Merge branch 'py3' of https://github.com/eli-schwartz/calibre

src/calibre/ebooks/metadata/rb.py

@@ -5,10 +5,11 @@ __copyright__ = '2008, Ashish Kulkarni <kulkarni.ashish@gmail.com>'
 
 import sys, struct
 
+from calibre import prints
 from calibre.ebooks.metadata import MetaInformation, string_to_authors
 from polyglot.builtins import unicode_type
 
-MAGIC = '\xb0\x0c\xb0\x0c\x02\x00NUVO\x00\x00\x00\x00'
+MAGIC = b'\xb0\x0c\xb0\x0c\x02\x00NUVO\x00\x00\x00\x00'
 
 
 def get_metadata(stream):
@@ -37,7 +38,7 @@ def get_metadata(stream):
             return mi
 
         stream.seek(offset)
-        info = stream.read(length).splitlines()
+        info = stream.read(length).decode('utf-8', 'replace').splitlines()
         for line in info:
             if '=' not in line:
                 continue
@@ -45,10 +46,9 @@ def get_metadata(stream):
             if key.strip() == 'TITLE':
                 mi.title = value.strip()
             elif key.strip() == 'AUTHOR':
-                mi.author = value
                 mi.authors = string_to_authors(value)
     except Exception as err:
         msg = u'Couldn\'t read metadata from rb: %s with error %s'%(mi.title, unicode_type(err))
-        print(msg.encode('utf8'), file=sys.stderr)
+        prints(msg, file=sys.stderr)
         raise
     return mi

src/calibre/ebooks/rb/__init__.py

@@ -6,7 +6,7 @@ __docformat__ = 'restructuredtext en'
 
 import os
 
-HEADER = '\xb0\x0c\xb0\x0c\x02\x00NUVO\x00\x00\x00\x00'
+HEADER = b'\xb0\x0c\xb0\x0c\x02\x00NUVO\x00\x00\x00\x00'
 
 
 class RocketBookError(Exception):

src/calibre/ebooks/rb/reader.py

@@ -64,7 +64,7 @@ class Reader(object):
 
         toc = RBToc()
         for i in range(pages):
-            name = unquote(self.stream.read(32).strip('\x00'))
+            name = unquote(self.stream.read(32).strip(b'\x00'))
             size, offset, flags = self.read_i32(), self.read_i32(), self.read_i32()
             toc.append(RBToc.Item(name=name, size=size, offset=offset, flags=flags))
 
@@ -90,7 +90,7 @@ class Reader(object):
         else:
             output += self.stream.read(toc_item.size).decode('cp1252' if self.encoding is None else self.encoding, 'replace')
 
-        with open(os.path.join(output_dir, toc_item.name), 'wb') as html:
+        with open(os.path.join(output_dir, toc_item.name.decode('utf-8')), 'wb') as html:
             html.write(output.replace('<TITLE>', '<TITLE> ').encode('utf-8'))
 
     def get_image(self, toc_item, output_dir):
@@ -100,7 +100,7 @@ class Reader(object):
         self.stream.seek(toc_item.offset)
         data = self.stream.read(toc_item.size)
 
-        with open(os.path.join(output_dir, toc_item.name), 'wb') as img:
+        with open(os.path.join(output_dir, toc_item.name.decode('utf-8')), 'wb') as img:
             img.write(data)
 
     def extract_content(self, output_dir):
@@ -109,13 +109,14 @@ class Reader(object):
         images = []
 
         for item in self.toc:
-            if item.name.lower().endswith('html'):
+            iname = item.name.decode('utf-8')
-                self.log.debug('HTML item %s found...' % item.name)
+            if iname.lower().endswith('html'):
-                html.append(item.name)
+                self.log.debug('HTML item %s found...' % iname)
+                html.append(iname)
                 self.get_text(item, output_dir)
-            if item.name.lower().endswith('png'):
+            if iname.lower().endswith('png'):
-                self.log.debug('PNG item %s found...' % item.name)
+                self.log.debug('PNG item %s found...' % iname)
-                images.append(item.name)
+                images.append(iname)
                 self.get_image(item, output_dir)
 
         opf_path = self.create_opf(output_dir, html, images)

src/calibre/ebooks/rb/writer.py

@@ -73,13 +73,13 @@ class RBWriter(object):
         out_stream.write(struct.pack('<I', page_count))
         offset = out_stream.tell() + (len(toc_items) * 44)
         for item in toc_items:
-            out_stream.write(item.name)
+            out_stream.write(item.name.encode('utf-8'))
             out_stream.write(struct.pack('<I', item.size))
             out_stream.write(struct.pack('<I', offset))
             out_stream.write(struct.pack('<I', item.flags))
             offset += item.size
 
-        out_stream.write(info[0][1])
+        out_stream.write(info[0][1].encode('utf-8'))
 
         self.log.debug('Writing compressed RB HTHML...')
         # Compressed text with proper heading
@@ -92,7 +92,10 @@ class RBWriter(object):
 
         self.log.debug('Writing images...')
         for item in hidx+images:
-            out_stream.write(item[1])
+            w = item[1]
+            if not isinstance(w, bytes):
+                w = w.encode('utf-8')
+            out_stream.write(w)
 
         total_size = out_stream.tell()
         out_stream.seek(0x1c)
@@ -104,7 +107,7 @@ class RBWriter(object):
         size = len(text)
 
         pages = []
-        for i in range(0, (len(text) + TEXT_RECORD_SIZE-1) / TEXT_RECORD_SIZE):
+        for i in range(0, (len(text) + TEXT_RECORD_SIZE-1) // TEXT_RECORD_SIZE):
             zobj = zlib.compressobj(9, zlib.DEFLATED, 13, 8, 0)
             pages.append(zobj.compress(text[i * TEXT_RECORD_SIZE : (i * TEXT_RECORD_SIZE) + TEXT_RECORD_SIZE]) + zobj.flush())