def test_bad_data(self):
"""
Ensure that bad data is skipped when it exists.
"""
# the first data record in this file is corrupted and will be ignored
# we expect the first 2 particles to be the metadata particle and the
# intrument particle from the data record after the corrupted one
with open(os.path.join(RESOURCE_PATH, '11079419_BAD_PPB_OCR.txt'), 'rU') as file_handle:
log.debug(self.exception_callback_value)
parser = SpkirAbjCsppParser(self._recov_config,
file_handle,
self.exception_callback)
particles = parser.get_records(2)
self.assert_particles(particles, 'bad_data_record_recov.yml', RESOURCE_PATH)
with open(os.path.join(RESOURCE_PATH, '11079419_BAD_PPB_OCR.txt'), 'rU') as file_handle:
log.debug(self.exception_callback_value)
parser = SpkirAbjCsppParser(self._telem_config,
file_handle,
self.exception_callback)
particles = parser.get_records(2)
self.assert_particles(particles, 'bad_data_record_telem.yml', RESOURCE_PATH)
def test_simple(self):
"""
Read test data and pull out data particles
Assert that the results are those we expected.
"""
with open(os.path.join(RESOURCE_PATH, '11079364_PPD_OCR.txt'), 'rU') as file_handle:
# Note: since the recovered and teelemetered parser and particles are common
# to each other, testing one is sufficient, will be completely tested
# in driver tests
parser = SpkirAbjCsppParser(self._recov_config,
file_handle,
self.exception_callback)
particles = parser.get_records(20)
log.debug("*** test_simple Num particles %s", len(particles))
self.assert_particles(particles, '11079364_PPD_OCR_recov.yml', RESOURCE_PATH)
with open(os.path.join(RESOURCE_PATH, '11079364_PPD_OCR.txt'), 'rU') as file_handle:
# Note: since the recovered and teelemetered parser and particles are common
# to each other, testing one is sufficient, will be completely tested
# in driver tests
parser = SpkirAbjCsppParser(self._telem_config,
file_handle,
self.exception_callback)
particles = parser.get_records(20)
log.debug("*** test_simple Num particles %s", len(particles))
self.assert_particles(particles, '11079364_PPD_OCR_telem.yml', RESOURCE_PATH)
def test_extra_data(self):
"""
Ensure that bad data is skipped when it exists.
"""
# the first 2 data record in this file are corrupted by adding additional
# data vlaues separated by tabs and will be ignored
# we expect the first 2 particles to be the metadata particle and the
# intrument particle from the data record after the corrupted one
with open(
os.path.join(RESOURCE_PATH, '11079364_EXTRA_DATA_PPD_OCR.txt'),
'r') as file_handle:
log.info(self.exception_callback_value)
parser = SpkirAbjCsppParser(self._recov_config, file_handle,
self.exception_callback)
particles = parser.get_records(2)
self.assertEquals(len(self.exception_callback_value), 2)
for exception in self.exception_callback_value:
self.assert_(isinstance(exception, RecoverableSampleException))
# expect to see a recoverable sample exception in the log
log.debug('TEST EXTRA DATA exception call back is %s',
self.exception_callback_value)
self.assert_particles(particles, 'extra_data_values_recov.yml',
RESOURCE_PATH)
self.exception_callback_value = []
with open(
os.path.join(RESOURCE_PATH, '11079364_EXTRA_DATA_PPD_OCR.txt'),
'r') as file_handle:
log.info(self.exception_callback_value)
parser = SpkirAbjCsppParser(self._telem_config, file_handle,
self.exception_callback)
particles = parser.get_records(2)
self.assertEquals(len(self.exception_callback_value), 2)
for exception in self.exception_callback_value:
self.assert_(isinstance(exception, RecoverableSampleException))
# expect to see a recoverable sample exception in the log
log.debug('TEST EXTRA DATA exception call back is %s',
self.exception_callback_value)
self.assert_particles(particles, 'extra_data_values_telem.yml',
RESOURCE_PATH)
def test_bad_data(self):
"""
Ensure that bad data is skipped when it exists.
"""
# the first data record in this file is corrupted and will be ignored
# we expect the first 2 particles to be the metadata particle and the
# intrument particle from the data record after the corrupted one
file_path = os.path.join(RESOURCE_PATH, '11079419_BAD_PPB_OCR.txt')
stream_handle = open(file_path, 'rb')
log.info(self.exception_callback_value)
parser = SpkirAbjCsppParser(
self.config.get(DataTypeKey.SPKIR_ABJ_CSPP_RECOVERED), None,
stream_handle, self.state_callback, self.pub_callback,
self.exception_callback)
particles = parser.get_records(2)
expected_results = self.get_dict_from_yml('bad_data_record.yml')
self.assertTrue(len(particles) == 2)
for i in range(len(particles)):
self.assert_result(expected_results['data'][i], particles[i])
stream_handle.close()
def test_get_many(self):
"""
Read test data and pull out multiple data particles at one time.
Assert that the results are those we expected.
