eReader Metadata: Extract cover.

src/calibre/ebooks/metadata/ereader.py

@@ -16,6 +16,18 @@ from calibre.ebooks.pdb.ereader.reader132 import HeaderRecord
 from calibre.ebooks.pdb.header import PdbHeaderBuilder
 from calibre.ebooks.pdb.header import PdbHeaderReader
 
+def get_cover(pheader, eheader):
+    cover_data = None
+
+    for i in range(eheader.image_count):
+        raw = pheader.section_data(eheader.image_data_offset + i)
+
+        if raw[4:4 + 32].strip('\x00') == 'cover.png':
+            cover_data = raw[62:]
+            break
+
+    return ('png', cover_data)
+
 def get_metadata(stream, extract_cover=True):
     """
     Return metadata as a L{MetaInfo} object
@@ -29,7 +41,7 @@ def get_metadata(stream, extract_cover=True):
     if len(pheader.section_data(0)) == 132:
         hr = HeaderRecord(pheader.section_data(0))
 
-        if hr.version in (2, 10) and hr.has_metadata == 1:
+        if hr.compression in (2, 10) and hr.has_metadata == 1:
             try:
                 mdata = pheader.section_data(hr.metadata_offset)
 
@@ -41,6 +53,9 @@ def get_metadata(stream, extract_cover=True):
             except:
                 pass
 
+            if extract_cover:
+                mi.cover_data = get_cover(pheader, hr)
+
     if not mi.title:
         mi.title = pheader.title if pheader.title else _('Unknown')
 
@@ -56,7 +71,7 @@ def set_metadata(stream, mi):
     sections = [pheader.section_data(x) for x in range(0, pheader.section_count())]
     hr = HeaderRecord(sections[0])
 
-    if hr.version not in (2, 10):
+    if hr.compression not in (2, 10):
         return
 
     # Create a metadata record for the file if one does not alreay exist

src/calibre/ebooks/pdb/ereader/reader132.py

@@ -29,12 +29,19 @@ class HeaderRecord(object):
     '''
 
     def __init__(self, raw):
-        self.version, = struct.unpack('>H', raw[0:2])
+        self.compression, = struct.unpack('>H', raw[0:2])
         self.non_text_offset, = struct.unpack('>H', raw[12:14])
+        self.chapter_count, = struct.unpack('>H', raw[14:16])
+        self.image_count, = struct.unpack('>H', raw[20:22])
+        self.link_count, = struct.unpack('>H', raw[22:24])
         self.has_metadata, = struct.unpack('>H', raw[24:26])
-        self.footnote_rec, = struct.unpack('>H', raw[28:30])
+        self.footnote_count, = struct.unpack('>H', raw[28:30])
-        self.sidebar_rec, = struct.unpack('>H', raw[30:32])
+        self.sidebar_count, = struct.unpack('>H', raw[30:32])
+        self.chapter_offset, = struct.unpack('>H', raw[32:34])
+        self.small_font_page_offset, = struct.unpack('>H', raw[36:38])
+        self.large_font_page_offset, = struct.unpack('>H', raw[38:40])
         self.image_data_offset, = struct.unpack('>H', raw[40:42])
+        self.link_offset, = struct.unpack('>H', raw[42:44])
         self.metadata_offset, = struct.unpack('>H', raw[44:46])
         self.footnote_offset, = struct.unpack('>H', raw[48:50])
         self.sidebar_offset, = struct.unpack('>H', raw[50:52])
@@ -58,11 +65,11 @@ class Reader132(FormatReader):
 
         self.header_record = HeaderRecord(self.section_data(0))
 
-        if self.header_record.version not in (2, 10):
+        if self.header_record.compression not in (2, 10):
-            if self.header_record.version in (260, 272):
+            if self.header_record.compression in (260, 272):
                 raise DRMError('eReader DRM is not supported.')
             else:
-                raise EreaderError('Unknown book version %i.' % self.header_record.version)
+                raise EreaderError('Unknown book compression %i.' % self.header_record.compression)
 
         from calibre.ebooks.metadata.pdb import get_metadata
         self.mi = get_metadata(stream, False)
@@ -71,9 +78,9 @@ class Reader132(FormatReader):
         return self.sections[number]
 
     def decompress_text(self, number):
-        if self.header_record.version == 2:
+        if self.header_record.compression == 2:
             return decompress_doc(self.section_data(number)).decode('cp1252' if self.encoding is None else self.encoding, 'replace')
-        if self.header_record.version == 10:
+        if self.header_record.compression == 10:
             return zlib.decompress(self.section_data(number)).decode('cp1252' if self.encoding is None else self.encoding, 'replace')
 
     def get_image(self, number):
@@ -115,19 +122,19 @@ class Reader132(FormatReader):
             pml += self.get_text_page(i)
         html += pml_to_html(pml)
 
