You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
97 lines
2.7 KiB
97 lines
2.7 KiB
import io
|
|
import logging
|
|
|
|
|
|
logger = logging.getLogger(__name__)
|
|
|
|
|
|
class CorruptDataError(Exception):
|
|
pass
|
|
|
|
class LZWDecoder:
|
|
|
|
def __init__(self, fp):
|
|
self.fp = fp
|
|
self.buff = 0
|
|
self.bpos = 8
|
|
self.nbits = 9
|
|
self.table = None
|
|
self.prevbuf = None
|
|
|
|
def readbits(self, bits):
|
|
v = 0
|
|
while 1:
|
|
# the number of remaining bits we can get from the current buffer.
|
|
r = 8-self.bpos
|
|
if bits <= r:
|
|
# |-----8-bits-----|
|
|
# |-bpos-|-bits-| |
|
|
# | |----r----|
|
|
v = (v<<bits) | ((self.buff>>(r-bits)) & ((1<<bits)-1))
|
|
self.bpos += bits
|
|
break
|
|
else:
|
|
# |-----8-bits-----|
|
|
# |-bpos-|---bits----...
|
|
# | |----r----|
|
|
v = (v<<r) | (self.buff & ((1<<r)-1))
|
|
bits -= r
|
|
x = self.fp.read(1)
|
|
if not x: raise EOFError
|
|
self.buff = ord(x)
|
|
self.bpos = 0
|
|
return v
|
|
|
|
def feed(self, code):
|
|
x = b''
|
|
if code == 256:
|
|
self.table = [bytes([i]) for i in range(256)] # 0-255
|
|
self.table.append(None) # 256
|
|
self.table.append(None) # 257
|
|
self.prevbuf = b''
|
|
self.nbits = 9
|
|
elif code == 257:
|
|
pass
|
|
elif not self.prevbuf:
|
|
try:
|
|
x = self.prevbuf = self.table[code]
|
|
except (TypeError, IndexError):
|
|
# TypeError: table is None
|
|
raise CorruptDataError()
|
|
else:
|
|
if code < len(self.table):
|
|
x = self.table[code]
|
|
self.table.append(self.prevbuf+x[:1])
|
|
elif code == len(self.table):
|
|
self.table.append(self.prevbuf+self.prevbuf[:1])
|
|
x = self.table[code]
|
|
else:
|
|
raise CorruptDataError()
|
|
l = len(self.table)
|
|
if l == 511:
|
|
self.nbits = 10
|
|
elif l == 1023:
|
|
self.nbits = 11
|
|
elif l == 2047:
|
|
self.nbits = 12
|
|
self.prevbuf = x
|
|
return x
|
|
|
|
def run(self):
|
|
while 1:
|
|
try:
|
|
code = self.readbits(self.nbits)
|
|
except EOFError:
|
|
break
|
|
try:
|
|
x = self.feed(code)
|
|
except CorruptDataError:
|
|
# just ignore corrupt data and stop yielding there
|
|
break
|
|
yield x
|
|
logger.debug('nbits=%d, code=%d, output=%r, table=%r', self.nbits, code, x, self.table)
|
|
|
|
# lzwdecode
|
|
def lzwdecode(data):
|
|
fp = io.BytesIO(data)
|
|
return b''.join(LZWDecoder(fp).run())
|