Allow "th" to occur in thead, tbody, and tfoot

Fixes https://gitlab.com/yorickpeterse/oga/issues/190

lib/oga/xml/lexer.rb

@@ -58,7 +58,11 @@ module Oga
 
       HTML_SCRIPT_ELEMENTS = Whitelist.new(%w{script template})
 
-      HTML_TABLE_ROW_ELEMENTS = Whitelist.new(%w{tr}) + HTML_SCRIPT_ELEMENTS
+      # The elements that may occur in a thead, tbody, or tfoot.
+      #
+      # Technically "th" is not allowed per the HTML5 spec, but it's so commonly
+      # used in these elements that we allow it anyway.
+      HTML_TABLE_ROW_ELEMENTS = Whitelist.new(%w{tr th}) + HTML_SCRIPT_ELEMENTS
 
       # Elements that should be closed automatically before a new opening tag is
       # processed.

spec/oga/html/lexer/closing_rules/tbody_spec.rb

@@ -53,5 +53,15 @@ describe Oga::XML::Lexer do
         [:T_ELEM_END, nil, 1]
       ])
     end
+
+    it 'lexes a <tbody> element containing a <th> element' do
+      expect(lex_html('<tbody><th>foo</th></tbody>')).to eq([
+        [:T_ELEM_NAME, 'tbody', 1],
+        [:T_ELEM_NAME, 'th', 1],
+        [:T_TEXT, 'foo', 1],
+        [:T_ELEM_END, nil, 1],
+        [:T_ELEM_END, nil, 1]
+      ])
+    end
   end
 end

spec/oga/html/lexer/closing_rules/tfoot_spec.rb

@@ -53,5 +53,15 @@ describe Oga::XML::Lexer do
         [:T_ELEM_END, nil, 1]
       ])
     end
+
+    it 'lexes a <tfoot> element containing a <th> element' do
+      expect(lex_html('<tfoot><th>foo</th></tfoot>')).to eq([
+        [:T_ELEM_NAME, 'tfoot', 1],
+        [:T_ELEM_NAME, 'th', 1],
+        [:T_TEXT, 'foo', 1],
+        [:T_ELEM_END, nil, 1],
+        [:T_ELEM_END, nil, 1]
+      ])
+    end
   end
 end

spec/oga/html/lexer/closing_rules/thead_spec.rb

@@ -53,5 +53,15 @@ describe Oga::XML::Lexer do
         [:T_ELEM_END, nil, 1]
       ])
     end
+
+    it 'lexes a <thead> element containing a <th> element' do
+      expect(lex_html('<thead><th>foo</th></thead>')).to eq([
+        [:T_ELEM_NAME, 'thead', 1],
+        [:T_ELEM_NAME, 'th', 1],
+        [:T_TEXT, 'foo', 1],
+        [:T_ELEM_END, nil, 1],
+        [:T_ELEM_END, nil, 1]
+      ])
+    end
   end
 end