gammo 0.1.0

data/lib/gammo/parser/insertion_mode/in_select_in_table.rb

@@ -0,0 +1,46 @@
+module Gammo
+  class Parser
+    # Section 12.2.6.4.17.
+    class InSelectInTable < InsertionMode
+      def start_tag_token(token)
+        case token.tag
+        when Tags::Caption, Tags::Table, Tags::Tbody, Tags::Tfoot, Tags::Thead, Tags::Tr, Tags::Td, Tags::Th
+          if token.instance_of?(Tokenizer::EndTagToken) && parser.element_in_scope?(TABLE_SCOPE, token.tag)
+            # ignore the token
+            halt true
+          end
+          parser.open_elements.reverse_each_with_index do |elm, i|
+            if elm.tag == Tags::Select
+              parser.open_elements = parser.open_elements.slice(0, i)
+              break
+            end
+          end
+          parser.reset_insertion_mode
+          halt false
+        end
+      end
+
+      def end_tag_token(token)
+        case token.tag
+        when Tags::Caption, Tags::Table, Tags::Tbody, Tags::Tfoot, Tags::Thead, Tags::Tr, Tags::Td, Tags::Th
+          if token.instance_of?(Tokenizer::EndTagToken) && !parser.element_in_scope?(TABLE_SCOPE, token.tag)
+            # ignore the token
+            halt true
+          end
+          parser.open_elements.reverse_each_with_index do |elm, i|
+            if elm.tag == Tags::Select
+              parser.open_elements = parser.open_elements.slice(0, i)
+              break
+            end
+          end
+          parser.reset_insertion_mode
+          halt false
+        end
+      end
+
+      def default(_)
+        halt InSelect.new(parser).process
+      end
+    end
+  end
+end

data/lib/gammo/parser/insertion_mode/in_table.rb

@@ -0,0 +1,114 @@
+module Gammo
+  class Parser
+    # Section 12.2.6.4.9.
+    class InTable < InsertionMode
+      def text_token(token)
+        token.data = token.data.gsub("\x00", "")
+        case parser.open_elements.last.tag
+        when Tags::Table, Tags::Tbody, Tags::Tfoot, Tags::Thead, Tags::Tr
+          if token.data.strip == ""
+            parser.add_text token.data
+            halt true
+          end
+        end
+      end
+
+      def start_tag_token(token)
+        case token.tag
+        when Tags::Caption
+          parser.clear_stack_to_context(TABLE_SCOPE)
+          parser.active_formatting_elements << Node::DEFAULT_SCOPE_MARKER
+          parser.add_element
+          parser.insertion_mode = InCaption
+          halt true
+        when Tags::Colgroup
+          parser.clear_stack_to_context(TABLE_SCOPE)
+          parser.add_element
+          parser.insertion_mode = InColumnGroup
+          halt true
+        when Tags::Col
+          parser.parse_implied_token(Tokenizer::StartTagToken, Tags::Colgroup, Tags::Colgroup.to_s)
+          halt false
+        when Tags::Tbody, Tags::Tfoot, Tags::Thead
+          parser.clear_stack_to_context(TABLE_SCOPE)
+          parser.add_element
+          parser.insertion_mode = InTableBody
+          halt true
+        when Tags::Td, Tags::Th, Tags::Tr
+          parser.parse_implied_token(Tokenizer::StartTagToken, Tags::Tbody, Tags::Tbody.to_s)
+          halt false
+        when Tags::Table
+          if parser.pop_until(TABLE_SCOPE, Tags::Table)
+            parser.reset_insertion_mode
+            halt false
+          end
+          # ignore the token
+          halt true
+        when Tags::Style, Tags::Script, Tags::Template
+          halt InHead.new(parser).process
+        when Tags::Input
+          token.attributes.each do |attr|
+            # skip setting frameset_ok = false
+            if attr.key == 'type' && attr.value.downcase == 'hidden'
+              parser.add_element
+              parser.open_elements.pop
+              halt true
+            end
+          end
+        when Tags::Form
+          # ignore the token
+          halt true if parser.form || parser.open_elements.any? { |open_element| open_element.tag == Tags::Template }
+          parser.add_element
+          parser.form = parser.open_elements.pop
+        when Tags::Select
+          parser.reconstruct_active_formatting_elements
+          case parser.top.tag
+          when Tags::Table, Tags::Tbody, Tags::Tfoot, Tags::Thead, Tags::Tr
+            parser.foster_parenting = true
+          end
+          parser.add_element
+          parser.foster_parenting = false
+          parser.frameset_ok = true
+          parser.insertion_mode = InSelectInTable
+          halt true
+        end
+      end
+
+      def end_tag_token(token)
+        case token.tag
+        when Tags::Table
+          parser.reset_insertion_mode if parser.pop_until(TABLE_SCOPE, Tags::Table)
+          # Ignore the token
+          halt true
+        when Tags::Body, Tags::Caption, Tags::Col, Tags::Colgroup, Tags::Html,
+          Tags::Tbody, Tags::Td, Tags::Tfoot, Tags::Th, Tags::Thead, Tags::Tr
+          # Ignore the token
+          halt true
+        when Tags::Template
+          halt InHead.new(parser).process
+        end
+      end
+
+      def comment_token(token)
+        parser.add_child(Node::Comment.new(data: token.data))
+        halt true
+      end
+
+      def doctype_token(token)
+        # Ignore the token
+        halt true
+      end
+
+      def error_token(token)
+        InBody.new(parser).process
+      end
+
+      def default(_)
+        parser.foster_parenting = true
+        result = InBody.new(parser).process
+        parser.foster_parenting = false
+        halt result
+      end
+    end
+  end
+end

