manual: Remove footer for manual pages using Python XML module

Using Python XML module instead of egrep for removing the lines of the footer.

Signed-off-by: Joseph Nuthalapati <njoseph@thoughtworks.com>
Reviewed-by: James Valleroy <jvalleroy@mailbox.org>

doc/post-processor

@@ -31,7 +31,7 @@ def parse_arguments():
 
     subparser = subparsers.add_parser('fix-wiki-urls',
                                       help='Fix wrongly formatted wiki urls')
-    subparser.add_argument('filename', help='Name of the XML file')
+    subparser.add_argument('filename', help='Name of the Docbook file')
 
     subparsers.required = True
     return parser.parse_args()
@@ -52,27 +52,36 @@ def subcommand_fix_wiki_urls(arguments):
 
 
 def subcommand_remove_footer(arguments):
+    """Remove the footer template from the given wiki page."""
     filename = arguments.filename
     tree = etree.parse(filename)
     root = tree.getroot()
-    informaltables = []
 
-    def recurse(elem):
-        for child in elem:
-            if child.tag == 'informaltable':
-                informaltables.append((elem, child))
-            else:
-                recurse(child)
+    # The footer will always be in the last <section>
+    def find_last_section(elem):
+        if elem.getchildren():
+            last_element = elem[-1]
+            if last_element.tag == 'section':
+                return find_last_section(last_element)
+        return elem
 
-    recurse(root)
+    last_section = find_last_section(root)
 
-    if informaltables:
-        parent, child = informaltables[-1]
-        parent.remove(child)
+    if last_section.getchildren():
+        # Remove all elements till <informaltable> is reached
+        while last_section[-1].tag != 'informaltable':
+            last_section.remove(last_section[-1])
+        # remove <informaltable> itself
+        last_section.remove(last_section[-1])
+
+        # Remove the line "Back to Features introduction or manual pages."
+        if last_section[-1].text.startswith('Back to'):
+            last_section.remove(last_section[-1])
 
     processed_xml = etree.tostring(root, encoding='utf-8').decode()
 
     with open(filename, 'r') as xml_file:
+        # <xml> and <DOCTYPE> elements which etree skips
         header = xml_file.readlines()[:2]
 
     with open(filename, 'w') as xml_file: