From 763ceacafb819a0647a85409f1ea50255d41493d Mon Sep 17 00:00:00 2001
From: "Marshall T. Vandegrift" <llasram@gmail.com>
Date: Wed, 28 Jan 2009 12:04:22 -0500
Subject: [PATCH 1/3] Fix #1720.  Fix HTML comments which aren't valid XML
 comments.

---
 src/calibre/ebooks/oeb/base.py | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/src/calibre/ebooks/oeb/base.py b/src/calibre/ebooks/oeb/base.py
index 8b505ae4aa..0cfdec6355 100644
--- a/src/calibre/ebooks/oeb/base.py
+++ b/src/calibre/ebooks/oeb/base.py
@@ -353,9 +353,13 @@ class Manifest(object):
                 try:
                     data = etree.fromstring(data)
                 except etree.XMLSyntaxError:
+                    # TODO: Factor out HTML->XML coercion
                     self.oeb.logger.warn('Parsing file %r as HTML' % self.href)
                     data = html.fromstring(data)
                     data.attrib.pop('xmlns', None)
+                    for elem in data.iter(tag=etree.Comment):
+                        if elem.text:
+                            elem.text = elem.text.strip('-')
                     data = etree.tostring(data, encoding=unicode)
                     data = etree.fromstring(data)
             # Force into the XHTML namespace

From 0c57045cf85de33a653bcda584373b2e4d81bdc3 Mon Sep 17 00:00:00 2001
From: "Marshall T. Vandegrift" <llasram@gmail.com>
Date: Wed, 28 Jan 2009 18:27:09 -0500
Subject: [PATCH 2/3] Fix #1716.  For real this time?

---
 src/calibre/ebooks/html.py | 33 +++++++++++----------------------
 1 file changed, 11 insertions(+), 22 deletions(-)

diff --git a/src/calibre/ebooks/html.py b/src/calibre/ebooks/html.py
index 32601320d4..0199bedf19 100644
--- a/src/calibre/ebooks/html.py
+++ b/src/calibre/ebooks/html.py
@@ -561,28 +561,17 @@ class Processor(Parser):
         for elem in self.detected_chapters:
             text = u' '.join([t.strip() for t in elem.xpath('descendant::text()')])
             self.log_info('\tDetected chapter: %s', text[:50])
-            if self.opts.chapter_mark != 'none':
-                hr = etree.Element('hr')
-                if elem.getprevious() is None:
-                    elem.getparent()[:0] = [hr]
-                elif elem.getparent() is not None:
-                    insert = None
-                    for i, c in enumerate(elem.getparent()):
-                        if c is elem:
-                            insert = i
-                            break
-                    elem.getparent()[insert:insert] = [hr]
-                if self.opts.chapter_mark != 'rule':
-                    hr.set('style', 'width:0pt;page-break-before:always')
-                    if self.opts.chapter_mark == 'both':
-                        hr2 = etree.Element('hr')
-                        hr2.tail = u'\u00a0'
-                        p = hr.getparent()
-                        i = p.index(hr)
-                        p[i:i] = [hr2]
-                
-                    
-        
+            chapter_mark = self.opts.chapter_mark
+            if chapter_mark == 'none':
+                continue
+            elif chapter_mark == 'rule':
+                mark = etree.Element('hr')
+            elif chapter_mark == 'pagebreak':
+                mark = etree.Element('div', style='page-break-after: always')
+            else: # chapter_mark == 'both':
+                mark = etree.Element('hr', style='page-break-before: always')
+            elem.addprevious(mark)
+    
     def save(self):
         style_path = os.path.splitext(os.path.basename(self.save_path()))[0]
         for i, sheet in enumerate([self.stylesheet, self.font_css, self.override_css]):

From 2fac1ba61c0c199b479b6a84cbb43d85d2138720 Mon Sep 17 00:00:00 2001
From: "Marshall T. Vandegrift" <llasram@gmail.com>
Date: Wed, 28 Jan 2009 18:51:34 -0500
Subject: [PATCH 3/3] Fix #1716.  Ah, allow breaks to always always work under
 ADE.

---
 src/calibre/ebooks/html.py | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/src/calibre/ebooks/html.py b/src/calibre/ebooks/html.py
index 0199bedf19..2de1b093ba 100644
--- a/src/calibre/ebooks/html.py
+++ b/src/calibre/ebooks/html.py
@@ -558,18 +558,20 @@ class Processor(Parser):
     
     def detect_chapters(self):
         self.detected_chapters = self.opts.chapter(self.root)
+        chapter_mark = self.opts.chapter_mark
+        page_break_before = 'display: block; page-break-before: always'
+        page_break_after = 'display: block; page-break-after: always'
         for elem in self.detected_chapters:
             text = u' '.join([t.strip() for t in elem.xpath('descendant::text()')])
             self.log_info('\tDetected chapter: %s', text[:50])
-            chapter_mark = self.opts.chapter_mark
             if chapter_mark == 'none':
                 continue
             elif chapter_mark == 'rule':
                 mark = etree.Element('hr')
             elif chapter_mark == 'pagebreak':
-                mark = etree.Element('div', style='page-break-after: always')
+                mark = etree.Element('div', style=page_break_after)
             else: # chapter_mark == 'both':
-                mark = etree.Element('hr', style='page-break-before: always')
+                mark = etree.Element('hr', style=page_break_before)
             elem.addprevious(mark)
     
     def save(self):