Fix usages of soup.prettify()

src/calibre/devices/kindle/driver.py

@@ -267,6 +267,7 @@ class KINDLE(USBMS):
 
     def add_annotation_to_library(self, db, db_id, annotation):
         from calibre.ebooks.metadata import MetaInformation
+        from calibre.ebooks.BeautifulSoup import prettify
 
         bm = annotation
         ignore_tags = {'Catalog', 'Clippings'}
@@ -289,9 +290,9 @@ class KINDLE(USBMS):
                     hrTag['class'] = 'annotations_divider'
                     user_notes_soup.insert(0, hrTag)
 
-                mi.comments += unicode_type(user_notes_soup.prettify())
+                mi.comments += prettify(user_notes_soup)
             else:
-                mi.comments = unicode_type(user_notes_soup.prettify())
+                mi.comments = prettify(user_notes_soup)
             # Update library comments
             db.set_comment(db_id, mi.comments)
 

src/calibre/devices/kobo/driver.py

@@ -1299,6 +1299,7 @@ class KOBO(USBMS):
         return ka_soup
 
     def add_annotation_to_library(self, db, db_id, annotation):
+        from calibre.ebooks.BeautifulSoup import prettify
         bm = annotation
         ignore_tags = {'Catalog', 'Clippings'}
 
@@ -1321,9 +1322,9 @@ class KOBO(USBMS):
                     hrTag['class'] = 'annotations_divider'
                     user_notes_soup.insert(0, hrTag)
 
-                mi.comments += unicode_type(user_notes_soup.prettify())
+                mi.comments += prettify(user_notes_soup)
             else:
-                mi.comments = unicode_type(user_notes_soup.prettify())
+                mi.comments = prettify(user_notes_soup)
             # Update library comments
             db.set_comment(db_id, mi.comments)
 

src/calibre/ebooks/BeautifulSoup.py

@@ -26,6 +26,13 @@ def parse_html(markup):
     return parse(markup, return_root=False)
 
 
+def prettify(soup):
+    ans = soup.prettify()
+    if isinstance(ans, bytes):
+        ans = ans.decode('utf-8')
+    return ans
+
+
 def BeautifulSoup(markup='', *a, **kw):
     return parse_html(markup)
 

src/calibre/library/catalogs/epub_mobi_builder.py

@@ -13,7 +13,7 @@ from calibre import (
 from calibre.constants import isosx, cache_dir
 from calibre.customize.conversion import DummyReporter
 from calibre.customize.ui import output_profiles
-from calibre.ebooks.BeautifulSoup import BeautifulSoup, BeautifulStoneSoup, NavigableString
+from calibre.ebooks.BeautifulSoup import BeautifulSoup, BeautifulStoneSoup, NavigableString, prettify
 from calibre.ebooks.chardet import substitute_entites
 from calibre.ebooks.metadata import author_to_author_sort
 from calibre.library.catalogs import AuthorSortMismatchException, EmptyCatalogException, \
@@ -1621,9 +1621,8 @@ class CatalogBuilder(object):
 
         # Write the generated file to content_dir
         outfile_spec = "%s/ByAlphaAuthor.html" % (self.content_dir)
-        outfile = open(outfile_spec, 'w')
+        with open(outfile_spec, 'wb') as outfile:
-        outfile.write(soup.prettify())
+            outfile.write(prettify(soup).encode('utf-8'))
-        outfile.close()
         self.html_filelist_1.append("content/ByAlphaAuthor.html")
 
     def generate_html_by_date_added(self):
@@ -1885,9 +1884,8 @@ class CatalogBuilder(object):
 
         # Write the generated file to content_dir
         outfile_spec = "%s/ByDateAdded.html" % (self.content_dir)
-        outfile = open(outfile_spec, 'w')
+        with open(outfile_spec, 'wb') as outfile:
-        outfile.write(soup.prettify())
+            outfile.write(prettify(soup).encode('utf-8'))
-        outfile.close()
         self.html_filelist_2.append("content/ByDateAdded.html")
 
     def generate_html_by_date_read(self):
@@ -2062,9 +2060,8 @@ class CatalogBuilder(object):
 
         # Write the generated file to content_dir
         outfile_spec = "%s/ByDateRead.html" % (self.content_dir)
-        outfile = open(outfile_spec, 'w')
+        with open(outfile_spec, 'wb') as outfile:
-        outfile.write(soup.prettify())
+            outfile.write(prettify(soup).encode('utf-8'))
-        outfile.close()
         self.html_filelist_2.append("content/ByDateRead.html")
 
     def generate_html_by_genres(self):
@@ -2309,9 +2306,8 @@ class CatalogBuilder(object):
             dtc += 1
 
         # Write the generated file to content_dir
-        outfile = open(outfile, 'w')
+        with open(outfile, 'wb') as outfile:
-        outfile.write(soup.prettify())
+            outfile.write(prettify(soup).encode('utf-8'))
-        outfile.close()
 
         if len(books) > 1:
             titles_spanned = [(books[0]['author'], books[0]['title']), (books[-1]['author'], books[-1]['title'])]
@@ -2469,9 +2465,8 @@ class CatalogBuilder(object):
 
         # Write the generated file to content_dir
         outfile_spec = "%s/BySeries.html" % (self.content_dir)
-        outfile = open(outfile_spec, 'w')
+        with open(outfile_spec, 'wb') as outfile:
-        outfile.write(soup.prettify())
+            outfile.write(prettify(soup).encode('utf-8'))
-        outfile.close()
         self.html_filelist_1.append("content/BySeries.html")
 
     def generate_html_by_title(self):
@@ -2624,9 +2619,8 @@ class CatalogBuilder(object):
 
         # Write the volume to content_dir
         outfile_spec = "%s/ByAlphaTitle.html" % (self.content_dir)
-        outfile = open(outfile_spec, 'w')
+        with open(outfile_spec, 'wb') as outfile:
-        outfile.write(soup.prettify())
+            outfile.write(prettify(soup).encode('utf-8'))
-        outfile.close()
         self.html_filelist_1.append("content/ByAlphaTitle.html")
 
     def generate_html_description_header(self, book):
@@ -2845,9 +2839,8 @@ class CatalogBuilder(object):
             soup = self.generate_html_description_header(title)
 
             # Write the book entry to content_dir
-            outfile = open("%s/book_%d.html" % (self.content_dir, int(title['id'])), 'w')
+            with open("%s/book_%d.html" % (self.content_dir, int(title['id'])), 'wb') as outfile:
-            outfile.write(soup.prettify())
+                outfile.write(prettify(soup).encode('utf-8'))
-            outfile.close()
 
     def generate_html_empty_header(self, title):
         """ Return a boilerplate HTML header.