-        if self.header_record.footnote_rec > 0:
+        if self.header_record.footnote_count > 0:
             html += '<br /><h1>%s</h1>' % _('Footnotes')
             footnoteids = re.findall('\w+(?=\x00)', self.section_data(self.header_record.footnote_offset).decode('cp1252' if self.encoding is None else self.encoding))
-            for fid, i in enumerate(range(self.header_record.footnote_offset + 1, self.header_record.footnote_offset + self.header_record.footnote_rec)):
+            for fid, i in enumerate(range(self.header_record.footnote_offset + 1, self.header_record.footnote_offset + self.header_record.footnote_count)):
                 self.log.debug('Extracting footnote page %i' % i)
                 html += '<dl>'
                 html += footnote_sidebar_to_html(footnoteids[fid], self.decompress_text(i))
                 html += '</dl>'
 
-        if self.header_record.sidebar_rec > 0:
+        if self.header_record.sidebar_count > 0:
             html += '<br /><h1>%s</h1>' % _('Sidebar')
             sidebarids = re.findall('\w+(?=\x00)', self.section_data(self.header_record.sidebar_offset).decode('cp1252' if self.encoding is None else self.encoding))
-            for sid, i in enumerate(range(self.header_record.sidebar_offset + 1, self.header_record.sidebar_offset + self.header_record.sidebar_rec)):
+            for sid, i in enumerate(range(self.header_record.sidebar_offset + 1, self.header_record.sidebar_offset + self.header_record.sidebar_count)):
                 self.log.debug('Extracting sidebar page %i' % i)
                 html += '<dl>'
                 html += footnote_sidebar_to_html(sidebarids[sid], self.decompress_text(i))

src/calibre/ebooks/pml/input.py

@@ -27,20 +27,20 @@ class PMLInput(InputFormatPlugin):
     def process_pml(self, pml_path, html_path, close_all=False):
         pclose = False
         hclose = False
-    
+
         if not hasattr(pml_path, 'read'):
             pml_stream = open(pml_path, 'rb')
             pclose = True
         else:
             pml_stream = pml_path
             pml_stream.seek(0)
-            
+
         if not hasattr(html_path, 'write'):
             html_stream = open(html_path, 'wb')
             hclose = True
         else:
             html_stream = html_path
-        
+
         ienc = pml_stream.encoding if pml_stream.encoding else 'cp1252'
         if self.options.input_encoding:
             ienc = self.options.input_encoding
@@ -95,12 +95,12 @@ class PMLInput(InputFormatPlugin):
             with TemporaryDirectory('_unpmlz') as tdir:
                 zf = ZipFile(stream)
                 zf.extractall(tdir)
-            
+
                 pmls = glob.glob(os.path.join(tdir, '*.pml'))
                 for pml in pmls:
                     html_name = os.path.splitext(os.path.basename(pml))[0]+'.html'
                     html_path = os.path.join(os.getcwd(), html_name)
-                    
+
                     pages.append(html_name)
                     log.debug('Processing PML item %s...' % pml)
                     ttoc = self.process_pml(pml, html_path)
@@ -119,7 +119,7 @@ class PMLInput(InputFormatPlugin):
         manifest_items = []
         for item in pages+images:
             manifest_items.append((item, None))
-        
+
         from calibre.ebooks.metadata.meta import get_metadata
         log.debug('Reading metadata from input file...')
         mi = get_metadata(stream, 'pml')
@@ -133,5 +133,5 @@ class PMLInput(InputFormatPlugin):
         with open('metadata.opf', 'wb') as opffile:
             with open('toc.ncx', 'wb') as tocfile:
                 opf.render(opffile, tocfile, 'toc.ncx')
-        
+
         return os.path.join(os.getcwd(), 'metadata.opf')