Read support for the Haodoo PDB ebook format by Kan-Ru Chen. Fixes #976478 (Add Haodoo.net PDB format support)

src/calibre/customize/builtins.py

@@ -289,7 +289,7 @@ class OPFMetadataReader(MetadataReaderPlugin):
 class PDBMetadataReader(MetadataReaderPlugin):
 
     name        = 'Read PDB metadata'
-    file_types  = set(['pdb'])
+    file_types  = set(['pdb', 'updb'])
     description = _('Read metadata from %s files') % 'PDB'
     author      = 'John Schember'
 

src/calibre/ebooks/__init__.py

@@ -27,7 +27,7 @@ class ParserError(ValueError):
     pass
 
 BOOK_EXTENSIONS = ['lrf', 'rar', 'zip', 'rtf', 'lit', 'txt', 'txtz', 'text', 'htm', 'xhtm',
-                   'html', 'htmlz', 'xhtml', 'pdf', 'pdb', 'pdr', 'prc', 'mobi', 'azw', 'doc',
+                   'html', 'htmlz', 'xhtml', 'pdf', 'pdb', 'updb', 'pdr', 'prc', 'mobi', 'azw', 'doc',
                    'epub', 'fb2', 'djv', 'djvu', 'lrx', 'cbr', 'cbz', 'cbc', 'oebzip',
                    'rb', 'imp', 'odt', 'chm', 'tpz', 'azw1', 'pml', 'pmlz', 'mbp', 'tan', 'snb',
                    'xps', 'oxps', 'azw4', 'book', 'zbf', 'pobi', 'docx', 'md',

src/calibre/ebooks/conversion/plugins/pdb_input.py

@@ -13,7 +13,7 @@ class PDBInput(InputFormatPlugin):
     name        = 'PDB Input'
     author      = 'John Schember'
     description = 'Convert PDB to HTML'
-    file_types  = set(['pdb'])
+    file_types  = set(['pdb', 'updb'])
 
     def convert(self, stream, options, file_ext, log,
                 accelerators):

src/calibre/ebooks/metadata/haodoo.py

@@ -0,0 +1,23 @@
+# -*- coding: utf-8 -*-
+
+'''
+Read meta information from Haodoo.net pdb files.
+'''
+
+__license__   = 'GPL v3'
+__copyright__ = '2012, Kan-Ru Chen <kanru@kanru.info>'
+__docformat__ = 'restructuredtext en'
+
+from calibre.ebooks.pdb.header import PdbHeaderReader
+from calibre.ebooks.pdb.haodoo.reader import Reader
+
+def get_metadata(stream, extract_cover=True):
+    '''
+    Return metadata as a L{MetaInfo} object
+    '''
+    stream.seek(0)
+
+    pheader = PdbHeaderReader(stream)
+    reader = Reader(pheader, stream, None, None)
+
+    return reader.get_metadata()

src/calibre/ebooks/metadata/pdb.py

@@ -14,11 +14,14 @@ from calibre.ebooks.metadata import MetaInformation
 from calibre.ebooks.pdb.header import PdbHeaderReader
 from calibre.ebooks.metadata.ereader import get_metadata as get_eReader
 from calibre.ebooks.metadata.plucker import get_metadata as get_plucker
+from calibre.ebooks.metadata.haodoo import get_metadata as get_Haodoo
 
 MREADER = {
     'PNPdPPrs' : get_eReader,
     'PNRdPPrs' : get_eReader,
     'DataPlkr' : get_plucker,
+    'BOOKMTIT' : get_Haodoo,
+    'BOOKMTIU' : get_Haodoo,
 }
 
 from calibre.ebooks.metadata.ereader import set_metadata as set_eReader

src/calibre/ebooks/pdb/__init__.py

@@ -16,6 +16,7 @@ def _import_readers():
     from calibre.ebooks.pdb.ztxt.reader import Reader as ztxt_reader
     from calibre.ebooks.pdb.pdf.reader import Reader as pdf_reader
     from calibre.ebooks.pdb.plucker.reader import Reader as plucker_reader
+    from calibre.ebooks.pdb.haodoo.reader import Reader as haodoo_reader
 
     FORMAT_READERS = {
         'PNPdPPrs': ereader_reader,
@@ -24,6 +25,8 @@ def _import_readers():
         'TEXtREAd': palmdoc_reader,
         '.pdfADBE': pdf_reader,
         'DataPlkr': plucker_reader,
+        'BOOKMTIT': haodoo_reader,
+        'BOOKMTIU': haodoo_reader,
     }
 
 ALL_FORMAT_WRITERS = {'doc', 'ztxt', 'ereader'}
@@ -47,6 +50,8 @@ IDENTITY_TO_NAME = {
     'TEXtREAd': 'PalmDOC',
     '.pdfADBE': 'Adobe Reader',
     'DataPlkr': 'Plucker',
+    'BOOKMTIT': 'Haodoo.net',
+    'BOOKMTIU': 'Haodoo.net',
 
