def testMultiDatasetExtraction(self):
"""This test demonstrates use case #402.
There are 2 datasets created, the first holds a datafile, and has a pull ingest occurring, along with
a configured custom script. The second dataset holds observation data, that will be extracted from the
datafile in the first dataset.
"""
temperature_schema = DataEntrySchema()
temperature_schema.addAttr(Double("Temperature"))
temperature_schema = self.ingester_platform.post(temperature_schema)
file_schema = DataEntrySchema()
file_schema.addAttr(FileDataType("file"))
file_schema = self.ingester_platform.post(file_schema)
location = self.ingester_platform.post(
Location(10.0, 11.0, "Test Site", 100))
temp_dataset = Dataset(location=None, schema=temperature_schema.id)
file_dataset = Dataset(
location=None,
schema=file_schema.id,
data_source=PullDataSource(
"http://test.com",
"file_handle",
processing_script=
"file://d:/processing_scripts/awsome_processing.py"))
def setup_datasets(self):
ds = Dataset()
ds.data_source = _DataSource()
ds.id = 1
ds.data_source.__xmlrpc_class__ = "mock_data_source"
ds.data_source.sampling = PeriodicSampling(10000)
self.datasets.append(ds)
def testPush(self):
"""This tests the push ingest by creating a test dir, populating it, then forcing the ingester to run
"""
staging = tempfile.mkdtemp()
self.todelete.append(staging)
# Create a temp file to ingest
f_name = str(int(time.time()))
f = open(os.path.join(staging, f_name), "a")
f.close()
# Check there is only 1 file here
self.assertEquals(1, len(os.listdir(staging)))
dataset = Dataset()
dataset.id = 1
dataset.data_source = PushDataSource(path=staging,
pattern='^(?P<timestamp>[0-9]+)$')
self.ingester.enqueue_ingress(dataset)
self.ingester.process_ingress_queue(True)
self.assertEquals(1, self.ingester._archive_queue.qsize())
# Check there are now no files
self.assertEquals(0, len(os.listdir(staging)))
def testPostProcessScript(self):
"""This test performs a complex data ingest, where the main data goes into dataset 1 and
the extracted data goes into dataset 2"""
script = """import os
import datetime
from jcudc24ingesterapi.models.data_entry import DataEntry, FileObject
def process(cwd, data_entry):
data_entry = data_entry[0]
ret = []
with open(os.path.join(cwd, data_entry["file1"].f_path)) as f:
for l in f.readlines():
l = l.strip().split(",")
if len(l) != 2: continue
new_data_entry = DataEntry(timestamp=datetime.datetime.now())
new_data_entry["a"] = FileObject(f_path=l[1].strip())
ret.append( new_data_entry )
return ret
"""
# Capture the ingests
data_entries = DataEntryListener()
self.ingester.service.register_observation_listener(data_entries)
dataset = Dataset(dataset_id=1)
dataset.data_source = _DataSource(processing_script=script)
dataset.data_source.__xmlrpc_class__ = "csv1"
self.ingester.enqueue_ingress(dataset)
self.ingester.process_ingress_queue(True)
self.assertEquals(0, self.ingester._ingress_queue.qsize())
self.assertEquals(1, self.ingester._archive_queue.qsize())
self.ingester.process_archive_queue(True)
self.assertEquals(0, self.ingester._archive_queue.qsize())
self.assertEquals(2, data_entries.count())
def test_dataset_roundtrip(self):
"""Attempt to round trip a dataset object"""
script_contents = """Some Script
More"""
dataset = Dataset(location=1,
schema=2,
data_source=PullDataSource(
"http://www.bom.gov.au/radar/IDR733.gif",
"file",
processing_script=script_contents),
location_offset=LocationOffset(0, 1, 2))
dataset_dict = self.marshaller.obj_to_dict(dataset)
dataset1 = self.marshaller.dict_to_obj(dataset_dict)
self.assertIsNotNone(dataset1, "Dataset should not be none")
self.assertEquals(dataset1.location, dataset.location,
"Location ID does not match")
self.assertEquals(
dataset1.schema, dataset.schema,
"schema does not match %d!=%d" % (dataset1.schema, dataset.schema))
self.assertEquals(dataset1.location_offset.x, 0)
self.assertEquals(dataset1.location_offset.y, 1)
