From 7e849659cd2e248e951afc4bcb8627e883a8716e Mon Sep 17 00:00:00 2001
From: Kovid Goyal <kovid@kovidgoyal.net>
Date: Tue, 3 Mar 2009 19:18:30 -0800
Subject: [PATCH] MOBI Input:Fix extraction of Table of Contents for MOBI files
 that have their TOC at the start of the book instead of the end

---
 src/calibre/ebooks/html.py                  | 4 +++-
 src/calibre/ebooks/lrf/html/convert_from.py | 2 +-
 src/calibre/ebooks/mobi/reader.py           | 8 +++++---
 3 files changed, 9 insertions(+), 5 deletions(-)

diff --git a/src/calibre/ebooks/html.py b/src/calibre/ebooks/html.py
index 40e882c2af..5329e8ed86 100644
--- a/src/calibre/ebooks/html.py
+++ b/src/calibre/ebooks/html.py
@@ -640,7 +640,7 @@ class Processor(Parser):
         name = self.htmlfile_map[self.htmlfile.path]
         href = 'content/'+name
         
-        # Add level 1 and level 2 TOC items
+        # Add level* TOC items
         counter = 0
         
         def elem_to_link(elem, href, counter):
@@ -711,6 +711,8 @@ class Processor(Parser):
                         
                     
             if len(toc) > 0:
+                # Detected TOC entries using --level* options
+                # so aborting all other toc processing
                 return
         # Add chapters to TOC
         if not self.opts.no_chapters_in_toc:
diff --git a/src/calibre/ebooks/lrf/html/convert_from.py b/src/calibre/ebooks/lrf/html/convert_from.py
index 48c2ffe993..2bd63d1d8f 100644
--- a/src/calibre/ebooks/lrf/html/convert_from.py
+++ b/src/calibre/ebooks/lrf/html/convert_from.py
@@ -479,7 +479,7 @@ class HTMLConverter(object, LoggingInterface):
             pprop.update(self.pseudo_css[tagname])
         if tag.has_key("class"):
             cls = tag["class"].lower()
-            for cls in cls.split():            
+            for cls in cls.split():
                 for classname in ["."+cls, tagname+"."+cls]:
                     if self.css.has_key(classname):
                         prop.update(self.css[classname])
diff --git a/src/calibre/ebooks/mobi/reader.py b/src/calibre/ebooks/mobi/reader.py
index df728e400e..81bd510694 100644
--- a/src/calibre/ebooks/mobi/reader.py
+++ b/src/calibre/ebooks/mobi/reader.py
@@ -361,7 +361,7 @@ class MobiReader(object):
                         continue
                     if reached and x.tag == 'a':
                         href = x.get('href', '')
-                        if href:
+                        if href and re.match('\w+://', href) is None:
                             try:
                                 text = u' '.join([t.strip() for t in \
                                                 x.xpath('descendant::text()')])
@@ -370,6 +370,8 @@ class MobiReader(object):
                             text = ent_pat.sub(entity_to_unicode, text)
                             tocobj.add_item(toc.partition('#')[0], href[1:], 
                                             text)
+                    if reached and x.get('class', None) == 'mbp_pagebreak':
+                        break
             if tocobj is not None:
                 opf.set_toc(tocobj)
         
@@ -435,7 +437,7 @@ class MobiReader(object):
     
     def replace_page_breaks(self):
         self.processed_html = self.PAGE_BREAK_PAT.sub(
-            '<div style="page-break-after: always; margin: 0; display: block" />',
+            '<div class="mbp_pagebreak" style="page-break-after: always; margin: 0; display: block" />',
             self.processed_html)
     
     def add_anchors(self):
@@ -521,7 +523,7 @@ def option_parser():
     parser = OptionParser(usage=_('%prog [options] myebook.mobi'))
     parser.add_option('-o', '--output-dir', default='.', 
                       help=_('Output directory. Defaults to current directory.'))
-    parser.add_option('--verbose', default=False, action='store_true',
+    parser.add_option('-v', '--verbose', default=False, action='store_true',
                       help='Useful for debugging.')
     return parser