From 754923ce07cbd268039b70bb9c8563f217b17730 Mon Sep 17 00:00:00 2001
From: John Schember <john@nachtimwald.com>
Date: Thu, 2 Apr 2009 20:44:48 -0400
Subject: [PATCH] pdf input to txt output giving correct output

---
 src/calibre/ebooks/pdf/input.py     | 2 +-
 src/calibre/ebooks/pdf/pdftohtml.py | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/src/calibre/ebooks/pdf/input.py b/src/calibre/ebooks/pdf/input.py
index 060b9f5367..6f55b71dd5 100644
--- a/src/calibre/ebooks/pdf/input.py
+++ b/src/calibre/ebooks/pdf/input.py
@@ -25,7 +25,7 @@ class PDFInput(InputFormatPlugin):
         html = pdftohtml(stream.name)
         
         with open('index.html', 'wb') as index:
-            index.write(html.encode('utf-8'))
+            index.write(html)
             
         #mi = metadata_from_formats([stream.name])
         mi = MetaInformation(_('Unknown'), _('Unknown'))
diff --git a/src/calibre/ebooks/pdf/pdftohtml.py b/src/calibre/ebooks/pdf/pdftohtml.py
index 168923ad1a..27cdb3f691 100644
--- a/src/calibre/ebooks/pdf/pdftohtml.py
+++ b/src/calibre/ebooks/pdf/pdftohtml.py
@@ -68,8 +68,8 @@ def pdftohtml(pdf_path):
                 raise DRMError()
         
             with open(index, 'rb') as i:
-                raw = i.read().decode('latin-1')
+                raw = i.read()
             if not '<br' in raw[:4000]:
                 raise ConversionError(os.path.basename(pdf_path) + _(' is an image based PDF. Only conversion of text based PDFs is supported.'), True)
 
-            return u'<!-- created by calibre\'s pdftohtml -->\n' + raw
+            return '<!-- created by calibre\'s pdftohtml -->\n' + raw