data/lib/gammo/parser/insertion_mode/in_table_body.rb

@@ -0,0 +1,55 @@
+module Gammo
+  class Parser
+    # Section 12.2.6.4.13.
+    class InTableBody < InsertionMode
+      def start_tag_token(token)
+        case token.tag
+        when Tags::Tr
+          parser.clear_stack_to_context(TABLE_BODY_SCOPE)
+          parser.add_element
+          parser.insertion_mode = InRow
+          halt true
+        when Tags::Td, Tags::Th
+          parser.parse_implied_token(Tokenizer::StartTagToken, Tags::Tr, Tags::Tr.to_s)
+          halt false
+        when Tags::Caption, Tags::Col, Tags::Colgroup, Tags::Tbody, Tags::Tfoot, Tags::Thead
+          # ignore the token
+          halt true unless parser.pop_until(TABLE_SCOPE, Tags::Tbody, Tags::Thead, Tags::Tfoot)
+          parser.insertion_mode = InTable
+          halt false
+        end
+      end
+
+      def end_tag_token(token)
+        case token.tag
+        when Tags::Tbody, Tags::Tfoot, Tags::Thead
+          if parser.element_in_scope?(TABLE_SCOPE, token.tag)
+            parser.clear_stack_to_context(TABLE_BODY_SCOPE)
+            parser.open_elements.pop
+            parser.insertion_mode = InTable
+          end
+          halt true
+        when Tags::Table
+          if parser.pop_until(TABLE_SCOPE, Tags::Tbody, Tags::Thead, Tags::Tfoot)
+            parser.insertion_mode = InTable
+            halt false
+          end
+          # ignore the token
+          halt true
+        when Tags::Body, Tags::Caption, Tags::Colgroup, Tags::Html, Tags::Td, Tags::Th, Tags::Tr
+          # ignore the token
+          halt true
+        end
+      end
+
+      def comment_token(token)
+        parser.add_child(Node::Comment.new(data: token.data))
+        halt true
+      end
+
+      def default(_)
+        halt InTable.new(parser).process
+      end
+    end
+  end
+end

