def execute(arguments):
_validate_arguments(arguments)
_enrich_arguments(arguments)
compress_value = arguments[_compress_command]
read_file_value = arguments[_read_file_parameter]
write_file_value = arguments[_write_file_parameter]
lzw_value = arguments[_lzw_option]
elias_value = arguments[_elias_option]
code_value = arguments[_code_option]
if compress_value:
if lzw_value:
lzw.compress(read_file_value, write_file_value)
elif elias_value:
elias.compress(read_file_value,
write_file_value,
code_type=code_value)
else:
if lzw_value:
lzw.decompress(read_file_value, write_file_value)
elif elias_value:
elias.decompress(read_file_value,
write_file_value,
code_type=code_value)
def calcular_distancia(buscados, comprimidos): for archivo in os.listdir(buscados): bytes_archivo = lzw.readbytes(os.path.abspath(buscados + "/" + archivo)) # X archivo_comprimido = lzw.compress(bytes_archivo) # C(X) for comprimido in os.listdir(comprimidos): bytes_comprimido = lzw.readbytes(os.path.abspath(comprimidos + "/" + comprimido)) # C(Y) bytes_descomprimido = lzw.decompress(bytes_comprimido) # C(Y)
def decompress(option, encoded_file, decoded_file):
global byteArr
global bitPosition
if option == "1":
print "no compression"
file = open(decoded_file, "w")
file.write(open(encoded_file, "r").read())
file.close()
if option == "2":
# read the whole input file into a byte array
fileSize = os.path.getsize(str(os.path.abspath((encoded_file))))
fi = open(encoded_file, "rb")
# byteArr = map(ord, fi.read(fileSize))
byteArr = bytearray(fi.read(fileSize))
fi.close()
fileSize = len(byteArr)
print "File size in bytes:", fileSize
bitPosition = 0
n = int(bitReader(8), 2) + 1 # first read the number of encoding tuples
# print 'Number of encoding tuples:', n
dic = dict()
for i in range(n):
# read the byteValue
byteValue = int(bitReader(8), 2)
# read 3-bit(len(encodingBitStr)-1) value
m = int(bitReader(3), 2) + 1
# read encodingBitStr
encodingBitStr = bitReader(m)
dic[encodingBitStr] = byteValue # add to the dictionary
# print 'The dictionary of encodingBitStr : byteValue pairs:'
# print dic
# print
# read 32-bit file size (number of encoded bytes) value
numBytes = long(bitReader(32), 2) + 1
print "Number of bytes to decode:", numBytes
# read the encoded data, decode it, write into the output file
fo = open(decoded_file, "wb")
for b in range(numBytes):
# read bits until a decoding match is found
encodingBitStr = ""
while True:
encodingBitStr += bitReader(1)
if encodingBitStr in dic:
byteValue = dic[encodingBitStr]
fo.write(chr(byteValue))
break
fo.close()
if option == "3":
newbytes = b"".join(lzw.decompress(lzw.readbytes(encoded_file)))
decoded = open(decoded_file, "w")
decoded.write(newbytes)
print "LZW decoding num of bytes: " + str(newbytes.__sizeof__())
if option == "4":
ar = arcode.ArithmeticCode(False)
ar.decode_file(encoded_file, decoded_file)
def verify_compressed_file(self, testfile=GIANT_FILE):
with tempfile.TemporaryFile("w+b") as compressedfile:
originalsize = 0
compressedsize = 0
uncompressedsize = 0
bigstream = lzw.readbytes(testfile)
compressed = lzw.compress(bigstream)
for bs in compressed:
compressedsize = compressedsize + 1
compressedfile.write(bs)
############################
compressedfile.flush()
compressedfile.seek(0)
checkstream = lzw.readbytes(testfile)
uncompressed = lzw.decompress(lzw.filebytes(compressedfile))