self.assertEquals(dataset1.location_offset.z, 2)
def test_unit_of_work_persistence(self):
unit = self.ingester_platform.createUnitOfWork()
loc = Location(10.0, 11.0, "Test Site", 100, None)
unit.insert(loc)
self.assertIsNotNone(loc.id)
file_schema = DataEntrySchema()
file_schema.name = "File Schema"
file_schema.addAttr(FileDataType("file"))
file_schema_id = unit.insert(file_schema)
self.assertIsNotNone(file_schema_id, "Schema ID should not be null")
dataset = Dataset(location=loc.id,
schema=file_schema.id,
data_source=PullDataSource(
"http://www.bom.gov.au/radar/IDR733.gif",
"file"))
unit.insert(dataset)
# Persist all the objects
unit.commit()
self.assertIsNotNone(loc, "Location should not be none")
self.assertIsNotNone(loc.id, "Location should not be none")
self.assertGreater(loc.id, 0, "Location ID not real")
self.assertEqual(loc.name, "Test Site", "Location name doesn't match")
self.assertIsNotNone(dataset, "dataset should not be none")
self.assertIsNotNone(dataset.id, "dataset should not be none")
self.assertGreater(dataset.id, 0, "dataset ID not real")
def test_dataset_data_source_unit(self):
"""This test creates a simple schema hierarchy, and tests updates, etc"""
unit = UnitOfWork(None)
schema1 = DataEntrySchema("base1")
schema1.addAttr(FileDataType("file"))
schema_id = unit.post(schema1)
loc = Location(10.0, 11.0)
loc.name = "Location"
loc_id = unit.post(loc)
dataset1 = Dataset()
dataset1.schema = schema_id
dataset1.location = loc_id
dataset1_id = unit.post(dataset1)
dataset2 = Dataset()
dataset2.schema = schema_id
dataset2.location = loc_id
dataset2.data_source = DatasetDataSource(dataset1_id, "")
dataset2_id = unit.post(dataset2)
ret = self.service.commit(unit, None)
found = False
for r in ret:
if isinstance(r, Dataset) and dataset1_id == r.correlationid:
dataset1_id = r.id
elif isinstance(r, Dataset) and dataset2_id == r.correlationid:
self.assertEquals(dataset1_id, r.data_source.dataset_id,
"Data source dataset_id was not updated")
found = True
self.assertTrue(
found, "Didn't find the dataset with the dataset data source")
def test_dataset_persistence(self):
loc = Location(10.0, 11.0, "Test Site", 100, None)
loc = self.ingester_platform.post(loc)
self.assertIsNotNone(loc, "Location should not be none")
self.assertIsNotNone(loc.id, "Location should not be none")
file_schema = DataEntrySchema()
file_schema.addAttr(FileDataType("file"))
file_schema = self.ingester_platform.post(file_schema)
script_contents = """Some Script
More"""
dataset = Dataset(location=loc.id,
schema=file_schema.id,
data_source=PullDataSource(
"http://www.bom.gov.au/radar/IDR733.gif",
"file",
processing_script=script_contents),
location_offset=LocationOffset(0, 1, 2))
dataset1 = self.ingester_platform.post(dataset)
self.assertIsNotNone(dataset1, "Dataset should not be none")
self.assertEquals(dataset1.location, dataset.location,
"Location ID does not match")
self.assertEquals(
dataset1.schema, dataset.schema,
"schema does not match %d!=%d" % (dataset1.schema, dataset.schema))
self.assertEquals(dataset1.location_offset.x, 0)
self.assertEquals(dataset1.location_offset.y, 1)
self.assertEquals(dataset1.location_offset.z, 2)
self.assertEquals(script_contents,
dataset1.data_source.processing_script)
datasets = self.ingester_platform.findDatasets()
self.assertEquals(1, len(datasets))
datasets = self.ingester_platform.findDatasets(location=loc.id)
self.assertEquals(1, len(datasets))
data_entry_schemas = self.ingester_platform.search(
DataEntrySchemaSearchCriteria(), 0, 10).results
self.assertEquals(1, len(data_entry_schemas))
datasets = self.ingester_platform.search(DatasetSearchCriteria(), 0,
10).results
self.assertEquals(1, len(datasets))
def test_parent_schemas(self):
"""This test creates a nested schema with attributes provided at 2
different levels. A data entry is saved, and then retrieved, and the
values tested.