data/lib/gammo/parser/insertion_mode/in_template.rb

@@ -0,0 +1,80 @@
+module Gammo
+  class Parser
+    # Section 12.2.6.4.18.
+    class InTemplate < InsertionMode
+      def text_token(token)
+        halt InBody.new(parser).process
+      end
+
+      def comment_token(token)
+        halt InBody.new(parser).process
+      end
+
+      def doctype_token(token)
+        halt InBody.new(parser).process
+      end
+
+      def start_tag_token(token)
+        case token.tag
+        when Tags::Base, Tags::Basefont, Tags::Bgsound, Tags::Link, Tags::Meta, Tags::Noframes, Tags::Script, Tags::Style, Tags::Template, Tags::Title
+          halt InHead.new(parser).process
+        when Tags::Caption, Tags::Colgroup, Tags::Tbody, Tags::Tfoot, Tags::Thead
+          parser.template_stack.pop
+          parser.template_stack << InTable
+          parser.insertion_mode = InTable
+          halt false
+        when Tags::Col
+          parser.template_stack.pop
+          parser.template_stack << InColumnGroup
+          parser.insertion_mode = InColumnGroup
+          halt false
+        when Tags::Tr
+          parser.template_stack.pop
+          parser.template_stack << InTableBody
+          parser.insertion_mode = InTableBody
+          halt false
+        when Tags::Td, Tags::Th
+          parser.template_stack.pop
+          parser.template_stack << InRow
+          parser.insertion_mode = InRow
+          halt false
+        else
+          parser.template_stack.pop
+          parser.template_stack << InBody
+          parser.insertion_mode = InBody
+          halt false
+        end
+      end
+
+      def end_tag_token(token)
+        case token.tag
+        when Tags::Template
+          halt InHead.new(parser).process
+        else
+          # ignore the token
+          halt true
+        end
+      end
+
+      def error_token(token)
+        halt true unless parser.open_elements.any? {|elm| elm.tag == Tags::Template }
+        # remove this divergence from the html5 spec
+        parser.generate_implied_end_tags
+        parser.open_elements.reverse_each_with_index do |elm, index|
+          if !elm.namespace && elm.tag == Tags::Template
+            parser.open_elements = parser.open_elements.slice(0, index)
+            break
+          end
+        end
+        parser.clear_active_formatting_elements
+        parser.template_stack.pop
+        parser.reset_insertion_mode
+        halt false
+      end
+
+      def default(_)
+        halt false
+      end
+    end
+  end
+end

