def read(self):
if self.cursor < len(self.data):
index = self.cursor
self.cursor = index + 1
self.current = convert_to_bytes(self.data[index])
else:
self.current = None
return self.current
def read(self):
if self.cursor < len(self.data):
index = self.cursor
self.cursor = index + 1
self.current = convert_to_bytes(self.data[index])
else:
self.current = None
return self.current
def test_unicode_input(unicrud):
unicrud = list(set(unicrud))
assume(len(unicrud) > 5)
# MPH is an entropy game, hence things with low-entropy will
# confuse the hash algorithms preventing convergence on a
# solution, making this test fail
assume(_entropy(unicrud) == -0.0)
mph = cmph.generate_hash(unicrud)
# ... break the encapsulation, knowing that we
# do this under the hood
test_strs = [convert_to_bytes(s) for s in unicrud]
for original, escaped in zip(unicrud, test_strs):
assert mph(escaped) == mph(original)
def test_unicode_input(unicrud):
unicrud = list(set(unicrud))
assume(len(unicrud) > 5)
# MPH is an entropy game, hence things with low-entropy will
# confuse the hash algorithms preventing convergence on a
# solution, making this test fail
assume(_entropy(unicrud) == -0.0)
mph = cmph.generate_hash(unicrud)
# ... break the encapsulation, knowing that we
# do this under the hood
test_strs = [convert_to_bytes(s) for s in unicrud]
for original, escaped in zip(unicrud, test_strs):
assert mph(escaped) == mph(original)
def lookup(self, key):
"""
Generate hash code for a key from the Minimal Perfect Hash (MPH)
Parameters
----------
Key : object
The item to generate a key for, this works best for keys that
are strings, or can be transformed fairly directly into bytes
Returns : int
The code for the given item
"""
assert self._mph
key = convert_to_bytes(key)
box = ffi.new('char[]', key)
try:
result = _cmph.cmph_search(self._mph, box, len(key))
return result
finally:
del box
def test_unicode_bytes(unicrud):
assert unicrud_type(convert_to_bytes(unicrud), 'utf8') == unicrud