"""
loc1 = self.ingester_platform.post(
Location(11.0, 11.0, "Test Site", 100))
temp_work = self.ingester_platform.createUnitOfWork()
temperature_schema = DataEntrySchema("Test Temp Schema")
temperature_schema.addAttr(Double("temperature"))
temp_work.post(temperature_schema)
temp_work.commit()
air_temperature_schema = DataEntrySchema("Air Temp Schema")
air_temperature_schema.extends = [temperature_schema.id]
air_temperature_schema = self.ingester_platform.post(
air_temperature_schema)
instrument_schema = DataEntrySchema("Instrument Schema")
instrument_schema.extends = [air_temperature_schema.id]
instrument_schema.addAttr(Double("var2"))
instrument_schema = self.ingester_platform.post(instrument_schema)
dataset = Dataset(location=loc1.id, schema=instrument_schema.id)
dataset = self.ingester_platform.post(dataset)
work = self.ingester_platform.createUnitOfWork()
data_entry = DataEntry(dataset.id, datetime.datetime.now())
data_entry["temperature"] = 10
data_entry["var2"] = 11
work.post(data_entry)
work.commit()
data_entry_ret = self.ingester_platform.getDataEntry(
dataset.id, data_entry.id)
self.assertEquals(data_entry["temperature"],
data_entry_ret["temperature"])
self.assertEquals(data_entry["var2"], data_entry_ret["var2"])
def testBasicIngest(self):
"""This test performs a simple data ingest"""
# Capture the ingests
data_entries = DataEntryListener()
self.ingester.service.register_observation_listener(data_entries)
dataset = Dataset()
dataset.id = 1
datasource = _DataSource()
datasource.__xmlrpc_class__ = "csv1"
dataset.data_source = datasource
self.ingester.enqueue_ingress(dataset)
self.assertEquals(1, self.ingester._ingress_queue.qsize())
self.ingester.process_ingress_queue(True)
self.assertEquals(1, self.ingester._archive_queue.qsize())
self.ingester.process_archive_queue(True)
self.assertEquals(1, data_entries.count())
def test_dataset_persist(self):
schema = DataEntrySchema("base1")
schema.addAttr(FileDataType("file"))
schema = self.service.persist(schema)
loc = Location(10.0, 11.0)
loc.name = "Location"
loc = self.service.persist(loc)
dataset = Dataset()
dataset.schema = schema.id
dataset.location = loc.id
dataset1 = self.service.persist(dataset)
self.assertEquals(1, dataset1.version)
dataset1.version = 0
self.assertRaises(StaleObjectError, self.service.persist, dataset1)
dataset1.version = 1
dataset2 = self.service.persist(dataset1)
self.assertEquals(2, dataset2.version)
def test_pull_ingest_functionality(self):
loc = Location(10.0, 11.0, "Test Site", 100, None)
loc = self.ingester_platform.post(loc)
file_schema = DataEntrySchema()
file_schema.addAttr(FileDataType("file"))
file_schema = self.ingester_platform.post(file_schema)
dataset = Dataset(location=loc.id,
schema=file_schema.id,
data_source=PullDataSource(
"http://www.bom.gov.au/radar/IDR733.gif",
"file",
sampling=PeriodicSampling(10000)))
dataset1 = self.ingester_platform.post(dataset)
self.assertEquals(dataset1.location, dataset.location,
"Location ID does not match")
self.assertEquals(dataset1.schema, dataset.schema,
"schema does not match")
self.ingester_platform.disableDataset(dataset1.id)
dataset1a = self.ingester_platform.getDataset(dataset1.id)
self.assertEquals(dataset1a.enabled, False)
def test_dataset(self):
# Basic instanciation
ds = Dataset()
ds = Dataset(data_source=PullDataSource())
