def calc_onehash(shingle, seed):
def c4_hash(shingle):
h = struct.unpack('<i',shingle)[0]
return h % ((sys.maxsize + 1) * 2)
if self.sh_type == 'c4':
return operator.xor(c4_hash(shingle), long(seed)) % self.modulo
else:
return operator.xor(compute_positive_hash(shingle), long(seed)) % self.modulo
def calc_onehash(sh_type, shingle, seed, modulo):
def c4_hash(shingle):
h = struct.unpack('<i', shingle)[0]
return h % ((sys.maxsize + 1) * 2)
if sh_type == 'c4':
return operator.xor(c4_hash(shingle), long(seed)) % modulo
else:
return operator.xor(compute_positive_hash(shingle),
long(seed)) % modulo
def _calculate_hash(self, obj):
"""
This method computes hash of object using a dynamic hashing function.
:param obj: object we are computing hash code for.
:return: hash code (long integer)
"""
# hash value to positive integer
h1 = compute_positive_hash(obj)
#calculate hash code: h(obj) mod 2^b (where b is num_bands)
return h1 % 2**self.num_bands
def _calculate_hash(self, obj):
"""
This method computes hash of object using a dynamic hashing function.
:param obj: object we are computing hash code for.
:return: hash code (long integer)
"""
# hash value to positive integer
h1 = compute_positive_hash(obj)
#calculate hash code: h(obj) mod 2^b (where b is num_bands)
return h1 % 2**self.num_bands
def calc_onehash(shingle, seed):
def c4_hash(shingle):
h = struct.unpack('<i',shingle)[0]
hash_val = h & max_mask
return hash_val
# hash_val = shingle_cache.get(shingle)
# if hash_val:
# return hash_val
# h = struct.unpack('<i',shingle)[0]
# hash_val = h & max_mask
# shingle_cache.set(shingle, hash_val)
# return hash_val
if self.sh_type == 'c4':
return operator.xor(c4_hash(shingle), long(seed)) % self.modulo
else:
return operator.xor(compute_positive_hash(shingle), long(seed)) % self.modulo
def calc_onehash(shingle, seed):
def c4_hash(shingle):
try:
h = struct.unpack('<i',shingle)[0]
except struct.error:
# We land here when the shingle has non-ascii characters in it.
size = 4
encoded = shingle.encode('utf-8')
int_hashes = [int(encoded[i:i + size].encode('hex'), 16) for i in range(len(encoded)/size)]
h = reduce(operator.xor, int_hashes)
hash_val = h & settings.max_mask
return hash_val
if self.sh_type == 'c4':
return operator.xor(c4_hash(shingle), long(seed)) % self.modulo
else:
return operator.xor(compute_positive_hash(shingle), long(seed)) % self.modulo
def calc_onehash(shingle, seed):
def c4_hash(shingle):
try:
h = struct.unpack('<i', shingle)[0]
except struct.error:
# We land here when the shingle has non-ascii characters in it.
size = 4
encoded = shingle.encode('utf-8')
int_hashes = [
int(encoded[i:i + size].encode('hex'), 16)
for i in range(len(encoded) / size)
]
h = reduce(operator.xor, int_hashes)
hash_val = h & settings.max_mask
return hash_val
if self.sh_type == 'c4':
return operator.xor(c4_hash(shingle),
long(seed)) % self.modulo
else:
return operator.xor(compute_positive_hash(shingle),
long(seed)) % self.modulo
def calc_hash(value):
return compute_positive_hash(value)
