From 5091859c49e9d24cffdb19acd1a665a50e88dee8 Mon Sep 17 00:00:00 2001
From: GRiker <griker@hotmail.com>
Date: Fri, 26 Mar 2010 04:44:52 -0700
Subject: [PATCH 1/2] GwR improved diagnostics for metadata, cover failures

---
 src/calibre/ebooks/metadata/topaz.py | 8 ++++----
 src/calibre/ebooks/mobi/reader.py    | 2 +-
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/src/calibre/ebooks/metadata/topaz.py b/src/calibre/ebooks/metadata/topaz.py
index df25172531..233ce78f25 100644
--- a/src/calibre/ebooks/metadata/topaz.py
+++ b/src/calibre/ebooks/metadata/topaz.py
@@ -4,7 +4,7 @@ __copyright__ = '2010, Greg Riker <griker@hotmail.com>'
 __docformat__ = 'restructuredtext en'
 
 ''' Read/write metadata from Amazon's topaz format '''
-import StringIO, sys
+import os, StringIO, sys
 from struct import pack
 
 from calibre.ebooks.metadata import MetaInformation
@@ -83,7 +83,7 @@ class MetadataUpdater(object):
 
         sig = self.data[:4]
         if not sig.startswith('TPZ'):
-            raise ValueError('Not a Topaz file')
+            raise ValueError("'%s': unexpected Topaz signature '%s'" % (os.path.basename(stream.name),self.data[:4]))
         offset = 4
 
         self.header_records, consumed = self.decode_vwi(self.data[offset:offset+4])
@@ -92,13 +92,13 @@ class MetadataUpdater(object):
 
         # First integrity test - metadata header
         if not 'metadata' in self.topaz_headers:
-            raise ValueError('Invalid Topaz format - no metadata record')
+            raise ValueError("'%s': Topaz metadata record missing" % os.path.basename(stream.name))
 
         # Second integrity test - metadata body
         md_offset = self.topaz_headers['metadata']['blocks'][0]['offset']
         md_offset += self.base
         if self.data[md_offset+1:md_offset+9] != 'metadata':
-            raise ValueError('Damaged metadata record')
+            raise ValueError("'%s': damaged Topaz metadata record" % os.path.basename(stream.name))
 
     def book_length(self):
         ''' convenience method for retrieving book length '''
diff --git a/src/calibre/ebooks/mobi/reader.py b/src/calibre/ebooks/mobi/reader.py
index c4829ec22f..ac98660e09 100644
--- a/src/calibre/ebooks/mobi/reader.py
+++ b/src/calibre/ebooks/mobi/reader.py
@@ -833,7 +833,7 @@ def get_metadata(stream):
     try:
         im = PILImage.open(buf)
     except:
-        log.exception('Failed to read MOBI cover')
+        log.exception("Failed to read MOBI cover: '%s'" % os.path.basename(stream.name))
     else:
         obuf = cStringIO.StringIO()
         im.convert('RGB').save(obuf, format='JPEG')

From 9c38a1b5b84afce49c1950cf366e5398530f30de Mon Sep 17 00:00:00 2001
From: GRiker <griker@hotmail.com>
Date: Fri, 26 Mar 2010 13:07:09 -0700
Subject: [PATCH 2/2] GwR fixes for tickets 2279, 4652, 5162

---
 src/calibre/ebooks/metadata/mobi.py | 13 ++++++++++++-
 src/calibre/ebooks/oeb/reader.py    |  2 +-
 2 files changed, 13 insertions(+), 2 deletions(-)

diff --git a/src/calibre/ebooks/metadata/mobi.py b/src/calibre/ebooks/metadata/mobi.py
index fd1a1c3947..7581b560cd 100644
--- a/src/calibre/ebooks/metadata/mobi.py
+++ b/src/calibre/ebooks/metadata/mobi.py
@@ -331,7 +331,7 @@ class MetadataUpdater(object):
             authors = mi.author_sort
             update_exth_record((100, authors.encode(self.codec, 'replace')))
         elif mi.authors:
-            authors = '; '.join(mi.authors)
+            authors = ';'.join(mi.authors)
             update_exth_record((100, authors.encode(self.codec, 'replace')))
         if mi.publisher:
             update_exth_record((101, mi.publisher.encode(self.codec, 'replace')))
@@ -349,6 +349,17 @@ class MetadataUpdater(object):
         if mi.tags:
             subjects = '; '.join(mi.tags)
             update_exth_record((105, subjects.encode(self.codec, 'replace')))
+
+            # >>> Begin patch for ticket #4652 <<<
+            kindle_doc_types = set([u'[kindle_ebok]',u'[kindle_pdoc]'])
+            doc_type = list(kindle_doc_types.intersection(set(mi.tags)))[0]
+            if doc_type:
+                if doc_type == '[kindle_ebok]':
+                    update_exth_record((501,str('EBOK')))
+                elif doc_type == '[kindle_pdoc]':
+                    update_exth_record((501, str('PDOC')))
+            # >>> End patch
+
         if mi.pubdate:
             update_exth_record((106, str(mi.pubdate).encode(self.codec, 'replace')))
         elif mi.timestamp:
diff --git a/src/calibre/ebooks/oeb/reader.py b/src/calibre/ebooks/oeb/reader.py
index 9043db97f1..e829891a45 100644
--- a/src/calibre/ebooks/oeb/reader.py
+++ b/src/calibre/ebooks/oeb/reader.py
@@ -359,7 +359,7 @@ class OEBReader(object):
             self._toc_from_navpoint(item, node, child)
 
     def _toc_from_ncx(self, item):
-        if item is None:
+        if (item is None) or (item.data is None):
             return False
         self.log.debug('Reading TOC from NCX...')
         ncx = item.data