self.assertRaises(TypeError, Dataset, data_source=1)
def test_data_types(self):
schema1 = DatasetMetadataSchema("schema1")
schema1.addAttr(FileDataType("file"))
schema1a = self.service.persist(schema1)
self.assertEquals(1, len(schema1a.attrs))
schema2 = DataEntrySchema("schema2")
schema2.addAttr(FileDataType("file"))
schema2.addAttr(Double("x"))
schema2a = self.service.persist(schema2)
loc = Location(10.0, 11.0)
loca = self.service.persist(loc)
dataset = Dataset()
dataset.schema = schema1a.id
dataset.location = loca.id
# We've trying to use a dataset_metadata schema, so this should fail
self.assertRaises(ValueError, self.service.persist, dataset)
dataset.schema = schema2a.id
# Now we're using the correct type of schema
dataset1a = self.service.persist(dataset)
dataset1b = self.service.get_dataset(dataset1a.id)
self.assertEquals(dataset1a.id, dataset1b.id)
self.assertDictEqual(dataset1a.__dict__, dataset1b.__dict__)
# Update and add a data source
dataset1b.data_source = PullDataSource(
"http://www.abc.net.au",
None,
recursive=False,
field="file",
processing_script="TEST",
sampling=PeriodicSampling(10000))
dataset1b.enabled = True
dataset1c = self.service.persist(dataset1b)
self.assertNotEqual(None, dataset1c.data_source)
self.assertEqual("TEST", dataset1c.data_source.processing_script)
self.assertNotEqual(None, dataset1c.data_source.sampling)
datasets = self.service.get_active_datasets()
self.assertEquals(1, len(datasets))
self.assertNotEqual(None, datasets[0].data_source)
self.assertEqual("TEST", datasets[0].data_source.processing_script)
self.assertNotEqual(None, datasets[0].data_source.sampling)
# Test with criteria
datasets = self.service.get_active_datasets(kind="pull_data_source")
self.assertEquals(1, len(datasets))
datasets = self.service.get_active_datasets(kind="push_data_source")
self.assertEquals(0, len(datasets))
schema1b = self.service.get_schema(schema1a.id)
self.assertEquals(schema1a.id, schema1b.id)
datasets = self.service.search("dataset")
self.assertEquals(1, len(datasets))
schemas = self.service.search("data_entry_schema")
self.assertEquals(1, len(schemas))
schemas = self.service.search("dataset_metadata_schema")
self.assertEquals(1, len(schemas))
locs = self.service.search("location")
self.assertEquals(1, len(locs))
# Test ingest
data_entry_1 = DataEntry(dataset1b.id, datetime.datetime.now())
data_entry_1['x'] = 27.8
data_entry_1 = self.service.persist(data_entry_1)
self.assertIsNotNone(data_entry_1.id)
file_schema.addAttr(FileDataType("file"))
file_schema = ingester_platform.post(file_schema)
temp_schema = DataEntrySchema("temperature_reading")
temp_schema.addAttr(Double("temp"))
temp_schema = ingester_platform.post(temp_schema)
# Setup the location
loc = Location(-19.34427, 146.784197, "Mt Stuart", 100, None)
loc = ingester_platform.post(loc)
# Create the dataset to store the data in
file_dataset = Dataset(location=loc.id,
schema=file_schema.id,
data_source=PullDataSource(
url=tree_url,
field="file",
recursive=True,
sampling=PeriodicSampling(10000)))
file_dataset.enabled = False
file_dataset = ingester_platform.post(file_dataset)
# This processing script is
processing_script = """
import os
import datetime
from dc24_ingester_platform.utils import *
def process(cwd, data_entries):
ret = []
started = False
def test_api_usage(self):
