palmdoc pdb input.

src/calibre/ebooks/pdb/__init__.py

@@ -7,13 +7,13 @@ __docformat__ = 'restructuredtext en'
 
 from calibre.ebooks.pdb.ereader.reader import Reader as eReader
 from calibre.ebooks.pdb.ztxt.reader import Reader as zTXT
-#from calibre.ebooks.pdb.palmdoc.reader import Reader as PalmDoc
+from calibre.ebooks.pdb.palmdoc.reader import Reader as PalmDoc
 
 FORMATS = {
     'PNPdPPrs' : eReader,
     'PNRdPPrs' : eReader,
     'zTXTGPlm' : zTXT,
-#    'TEXtREAd' : PalmDoc,
+    'TEXtREAd' : PalmDoc,
 }
 
 IDENTITY_TO_NAME = {

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

@@ -0,0 +1,70 @@
+# -*- coding: utf-8 -*-
+
+'''
+Read content from palmdoc pdb file.
+'''
+
+__license__   = 'GPL v3'
+__copyright__ = '2009, John Schember <john@nachtimwald.com>'
+__docformat__ = 'restructuredtext en'
+
+import os, struct, zlib
+
+from calibre.ebooks.pdb.formatreader import FormatReader
+from calibre.ebooks.mobi.palmdoc import decompress_doc
+from calibre.ebooks.txt.processor import txt_to_markdown, opf_writer
+
+class HeaderRecord(object):
+    '''
+    The first record in the file is always the header record. It holds
+    information related to the location of text, images, and so on
+    in the file. This is used in conjunction with the sections
+    defined in the file header.
+    '''
+
+    def __init__(self, raw):
+        self.compression, = struct.unpack('>H', raw[0:2])
+        self.num_records, = struct.unpack('>H', raw[8:10])
+        
+    
+class Reader(FormatReader):
+    
+    def __init__(self, header, stream, log, encoding=None):
+        self.stream = stream
+        self.log = log
+        self.encoding = encoding
+    
+        self.sections = []
+        for i in range(header.num_sections):
+            self.sections.append(header.section_data(i))
+
+        self.header_record = HeaderRecord(self.section_data(0))
+
+    def section_data(self, number):
+        return self.sections[number]
+
+    def decompress_text(self, number):
+        if self.header_record.compression == 1:
+            return self.section_data(number).decode('cp1252' if self.encoding is None else self.encoding)
+        if self.header_record.compression == 2:
+            return decompress_doc(self.section_data(number)).decode('cp1252' if self.encoding is None else self.encoding)
+        return ''
+
+    def extract_content(self, output_dir):
+        txt = ''
+        
+        for i in range(1, self.header_record.num_records + 1):
+            txt += self.decompress_text(i)
+
+        html = txt_to_markdown(txt)
+        with open(os.path.join(output_dir, 'index.html'), 'wb') as index:
+            index.write(html.encode('utf-8'))
+                        
+        from calibre.ebooks.metadata.meta import get_metadata
+        mi = get_metadata(self.stream, 'pdb')
+        manifest = [('index.html', None)]
+        spine = ['index.html']
+        opf_writer(output_dir, 'metadata.opf', manifest, spine, mi)
+        
+        return os.path.join(output_dir, 'metadata.opf')
+

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

@@ -8,11 +8,10 @@ __license__   = 'GPL v3'
 __copyright__ = '2009, John Schember <john@nachtimwald.com>'
 __docformat__ = 'restructuredtext en'
 
-import StringIO, os, struct, zlib
+import os, struct, zlib
 
 from calibre.ebooks.pdb.formatreader import FormatReader
 from calibre.ebooks.pdb.ztxt import zTXTError
-from calibre.ebooks.metadata import MetaInformation
 from calibre.ebooks.txt.processor import txt_to_markdown, opf_writer
 
 class HeaderRecord(object):