def test_get_many(self):
"""
This test exercises retrieving 20 particles, verifying the 20th particle, then retrieves 30 particles
and verifies the 30th particle.
"""
file_path = os.path.join(RESOURCE_PATH, 'ctd_1_20131124T005004_458.mpk')
stream_handle = open(file_path, 'rb')
state = {StateKey.PARTICLES_RETURNED: 0}
parser = CtdpfCklMmpCdsParser(self.config, state, stream_handle,
self.state_callback, self.pub_callback)
particles = parser.get_records(20)
# Should end up with 20 particles
self.assertTrue(len(particles) == 20)
test_data = self.get_dict_from_yml('get_many_one.yml')
self.assert_result(test_data['data'][0], particles[19])
particles = parser.get_records(30)
# Should end up with 30 particles
self.assertTrue(len(particles) == 30)
test_data = self.get_dict_from_yml('get_many_two.yml')
self.assert_result(test_data['data'][0], particles[29])
stream_handle.close()
def test_mid_state_start(self):
"""
This test exercises setting the state past one chunk, retrieving particles and verify the result of one
of the particles.
"""
# Using two concatenated msgpack files to simulate two chunks.
file_path = os.path.join(RESOURCE_PATH, 'set_state.mpk')
stream_handle = open(file_path, 'rb')
stat_info = os.stat(file_path)
# Moving the file position to the end of the first chunk
state = {StateKey.PARTICLES_RETURNED: 20}
parser = CtdpfCklMmpCdsParser(self.config, state, stream_handle,
self.state_callback, self.pub_callback)
particles = parser.get_records(4)
log.info(len(particles))
# Should end up with 4 particles
self.assertTrue(len(particles) == 4)
test_data = self.get_dict_from_yml('set_state.yml')
self.assert_result(test_data['data'][23], particles[3])
stream_handle.close()
def test_bad_data_two(self):
"""
This test verifies that a SampleException is raised when an entire msgpack buffer is not msgpack.
"""
file_path = os.path.join(RESOURCE_PATH, 'not-msg-pack.mpk')
stream_handle = open(file_path, 'rb')
state = {StateKey.PARTICLES_RETURNED: 0}
parser = CtdpfCklMmpCdsParser(self.config, state, stream_handle,
self.state_callback, self.pub_callback)
with self.assertRaises(SampleException):
parser.get_records(1)
stream_handle.close()
def test_bad_data_one(self):
"""
This test verifies that a SampleException is raised when msgpack data is malformed.
"""
file_path = os.path.join(RESOURCE_PATH, 'ctd_1_20131124T005004_BAD.mpk')
stream_handle = open(file_path, 'rb')
state = {StateKey.PARTICLES_RETURNED: 0}
parser = CtdpfCklMmpCdsParser(self.config, state, stream_handle,
self.state_callback, self.pub_callback)
with self.assertRaises(SampleException):
parser.get_records(1)
stream_handle.close()
def test_long_stream(self):
"""
This test exercises retrieve approximately 200 particles.
"""
# Using two concatenated msgpack files to simulate two chunks to get more particles.
file_path = os.path.join(RESOURCE_PATH, 'ctd_concat.mpk')
stream_handle = open(file_path, 'rb')
state = {StateKey.PARTICLES_RETURNED: 0}
parser = CtdpfCklMmpCdsParser(self.config, state, stream_handle,
self.state_callback, self.pub_callback)
# Attempt to retrieve 200 particles, but we will retrieve less
particles = parser.get_records(200)
# Should end up with 172 particles
self.assertEqual(len(particles), 172)
stream_handle.close()
def test_simple(self):
"""
This test reads in a small number of particles and verifies the result of one of the particles.
