From d253544a1f311aa692e78e5ff333af6d870fece3 Mon Sep 17 00:00:00 2001
From: Kovid Goyal <kovid@kovidgoyal.net>
Date: Sat, 25 Apr 2009 14:38:23 -0700
Subject: [PATCH] Implement a --page-breaks-before option

---
 src/calibre/ebooks/conversion/cli.py           | 2 +-
 src/calibre/ebooks/conversion/plumber.py       | 8 ++++++++
 src/calibre/ebooks/oeb/transforms/structure.py | 8 ++++++++
 3 files changed, 17 insertions(+), 1 deletion(-)

diff --git a/src/calibre/ebooks/conversion/cli.py b/src/calibre/ebooks/conversion/cli.py
index ae0af532ab..e12686a36c 100644
--- a/src/calibre/ebooks/conversion/cli.py
+++ b/src/calibre/ebooks/conversion/cli.py
@@ -128,7 +128,7 @@ def add_pipeline_options(parser, plumber):
                   [
                       'dont_split_on_page_breaks', 'chapter', 'chapter_mark',
                       'prefer_metadata_cover', 'remove_first_image',
-                      'insert_comments',
+                      'insert_comments', 'page_breaks_before',
                   ]
                   ),
 
diff --git a/src/calibre/ebooks/conversion/plumber.py b/src/calibre/ebooks/conversion/plumber.py
index f55d677d08..da9c9f11e2 100644
--- a/src/calibre/ebooks/conversion/plumber.py
+++ b/src/calibre/ebooks/conversion/plumber.py
@@ -227,6 +227,14 @@ OptionRecommendation(name='extra_css',
                 'rules.')
         ),
 
+OptionRecommendation(name='page_breaks_before',
+            recommended_value="//*[name()='h1' or name()='h2']",
+            level=OptionRecommendation.LOW,
+            help=_('An XPath expression. Page breaks are inserted '
+            'before the specified elements.')
+        ),
+
+
 OptionRecommendation(name='margin_top',
         recommended_value=5.0, level=OptionRecommendation.LOW,
         help=_('Set the top margin in pts. Default is %default')),
diff --git a/src/calibre/ebooks/oeb/transforms/structure.py b/src/calibre/ebooks/oeb/transforms/structure.py
index 605cdaa7cf..8ec3c7737a 100644
--- a/src/calibre/ebooks/oeb/transforms/structure.py
+++ b/src/calibre/ebooks/oeb/transforms/structure.py
@@ -45,6 +45,14 @@ class DetectStructure(object):
                 if not node.title or regexp.search(node.title) is not None:
                     self.oeb.toc.remove(node)
 
+        if opts.page_breaks_before is not None:
+            pb_xpath = XPath(opts.page_breaks_before)
+            for item in oeb.spine:
+                for elem in pb_xpath(item.data):
+                    style = elem.get('style', '')
+                    if style:
+                        style += '; '
+                    elem.set('style', style+'page-break-before:always')
 
     def detect_chapters(self):
         self.detected_chapters = []