def test_mixed_run_report(self):
"""
Test that passing and failing files are written to the run report
when a mixture of the two is processed
"""
infile_root = resolve_test_data_path('data/general/pass_and_fail')
agency = 'MSC'
expected_passes = set()
expected_fails = set()
with report.OperatorReport() as error_bank:
run_report = report.RunReport(SANDBOX_DIR)
for infile in os.listdir(infile_root):
fullpath = os.path.join(infile_root, infile)
try:
contents = util.read_file(fullpath)
ecsv = parser.ExtendedCSV(contents, error_bank)
except (parser.MetadataValidationError,
parser.NonStandardDataError):
expected_fails.add(fullpath)
run_report.write_failing_file(fullpath, agency)
continue
try:
ecsv.validate_metadata_tables()
ecsv.validate_dataset_tables()
data_record = models.DataRecord(ecsv)
data_record.filename = infile
expected_passes.add(fullpath)
run_report.write_passing_file(fullpath, agency)
except (parser.MetadataValidationError,
parser.NonStandardDataError):
expected_fails.add(fullpath)
run_report.write_failing_file(fullpath, agency)
self.assertEquals(len(expected_passes), 6)
self.assertEquals(len(expected_fails), 4)
output_path = os.path.join(SANDBOX_DIR, 'run_report')
self.assertTrue(os.path.exists(output_path))
with open(output_path) as output:
lines = output.read().splitlines()
self.assertEquals(lines[0], agency)
self.assertEquals(len(lines),
len(expected_passes) + len(expected_fails) + 1)
for line in lines[1:]:
if line.startswith('Pass'):
target = line[6:].strip()
self.assertIn(target, expected_passes)
elif line.startswith('Fail'):
target = line[6:].strip()
self.assertIn(target, expected_fails)
def test_passing_run_report(self):
"""Test that a passing file is written to the run report"""
filename = '20080101.Kipp_Zonen.UV-S-E-T.000560.PMOD-WRC.csv'
infile = resolve_test_data_path('data/general/{}'.format(filename))
contents = util.read_file(infile)
run_report = report.RunReport(SANDBOX_DIR)
with report.OperatorReport() as error_bank:
ecsv = parser.ExtendedCSV(contents, error_bank)
ecsv.validate_metadata_tables()
ecsv.validate_dataset_tables()
data_record = models.DataRecord(ecsv)
data_record.filename = filename
agency = ecsv.extcsv['DATA_GENERATION']['Agency']
output_path = os.path.join(SANDBOX_DIR, 'run_report')
run_report.write_passing_file(infile, agency)
self.assertTrue(os.path.exists(output_path))
with open(output_path) as output:
lines = output.read().splitlines()
self.assertEquals(len(lines), 2)
self.assertEquals(lines[0], agency)
self.assertEquals(lines[1], 'Pass: {}'.format(infile))
def test_warning_operator_report(self):
"""Test that file warnings are written in the operator report"""
filename = 'ecsv-trailing-commas.csv'
infile = resolve_test_data_path('data/general/{}'.format(filename))
contents = util.read_file(infile)
with report.OperatorReport(SANDBOX_DIR) as op_report:
ecsv = parser.ExtendedCSV(contents, op_report)
# Some warnings are encountered during parsing.
ecsv.validate_metadata_tables()
ecsv.validate_dataset_tables()
data_record = models.DataRecord(ecsv)
data_record.filename = filename
agency = ecsv.extcsv['DATA_GENERATION']['Agency']
today = datetime.now().strftime('%Y-%m-%d')
output_path = os.path.join(SANDBOX_DIR,
'operator-report-{}.csv'.format(today))
op_report.add_message(200) # File passes validation
op_report.write_passing_file(infile, ecsv, data_record)
self.assertTrue(os.path.exists(output_path))
with open(output_path) as output:
reader = csv.reader(output)
next(reader)
expected_warnings = len(ecsv.warnings)
for _ in range(expected_warnings):
report_line = next(reader)
self.assertEquals(report_line[0], 'P')
self.assertEquals(report_line[1], 'Warning')
self.assertIn(agency, report_line)
self.assertIn(os.path.basename(infile), report_line)
report_line = next(reader)
self.assertEquals(report_line[0], 'P')
self.assertEquals(report_line[1], 'Warning')
self.assertEquals(report_line[2], '200')
self.assertIn(agency, report_line)
self.assertIn(os.path.basename(infile), report_line)
with self.assertRaises(StopIteration):
next(reader)
def test_passing_operator_report(self):
"""Test that a passing file is written in the operator report"""
filename = '20080101.Kipp_Zonen.UV-S-E-T.000560.PMOD-WRC.csv'
infile = resolve_test_data_path('data/general/{}'.format(filename))
contents = util.read_file(infile)
with report.OperatorReport(SANDBOX_DIR) as op_report:
ecsv = parser.ExtendedCSV(contents, op_report)
ecsv.validate_metadata_tables()
ecsv.validate_dataset_tables()
data_record = models.DataRecord(ecsv)
data_record.filename = filename
agency = ecsv.extcsv['DATA_GENERATION']['Agency']
today = datetime.now().strftime('%Y-%m-%d')
output_path = os.path.join(SANDBOX_DIR,
'operator-report-{}.csv'.format(today))
op_report.add_message(200) # File passes validation
op_report.write_passing_file(infile, ecsv, data_record)
self.assertTrue(os.path.exists(output_path))
with open(output_path) as output:
reader = csv.reader(output)
next(reader)
report_line = next(reader)
self.assertEquals(report_line[0], 'P')
self.assertEquals(report_line[2], '200')
self.assertIn(agency, report_line)
self.assertIn(os.path.basename(infile), report_line)
with self.assertRaises(StopIteration):
next(reader)
def test_run_report_multiple_agencies(self):
"""Test that files in the run report are grouped by agency"""
infile_root = resolve_test_data_path('data/general/agencies')
expected_passes = {}
expected_fails = {}
agency_aliases = {
'msc': 'MSC',
'casiap': 'CAS-IAP',
'mlcd-lu': 'MLCD-LU',
'dwd-mohp': 'DWD-MOHp'
}
with report.OperatorReport() as error_bank:
run_report = report.RunReport(SANDBOX_DIR)
for dirpath, dirnames, filenames in os.walk(infile_root):
for infile in filenames:
fullpath = os.path.join(dirpath, infile)
