pdb/eReader metadata writer

src/calibre/customize/builtins.py

@@ -297,6 +297,17 @@ class PDFMetadataWriter(MetadataWriterPlugin):
         from calibre.ebooks.metadata.pdf import set_metadata
         set_metadata(stream, mi)
 
+class PDBMetadataWriter(MetadataWriterPlugin):
+
+    name        = 'Set PDB metadata'
+    file_types  = set(['pdb'])
+    description = _('Set metadata from %s files') % 'PDB'
+    author      = 'John Schember'
+
+    def set_metadata(self, stream, mi, type):
+        from calibre.ebooks.metadata.pdb import set_metadata
+        set_metadata(stream, mi)
+
 
 from calibre.ebooks.epub.input import EPUBInput
 from calibre.ebooks.mobi.input import MOBIInput

src/calibre/ebooks/metadata/ereader.py

@@ -10,8 +10,8 @@ __docformat__ = 'restructuredtext en'
 
 import re
 
-from calibre.ebooks.metadata import MetaInformation
-from calibre.ebooks.pdb.header import PdbHeaderReader
+from calibre.ebooks.metadata import MetaInformation, authors_to_string
+from calibre.ebooks.pdb.header import PdbHeaderReader, PdbHeaderBuilder
 from calibre.ebooks.pdb.ereader.reader import HeaderRecord
 
 def get_metadata(stream, extract_cover=True):
@@ -24,8 +24,8 @@ def get_metadata(stream, extract_cover=True):
     pheader = PdbHeaderReader(stream)
     hr = HeaderRecord(pheader.section_data(0))
         
-    if hr.version in (2, 10):
-        try:    
+    if hr.version in (2, 10) and hr.has_metadata == 1:
+        try:
             mdata = pheader.section_data(hr.metadata_offset)
     
             mdata = mdata.split('\x00')
@@ -41,3 +41,42 @@ def get_metadata(stream, extract_cover=True):
 
     return mi
 
+def set_metadata(stream, mi):
+    pheader = PdbHeaderReader(stream)
+    sections = [pheader.section_data(x) for x in range(0, pheader.section_count())]
+    hr = HeaderRecord(sections[0])
+    
+    if hr.version not in (2, 10):
+        return
+    
+    # Create a metadata record for the file if one does not alreay exist
+    if not hr.has_metadata:
+        sections += ['', 'MeTaInFo\x00']
+        last_data = len(sections) - 1
+        
+        for i in range(0, 132, 2):
+            val, = struct.unpack('>H', sections[0][i:i+2])
+            if val >= hr.last_data_offset:
+                sections[0][i:i+2] = struct.pack('>H', last_data)
+            
+        sections[0][24:26] = struct.pack('>H', 1) # Set has metadata
+        sections[0][44:46] = struct.pack('>H', last_data - 1) # Set location of metadata
+        sections[0][52:54] = struct.pack('>H', last_data) # Ensure last data offset is updated
+    
+    # Merge the metadata into the file
+    file_mi = get_metadata(stream, False)
+    file_mi.smart_update(mi)
+    sections[hr.metadata_offset] = '%s\x00%s\x00%s\x00%s\x00%s\x00' % \
+        (file_mi.title, authors_to_string(file_mi.authors), '', file_mi.publisher, file_mi.isbn)
+
+    # Rebuild the PDB wrapper because the offsets have changed due to the
+    # new metadata.
+    pheader_builder = PdbHeaderBuilder(pheader.ident, pheader.title)
+    stream.seek(0)
+    stream.truncate(0)
+    pheader_builder.build_header([len(x) for x in sections], stream)
+
+    # Write the data back to the file
+    for item in sections:
+        stream.write(item)
+

src/calibre/ebooks/metadata/pdb.py

@@ -12,11 +12,18 @@ import re
 
 from calibre.ebooks.metadata import MetaInformation
 from calibre.ebooks.pdb.header import PdbHeaderReader
-from calibre.ebooks.metadata.ereader import get_metadata as eReader
+from calibre.ebooks.metadata.ereader import get_metadata as get_eReader
 
-MREADER = {    
-    'PNPdPPrs' : eReader,
-    'PNRdPPrs' : eReader,
+MREADER = {
+    'PNPdPPrs' : get_eReader,
+    'PNRdPPrs' : get_eReader,
+}
+
+from calibre.ebooks.metadata.ereader import set_metadata as set_eReader
+
+MWRITER = {
+    'PNPdPPrs' : set_eReader,
+    'PNRdPPrs' : set_eReader,
 }
 
 def get_metadata(stream, extract_cover=True):
@@ -34,3 +41,16 @@ def get_metadata(stream, extract_cover=True):
     
     return MetadataReader(stream, extract_cover)
     
+def set_metadata(stream, mi):
+    stream.seek(0)
+    
+    pheader = PdbHeaderReader(stream)
+    
+    MetadataWriter = MWRITER.get(pheader.ident, None)
+    
+    if MetadataWriter:
+        MetadataWriter(stream, mi)
+
+    stream.seek(0)
+    stream.write(re.sub('[^-A-Za-z0-9]+', '_', mi.title).ljust(32, '\x00')[:32])
+

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

@@ -12,7 +12,6 @@ import os, re, struct, zlib
 
 from calibre import CurrentDir
 from calibre.ebooks import DRMError
-from calibre.ebooks.metadata import MetaInformation
 from calibre.ebooks.pdb.formatreader import FormatReader
 from calibre.ebooks.pdb.ereader import EreaderError
 from calibre.ebooks.pml.pmlconverter import pml_to_html, \
@@ -31,6 +30,7 @@ class HeaderRecord(object):
     def __init__(self, raw):
         self.version, = struct.unpack('>H', raw[0:2])
         self.non_text_offset, = struct.unpack('>H', raw[12:14])
+        self.has_metadata, = struct.unpack('>H', raw[24:26])
         self.footnote_rec, = struct.unpack('>H', raw[28:30])
         self.sidebar_rec, =  struct.unpack('>H', raw[30:32])
         self.bookmark_offset, = struct.unpack('>H', raw[32:34])
@@ -62,6 +62,9 @@ class Reader(FormatReader):
             else:
                 raise EreaderError('Unknown book version %i.' % self.header_record.version)
 
+        from calibre.ebooks.metadata.pdb import get_metadata
+        self.mi = get_metadata(stream, False)
+
     def section_data(self, number):
         return self.sections[number]
 
@@ -144,10 +147,8 @@ class Reader(FormatReader):
         return opf_path
 
     def create_opf(self, output_dir, images):
-        mi = MetaInformation(None, None)
-
         with CurrentDir(output_dir):
-            opf = OPFCreator(output_dir, mi)
+            opf = OPFCreator(output_dir, self.mi)
 
             manifest = [('index.html', None)]
 

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

@@ -18,11 +18,7 @@ from calibre.ebooks.pdb.header import PdbHeaderBuilder
 from calibre.ebooks.pdb.ereader import image_name
 from calibre.ebooks.pml.pmlconverter import html_to_pml
 
-# We are using the older identity because we do not user newer features
-# (sidebar, footnotes). This will ensure compatibility with older readers.
-# If newer features are used (anything supported by dropbook but not by makebook
-# change the identity to the newer PNRdPPrs.
-IDENTITY = 'PNPdPPrs'
+IDENTITY = 'PNRdPPrs'
 
 # This is an arbitrary number that is small enough to work. The actual maximum
 # record size is unknown.