From 1e0b7feb026d95f2b04706391a868d64b7e5de6e Mon Sep 17 00:00:00 2001
From: Yorick Peterse <yorickpeterse@gmail.com>
Date: Tue, 12 May 2015 00:35:00 +0200
Subject: [PATCH] Recursively closing of parent HTML elements

When closing certain HTML elements the lexer should also close whatever
parent elements remain. For example, consider the following HTML:

    <table>
        <thead>
            <tr>
                <th>Foo
                <th>Bar
        <tbody>
            ...
        </tbody>
    </table>

Here the "<tbody>" element shouldn't only close the "<th>Bar" element
but also the parent "<tr>" and "<thead>" elements. This ensures we'd end
up with the following HTML:

    <table>
        <thead>
            <tr>
                <th>Foo</th>
                <th>Bar</th>
            </tr>
        </thead>
        <tbody>
            ...
        </tbody>
    </table>

Instead of garbage along the lines of this:

    <table>
        <thead>
            <tr>
                <th>Foo</th>
                <th>Bar</th>
        <tbody>
            ...
        </tbody>
    </table></tr></thead>

Fixes #99 (hopefully for good this time)
---
 lib/oga/xml/lexer.rb                          | 25 +++++++++++-----
 .../lexer/html_closing_rules/optgroup_spec.rb | 22 ++++++++++++++
 .../lexer/html_closing_rules/table_spec.rb    | 30 +++++++++++++++++++
 3 files changed, 70 insertions(+), 7 deletions(-)
 create mode 100644 spec/oga/xml/lexer/html_closing_rules/optgroup_spec.rb
 create mode 100644 spec/oga/xml/lexer/html_closing_rules/table_spec.rb

diff --git a/lib/oga/xml/lexer.rb b/lib/oga/xml/lexer.rb
index 6fbb583..3fbb00b 100644
--- a/lib/oga/xml/lexer.rb
+++ b/lib/oga/xml/lexer.rb
@@ -66,14 +66,14 @@ module Oga
         'rp'       => NodeNameSet.new(%w{rb rt rtc rp}),
         'optgroup' => NodeNameSet.new(%w{optgroup}),
         'option'   => NodeNameSet.new(%w{option optgroup}),
-        'colgrop'  => NodeNameSet.new(%w{thead tbody tfoot}),
+        'colgroup' => NodeNameSet.new(%w{thead tbody tfoot}),
         'caption'  => NodeNameSet.new(%w{thead tbody tfoot}),
-        'thead'    => NodeNameSet.new(%w{tbody tfoot}),
-        'tbody'    => NodeNameSet.new(%w{tbody tfoot}),
-        'tfoot'    => NodeNameSet.new(%w{tbody}),
-        'tr'       => NodeNameSet.new(%w{tr}),
-        'td'       => NodeNameSet.new(%w{td th}),
-        'th'       => NodeNameSet.new(%w{td th}),
+        'thead'    => NodeNameSet.new(%w{thead tbody tfoot}),
+        'tbody'    => NodeNameSet.new(%w{thead tbody tfoot}),
+        'tfoot'    => NodeNameSet.new(%w{thead tbody tfoot}),
+        'tr'       => NodeNameSet.new(%w{tr tbody thead tfoot}),
+        'td'       => NodeNameSet.new(%w{td th tbody thead tfoot tr}),
+        'th'       => NodeNameSet.new(%w{td th tbody thead tfoot tr}),
         'p'        => NodeNameSet.new(%w{
           address article aside blockquote div dl fieldset footer form h1 h2 h3
           h4 h5 h6 header hgroup hr main nav ol p pre section table ul
@@ -427,6 +427,17 @@ module Oga
         if close_current and close_current.include?(name)
           on_element_end
         end
+
+        # Close remaining parent elements. This for example ensures that a
+        # "<tbody>" not only closes an unclosed "<th>" but also the surrounding,
+        # unclosed "<tr>".
+        while close_current = HTML_CLOSE_SELF[current_element]
+          if close_current.include?(name)
+            on_element_end
+          else
+            break
+          end
+        end
       end
 
       ##
diff --git a/spec/oga/xml/lexer/html_closing_rules/optgroup_spec.rb b/spec/oga/xml/lexer/html_closing_rules/optgroup_spec.rb
new file mode 100644
index 0000000..913807a
--- /dev/null
+++ b/spec/oga/xml/lexer/html_closing_rules/optgroup_spec.rb
@@ -0,0 +1,22 @@
+require 'spec_helper'
+
+describe Oga::XML::Lexer do
+  describe 'HTML optgroup elements' do
+    describe 'with unclosed <optgroup> tags' do
+      it 'lexes an <option> tag followed by a <optgroup> tag' do
+        lex_html('<optgroup><option>foo<optgroup><option>bar').should == [
+          [:T_ELEM_NAME, 'optgroup', 1],
+          [:T_ELEM_NAME, 'option', 1],
+          [:T_TEXT, 'foo', 1],
+          [:T_ELEM_END, nil, 1],
+          [:T_ELEM_END, nil, 1],
+          [:T_ELEM_NAME, 'optgroup', 1],
+          [:T_ELEM_NAME, 'option', 1],
+          [:T_TEXT, 'bar', 1],
+          [:T_ELEM_END, nil, 1],
+          [:T_ELEM_END, nil, 1]
+        ]
+      end
+    end
+  end
+end
diff --git a/spec/oga/xml/lexer/html_closing_rules/table_spec.rb b/spec/oga/xml/lexer/html_closing_rules/table_spec.rb
new file mode 100644
index 0000000..5a7f437
--- /dev/null
+++ b/spec/oga/xml/lexer/html_closing_rules/table_spec.rb
@@ -0,0 +1,30 @@
+require 'spec_helper'
+
+describe Oga::XML::Lexer do
+  describe 'HTML tables' do
+    describe 'with unclosed <tr> tags' do
+      it 'lexes a <tr> tag followed by a <tbody> tag' do
+        lex_html('<tr>foo<tbody></tbody>').should == [
+          [:T_ELEM_NAME, 'tr', 1],
+          [:T_TEXT, 'foo', 1],
+          [:T_ELEM_END, nil, 1],
+          [:T_ELEM_NAME, 'tbody', 1],
+          [:T_ELEM_END, nil, 1]
+        ]
+      end
+
+      it 'lexes an unclosed <th> tag followed by a <tbody> tag' do
+        lex_html('<tr><th>foo<tbody>bar</tbody>').should == [
+          [:T_ELEM_NAME, 'tr', 1],
+          [:T_ELEM_NAME, 'th', 1],
+          [:T_TEXT, 'foo', 1],
+          [:T_ELEM_END, nil, 1],
+          [:T_ELEM_END, nil, 1],
+          [:T_ELEM_NAME, 'tbody', 1],
+          [:T_TEXT, 'bar', 1],
+          [:T_ELEM_END, nil, 1]
+        ]
+      end
+    end
+  end
+end