# Agency inferred from directory name.
agency = dirpath.split('/')[-1]
try:
contents = util.read_file(fullpath)
ecsv = parser.ExtendedCSV(contents, error_bank)
except (parser.MetadataValidationError,
parser.NonStandardDataError):
if agency not in expected_passes:
expected_passes[agency] = set()
if agency not in expected_fails:
expected_fails[agency] = set()
expected_fails[agency].add(fullpath)
run_report.write_failing_file(fullpath, agency)
continue
try:
ecsv.validate_metadata_tables()
agency = ecsv.extcsv['DATA_GENERATION']['Agency']
if agency not in expected_passes:
expected_passes[agency] = set()
if agency not in expected_fails:
expected_fails[agency] = set()
ecsv.validate_dataset_tables()
data_record = models.DataRecord(ecsv)
data_record.filename = infile
expected_passes[agency].add(fullpath)
run_report.write_passing_file(fullpath, agency)
except (parser.MetadataValidationError,
parser.NonStandardDataError):
agency = agency_aliases[agency]
if agency not in expected_passes:
expected_passes[agency] = set()
if agency not in expected_fails:
expected_fails[agency] = set()
expected_fails[agency].add(fullpath)
run_report.write_failing_file(fullpath, agency)
self.assertEquals(len(expected_passes['CAS-IAP']), 1)
self.assertEquals(len(expected_passes['DWD-MOHp']), 2)
self.assertEquals(len(expected_passes['MLCD-LU']), 3)
self.assertEquals(len(expected_passes['MSC']), 4)
self.assertEquals(len(expected_fails['CAS-IAP']), 0)
self.assertEquals(len(expected_fails['DWD-MOHp']), 1)
self.assertEquals(len(expected_fails['MLCD-LU']), 0)
self.assertEquals(len(expected_fails['MSC']), 1)
output_path = os.path.join(SANDBOX_DIR, 'run_report')
self.assertTrue(os.path.exists(output_path))
with open(output_path) as output:
lines = output.read().splitlines()
curr_agency = None
for line in lines:
if line.startswith('Pass'):
target = line[6:]
self.assertIn(target, expected_passes[curr_agency])
elif line.startswith('Fail'):
target = line[6:]
self.assertIn(target, expected_fails[curr_agency])
elif line.strip() != '':
curr_agency = line.strip()
self.assertIn(line, agency_aliases.values())
def test_mixed_operator_report(self):
"""
Test that passing and failing files are written to the operator report
when a mixture of the two is processed
"""
infile_root = resolve_test_data_path('data/general/pass_and_fail')
warnings = {}
errors = {}
expected_warnings = {}
expected_errors = {}
agency = 'UNKNOWN'
with report.OperatorReport(SANDBOX_DIR) as op_report:
for infile in os.listdir(infile_root):
fullpath = os.path.join(infile_root, infile)
warnings[fullpath] = 0
errors[fullpath] = 0
try:
contents = util.read_file(fullpath)
ecsv = parser.ExtendedCSV(contents, op_report)
except (parser.MetadataValidationError,
parser.NonStandardDataError) as err:
expected_errors[fullpath] = len(err.errors)
op_report.add_message(209)
op_report.write_failing_file(fullpath, agency)
continue
try:
ecsv.validate_metadata_tables()
agency = ecsv.extcsv['DATA_GENERATION']['Agency']
ecsv.validate_dataset_tables()
data_record = models.DataRecord(ecsv)
data_record.filename = infile
expected_warnings[fullpath] = len(ecsv.warnings)
expected_errors[fullpath] = 0
op_report.write_passing_file(fullpath, ecsv, data_record)
except (parser.MetadataValidationError,
parser.NonStandardDataError):
expected_warnings[fullpath] = len(ecsv.warnings)
expected_errors[fullpath] = len(ecsv.errors)
op_report.add_message(209)
op_report.write_failing_file(fullpath, agency, ecsv)
today = datetime.now().strftime('%Y-%m-%d')
output_path = os.path.join(SANDBOX_DIR,
'operator-report-{}.csv'.format(today))
self.assertTrue(os.path.exists(output_path))
with open(output_path) as output:
reader = csv.reader(output)
next(reader)
for line in reader:
if expected_errors[line[12]] == 0:
self.assertEquals(line[0], 'P')
self.assertEquals(line[1], 'Warning')
else:
self.assertEquals(line[0], 'F')
if line[2] == '200':
self.assertEquals(expected_errors[line[12]], 0)
elif line[2] == '209':
self.assertGreater(expected_errors[line[12]], 0)
elif line[1] == 'Warning':
warnings[line[12]] += 1
elif line[1] == 'Error':
errors[line[12]] += 1
self.assertEquals(warnings, expected_warnings)
self.assertEquals(errors, expected_errors)