Don't add footnotes to table of contents.

[wolnelektury.git] / lib / librarian / html.py

diff --git a/lib/librarian/html.py b/lib/librarian/html.py
index ea71bbb..13cef65 100644 (file)
--- a/lib/librarian/html.py
+++ b/lib/librarian/html.py
@@ -53,9 +53,13 @@ def transform(input_filename, output_filename):
     doc = etree.parse(doc_file, parser)
 
     result = doc.xslt(style)
-    add_anchors(result.getroot())
-    add_table_of_contents(result.getroot())
-    result.write(output_filename, xml_declaration=True, pretty_print=True, encoding='utf-8')
+    if result.find('//h1') is not None:
+        add_anchors(result.getroot())
+        add_table_of_contents(result.getroot())
+        result.write(output_filename, xml_declaration=False, pretty_print=True, encoding='utf-8')
+        return True
+    else:
+        return False
 
 
 class Fragment(object):
@@ -109,7 +113,7 @@ def extract_fragments(input_filename):
 
     for event, element in etree.iterparse(input_filename, events=('start', 'end')):
         # Process begin and end elements
-        if element.tag == 'span' and element.get('class', '') in ('theme-begin', 'theme-end'):
+        if element.get('class', '') in ('theme-begin', 'theme-end'):
             if not event == 'end': continue # Process elements only once, on end event
 
             # Open new fragment
@@ -117,9 +121,9 @@ def extract_fragments(input_filename):
                 fragment = Fragment(id=element.get('fid'), themes=element.text)
 
                 # Append parents
-                if element.getparent().tag != 'body':
+                if element.getparent().get('id', None) != 'book-text':
                     parents = [element.getparent()]
-                    while parents[-1].getparent().tag != 'body':
+                    while parents[-1].getparent().get('id', None) != 'book-text':
                         parents.append(parents[-1].getparent())
 
                     parents.reverse()
@@ -171,12 +175,18 @@ def add_anchor(element, number):
     element.insert(0, anchor_target)
 
 
+def any_ancestor(element, test):
+    for ancestor in element.iterancestors():
+        if test(ancestor):
+            return True
+    return False
+
+
 def add_anchors(root):
     counter = 1
     for element in root.iterdescendants():
-        if element.getparent().tag in 'div' and 'note' in element.getparent().get('class', ''):
-            continue
-        if element.getparent().tag in 'blockquote':
+        if any_ancestor(element, lambda e: e.get('class') in ('note', 'motto', 'motto_podpis', 'dedication')
+        or e.tag == 'blockquote'):
             continue
         
         if element.tag == 'p' and 'verse' in element.get('class', ''):
@@ -193,6 +203,9 @@ def add_table_of_contents(root):
 
     for element in root.iterdescendants():
         if element.tag in ('h2', 'h3'):
+            if any_ancestor(element, lambda e: e.get('id') in ('footnotes')):
+                continue
+            
             if element.tag == 'h3' and len(sections) and sections[-1][0] == 'h2':
                 sections[-1][2].append((element.tag, ''.join(element.xpath('descendant-or-self::text()')), []))
             else: