Move from Beautiful Soup to lxml in txt and fb2 output.

src/calibre/ebooks/fb2/fb2ml.py

@@ -12,14 +12,14 @@ import os
 import re
 from base64 import b64encode
 
+from lxml import etree
+
 from calibre import entity_to_unicode
 from calibre.ebooks.oeb.base import XHTML, XHTML_NS, barename, namespace
 from calibre.ebooks.oeb.stylizer import Stylizer
 from calibre.ebooks.oeb.base import OEB_IMAGES
 from calibre.constants import __appname__, __version__
 
-from BeautifulSoup import BeautifulSoup
-
 TAG_MAP = {
     'b' : 'strong',
     'i' : 'emphasis',
@@ -57,11 +57,10 @@ class FB2MLizer(object):
         output += self.fb2mlize_images()
         output += self.fb2_footer()
         output = self.clean_text(output)
-        return BeautifulSoup(output.encode('utf-8')).prettify()
+        return etree.tostring(etree.fromstring(output), encoding=unicode, pretty_print=True)
 
     def fb2_header(self):
-        return u'<?xml version="1.0" encoding="utf-8"?> ' \
-        '<FictionBook xmlns:xlink="http://www.w3.org/1999/xlink" ' \
+        return u'<FictionBook xmlns:xlink="http://www.w3.org/1999/xlink" ' \
         'xmlns="http://www.gribuser.ru/xml/fictionbook/2.0"> ' \
         '<description><title-info><book-title>%s</book-title> ' \
         '</title-info><document-info> ' \

src/calibre/ebooks/fb2/output.py

@@ -30,7 +30,7 @@ class FB2Output(OutputFormatPlugin):
         
         out_stream.seek(0)
         out_stream.truncate()
-        out_stream.write(fb2_content)
+        out_stream.write(fb2_content.encode('utf-8'))
         
         if close:
             out_stream.close()

src/calibre/ebooks/txt/writer.py

@@ -11,9 +11,10 @@ Write content to TXT.
 import os
 import re
 
-from calibre import entity_to_unicode
+from lxml import etree
 
-from BeautifulSoup import BeautifulSoup
+from calibre import entity_to_unicode
+from calibre.ebooks.oeb.base import XHTML
 
 class TxtWriter(object):
     def __init__(self, newline, log):
@@ -23,10 +24,8 @@ class TxtWriter(object):
     def dump(self, spine):
         out = u''
         for item in spine:
-            content = unicode(item)
-            # Convert newlines to unix style \n for processing. These
-            # will be changed to the specified type later in the process.
-            content = self.unix_newlines(content)
+            content = unicode(etree.tostring(item.data.find(XHTML('body')), encoding=unicode))
+            content = self.remove_newlines(content)
             content = self.strip_html(content)
             content = self.replace_html_symbols(content)
             content = self.cleanup_text(content)
@@ -40,12 +39,9 @@ class TxtWriter(object):
 
         return out
 
-    def strip_html(self, html):
+    def strip_html(self, text):
         stripped = u''
 
-        for dom_tree in BeautifulSoup(html).findAll('body'):
-            text = unicode(dom_tree)
-            
         # Remove unnecessary tags
         for tag in ['script', 'style']:
             text = re.sub('(?imu)<[ ]*%s[ ]*.*?>(.*)</[ ]*%s[ ]*>' % (tag, tag), '', text)
@@ -117,7 +113,7 @@ class TxtWriter(object):
 
         return text
 
-    def unix_newlines(self, text):
+    def remove_newlines(self, text):
         text = text.replace('\r\n', ' ')
         text = text.replace('\n', ' ')
         text = text.replace('\r', ' ')