"""
file_path = os.path.join(RESOURCE_PATH, '11079419_PPB_OCR.txt')
stream_handle = open(file_path, 'r')
# Note: since the recovered and teelemetered parser and particles are common
# to each other, testing one is sufficient, will be completely tested
# in driver tests
parser = SpkirAbjCsppParser(
self.config.get(DataTypeKey.SPKIR_ABJ_CSPP_RECOVERED), None,
stream_handle, self.state_callback, self.pub_callback,
self.exception_callback)
# try to get 2000 particles, there are only 1623 data records
# so should get 1624 including the meta data
particles = parser.get_records(2000)
log.debug("*** test_get_many Num particles %s", len(particles))
self.assertEqual(len(particles), 1624)
stream_handle.close()
def test_simple(self):
"""
Read test data and pull out data particles
Assert that the results are those we expected.
"""
file_path = os.path.join(RESOURCE_PATH, '11079419_PPB_OCR.txt')
stream_handle = open(file_path, 'r')
# Note: since the recovered and teelemetered parser and particles are common
# to each other, testing one is sufficient, will be completely tested
# in driver tests
parser = SpkirAbjCsppParser(
self.config.get(DataTypeKey.SPKIR_ABJ_CSPP_RECOVERED), None,
stream_handle, self.state_callback, self.pub_callback,
self.exception_callback)
particles = parser.get_records(20)
log.debug("*** test_simple Num particles %s", len(particles))
# check the first particle, which should be the metadata particle (recovered)
test_data = self.get_dict_from_yml('11079419_PPB_OCR_recov.yml')
# check all the values against expected results.
for i in range(len(particles)):
self.assert_result(test_data['data'][i], particles[i])
stream_handle.close()
def test_get_many(self):
"""
Read test data and pull out multiple data particles at one time.
Assert that the results are those we expected.
"""
with open(os.path.join(RESOURCE_PATH, '11079419_PPB_OCR.txt'),
'r') as file_handle:
# Note: since the recovered and teelemetered parser and particles are common
# to each other, testing one is sufficient, will be completely tested
# in driver tests
parser = SpkirAbjCsppParser(self._recov_config, file_handle,
self.exception_callback)
# try to get 2000 particles, there are only 1623 data records
# so should get 1624 including the meta data
particles = parser.get_records(2000)
log.debug("*** test_get_many Num particles %s", len(particles))
self.assert_particles(particles, '11079419_PPB_OCR_recov.yml',
RESOURCE_PATH)
with open(os.path.join(RESOURCE_PATH, '11079419_PPB_OCR.txt'),
'r') as file_handle:
# Note: since the recovered and teelemetered parser and particles are common
# to each other, testing one is sufficient, will be completely tested
# in driver tests
parser = SpkirAbjCsppParser(self._telem_config, file_handle,
self.exception_callback)
# try to get 2000 particles, there are only 1623 data records
# so should get 1624 including the meta data
particles = parser.get_records(2000)
log.debug("*** test_get_many Num particles %s", len(particles))
self.assert_particles(particles, '11079419_PPB_OCR_telem.yml',
RESOURCE_PATH)
def _build_parser(self, parser_state, stream_in, data_key):
"""
Build and return the parser
"""
config = self._parser_config.get(data_key)
if config is None:
log.warn(
'Parser config does not exist for key = %s. Not building parser',
data_key)
raise ConfigurationException
if data_key == DataTypeKey.SPKIR_ABJ_CSPP_TELEMETERED:
config.update({
DataSetDriverConfigKeys.PARTICLE_MODULE:
'mi.dataset.parser.spkir_abj_cspp',
DataSetDriverConfigKeys.PARTICLE_CLASS: None,
DataSetDriverConfigKeys.PARTICLE_CLASSES_DICT: {
METADATA_PARTICLE_CLASS_KEY:
SpkirAbjCsppMetadataTelemeteredDataParticle,
DATA_PARTICLE_CLASS_KEY:
SpkirAbjCsppInstrumentTelemeteredDataParticle,
}
})
elif data_key == DataTypeKey.SPKIR_ABJ_CSPP_RECOVERED:
config.update({
DataSetDriverConfigKeys.PARTICLE_MODULE:
'mi.dataset.parser.spkir_abj_cspp',
DataSetDriverConfigKeys.PARTICLE_CLASS: None,
DataSetDriverConfigKeys.PARTICLE_CLASSES_DICT: {
METADATA_PARTICLE_CLASS_KEY:
SpkirAbjCsppMetadataRecoveredDataParticle,
DATA_PARTICLE_CLASS_KEY:
SpkirAbjCsppInstrumentRecoveredDataParticle,
}
})
else:
log.warn('Invalid Data_Key %s. Not building parser', data_key)
raise ConfigurationException
log.debug("_build_parser Config: %s", config)
parser = SpkirAbjCsppParser(
config, parser_state, stream_in,
lambda state, ingested: self._save_parser_state(
state, data_key, ingested), self._data_callback,
self._sample_exception_callback)
return parser
def _build_parser(self, stream_handle):
parser_config = {
DataSetDriverConfigKeys.PARTICLE_MODULE: 'mi.dataset.parser.spkir_abj_cspp',
DataSetDriverConfigKeys.PARTICLE_CLASS: None,
