def read_bits_s(self, count):
if count > 32:
raise LsdError("Many bits for read: %d" % count)
res = 0
for i in range(count):
res <<= 1
res += self.read_bit()
return res
def read_some(self, length):
if length == 1:
return self.read_byte()
elif length == 2:
return self.read_word()
elif length == 4:
return self.read_int()
else:
raise LsdError('Allow read byte, word and int length: %i' % length)
def read_bits_o(self, count):
if count > 32:
raise LsdError("Many bits for read: %d" % count)
masks = (1, 3, 7, 0xF, 0x1F, 0x3F, 0x7F, 0xFF)
count_bytes = (count + self.in_byte_pos) // 8
if count + self.in_byte_pos - 8 * count_bytes > 0:
count_bytes += 1
# if in single raw byt
if count_bytes == 1:
if (self.in_byte_pos + count) < 8:
byte = self.record[self.pos]
byte >>= 7 - self.in_byte_pos - count + 1
byte &= masks[count - 1]
self.in_byte_pos += count
return byte
# many raw bytes
# inBitPos
# | count = 13 |
# 01234567 | 01234567 | 0123456
#
# inBitPos = 5 count_first = 3 count_las = 2
#
p = self.pos
count_last = (count + self.in_byte_pos) % 8
count_first = 8 - self.in_byte_pos
byte_first = self.record[p]
p += 1
byte_first &= masks[count_first - 1]
res = byte_first
# full bytes
full_bytes = (count - count_first) // 8
if full_bytes > 0:
for i in range(full_bytes):
res <<= 8
res += self.record[p]
p += 1
# last byte
if count_last > 0:
byte = self.record[p]
byte >>= 8 - count_last
res <<= count_last
res += byte
self.in_byte_pos = count_last
self.pos = p
return res
def parse(self):
if not self.readed:
self.read()
if self.verbose:
print("decoding overlay..")
self.overlay = OverlayReader(self.bstr, self.overlay_data)
if self.verbose:
print("decoding headings: %d" % self.header.entries_count)
self.read_headings()
if self.headings.appended != self.header.entries_count:
raise LsdError("Decoded not all entries %d != %d" % (self.headings.appended, self.header.entries_count))
# merge multititle headings
# self.headings = self.merge_headings()
if self.verbose:
print("decoding articles: %d" % len(self.headings))
for h in self.headings:
# h.dump()
self.dict.append((h, self.read_article(h)))
self._parsed = True
if self.verbose:
print("OK")
def __init__(self, dict_file, verbose=False):
self.filename = dict_file
self._readed = False
self._parsed = False
self.verbose = verbose
with open(dict_file, 'rb') as fp:
self.bstr = BitStream(bytearray(fp.read()))
self.overlay = None
self.headings = ArticleHeadingList()
self.dict = []
self.header = Header(self.bstr)
# check magic
if self.header.magic != u'LingVo':
raise LsdError('Allow only Lsd "LingVo" ident: %s' % repr(self.header.magic))
# initialize decoder
self.decoder = None
hi_version = self.header.hi_version
version = self.header.version
if hi_version == 0x11: # lingvo 11 dictionary: 0x11001
self.decoder = decoder.UserDictionaryDecoder(self.bstr)
elif hi_version == 0x12: # lingvo 12 dictionary: 0x12001
self.decoder = decoder.UserDictionaryDecoder(self.bstr)
elif hi_version == 0x13: # x3 dictionary: 0x131001 and 0x132001 if pages count > 1000
self.decoder = decoder.SystemDictionaryDecoder13(self.bstr)
elif hi_version == 0x14: # x5 dictionary
if version == 0x142001: # user dictionaries
self.decoder = decoder.UserDictionaryDecoder(self.bstr)
elif version == 0x141004: # system dictionaries
self.decoder = decoder.SystemDictionaryDecoder14(self.bstr)
elif version == 0x145001: # abbreviation dictionaries
self.decoder = decoder.AbbreviationDictionaryDecoder(self.bstr)
elif hi_version == 0x15: # x6 dictionary
if version == 0x152001: # user dictionaries
self.decoder = decoder.UserDictionaryDecoder(self.bstr)
elif version == 0x151005: # system dictionaries
# xor dictionary
self.xor_block_x6(self.header.dictionary_encoder_offset, self.header.articles_offset)
self.decoder = decoder.SystemDictionaryDecoder14(self.bstr)
elif version == 0x155001: # abbreviation dictionaries
self.decoder = decoder.AbbreviationDictionaryDecoder(self.bstr)
if self.decoder is None:
self.dump()
print("Not supported dictionary version: %s" % hex(self.header.version))
exit(1)
# raise LsdError("Not supported dict version %s" % hex(self.header.version))
name_len = self.bstr.read_some(1)
self.name = self.bstr.read_unicode(name_len, False)
self.first_heading = self.bstr.read_unicode(self.bstr.read_byte(), False)
self.last_heading = self.bstr.read_unicode(self.bstr.read_byte(), False)
capitals_len = reverse32(self.bstr.read_int())
self.capitals = self.bstr.read_unicode(capitals_len, False)
# icon v12+
if self.header.version > 0x120000:
self.icon_size = reverse16(self.bstr.read_word())
self.icon = self.bstr.read(self.icon_size)
else:
self.icon_size = 0
self.icon = None
if self.header.version > 0x140000:
self.header_checksum = reverse32(self.bstr.read_int())
else:
self.header_checksum = 0
if self.header.version > 0x120000:
self.pages_end = reverse32(self.bstr.read_int())
self.overlay_data = reverse32(self.bstr.read_int())
else:
self.pages_end = self.bstr.length
self.overlay_data = self.bstr.length # no overlay
if self.header.version > 0x140000:
self.dummy1 = reverse32(self.bstr.read_int())
self.dummy2 = reverse32(self.bstr.read_int())
else:
self.dummy1 = 0
self.dummy2 = 0
# set bstr pos for decoding
self.bstr.seek(self.header.dictionary_encoder_offset)