def camunzip(filename):
if (filename[-1] == 'h'):
method = 'huffman'
elif (filename[-1] == 's'):
method = 'shannon_fano'
elif (filename[-2] + filename[-1] == 'ar'):
print("dfass")
method = 'arithmetic'
elif (filename[-2] + filename[-1] == 'ad'):
method = 'arithmetic_adaptive'
elif (filename[-2] + filename[-1] == 'ca'):
method = 'context_adaptive'
else:
raise NameError('Unknown compression method')
with open(filename, 'rb') as fin:
y = fin.read()
y = bytes2bits(y)
if method == 'arithmetic' or method == 'arithmetic_adaptive' or method == 'context_adaptive':
pfile = filename[:-2] + 'p'
else:
pfile = filename[:-1] + 'p'
with open(pfile, 'r') as fp:
frequencies = load(fp)
n = sum([frequencies[a] for a in frequencies])
p = dict([(a, frequencies[a] / n) for a in frequencies])
if method == 'huffman' or method == 'shannon_fano':
if (method == 'huffman'):
xt = huffman(p)
c = xtree2code(xt)
else:
c = shannon_fano(p)
xt = code2xtree(c)
x = vl_decode(y, xt)
elif method == 'arithmetic_adaptive':
x = arithmetic_ftr_adaptive.decode(y)
elif method == 'arithmetic':
x = arithmetic.decode(y, p, n)
elif method == 'context_adaptive':
x = arithmetic_ftr.decode(y)
else:
raise NameError('This will never happen (famous last words)')
#'.cuz' for Cam UnZipped (don't want to overwrite the original file...)
if method == 'arithmetic' or method == 'arithmetic_adaptive' or method == 'context_adaptive':
outfile = filename[:-5] + '.cuz'
else:
outfile = filename[:-4] + '.cuz'
with open(outfile, 'w') as fout:
for c in x:
fout.write(c)
def unzip(self, filename):
"Decompress a file compressed by camzip"
filename = self.content_dir + filename
if filename[-1] == 'h':
method = 'huffman'
elif filename[-1] == 's':
method = 'shannon_fano'
elif filename[-1] == 'a':
method = 'arithmetic'
else:
raise NameError('Unknown compression method')
with open(filename, 'rb') as fin:
y = fin.read()
y = bytes2bits(y)
# Load frequencies file
pfile = filename[:-1] + 'p'
with open(pfile, 'r') as fp:
frequencies = load(fp)
tot = sum(list(frequencies.values()))
p = dict([(a, frequencies[a] / tot) for a in frequencies])
# Initial conditions
cp, f_initial = self._build_conditional_pdf(frequencies)
tot_initial = sum(list(f_initial.values()))
p_initial = dict([(a, f_initial[a] / tot_initial) for a in f_initial])
if method == 'huffman' or method == 'shannon_fano':
c = {}
xt = {}
init_key = '$' # unused character
if len(cp.keys()) == 1:
c[init_key], xt[init_key] = self._build_structures(method, p)
else:
c[init_key], xt[init_key] = self._build_structures(
method, p_initial)
for key in cp.keys():
c[key], xt[key] = self._build_structures(method, cp[key])
x = vl_decode(y, xt)
elif method == 'arithmetic':
x = arithmetic.decode(y, p, tot)
else:
raise NameError('This will never happen (famous last words)')
