def create_validator():
"""Create an example CSV validator for patient demographic data."""
field_names = ('study_id', 'patient_id', 'gender', 'age_years',
'age_months', 'date_inclusion')
validator = CSVValidator(field_names)
# basic header and record length checks
validator.add_header_check('EX1', 'bad header')
validator.add_record_length_check('EX2', 'unexpected record length')
# some simple value checks
validator.add_value_check('study_id', int, 'EX3',
'study id must be an integer')
validator.add_value_check('patient_id', int, 'EX4',
'patient id must be an integer')
validator.add_value_check('gender', enumeration('M', 'F'), 'EX5',
'invalid gender')
validator.add_value_check('age_years', number_range_inclusive(0, 120, int),
'EX6', 'invalid age in years')
validator.add_value_check('date_inclusion', datetime_string('%Y-%m-%d'),
'EX7', 'invalid date')
# a more complicated record check
def check_age_variables(r):
age_years = int(r['age_years'])
age_months = int(r['age_months'])
valid = (age_months >= age_years * 12 and age_months % age_years < 12)
if not valid:
raise RecordError('EX8', 'invalid age variables')
validator.add_record_check(check_age_variables)
return validator
def __init__(self, model):
indicator_names = IndicatorType.objects.all()\
.values_list('nombre', flat=True)
self.value_checks = [
('fecha', datetime_string('%Y-%m-%d')),
('indicador_tipo', enumeration(list(indicator_names))),
('indicador_valor', str),
]
self.unique_checks = ('fecha',
'indicador_tipo', )
if issubclass(model, AbstractIndicator):
self.value_checks = self.value_checks + [
('jurisdiccion_nombre', str),
('jurisdiccion_id', str), ]
self.unique_checks = self.unique_checks + ('jurisdiccion_id', )
self.field_names = [check[0] for check in self.value_checks]
def sampleValidation():
"""
Sample code to validate a CSV file structure
"""
data_path = "data.csv"
report_path = "csv_validation_problems_report.txt"
field_names = ('date','','','type','reach','clicks','reactions')
validator = cv.CSVValidator(field_names)
validator.add_value_check('date',cv.datetime_string('%m/%d/%Y'),'EX1','invalid date')
validator.add_value_check('type',cv.enumeration('Video','Photo','Link'),"EX2","invalid type found")
validator.add_value_check('reach',int,"EX3","reach should be an integer")
validator.add_value_check('clicks',int,"EX4","clicks should be an integer")
validator.add_value_check('reactions',int,"EX4","reactions should be an integer")
validate_csv_file(data_path,report_path,validator)
def create_validator():
"""Create an example CSV validator for patient demographic data."""
field_names = (
'study_id',
'patient_id',
'gender',
'age_years',
'age_months',
'date_inclusion'
)
validator = CSVValidator(field_names)
# basic header and record length checks
validator.add_header_check('EX1', 'bad header')
validator.add_record_length_check('EX2', 'unexpected record length')
# some simple value checks
validator.add_value_check('study_id', int,
'EX3', 'study id must be an integer')
validator.add_value_check('patient_id', int,
'EX4', 'patient id must be an integer')
validator.add_value_check('gender', enumeration('M', 'F'),
'EX5', 'invalid gender')
validator.add_value_check('age_years', number_range_inclusive(0, 120, int),
'EX6', 'invalid age in years')
validator.add_value_check('date_inclusion', datetime_string('%Y-%m-%d'),
'EX7', 'invalid date')
# a more complicated record check
def check_age_variables(r):
age_years = int(r['age_years'])
age_months = int(r['age_months'])
valid = (age_months >= age_years * 12 and
age_months % age_years < 12)
if not valid:
raise RecordError('EX8', 'invalid age variables')
validator.add_record_check(check_age_variables)
return validator
def test_value_checks_datetime():
"""Test value checks with datetimes."""
field_names = ('foo', 'bar')
validator = CSVValidator(field_names)
validator.add_value_check('bar', datetime_string('%Y-%m-%d'))
data = (
('foo', 'bar'),
('A', '1999-09-09'), # valid
('B', '1999-13-09'), # invalid month
('C', '1999-09-32'), # invalid day
('D', '1999-09-09ss') # invalid string
)
problems = validator.validate(data)
assert len(problems) == 3, problems
for p in problems:
assert p['code'] == VALUE_CHECK_FAILED
assert problems[0]['row'] == 3 and problems[0]['field'] == 'bar'
assert problems[1]['row'] == 4 and problems[1]['field'] == 'bar'
assert problems[2]['row'] == 5 and problems[2]['field'] == 'bar'
def test_value_checks_datetime():
"""Test value checks with datetimes."""
field_names = ('foo', 'bar')
validator = CSVValidator(field_names)
validator.add_value_check('bar', datetime_string('%Y-%m-%d'))
data = (
('foo', 'bar'),
('A', '1999-09-09'), # valid
('B', '1999-13-09'), # invalid month
('C', '1999-09-32'), # invalid day
('D', '1999-09-09ss') # invalid string
)
problems = validator.validate(data)
assert len(problems) == 3, problems
for p in problems:
assert p['code'] == VALUE_CHECK_FAILED
assert problems[0]['row'] == 3 and problems[0]['field'] == 'bar'
assert problems[1]['row'] == 4 and problems[1]['field'] == 'bar'
assert problems[2]['row'] == 5 and problems[2]['field'] == 'bar'
