def test_missing_column_names(self):
column_names = [
'given name', 'surname', 'email address', 'join date',
'account type'
] # missing 'age'
msg = 'Expected missing column name to throw FormatError.'
with self.assertRaises(FormatError, msg=msg):
validate_header(self.fields, column_names)
def test_invalid_column_names(self):
column_names = [
'given name', 'surname', 'email address', 'age', 'join date',
'nonexistent field'
]
msg = 'Expected invalid column name to throw FormatError.'
with self.assertRaises(FormatError, msg=msg):
validate_header(self.fields, column_names)
def generate_clk_from_csv(
input_f, # type: TextIO
keys, # type: Tuple[AnyStr, AnyStr]
schema, # type: Schema
validate=True, # type: bool
header=True, # type: Union[bool, AnyStr]
progress_bar=True # type: bool
):
# type: (...) -> List[str]
""" Generate Bloom filters from CSV file, then serialise them.
This function also computes and outputs the Hamming weight
(a.k.a popcount -- the number of bits set to high) of the
generated Bloom filters.
:param input_f: A file-like object of csv data to hash.
:param keys: A tuple of two lists of secret keys.
:param schema: Schema specifying the record formats and
hashing settings.
:param validate: Set to `False` to disable validation of
data against the schema. Note that this will silence
warnings whose aim is to keep the hashes consistent between
data sources; this may affect linkage accuracy.
:param header: Set to `False` if the CSV file does not have
a header. Set to `'ignore'` if the CSV file does have a
header but it should not be checked against the schema.
:param bool progress_bar: Set to `False` to disable the progress
bar.
:return: A list of serialized Bloom filters and a list of
corresponding popcounts.
"""
if header not in {False, True, 'ignore'}:
raise ValueError(
"header must be False, True or 'ignore' but is {}.".format(header))
log.info("Hashing data")
# Read from CSV file
reader = unicode_reader(input_f)
if header:
column_names = next(reader)
if header != 'ignore':
validate_header(schema.fields, column_names)
start_time = time.time()
# Read the lines in CSV file and add it to PII
pii_data = []
for line in reader:
pii_data.append(tuple(element.strip() for element in line))
validate_row_lengths(schema.fields, pii_data)
if progress_bar:
stats = OnlineMeanVariance()
with tqdm(desc="generating CLKs",
total=len(pii_data),
unit='clk',
unit_scale=True,
postfix={
'mean': stats.mean(),
'std': stats.std()
}) as pbar:
def callback(tics, clk_stats):
stats.update(clk_stats)
pbar.set_postfix(mean=stats.mean(),
std=stats.std(),
refresh=False)
pbar.update(tics)
results = generate_clks(pii_data,
schema,
keys,
validate=validate,
callback=callback)
else:
results = generate_clks(pii_data, schema, keys, validate=validate)
log.info("Hashing took {:.2f} seconds".format(time.time() - start_time))
return results
def test_good_column_names(self):
column_names = [
'given name', 'surname', 'email address', 'age', 'join date',
'account type'
]
validate_header(self.fields, column_names) # This should not throw