# '.cuz' for Cam UnZipped (don't want to overwrite the original file...)
outfile = filename[:-4] + '.cuz'
with open(outfile, 'w') as fout:
fout.write(x)
self._convert_to_LF(outfile)
def main() -> None:
"""Main function"""
string = input("Enter the string: ")
freq_dict, code = arith.encode(string)
print(f"\nEncoded message:")
print(f"(common) {code.numerator} / {code.denominator}")
print(f"(decimal) {float(code):.30e}\n")
decoded_string = arith.decode(code, freq_dict)
print(f"Decoded message is '{decoded_string}'")
def fetch_info(self, username):
profiles = open("profiles.txt", "r")
blob = profiles.read()
blob_decoded = codify.decode(blob, 58)
xmlfile = etree.fromstring(blob_decoded)
result = ""
for element in etree.iterparse(xmlfile):
if element.text == username:
self.is_logged_in = YES
return element.root
else:
pass
def camunzip(filename, length_of_LSTM_context=30):
if (filename[-1] == 'h'):
method = 'huffman'
elif (filename[-1] == 's'):
method = 'shannon_fano'
elif (filename[-1] == 'a'):
method = 'arithmetic'
elif (filename[-1] == "c"):
method = "conditional-arithmetic"
else:
raise NameError('Unknown compression method')
with open(filename, 'rb') as fin:
y = fin.read()
y = bytes2bits(y)
pfile = filename[:-1] + 'p'
with open(pfile, 'r') as fp:
frequencies = load(fp)
n = sum([frequencies[a] for a in frequencies])
p = dict([(int(a), frequencies[a] / n) for a in frequencies])
if method == 'huffman' or method == 'shannon_fano':
if (method == 'huffman'):
xt = huffman(p)
c = xtree2code(xt)
else:
c = shannon_fano(p)
xt = code2xtree(c)
x = vl_decode(y, xt)
elif method == 'arithmetic':
x = arithmetic.decode(y, p, n)
elif method == "conditional-arithmetic":
x = con_ari.decode(y, p, n, length_of_LSTM_context)
else:
raise NameError('This will never happen (famous last words)')
# '.cuz' for Cam UnZipped (don't want to overwrite the original file...)
outfile = filename[:-4] + '.cuz'
with open(outfile, 'wb') as fout:
fout.write(bytes(x))
return x
def camunzip(filename,
cond_prob_dict_filename=' ',
cum_prob_dict_filename=' ',
orig_message_filename=' '):
if not filename[-1] == 'z':
if filename[-1] == 'h':
method = 'huffman'
elif filename[-1] == 's':
method = 'shannon_fano'
elif filename[-1] == 'a':
method = 'arithmetic'
else:
raise NameError('Unknown compression method')
with open(filename, 'rb') as fin:
y = fin.read()
y = bytes2bits(y)
pfile = filename[:-1] + 'p'
with open(pfile, 'r') as fp:
frequencies = load(fp)
n = sum([frequencies[a] for a in frequencies])
p = dict([(int(a), frequencies[a] / n) for a in frequencies])
if method == 'huffman' or method == 'shannon_fano':
if method == 'huffman':
xt = huffman(p)
c = xtree2code(xt)
else:
c = shannon_fano(p)
xt = code2xtree(c)
x = vl_decode(y, xt)
elif method == 'arithmetic':
x = arithmetic.decode(y, p, n)
else:
raise NameError('This will never happen (famous last words)')