for oldbyte, newbyte in six.moves.zip_longest(checkstream, uncompressed):
uncompressedsize = uncompressedsize + 1
if oldbyte != newbyte:
msg = "Corrupted byte at {0}, original {1} != {2}".format(uncompressedsize, oldbyte, newbyte)
self.assertEquals(oldbyte, newbyte, msg)
def verify_compressed_file(self, testfile=GIANT_FILE):
with tempfile.TemporaryFile("w+b") as compressedfile:
originalsize = 0
compressedsize = 0
uncompressedsize = 0
bigstream = lzw.readbytes(testfile)
compressed = lzw.compress(bigstream)
for bs in compressed:
compressedsize = compressedsize + 1
compressedfile.write(bs)
############################
compressedfile.flush()
compressedfile.seek(0)
checkstream = lzw.readbytes(testfile)
uncompressed = lzw.decompress(lzw.filebytes(compressedfile))
for oldbyte, newbyte in six.moves.zip_longest(
checkstream, uncompressed):
uncompressedsize = uncompressedsize + 1
if oldbyte != newbyte:
msg = "Corrupted byte at {0}, original {1} != {2}".format(
uncompressedsize, oldbyte, newbyte)
self.assertEquals(oldbyte, newbyte, msg)
def pic_from_data(self, data):
self.width, self.height, self.mode = unpack('HHB', data[0:5])
rle_data = lzw.decompress(map(ord, data[5:]), self.mode)
data = rle.decode(rle_data)
self.pic = []
for y in xrange(0, self.height):
self.pic.append(data[y*self.width:y*self.width + self.width])
print 'Pic read', self.width, 'x', self.height, '(', self.mode, ')'
def test_compressdecompress(self):
english = self.english
gibberish = self.gibberish
compressed = lzw.compress(english)
compressed = [ b for b in compressed ]
decompressed = b"".join(lzw.decompress(compressed))
self.assertEqual(english, decompressed)
compressed = lzw.compress(gibberish)
compressed = [ b for b in compressed ]
decompressed = b"".join(lzw.decompress(compressed))
self.assertEqual(gibberish, decompressed)
def test_compressdecompress(self):
english = self.english
gibberish = self.gibberish
compressed = lzw.compress(english)
compressed = [ b for b in compressed ]
decompressed = b"".join(lzw.decompress(compressed))
self.assertEqual(english, decompressed)
compressed = lzw.compress(gibberish)
compressed = [ b for b in compressed ]
decompressed = b"".join(lzw.decompress(compressed))
self.assertEqual(gibberish, decompressed)
def post_dissect(self, s):
if not conf.contribs["http"]["auto_compression"]:
return s
encodings = self._get_encodings()
# Un-chunkify
if "chunked" in encodings:
data = b""
while s:
length, _, body = s.partition(b"\r\n")
try:
length = int(length, 16)
except ValueError:
# Not a valid chunk. Ignore
break
else:
load = body[:length]
if body[length:length + 2] != b"\r\n":
# Invalid chunk. Ignore
break
s = body[length + 2:]
data += load
if not s:
s = data
# Decompress
try:
if "deflate" in encodings:
import zlib
s = zlib.decompress(s)
elif "gzip" in encodings:
s = gzip_decompress(s)
elif "compress" in encodings:
import lzw
s = lzw.decompress(s)
elif "br" in encodings:
if _is_brotli_available:
s = brotli.decompress(s)
else:
log_loading.info(
"Can't import brotli. brotli decompression "
"will be ignored !")
elif "zstd" in encodings:
if _is_zstd_available:
# Using its streaming API since its simple API could handle
# only cases where there is content size data embedded in
# the frame
bio = io.BytesIO(s)
reader = zstandard.ZstdDecompressor().stream_reader(bio)
s = reader.read()
else:
log_loading.info(
"Can't import zstandard. zstd decompression "
"will be ignored !")
