Docserver: Include <h4> tags in the table of contents, it's needed for the API

chrome/common/extensions/docs/server2/document_parser_test.py

 #!/usr/bin/env python
 # Copyright 2013 The Chromium Authors. All rights reserved.
 # Use of this source code is governed by a BSD-style license that can be
 # found in the LICENSE file.
 
 import unittest
 
 from document_parser import ParseDocument, RemoveTitle
 
 
@@ skipping 18 matching lines @@
 <h2>Grapefruit</h3>
 Grapefruit closed a h2 with a h3. This should be a warning.
 
 <h1 id='not-main'>Not the main header</h1>
 But it should still show up in the TOC as though it were an h2.
 
 <h2>Not <h3>a banana</h2>
 The embedded h3 should be ignored.
 
 <h4>It's a h4</h4>
-h4 are not considered part of the document structure.
+h4 are part of the document structure, but this is not inside a h3.
 
 <h3>Plantains</h3>
 Now I'm just getting lazy.
+
+<h4>Another h4</h4>
+This h4 is inside a h3 so will show up.
+
+<h5>Header 5</h5>
+Header 5s are not parsed.
 '''
 
 
 _WHOLE_DOCUMENT_WITHOUT_TITLE = '''
 Preamble before heading.
 
 
 Some intro to the content.
 
 <h2 id='banana' class='header'>Bananas</h2>
@@ skipping 11 matching lines @@
 <h2>Grapefruit</h3>
 Grapefruit closed a h2 with a h3. This should be a warning.
 
 <h1 id='not-main'>Not the main header</h1>
 But it should still show up in the TOC as though it were an h2.
 
 <h2>Not <h3>a banana</h2>
 The embedded h3 should be ignored.
 
 <h4>It's a h4</h4>
-h4 are not considered part of the document structure.
+h4 are part of the document structure, but this is not inside a h3.
 
 <h3>Plantains</h3>
 Now I'm just getting lazy.
+
+<h4>Another h4</h4>
+This h4 is inside a h3 so will show up.
+
+<h5>Header 5</h5>
+Header 5s are not parsed.
 '''
 
 
 class DocumentParserUnittest(unittest.TestCase):
 
   def testEmptyDocument(self):
     self.assertEqual(('', 'No opening <h1> was found'), RemoveTitle(''))
 
     result = ParseDocument('')
     self.assertEqual(None, result.title)
@@ skipping 49 matching lines @@
     self.assertEqual((_WHOLE_DOCUMENT_WITHOUT_TITLE, None),
                      RemoveTitle(_WHOLE_DOCUMENT))
     result = ParseDocument(_WHOLE_DOCUMENT, expect_title=True)
     self.assertEqual('Main header', result.title)
     self.assertEqual({'id': 'main', 'class': 'header'}, result.title_attributes)
     self.assertEqual([
       'Found closing </h3> while processing a <h2> (line 19, column 15)',
       'Found multiple <h1> tags. Subsequent <h1> tags will be classified as '
           '<h2> for the purpose of the structure (line 22, column 1)',
       'Found <h3> in the middle of processing a <h2> (line 25, column 9)',
+      # TODO(kalman): Re-enable this warning once the reference pages have
+      # their references fixed.
+      #'Found <h4> without any preceding <h3> (line 28, column 1)',
     ], result.warnings)
 
     # The non-trivial table of contents assertions...
     self.assertEqual(1, len(result.sections))
     entries = result.sections[0].structure
 
     self.assertEqual(5, len(entries), entries)
     entry0, entry1, entry2, entry3, entry4 = entries
 
     self.assertEqual('Bananas', entry0.name)
@@ skipping 15 matching lines @@
     self.assertEqual('Grapefruit', entry2.name)
     self.assertEqual({}, entry2.attributes)
     self.assertEqual([], entry2.entries)
 
     self.assertEqual('Not the main header', entry3.name)
     self.assertEqual({'id': 'not-main'}, entry3.attributes)
     self.assertEqual([], entry3.entries)
 
     self.assertEqual('Not a banana', entry4.name)
     self.assertEqual({}, entry4.attributes)
-    self.assertEqual(1, len(entry4.entries))
-    entry4_1, = entry4.entries
+    self.assertEqual(2, len(entry4.entries))
+    entry4_1, entry4_2 = entry4.entries
 
-    self.assertEqual('Plantains', entry4_1.name)
+    self.assertEqual('It\'s a h4', entry4_1.name)
     self.assertEqual({}, entry4_1.attributes)
     self.assertEqual([], entry4_1.entries)
 
+    self.assertEqual('Plantains', entry4_2.name)
+    self.assertEqual({}, entry4_2.attributes)
+    self.assertEqual(1, len(entry4_2.entries))
+    entry4_2_1, = entry4_2.entries
+
+    self.assertEqual('Another h4', entry4_2_1.name)
+    self.assertEqual({}, entry4_2_1.attributes)
+    self.assertEqual([], entry4_2_1.entries)
+
   def testSingleExplicitSection(self):
     def test(document):
       result = ParseDocument(document, expect_title=True)
       self.assertEqual([], result.warnings)
       self.assertEqual('Header', result.title)
       self.assertEqual(1, len(result.sections))
       section0, = result.sections
       entry0, = section0.structure
       self.assertEqual('An inner header', entry0.name)
     # A single section, one with the title inside the section, the other out.
@@ skipping 29 matching lines @@
     def assert_single_header(section, name):
       self.assertEqual(1, len(section.structure))
       self.assertEqual(name, section.structure[0].name)
     assert_single_header(section0, 'First header')
     assert_single_header(section1, 'Second header')
     assert_single_header(section2, 'Third header')
 
 
 if __name__ == '__main__':
   unittest.main()