@ -493,45 +502,47 @@ module URI
         @pos -= 1 if c != "/" && c != "\\"
         @state = :path_state
       elsif !@state_override && c == "?"
+        @parse_result[:query] = +""
         @state = :query_state
       elsif !@state_override && c == "#"
+        @parse_result[:fragment] = +""
         @state = :fragment_state
       elsif c != nil
         @pos -= 1 if c != "/"
         @state = :path_state
       elsif @state_override && @parse_result[:host].nil?
-        @paths ||= []
-        @paths << ""
+        @path ||= []
+        @path << ""
       end
     end
 
     def path_state(c)
-      @paths ||= []
+      @path ||= []
 
       if (c.nil? || c == "/") || (@special_url && c == "\\") || (!@state_override && (c == "?" || c == "#"))
         if double_dot_path_segments?(@buffer)
           shorten_url_path
 
           if c != "/" && !(@special_url && c == "\\")
-            @paths << ""
+            @path << ""
           end
         elsif single_dot_path_segments?(@buffer) && c != "/" && !((@special_url && c == "\\"))
-          @paths << ""
+          @path << ""
         elsif !single_dot_path_segments?(@buffer)
-          if @parse_result[:scheme] == "file" && @paths.empty? && windows_drive_letter?(@buffer)
+          if @parse_result[:scheme] == "file" && @path.empty? && windows_drive_letter?(@buffer)
             @buffer[1] = ":"
           end
 
-          @paths << @buffer
+          @path << @buffer
         end
 
         @buffer = +""
 
         if c == "?"
-          @parse_result[:query] = nil
+          @parse_result[:query] = +""
           @state = :query_state
         elsif c == "#"
-          @parse_result[:frament] = nil
+          @parse_result[:fragment] = +""
           @state = :fragment_state
         end
       else
@@ -541,20 +552,20 @@ module URI
 
     def opaque_path_state(c)
       if c == "?"
-        @parse_result[:query] = nil
+        @parse_result[:query] = +""
         @state = :query_state
       elsif c == "#"
-        @parse_result[:fragment] = nil
+        @parse_result[:fragment] = +""
         @state = :fragment_state
       elsif c == " "
         first_of_rest = @input_chars[@pos + 1]
         if first_of_rest == "?" || first_of_rest == "#"
-          @parse_result[:opaque] << "%20"
+          @path += "%20"
         else
-          @parse_result[:opaque] << " "
+          @path += " "
         end
       elsif !c.nil?
-        @parse_result[:opaque] << utf8_percent_encode(c, C0_CONTROL_PERCENT_ENCODE_SET)
+        @path += utf8_percent_encode(c, C0_CONTROL_PERCENT_ENCODE_SET)
       end
     end
 
@@ -564,7 +575,10 @@ module URI
         # TODO: We need to consider encoding here.
         @parse_result[:query] = utf8_percent_encode_string(@buffer, query_percent_encode_set)
         @buffer.clear
-        @state = :fragment_state if c == "#"
+        if c == "#"
+          @parse_result[:fragment] = +""
+          @state = :fragment_state
+        end
       elsif !c.nil?
         @buffer << c
       end
@@ -580,7 +594,9 @@ module URI
     end
 
     def starts_with_windows_drive_letter?(str)
-      STARTS_WITH_WINDOWS_DRIVE_LETTER.match?(str)
+      return false if str.length < 2
+      return false unless windows_drive_letter?(str[0, 2])
+      str.length == 2 || FILE_OTHERWISE_CODE_POINTS.include?(str[2])
     end
 
     def normalized_windows_drive_letter?(str)
@@ -600,9 +616,9 @@ module URI
     end
 
     def shorten_url_path
-      return if @paths.nil?
-      return if @parse_result[:scheme] == "file" && @paths.length == 1 && normalized_windows_drive_letter?(@paths.first)
-      @paths.pop
+      return if @path.nil? || @path.is_a?(String)
+      return if @parse_result[:scheme] == "file" && @path.length == 1 && normalized_windows_drive_letter?(@path.first)
+      @path.pop
     end
 
     def includes_credentials?
@@ -610,7 +626,7 @@ module URI
     end
 
     def rest
-      @input_chars[@pos + 1..]&.join
+      @input_chars[@pos..]&.join
     end
 
     def convert_to_uri(uri)
@@ -638,6 +654,10 @@ module URI
         end
       end
     end
+
+    def has_opaque_path?(path)
+      path.is_a?(String)
+    end
   end
 
   WHATWG_PARSER = URI::WhatwgParser.new

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: uri-whatwg_parser
 version: !ruby/object:Gem::Version
-  version: 0.2.1
+  version: 0.3.1
 platform: ruby
 authors:
 - Yuji Yaginuma
@@ -66,14 +66,14 @@ required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
-      version: 3.2.0
+      version: 3.0.0
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 4.0.3
+rubygems_version: 4.0.10
 specification_version: 4
 summary: Ruby implementation of the WHATWG URL Living Standard
 test_files: []