except Exception:
# Cannot decompress - probably incomplete data
pass
return s
def extract_config(lzw_file, zyxel_file, r):
# todo
header_size = 60
print("[+] Stripping header")
with open(lzw_file, 'wb') as f:
r.raw.decode_content = True
f.write(r.content[header_size:])
try:
mybytes = lzw.readbytes(lzw_file)
lbytes = lzw.decompress(mybytes)
lzw.writebytes(zyxel_file, lbytes)
except:
pass
def post_dissect(self, s):
if not conf.contribs["http"]["auto_compression"]:
return s
encodings = self._get_encodings()
# Un-chunkify
if "chunked" in encodings:
data = b""
while s:
length, _, body = s.partition(b"\r\n")
try:
length = int(length, 16)
except ValueError:
# Not a valid chunk. Ignore
break
else:
load = body[:length]
if body[length:length + 2] != b"\r\n":
# Invalid chunk. Ignore
break
s = body[length + 2:]
data += load
if not s:
s = data
# Decompress
try:
if "deflate" in encodings:
import zlib
s = zlib.decompress(s)
elif "gzip" in encodings:
s = gzip_decompress(s)
elif "compress" in encodings:
import lzw
s = lzw.decompress(s)
elif "br" in encodings:
if _is_brotli_available:
s = brotli.decompress(s)
else:
log_loading.info(
"Can't import brotli. brotli decompression "
"will be ignored !"
)
except Exception:
# Cannot decompress - probably incomplete data
pass
return s
def test_compress_decompress_2():
s = "rererere"
cmp_s, _, dico = compress(s)
res = decompress(cmp_s, dico)
assert res == s
def read_tiff(filename):
with open(filename, "rb") as tifffile:
header = tifffile.read(4)
assert header == TIFF_HEADER, "TIFF header not found"
directorystart = read_uint32(tifffile)
log.debug("directory start at 0x%x", directorystart)
tifffile.seek(directorystart)
directorylength = read_uint16(tifffile)
log.debug("directory length: %d", directorylength)
directory = {}
for i in range(directorylength):
tifffile.seek(directorystart + 2 + DIRECTORY_ENTRY_LENGTH * i)
tag = read_uint16(tifffile)
valuetype = read_uint16(tifffile)
length = read_uint32(tifffile)
typelen = VALUETYPE[valuetype][1]
if length * typelen > 4:
pointer = read_uint32(tifffile)
tifffile.seek(pointer)
raw_values = struct.unpack("<" + length * VALUETYPE[valuetype][0],
tifffile.read(length * typelen))
values = VALUETYPE[valuetype][2](raw_values)
assert tag in __FIELD_BY_ID_MAP, "Tag 0x%x not found" % tag
fielddata = __FIELD_BY_ID_MAP[tag]
value = fielddata[2](values)
if len(fielddata) == 4:
assert value in fielddata[3], \
"Value not acceptable: %s %s" % (fielddata[0], repr(value))
directory[fielddata[0]] = value
log.debug("Found tag: %s = %s", fielddata[0], repr(value))
assert len(directory["bitspersample"]) == directory["samplesperpixel"]
assert len(directory["sampleformat"]) == directory["samplesperpixel"]
nrstrips = int(math.ceil(float(directory["height"]) /
directory["rowsperstrip"]))
assert len(directory["stripoffsets"]) == nrstrips
assert len(directory["stripbytecounts"]) == nrstrips
notpresentfields = set([x[0] for x in FIELD]) - set(directory.keys())
for fieldname in notpresentfields:
assert fieldname in optionalfields, "Field %s missing" % fieldname
imageasstring = ""
# read the strips
for stripnr in range(nrstrips):
nrrows = min(directory["height"] -
directory["rowsperstrip"] * stripnr,
directory["rowsperstrip"])
nrpixels = directory["width"] * nrrows
nrbytes = nrpixels * sum(directory["bitspersample"]) / 8
tifffile.seek(directory["stripoffsets"][stripnr])
assert directory["compression"] == 5 # lzw
lzwstrip = tifffile.read(directory["stripbytecounts"][stripnr])
predictedstrip = numpy.fromstring(lzw.decompress(lzwstrip),
dtype=numpy.uint8)
assert len(predictedstrip) == nrbytes
assert directory["predictor"] == 3 # float predictor
# undo prediction in two steps:
# first add the value or the previous 4 columns to the next
# over the whole strip (should probably optimise)
width = directory["width"]
nrchannels = directory["samplesperpixel"]
bytespersample = directory["bitspersample"][0] / 8
assert len(set(directory["bitspersample"])) == 1, \
"Images with different number of bits per sample per channel" \
" not supported"
cumsummedstrip = (predictedstrip.reshape(
(nrrows, width * bytespersample, nrchannels)).