DataSetDriverConfigKeys.PARTICLE_CLASSES_DICT: {
METADATA_PARTICLE_CLASS_KEY: SpkirAbjCsppMetadataRecoveredDataParticle,
DATA_PARTICLE_CLASS_KEY: SpkirAbjCsppInstrumentRecoveredDataParticle,
}
}
parser = SpkirAbjCsppParser(parser_config, stream_handle,
self._exception_callback)
return parser
def test_set_state(self):
"""
Test changing to a new state after initializing the parser and
reading data, as if new data has been found and the state has
changed
"""
file_path = os.path.join(RESOURCE_PATH, '11079419_PPB_OCR.txt')
stream_handle = open(file_path, 'r')
# 11079419_PPB_OCR_20.yml has the metadata and the first 19
# instrument particles in it
expected_results = self.get_dict_from_yml('11079419_PPB_OCR_recov.yml')
parser = SpkirAbjCsppParser(
self.config.get(DataTypeKey.SPKIR_ABJ_CSPP_RECOVERED), None,
stream_handle, self.state_callback, self.pub_callback,
self.exception_callback)
particles = parser.get_records(2)
log.debug("Num particles: %s", len(particles))
self.assertTrue(len(particles) == 2)
for i in range(len(particles)):
self.assert_result(expected_results['data'][i], particles[i])
# position 3656 is the byte at the start of the 18th data record
new_state = {
StateKey.POSITION: 3769,
StateKey.METADATA_EXTRACTED: True
}
parser.set_state(new_state)
particles = parser.get_records(2)
self.assertTrue(len(particles) == 2)
# offset in the expected results
offset = 18
for i in range(len(particles)):
self.assert_result(expected_results['data'][i + offset],
particles[i])
stream_handle.close()
def test_extra_data(self):
"""
Ensure that bad data is skipped when it exists.
"""
# the first 2 data record in this file are corrupted by adding additional
# data vlaues separated by tabs and will be ignored
# we expect the first 2 particles to be the metadata particle and the
# intrument particle from the data record after the corrupted one
file_path = os.path.join(RESOURCE_PATH,
'11079364_EXTRA_DATA_PPD_OCR.txt')
stream_handle = open(file_path, 'rb')
log.info(self.exception_callback_value)
parser = SpkirAbjCsppParser(
self.config.get(DataTypeKey.SPKIR_ABJ_CSPP_RECOVERED), None,
stream_handle, self.state_callback, self.pub_callback,
self.exception_callback)
particles = parser.get_records(2)
self.assertTrue(self.exception_callback_value != None)
self.assert_(
isinstance(self.exception_callback_value,
RecoverableSampleException))
# expect to see a recoverable sample exception in the log
log.debug('TEST EXTRA DATA exception call back is %s',
self.exception_callback_value)
expected_results = self.get_dict_from_yml('extra_data_values.yml')
self.assertTrue(len(particles) == 2)
# since the first two records were corrupted the first records recieved
# should be metadata particle with the timestamp of the 3rd data row
# and the insturment particle from the 3rd row
for i in range(len(particles)):
self.assert_result(expected_results['data'][i], particles[i])
stream_handle.close()
def create_yml(self):
"""
This utility creates a yml file
"""
#ADCP_data_20130702.PD0 has one record in it
fid = open(os.path.join(RESOURCE_PATH, '11079419_PPB_OCR.txt'), 'r')
stream_handle = fid
parser = SpkirAbjCsppParser(
self.config.get(DataTypeKey.SPKIR_ABJ_CSPP_RECOVERED), None,
stream_handle, self.state_callback, self.pub_callback,
self.exception_callback)
particles = parser.get_records(20)
self.particle_to_yml(particles, '11079419_PPB_OCR_recov.yml')
fid.close()
def test_mid_state_start(self):
"""
This test makes sure that we retrieve the correct particles upon starting with an offset state.
"""
file_path = os.path.join(RESOURCE_PATH, '11079419_PPB_OCR.txt')
stream_handle = open(file_path, 'rb')
# position 1410 is the end of the frist data record, which would have produced the
# metadata particle and the first instrument particle
initial_state = {
StateKey.POSITION: 1410,
StateKey.METADATA_EXTRACTED: True
}
parser = SpkirAbjCsppParser(
self.config.get(DataTypeKey.SPKIR_ABJ_CSPP_RECOVERED),
initial_state, stream_handle, self.state_callback,
self.pub_callback, self.exception_callback)
#expect to get the 2nd and 3rd instrument particles next
particles = parser.get_records(2)
log.debug("Num particles: %s", len(particles))
self.assertTrue(len(particles) == 2)
expected_results = self.get_dict_from_yml('mid_state_start.yml')
for i in range(len(particles)):
self.assert_result(expected_results['data'][i], particles[i])
# now expect the state to be the end of the 4 data record and metadata sent
the_new_state = {
StateKey.POSITION: 1704,
StateKey.METADATA_EXTRACTED: True
}
log.debug("********** expected state: %s", the_new_state)
log.debug("******** new parser state: %s", parser._state)
self.assertTrue(parser._state == the_new_state)
stream_handle.close()