From ffe8fe5fd23a721af0fe1d07df109d78a39c743c Mon Sep 17 00:00:00 2001
From: Kovid Goyal <kovid@kovidgoyal.net>
Date: Fri, 17 Sep 2010 10:42:11 -0600
Subject: [PATCH] Fix use of UTF-8 raw string

---
 src/calibre/ebooks/conversion/preprocess.py | 1 -
 src/calibre/ebooks/conversion/utils.py      | 6 +++---
 2 files changed, 3 insertions(+), 4 deletions(-)

diff --git a/src/calibre/ebooks/conversion/preprocess.py b/src/calibre/ebooks/conversion/preprocess.py
index 3e5de26766..03a0047927 100644
--- a/src/calibre/ebooks/conversion/preprocess.py
+++ b/src/calibre/ebooks/conversion/preprocess.py
@@ -144,7 +144,6 @@ class HTMLPreProcessor(object):
     # Fix pdftohtml markup
     PDFTOHTML  = [
                   # Fix umlauts
-                  # ¨
                   (re.compile(u'¨\s*(<br.*?>)*\s*a', re.UNICODE), lambda match: u'ä'),
                   (re.compile(u'¨\s*(<br.*?>)*\s*A', re.UNICODE), lambda match: u'Ä'),
                   (re.compile(u'¨\s*(<br.*?>)*\s*e', re.UNICODE), lambda match: u'ë'),
diff --git a/src/calibre/ebooks/conversion/utils.py b/src/calibre/ebooks/conversion/utils.py
index 8588ff65ad..37fd169cb1 100644
--- a/src/calibre/ebooks/conversion/utils.py
+++ b/src/calibre/ebooks/conversion/utils.py
@@ -176,8 +176,8 @@ class PreProcessor(object):
         self.log("*** Median line length is " + str(length) + ", calculated with " + format + " format ***")
         #
         # Unwrap and/or delete soft-hyphens, hyphens
-        html = re.sub(u'­\s*(</span>\s*(</[iubp]>\s*<[iubp][^>]*>\s*)?<span[^>]*>|</[iubp]>\s*<[iubp][^>]*>)?\s*', '', html)
-        html = re.sub(u'(?<=[-–—])\s*(?=<)(</span>\s*(</[iubp]>\s*<[iubp][^>]*>\s*)?<span[^>]*>|</[iubp]>\s*<[iubp][^>]*>)?\s*(?=[[a-z\d])', '', html)
+        html = re.sub(u'\xad\s*(</span>\s*(</[iubp]>\s*<[iubp][^>]*>\s*)?<span[^>]*>|</[iubp]>\s*<[iubp][^>]*>)?\s*', '', html)
+        html = re.sub(u'(?<=[-\u2013\u2014])\s*(?=<)(</span>\s*(</[iubp]>\s*<[iubp][^>]*>\s*)?<span[^>]*>|</[iubp]>\s*<[iubp][^>]*>)?\s*(?=[[a-z\d])', '', html)
 
         # Unwrap lines using punctation and line length
         unwrap = re.compile(r"(?<=.{%i}([a-z,;):\IA]|(?<!\&\w{4});))\s*</(span|p|div)>\s*(</(p|span|div)>)?\s*(?P<up2threeblanks><(p|span|div)[^>]*>\s*(<(p|span|div)[^>]*>\s*</(span|p|div)>\s*)</(span|p|div)>\s*){0,3}\s*<(span|div|p)[^>]*>\s*(<(span|div|p)[^>]*>)?\s*" % length, re.UNICODE)
@@ -195,6 +195,6 @@ class PreProcessor(object):
         html = doubleheading.sub('\g<firsthead>'+'\n<h3'+'\g<secondhead>'+'</h3>', html)
 
         # put back non-breaking spaces in empty paragraphs to preserve original formatting
-        html = blankreg.sub('\n'+'\g<openline>'+' '+'\g<closeline>', html)
+        html = blankreg.sub('\n'+r'\g<openline>'+u'\u00a0'+r'\g<closeline>', html)
 
         return html