data/lib/gammo/parser/insertion_mode/initial.rb

@@ -0,0 +1,152 @@
+require 'gammo/attribute'
+
+module Gammo
+  class Parser
+    # Section 12.2.6.4.1
+    class Initial < InsertionMode
+      def text_token(token)
+        token.data = token.data.lstrip
+        # it's all whitespace so ignore it.
+        halt true if token.data.length.zero?
+      end
+
+      def comment_token(token)
+        parser.document.append_child(Node::Comment.new(data: token.data))
+        halt true
+      end
+
+      def doctype_token(token)
+        n, quirks = parse_doctype(token.data)
+        parser.document.append_child(n)
+        parser.quirks = quirks
+        parser.insertion_mode = BeforeHTML
+        halt true
+      end
+
+      def default(_)
+        parser.quirks = true
+        parser.insertion_mode = BeforeHTML
+        halt false
+      end
+
+      QUIRKY_IDS = [
+        "+//silmaril//dtd html pro v0r11 19970101//",
+        "-//advasoft ltd//dtd html 3.0 aswedit + extensions//",
+        "-//as//dtd html 3.0 aswedit + extensions//",
+        "-//ietf//dtd html 2.0 level 1//",
+        "-//ietf//dtd html 2.0 level 2//",
+        "-//ietf//dtd html 2.0 strict level 1//",
+        "-//ietf//dtd html 2.0 strict level 2//",
+        "-//ietf//dtd html 2.0 strict//",
+        "-//ietf//dtd html 2.0//",
+        "-//ietf//dtd html 2.1e//",
+        "-//ietf//dtd html 3.0//",
+        "-//ietf//dtd html 3.2 final//",
+        "-//ietf//dtd html 3.2//",
+        "-//ietf//dtd html 3//",
+        "-//ietf//dtd html level 0//",
+        "-//ietf//dtd html level 1//",
+        "-//ietf//dtd html level 2//",
+        "-//ietf//dtd html level 3//",
+        "-//ietf//dtd html strict level 0//",
+        "-//ietf//dtd html strict level 1//",
+        "-//ietf//dtd html strict level 2//",
+        "-//ietf//dtd html strict level 3//",
+        "-//ietf//dtd html strict//",
+        "-//ietf//dtd html//",
+        "-//metrius//dtd metrius presentational//",
+        "-//microsoft//dtd internet explorer 2.0 html strict//",
+        "-//microsoft//dtd internet explorer 2.0 html//",
+        "-//microsoft//dtd internet explorer 2.0 tables//",
+        "-//microsoft//dtd internet explorer 3.0 html strict//",
+        "-//microsoft//dtd internet explorer 3.0 html//",
+        "-//microsoft//dtd internet explorer 3.0 tables//",
+        "-//netscape comm. corp.//dtd html//",
+        "-//netscape comm. corp.//dtd strict html//",
+        "-//o'reilly and associates//dtd html 2.0//",
+        "-//o'reilly and associates//dtd html extended 1.0//",
+        "-//o'reilly and associates//dtd html extended relaxed 1.0//",
+        "-//softquad software//dtd hotmetal pro 6.0::19990601::extensions to html 4.0//",
+        "-//softquad//dtd hotmetal pro 4.0::19971010::extensions to html 4.0//",
+        "-//spyglass//dtd html 2.0 extended//",
+        "-//sq//dtd html 2.0 hotmetal + extensions//",
+        "-//sun microsystems corp.//dtd hotjava html//",
+        "-//sun microsystems corp.//dtd hotjava strict html//",
+        "-//w3c//dtd html 3 1995-03-24//",
+        "-//w3c//dtd html 3.2 draft//",
+        "-//w3c//dtd html 3.2 final//",
+        "-//w3c//dtd html 3.2//",
+        "-//w3c//dtd html 3.2s draft//",
+        "-//w3c//dtd html 4.0 frameset//",
+        "-//w3c//dtd html 4.0 transitional//",
+        "-//w3c//dtd html experimental 19960712//",
+        "-//w3c//dtd html experimental 970421//",
+        "-//w3c//dtd w3 html//",
+        "-//w3o//dtd w3 html 3.0//",
+        "-//webtechs//dtd mozilla html 2.0//",
+        "-//webtechs//dtd mozilla html//"
+      ].freeze
+
+      def parse_doctype(s)
+        node = Node::Doctype.new
+        pos = s.index(?\s)
+        pos = s.length unless pos
+        node.data = s.slice(0, pos)
+        quirks = false
+        quirks = true if node.data != 'html'
+        node.data = node.data.downcase
+        s = s.slice(pos..-1).lstrip
+        return [node, quirks || s != ''] if s.length < 6
+
+        key = s.slice(0, 6).downcase
+        s = s.slice(6..-1)
+        while key == 'public' || key == 'system'
+          s = s.lstrip
+          break if s.empty?
+          quote = s[0]
+          break if quote != ?" && quote != ?'
+          s = s.slice(1..-1)
+          id = ''
+          q = s.index(quote)
+          if q
+            id = s.slice(0, q)
+            s = s.slice((q + 1)..-1)
+          else
+            id = s
+            s = ''
+          end
+          node.attributes << Attribute.new(key: key, value: id)
+          key = key == 'public' ? 'system' : ''
+          if key != '' || s != ''
+            quirks = true
+          elsif node.attributes.length > 0
+            if node.attributes.first.key == 'public'
+              pub = node.attributes.first.value.downcase
+              case pub
+              when '-//w3o//dtd w3 html strict 3.0//en//', '-/w3d/dtd html 4.0 transitional/en', 'html'
+                quirks = true
+              else
+                QUIRKY_IDS.each do |quirky|
+                  if pub.start_with?(quirky)
+                    quirks = true
+                    break
+                  end
+                end
+              end
+              if node.attributes.length == 1 && pub.start_with?('-//w3c//dtd html 4.01 frameset//') || pub.start_with?('-//w3c//dtd html 4.01 transitional//')
+                quirks = true
+              end
+            end
+            last = node.attributes.last
+            if last.key == 'system' && last.value.downcase == 'http://www.ibm.com/data/dtd/v11/ibmxhtml1-transitional.dtd'
+              quirks = true
+            end
+          end
+        end
+        [node, quirks]
+      end
+
+      private :parse_doctype
+    end
+  end
+end