     'BVokBDIC': 'BDicty',
     'DB99DBOS': 'DB (Database program)',

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

@@ -0,0 +1,151 @@
+# -*- coding: utf-8 -*-
+
+'''
+Read content from Haodoo.net pdb file.
+'''
+
+__license__   = 'GPL v3'
+__copyright__ = '2012, Kan-Ru Chen <kanru@kanru.info>'
+__docformat__ = 'restructuredtext en'
+
+
+import struct
+import os
+
+from calibre import prepare_string_for_xml
+from calibre.ebooks.pdb.formatreader import FormatReader
+from calibre.ebooks.metadata import MetaInformation
+from calibre.ebooks.txt.processor import opf_writer, HTML_TEMPLATE
+
+BPDB_IDENT = b'BOOKMTIT'
+UPDB_IDENT = b'BOOKMTIU'
+
+punct_table = {
+    u"︵": u"（",
+    u"︶": u"）",
+    u"︷": u"｛",
+    u"︸": u"｝",
+    u"︹": u"〔",
+    u"︺": u"〕",
+    u"︻": u"【",
+    u"︼": u"】",
+    u"︗": u"〖",
+    u"︘": u"〗",
+    u"﹇": u"［］",
+    u"﹈": u"［］",
+    u"︽": u"《",
+    u"︾": u"》",
+    u"︿": u"〈",
+    u"﹀": u"〉",
+    u"﹁": u"「",
+    u"﹂": u"」",
+    u"﹃": u"『",
+    u"﹄": u"』",
+    u"｜": u"—",
+    u"︙": u"…",
+    u"ⸯ": u"～",
+    u"│": u"…",
+    u"￤": u"…",
+    u"　": u"  ",
+    }
+
+def fix_punct(line):
+    for (key, value) in punct_table.items():
+        line = line.replace(key, value)
+    return line
+
+class LegacyHeaderRecord(object):
+
+    def __init__(self, raw):
+        fields = raw.lstrip().replace(b'\x1b\x1b\x1b', b'\x1b').split(b'\x1b')
+        self.title = fix_punct(fields[0].decode('cp950', 'replace'))
+        self.num_records = int(fields[1])
+        self.chapter_titles = map(
+            lambda x: fix_punct(x.decode('cp950', 'replace').rstrip(b'\x00')),
+            fields[2:])
+
+class UnicodeHeaderRecord(object):
+
+    def __init__(self, raw):
+        fields = raw.lstrip().replace(b'\x1b\x00\x1b\x00\x1b\x00',
+                b'\x1b\x00').split(b'\x1b\x00')
+        self.title = fix_punct(fields[0].decode('utf_16_le', 'ignore'))
+        self.num_records = int(fields[1])
+        self.chapter_titles = map(
+            lambda x: fix_punct(x.decode('utf_16_le', 'replace').rstrip(b'\x00')),
+            fields[2].split(b'\r\x00\n\x00'))
+
+class Reader(FormatReader):
+
+    def __init__(self, header, stream, log, options):
+        self.stream = stream
+        self.log = log
+
+        self.sections = []
+        for i in range(header.num_sections):
+            self.sections.append(header.section_data(i))
+
+        if header.ident == BPDB_IDENT:
+            self.header_record = LegacyHeaderRecord(self.section_data(0))
+            self.encoding = 'cp950'
+        else:
+            self.header_record = UnicodeHeaderRecord(self.section_data(0))
+            self.encoding = 'utf_16_le'
+
+    def author(self):
+        self.stream.seek(35)
+        version = struct.unpack(b'>b', self.stream.read(1))[0]
+        if version == 2:
+            self.stream.seek(0)
+            author = self.stream.read(35).rstrip(b'\x00').decode(self.encoding, 'replace')
+            return author
+        else:
+            return u'Unknown'
+
+    def get_metadata(self):
+        mi = MetaInformation(self.header_record.title,
+                             [self.author()])
+        mi.language = u'zh-tw'
+
+        return mi
+
+    def section_data(self, number):
+        return self.sections[number]
+
+    def decompress_text(self, number):
+        return self.section_data(number).decode(self.encoding,
+                'replace').rstrip(b'\x00')
+
+    def extract_content(self, output_dir):
+        txt = u''
+
+        self.log.info(u'Decompressing text...')
+        for i in range(1, self.header_record.num_records + 1):
+            self.log.debug(u'\tDecompressing text section %i' % i)
+            title = self.header_record.chapter_titles[i-1]
+            lines = []
+            title_added = False
+            for line in self.decompress_text(i).splitlines():
+                line = fix_punct(line)
+                line = line.strip()
+                if not title_added and title in line:
+                    line = u'<h1 class="chapter">' + line + u'</h1>\n'
+                    title_added = True
+                else:
+                    line = prepare_string_for_xml(line)
+                lines.append(u'<p>%s</p>' % line)
+            if not title_added:
+                lines.insert(0, u'<h1 class="chapter">' + title + u'</h1>\n')
+            txt += u'\n'.join(lines)
+
+        self.log.info(u'Converting text to OEB...')
+        html = HTML_TEMPLATE % (self.header_record.title, txt)
+        with open(os.path.join(output_dir, u'index.html'), 'wb') as index:
+            index.write(html.encode('utf-8'))
+
+        mi = self.get_metadata()
+        manifest = [(u'index.html', None)]
+        spine = [u'index.html']
+        opf_writer(output_dir, u'metadata.opf', manifest, spine, mi)
+
+        return os.path.join(output_dir, u'metadata.opf')