cumsum(1, dtype=numpy.uint8))
# then re-arrange the value. if the width is 640 and 4 float32 per
# pixel
strip = (cumsummedstrip.reshape(
# reverse the order of columns, to go from bit to little endian
(nrrows, bytespersample, width * nrchannels))[:, ::-1, :].
# transpose in the second dimension to descramble the parts
# of the individual floats
transpose(0, 2, 1).flatten().tostring())
imageasstring += strip
flatimage = numpy.fromstring(imageasstring, dtype=numpy.float32)
return flatimage.reshape((directory["height"], directory["width"],
directory["samplesperpixel"]))
def lzwDecode(stream, parameters):
decodedStream = ''
try:
generator = lzw.decompress(stream)
for c in generator:
decodedStream += c
except:
return (-1,'Error decompressing string')
if parameters == None or parameters == {}:
return (0,decodedStream)
else:
if parameters.has_key('/Predictor'):
predictor = parameters['/Predictor'].getRawValue()
else:
predictor = None
if parameters.has_key('/Columns'):
columns = parameters['/Columns'].getRawValue()
else:
columns = None
if parameters.has_key('/Colors'):
colors = parameters['/Colors'].getRawValue()
else:
colors = None
if parameters.has_key('/BitsPerComponent'):
bits = parameters['/BitsPerComponent'].getRawValue()
else:
bits = None
if parameters.has_key('/EarlyChange'):
earlyChange = parameters['/EarlyChange'].getRawValue()
else:
earlyChange = None
if predictor != None and predictor != 1:
# PNG prediction:
if predictor >= 10 and predictor <= 15:
output = ''
# PNG prediction can vary from row to row
rowlength = columns + 1
prev_rowdata = (0,) * rowlength
for row in xrange(len(decodedStream) / rowlength):
rowdata = [ord(x) for x in decodedStream[(row*rowlength):((row+1)*rowlength)]]
filterByte = rowdata[0]
if filterByte == 0:
pass
elif filterByte == 1:
for i in range(2, rowlength):
rowdata[i] = (rowdata[i] + rowdata[i-1]) % 256
elif filterByte == 2:
for i in range(1, rowlength):
rowdata[i] = (rowdata[i] + prev_rowdata[i]) % 256
else:
# unsupported PNG filter
#sys.exit("Unsupported PNG filter %r" % filterByte)
return (-1,'Unsupported parameters')
prev_rowdata = rowdata
output += (''.join([chr(x) for x in rowdata[1:]]))
return (0,output)
else:
# unsupported predictor
#sys.exit("Unsupported flatedecode predictor %r" % predictor)
return (-1,'Unsupported parameters')
def test_compress_decompress_3():
s = "coucou"
cmp_s, _, dico = compress(s)
res = decompress(cmp_s, dico)
assert res == s
def decode(self, data):
assert self.getParams()['EarlyChange']==1
data = lzw.decompress(data)
data = Predictor(p['Predictor'],p['Columns'],p['BitsPerComponent']).decode(data)
return data
def test_compress_decompress_4():
s = "pourquoi pas"
cmp_s, _, dico = compress(s)
res = decompress(cmp_s, dico)
assert res == s
import lzw
infile = lzw.readbytes("3_1.spc")
compressed = lzw.compress(infile)
lzw.writebytes("3_1.spc.compressed", compressed)
infile.close()
infile = lzw.readbytes("3_1.spc.compressed")
uncompressed = lzw.decompress(infile)
lzw.writebytes("3_1.spc.decompressed", uncompressed)
infile.close()
import lzw
mybytes = lzw.readbytes("ElQuijote.txt")
lessbytes = lzw.compress(mybytes)
outFile = open("Compressed.txt", 'wb')
outFile.write(b"".join(lessbytes))