# '.cuz' for Cam UnZipped (don't want to overwrite the original file...)
outfile = filename[:-4] + '.cuz'
else:
with open('encoded_messages/' + filename + '_zipped.cz',
'r') as zipped_file:
with open('cond_prob_models/' + cond_prob_dict_filename,
'r') as cond_prob_file:
context_dict = json.load(cond_prob_file)
with open('cond_prob_models/' + cum_prob_dict_filename,
'r') as cum_prob_file:
cumulative_dict = json.load(cum_prob_file)
with open('text_files/' + orig_message_filename,
'r',
encoding='utf-8-sig') as file:
original_message = file.read()
x = contextual_arithmetic.decode(bytes2bits(zipped_file.read()),
context_dict, cumulative_dict,
len(original_message))
with open(outfile, 'wb') as fout:
fout.write(bytes(x))
def camunzip(filename, b=0.1, num=0, scale=(100000, 1), pr=0, pc=0):
if (filename[-1] == 'h'):
method = 'huffman'
elif (filename[-1] == 's'):
method = 'shannon_fano'
elif (filename[-1] == 'a'):
method = 'arithmetic'
elif (filename[-1] == 'c'):
method = 'carithmeticac'
elif (filename[-1] == 'i'):
method = 'iadhuffman'
elif (filename[-1] == 'f'):
method = 'fcondarithmetic'
elif (filename[-1] == 'g'):
method = 'gadconarithmetic'
elif (filename[-1] == 'j'):
method = 'jadconarithmetic'
else:
raise NameError('Unknown compression method')
with open(filename, 'rb') as fin: #*
y = fin.read()
y = bytes2bits(y)
pfile = filename[:-1] + 'p'
with open(pfile, 'r') as fp:
frequencies = load(fp)
n = sum([frequencies[a] for a in frequencies])
p = dict([(int(a), frequencies[a] / n) for a in frequencies]) #*
if method == 'huffman' or method == 'shannon_fano':
if (method == 'huffman'):
xt = huffman(p)
c = xtree2code(xt)
else:
c = shannon_fano(p)
xt = code2xtree(c)
x = vl_decode(y, xt)
elif method == 'arithmetic':
x = arithmetic.decode(y, p, n)
elif method == 'carithmeticac':
x = arithmeticac.decode(y, b, n, num, scale)
elif method == 'iadhuffman':
x = adhuffmandec(y, pr)
elif method == 'fcondarithmetic':
x = condarithmetic.decode(y, pc, n)
elif method == 'gadconarithmetic':
x = adconarithmetic.decode(y, b, n)
elif method == 'jadconarithmetic':
x = adconarithmetic2.decode(y, b, n)
else:
raise NameError('This will never happen (famous last words)')
# '.cuz' for Cam UnZipped (don't want to overwrite the original file...)
outfile = filename[:-4] + '.cuz'
with open(outfile, 'wb') as fout:
fout.write(bytes(x)) #*
import arithmetic as arith from vl_codes import bytes2bits, bits2bytes from os import stat from itertools import groupby filename = 'hamlet.txt' Nin = stat(filename).st_size f = open(filename, 'r') hamlet = f.read() frequencies = dict([(key, len(list(group))) for key, group in groupby(sorted(hamlet))]) p = dict([(a,frequencies[a]/Nin) for a in frequencies]) f.close() hamlet = hamlet * 10 Nin = Nin * 10 arith_encoded = arith.encode(hamlet, p, probability_on_the_go=False) arith_decoded = arith.decode(arith_encoded, p, Nin, probability_on_the_go=False) hamlet_zipped = bits2bytes(arith_encoded) Nout = len(hamlet_zipped) print(Nout/Nin) print(8 * Nout/Nin) arith_encoded = arith.encode(hamlet, p, probability_on_the_go=True) arith_decoded = arith.decode(arith_encoded, p, Nin, probability_on_the_go=True) hamlet_zipped = bits2bytes(arith_encoded) Nout = len(hamlet_zipped) print(Nout/Nin) print(8 * Nout/Nin)
from vl_codes import bytes2bits, bits2bytes
import arithmetic as arith
f = open('hamlet.txt', 'r')
hamlet = f.read()
f.close()
from itertools import groupby
frequencies = dict([(key, len(list(group)))
for key, group in groupby(sorted(hamlet))])
Nin = sum([frequencies[a] for a in frequencies])
p = dict([(a, frequencies[a] / Nin) for a in frequencies])
print(f'File length: {Nin}')
arith_encoded = arith.encode(hamlet, p)
arith_decoded = arith.decode(arith_encoded, p, Nin)
#==============================================================================
# c = huffman(p)
#==============================================================================
#print(xtree2newick(code2xtree(c)))
#==============================================================================
# hamlet_sf = vl_encode(hamlet,c);
# print(f'Length of binary sequence: {len(hamlet_sf)}')
#==============================================================================
#==============================================================================
# x = bits2bytes([0,1])
# print([format(a, '08b') for a in x])