def _parse_ereader_header132(raw):
h = Storage()
h.compression, _unknown1, h.encoding, h.number_small_pages, h.number_large_pages, h.non_text_records, h.number_chapters, h.number_small_index, h.number_large_index, h.number_images, h.number_links, h.metadata_available, _unknown2, h.number_footnotes, h.number_sidebars, h.chapter_index_records, h.magic_2560, h.small_page_index_record, h.large_page_index_record, h.image_data_record, h.links_record, h.metadata_record, _unknown3, h.footnote_record, h.sidebar_record, h.last_data_record, = struct.unpack(
">HLHHHHHHHHHHHHHHHHHHHHHHH", raw[:54]
)
return h
def _parse_palmdoc_header(raw):
h = Storage()
h.compression, _unused, h.text_length, h.record_count, h.record_size, h.current_position, = struct.unpack(
">HHLHHL", raw[0:0x10]
)
return h
def parse_opf_xml (rawxml):
rawxml, encoding = xml_to_unicode(rawxml, strip_encoding_pats=True, resolve_entities=True, assume_utf8=True)
rawxml = rawxml[rawxml.find('<'):]
tree = etree.fromstring(rawxml, etree.XMLParser(recover=True))
opf = Storage()
for section in ('metadata', 'manifest', 'spine', 'guide'):
subtree = tree.find('opf:%s' % section, namespaces=NAMESPACES)
if subtree is not None:
for el in subtree.getchildren():
opf.setdefault(section, []).append((el.tag, el.attrib, el.text))
return opf
def _parse_plucker_header(raw):
h = Storage()
h.uid, h.compression, h.records, = struct.unpack(">HHH", raw[0:6])
h.home_html = None
reserved = {}
for i in xrange(h.records):
adv = 4 * i
name, = struct.unpack(">H", raw[6 + adv : 8 + adv])
id, = struct.unpack(">H", raw[8 + adv : 10 + adv])
reserved[id] = name
if name == 0:
h.home_html = id
h.reserved = reserved
return h
def _parse_pdb_header(stream):
pdbheader = Storage()
# PDB fields
pdbheader.name, pdbheader.attributes, pdbheader.version, pdbheader.creation_timestamp, pdbheader.modification_timestamp, pdbheader.last_backup_timestamp, pdbheader.modification_number, pdbheader.appinfo_offset, pdbheader.sortinfo_offset, pdbheader.type, pdbheader.creator, pdbheader.uniqueidseed, pdbheader.nextrecordlistid, pdbheader.num_records, = struct.unpack(
">32sHHLLLLLL4s4sLLH", stream.read(78)
)
# record offsets and lengths
records = []
start = struct.unpack(">LBBBB", stream.read(8))[0]
for n in range(1, pdbheader.num_records):
next_start = struct.unpack(">LBBBB", stream.read(8))[0]
records.append((start, next_start - start))
start = next_start
stream.seek(0, 2)
end = stream.tell()
records.append((start, end - start))
pdbheader.records = records
# Clean up some of the fields
pdbheader.name = re.sub("[^-A-Za-z0-9'\";:,. ]+", "_", pdbheader.name.replace("\x00", ""))
return pdbheader
def _parse_exth_header (raw):
exth = Storage()
exth.identifier, \
exth.header_length, \
exth.record_count, \
= struct.unpack('>4sLL', raw[:12])
exthdata = raw[12:]
pos = 0
records = []
records_left = exth.record_count
while records_left > 0:
records_left -= 1
record = Storage()
record.type, \
record.length, \
= struct.unpack('>LL', exthdata[pos:pos + 8])
record.data = exthdata[pos+8:pos+record.length]
pos += record.length
records.append(record)
exth.records = records
return exth
def _parse_ztxt_header(raw):
h = Storage()
h.version, h.record_count, h.data_size, h.record_size, h.number_bookmarks, h.bookmark_record, h.number_annotations, h.annotation_record, h.flags, _reserved, h.crc32, = struct.unpack(
">HHLHHHHHBBL", raw[0:24]
)
return h
def _parse_ereader_header202(raw):
# Unfortunately, this header format is mostly unknown
h = Storage()
h.version, _unknown, h.non_text_records, = struct.unpack(">H6sH", raw[:10])
return h
def _parse_mobi_header (raw):
mobiheader = Storage()
mobiheader.compression, \
_unused, \
mobiheader.text_length, \
mobiheader.record_count, \
mobiheader.record_size, \
mobiheader.encryption, \
_unknown, \
= struct.unpack('>HHLHHHH', raw[0:0x10])
# Some ancient MOBI files have no more metadata than this
if len(raw) <= 16:
return mobiheader
mobiheader.identifier, \
mobiheader.header_length, \
mobiheader.mobi_type, \
mobiheader.text_encoding, \
mobiheader.unique_id, \
mobiheader.file_version, \
mobiheader.ortographic_index_record, \
mobiheader.inflection_index_record, \
mobiheader.index_names_record, \
mobiheader.index_keys_record, \
mobiheader.extra_index0_record, \
mobiheader.extra_index1_record, \
mobiheader.extra_index2_record, \
mobiheader.extra_index3_record, \
mobiheader.extra_index4_record, \
mobiheader.extra_index5_record, \
mobiheader.first_nonbook_record, \
mobiheader.fullname_offset, \
mobiheader.fullname_length, \
mobiheader.locale, \
mobiheader.dictionary_input_language, \
mobiheader.dictionary_output_language, \
mobiheader.min_version, \
mobiheader.first_image_record, \
mobiheader.huffman_record, \
mobiheader.huffman_record_count, \
mobiheader.huffman_table_record, \
mobiheader.huffman_table_length, \
mobiheader.exth_flags, \
= struct.unpack('>4sLLLLLLLLLLLLLLLLLLLLLLLLLLLL', raw[0x10:0x84])
if len(raw) >= 0xb4:
mobiheader.drm_offset, \
mobiheader.drm_count, \
mobiheader.drm_size, \
mobiheader.drm_flags, \
= struct.unpack('>LLLL', raw[0xa4:0xb4])
if mobiheader.header_length < 0xe4 or \
mobiheader.header_length > 0xf8:
mobiheader.extra_flags = 0
else:
mobiheader.extra_flags, = struct.unpack('>H', raw[0xf2:0xf4])
fullname_end = mobiheader.fullname_offset + mobiheader.fullname_length
if fullname_end < len(raw):
mobiheader.fullname = raw[mobiheader.fullname_offset:fullname_end]
else:
mobiheader.fullname = None
if mobiheader.exth_flags & 0x40:
mobiheader.exth = _parse_exth_header(raw[16 + mobiheader.header_length:])
return mobiheader