"""
file_path = os.path.join(RESOURCE_PATH, 'ctd_1_20131124T005004_458.mpk')
stream_handle = open(file_path, 'rb')
state = {StateKey.PARTICLES_RETURNED: 0}
parser = CtdpfCklMmpCdsParser(self.config, state, stream_handle,
self.state_callback, self.pub_callback)
particles = parser.get_records(6)
for particle in particles:
print particle.generate_dict()
test_data = self.get_dict_from_yml('good.yml')
self.assert_result(test_data['data'][0], particles[5])
stream_handle.close()
def _build_parser(self, stream_handle):
parser_config = {
DataSetDriverConfigKeys.PARTICLE_MODULE:
'mi.dataset.parser.ctdpf_ckl_mmp_cds',
DataSetDriverConfigKeys.PARTICLE_CLASS:
'CtdpfCklMmpCdsParserDataParticle'
}
parser = CtdpfCklMmpCdsParser(parser_config, None, stream_handle,
lambda state, ingested: None,
lambda data: None)
return parser
def _build_parser(self, parser_state, infile):
"""
Build and return the parser
"""
config = self._parser_config
config.update({
DataSetDriverConfigKeys.PARTICLE_MODULE: 'mi.dataset.parser.ctdpf_ckl_mmp_cds',
DataSetDriverConfigKeys.PARTICLE_CLASS: 'CtdpfCklMmpCdsParserDataParticle'
})
log.debug("My Config: %s", config)
self._parser = CtdpfCklMmpCdsParser(
config,
parser_state,
infile,
self._save_parser_state,
self._data_callback,
self._sample_exception_callback
)
return self._parser
def test_set_state(self):
"""
This test exercises setting the state past one chunk, retrieving particles, verifying one
of the particles, and then setting the state back to the beginning, retrieving a few particles, and
verifying one of the particles.
"""
# Using the default mspack test file.
file_path = os.path.join(RESOURCE_PATH, 'set_state.mpk')
stream_handle = open(file_path, 'rb')
# Moving the file position to the beginning
state = {StateKey.PARTICLES_RETURNED: 0}
parser = CtdpfCklMmpCdsParser(self.config, state, stream_handle,
self.state_callback, self.pub_callback)
particles = parser.get_records(4)
# Should end up with 4 particles
self.assertTrue(len(particles) == 4)
log.info(parser._state)
stat_info = os.stat(file_path)
test_data = self.get_dict_from_yml('set_state.yml')
self.assert_result(test_data['data'][3], particles[3])
state = copy.copy(parser._state)
log.info(state)
parser = CtdpfCklMmpCdsParser(self.config, state, stream_handle,
self.state_callback, self.pub_callback)
particles = parser.get_records(4)
# Should end up with 4 particles
self.assertTrue(len(particles) == 4)
self.assert_result(test_data['data'][7], particles[3])
# Give a bad position which will be ignored
state = {StateKey.PARTICLES_RETURNED: 0}
parser = CtdpfCklMmpCdsParser(self.config, state, stream_handle,
self.state_callback, self.pub_callback)
particles = parser.get_records(1)
self.assertTrue(len(particles) == 1)
# Give a bad position which will be ignored
state = {StateKey.PARTICLES_RETURNED: 0}
parser = CtdpfCklMmpCdsParser(self.config, state, stream_handle,
self.state_callback, self.pub_callback)
particles = parser.get_records(1000)
self.assertTrue(len(particles) == 30)
self.assert_result(test_data['data'][29], particles[29])
# Provide a bad particles returned
state = {StateKey.PARTICLES_RETURNED: 80}
parser = CtdpfCklMmpCdsParser(self.config, state, stream_handle,
self.state_callback, self.pub_callback)
particles = parser.get_records(1)
self.assertTrue(len(particles) == 0)
stream_handle.close()
def parse(basePythonCodePath, sourceFilePath, particleDataHdlrObj):
try:
if basePythonCodePath is not None:
pass
except NameError:
basePythonCodePath = os.curdir
sys.path.append(basePythonCodePath)
from mi.logging import config
config.add_configuration(os.path.join(basePythonCodePath, 'res', 'config', 'mi-logging.yml'))
from mi.core.log import get_logger
log = get_logger()
from mi.dataset.dataset_driver import DataSetDriver, ParticleDataHandler
from mi.dataset.parser.ctdpf_ckl_mmp_cds import CtdpfCklMmpCdsParser
from mi.dataset.dataset_parser import DataSetDriverConfigKeys
config = {
DataSetDriverConfigKeys.PARTICLE_MODULE: 'mi.dataset.parser.ctdpf_ckl_mmp_cds',
DataSetDriverConfigKeys.PARTICLE_CLASS: 'CtdpfCklMmpCdsParserDataParticle'
}
try:
if particleDataHdlrObj is not None:
pass
except NameError:
particleDataHdlrObj = ParticleDataHandler()
try:
if sourceFilePath is not None:
pass
except NameError:
try:
sourceFilePath = sys.argv[1]
except IndexError:
print "Need a source file path"
sys.exit(1)
def state_callback(state, ingested):
pass
def pub_callback(data):
log.trace("Found data: %s", data)
stream_handle = open(sourceFilePath, 'rb')
try:
parser = CtdpfCklMmpCdsParser(config, None, stream_handle,
state_callback, pub_callback)
driver = DataSetDriver(parser, particleDataHdlrObj)
driver.processFileStream()
finally:
stream_handle.close()
return particleDataHdlrObj