eReader metadata: extract cover and add a plugin to automatically convert PML to PMLZ on import

src/calibre/customize/builtins.py

@@ -1,3 +1,4 @@
+import os.path
 __license__   = 'GPL v3'
 __copyright__ = '2008, Kovid Goyal <kovid at kovidgoyal.net>'
 
@@ -49,6 +50,43 @@ every time you add an HTML file to the library.\
         'include: cp1252, latin1, iso-8859-1 and utf-8.')
 
 
+class PML2PMLZ(FileTypePlugin):
+    name = 'PML to PMLZ'
+    author = 'John Schember'
+    description = textwrap.dedent(_('''\
+        Create a PMLZ archive containing the PML file \
+        and all images in the directory pmlname_img or images \
+        file containing all linked files. This plugin is run \
+        every time you add an PML file to the library.\
+    '''))
+    version = numeric_version
+    file_types = set(['pml'])
+    supported_platforms = ['windows', 'osx', 'linux']
+    on_import = True
+
+    def run(self, pmlfile):
+        import zipfile
+        from calibre.ptempfile import TemporaryDirectory
+
+        with TemporaryDirectory('_plugin_pml2pmlz') as tdir:
+            name = os.path.join(tdir, '_plugin_pml2pmlz.pmlz')
+            pmlz = zipfile.ZipFile(name, 'w')
+            pmlz.write(pmlfile)
+
+            pml_img = os.path.basename(pmlfile)[0] + '_img'
+            img_dir = pml_img if os.path.exists(pml_img) else 'images' if \
+            os.path.exists('images') else ''
+            if img_dir:
+                for image in glob.glob(os.path.join(img_dir, '*.png')):
+                    pmlz.write(image)
+            pmlz.close()
+
+        return name
+
+    def customization_help(self, gui=False):
+        return _('Character encoding for the input PML files. Should ways be: cp1252.')
+
+
 class ComicMetadataReader(MetadataReaderPlugin):
 
     name = 'Read comic metadata'
@@ -387,7 +425,7 @@ from calibre.devices.nuut2.driver import NUUT2
 from calibre.devices.iriver.driver import IRIVER_STORY
 
 from calibre.ebooks.metadata.fetch import GoogleBooks, ISBNDB, Amazon
-plugins = [HTML2ZIP, GoogleBooks, ISBNDB, Amazon]
+plugins = [HTML2ZIP, PML2PMLZ, GoogleBooks, ISBNDB, Amazon]
 plugins += [
     ComicInput,
     EPUBInput,

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))
@@ -157,12 +164,15 @@ class Reader132(FormatReader):
 
     def create_opf(self, output_dir, images):
         with CurrentDir(output_dir):
+            if 'cover.png' in images:
+                self.mi.cover = os.path.join('images', 'cover.png')
+
             opf = OPFCreator(output_dir, self.mi)
 
             manifest = [('index.html', None)]
 
             for i in images:
-                manifest.append((os.path.join('images/', i), None))
+                manifest.append((os.path.join('images', i), None))
 
             opf.create_manifest(manifest)
             opf.create_spine(['index.html'])

src/calibre/ebooks/pml/input.py

@@ -123,6 +123,8 @@ class PMLInput(InputFormatPlugin):
         from calibre.ebooks.metadata.meta import get_metadata
         log.debug('Reading metadata from input file...')
         mi = get_metadata(stream, 'pml')
+        if 'images/cover.png' in images:
+            mi.cover = 'images/cover.png'
         opf = OPFCreator(os.getcwd(), mi)
         log.debug('Generating manifest...')
         opf.create_manifest(manifest_items)