outFile.close()
newbytes = b"".join(lzw.decompress(lessbytes))
oldbytes = b"".join(lzw.readbytes("ElQuijote.txt"))
print(oldbytes == newbytes)
def test_compress_decompress_1():
s = "ab*cde*fgh*"
cmp_s, _, dico = compress(s)
res = decompress(cmp_s, dico)
assert res == s
def decode(self, data):
assert self.getParams()['EarlyChange'] == 1
data = lzw.decompress(data)
data = Predictor(p['Predictor'], p['Columns'],
p['BitsPerComponent']).decode(data)
return data
def decode(data):
assert self.getParams()["EarlyChange"] == 1
data = lzw.decompress(data)
data = Predictor(p["Predictor"], p["Columns"], p["BitsPerComponent"]).decode(data)
return data
def lzwDecode(stream, parameters):
decodedStream = ''
try:
generator = lzw.decompress(stream)
for c in generator:
decodedStream += c
except:
return (-1, 'Error decompressing string')
if parameters == None or parameters == {}:
return (0, decodedStream)
else:
if parameters.has_key('/Predictor'):
predictor = parameters['/Predictor'].getRawValue()
else:
predictor = None
if parameters.has_key('/Columns'):
columns = parameters['/Columns'].getRawValue()
else:
columns = None
if parameters.has_key('/Colors'):
colors = parameters['/Colors'].getRawValue()
else:
colors = None
if parameters.has_key('/BitsPerComponent'):
bits = parameters['/BitsPerComponent'].getRawValue()
else:
bits = None
if parameters.has_key('/EarlyChange'):
earlyChange = parameters['/EarlyChange'].getRawValue()
else:
earlyChange = None
if predictor != None and predictor != 1:
# PNG prediction:
if predictor >= 10 and predictor <= 15:
output = ''
# PNG prediction can vary from row to row
rowlength = columns + 1
prev_rowdata = (0, ) * rowlength
for row in xrange(len(decodedStream) / rowlength):
rowdata = [
ord(x)
for x in decodedStream[(row * rowlength):((row + 1) *
rowlength)]
]
filterByte = rowdata[0]
if filterByte == 0:
pass
elif filterByte == 1:
for i in range(2, rowlength):
rowdata[i] = (rowdata[i] + rowdata[i - 1]) % 256
elif filterByte == 2:
for i in range(1, rowlength):
rowdata[i] = (rowdata[i] + prev_rowdata[i]) % 256
else:
# unsupported PNG filter
#sys.exit("Unsupported PNG filter %r" % filterByte)
return (-1, 'Unsupported parameters')
prev_rowdata = rowdata
output += (''.join([chr(x) for x in rowdata[1:]]))
return (0, output)
else:
# unsupported predictor
#sys.exit("Unsupported flatedecode predictor %r" % predictor)
return (-1, 'Unsupported parameters')
def test_compress_decompress_5():
s = "abcd*dccacbdda*aaddcba*"
cmp_s, _, dico = compress(s)
res = decompress(cmp_s, dico)
assert res == s
import lzw
print("Compressing text...")
print()
compressed = lzw.compress('darth_plagueis.txt')
print("Compressed text: ")
print(compressed)
print()
print("Decompressing text...")
lzw.decompress(compressed, 'darth_plagueis_out.txt')
def decompress():
lzw.decompress(
r'D:\workspace.python\data-compressor\files\wap_compressed.txt',
r'D:\workspace.python\data-compressor\files\wap_compressed_decompressed.txt',
)
def main():
fn_in = sys.argv[1]
fn_out = sys.argv[2]
print("Decompressing from " + fn_in + " to " + fn_out)
compr = string_to_compressed_code(read_file(fn_in))
write_file(fn_out, decompress(compr))
def test_property(x):
assert lzw.decompress(lzw.compress(x)) == x