# User data that is created by filling out the provisioning interface workflow steps.
# General
title = "Test project"
data_manager = "A Person"
project_lead = "Another Person"
# Metadata
project_region = Region("Test Region",
((1, 1), (2, 2), (2, 1), (1, 1)))
# Methods & Datasets
loc1 = Location(11.0, 11.0, "Test Site", 100)
loc2 = Location(11.0, 11.0, "Test Site", 100)
loc3 = Location(12.0, 11.0, "Test Site", 100)
temp_work = self.ingester_platform.createUnitOfWork()
temperature_schema = DataEntrySchema("Test Temp Schema")
temperature_schema.addAttr(Double("temperature"))
temp_work.post(temperature_schema)
temp_work.commit()
air_temperature_schema = DataEntrySchema("Air Temp Schema")
air_temperature_schema.extends = [temperature_schema.id]
air_temperature_schema = self.ingester_platform.post(
air_temperature_schema)
second_level_inheritence_schema = DataEntrySchema("Second Inheritence")
second_level_inheritence_schema.extends = [air_temperature_schema.id]
second_level_inheritence_schema = self.ingester_platform.post(
second_level_inheritence_schema)
# Check the name is set
temperature_schema_1 = self.ingester_platform.getSchema(
temperature_schema.id)
self.assertIsNotNone(temperature_schema.name)
self.assertEquals(temperature_schema.name, temperature_schema_1.name)
file_schema = DataEntrySchema()
file_schema.addAttr(FileDataType("file"))
file_schema = self.ingester_platform.post(file_schema)
dataset1 = Dataset(location=None, schema=temperature_schema.id)
dataset2 = Dataset(
location=None,
schema=file_schema.id,
data_source=PullDataSource(
"http://test.com",
"file_handle",
processing_script=
"file://d:/processing_scripts/awsome_processing.py"))
# dataset3 = Dataset(None, file_schema, PullDataSource("http://test.com", "file_handle"), CustomSampling("file://d:/sampling_scripts/awsome_sampling.py"), "file://d:/processing_scripts/awsome_processing.py")
self.cleanup_files.append(dataset2.data_source.processing_script)
# self.cleanup_files.push(dataset3.sampling.script)
# self.cleanup_files.push(dataset3.processing_script)
# Provisioning admin accepts the submitted project
work = self.ingester_platform.createUnitOfWork()
work.post(project_region) # Save the region
loc1.region = project_region.id # Set the datasets location to use the projects region
work.post(loc1) # Save the location
dataset1.location = loc1.id # Set the datasets location
work.post(dataset1) # Save the dataset
loc2.region = project_region.id
work.post(loc2)
dataset2.location = loc2.id
work.post(dataset2)
work.commit()
# Region, location and dataset id's will be saved to the project within the provisioning system in some way
# User searches for datasets
# TODO: Nigel? - Define searching api
found_dataset_id = dataset1.id # The dataset that has an extended file schema
# User manually enters data
timestamp = datetime.datetime.now()
data_entry_1 = DataEntry(found_dataset_id, timestamp)
data_entry_1['temperature'] = 27.8 # Add the extended schema items
data_entry_1 = self.ingester_platform.post(data_entry_1)
self.assertIsNotNone(data_entry_1.id)
timestamp2 = timestamp + datetime.timedelta(seconds=1)
data_entry_2 = DataEntry(found_dataset_id, timestamp2)
data_entry_2['temperature'] = 27.8 # Add the extended schema items
data_entry_2 = self.ingester_platform.post(data_entry_2)
self.assertEquals(
2,
len(
self.ingester_platform.search(
DataEntrySearchCriteria(found_dataset_id), 0, 10).results))
result = self.ingester_platform.search(
DataEntrySearchCriteria(found_dataset_id), 0, 1)
self.assertEquals(2, result.count)
self.assertEquals(1, len(result.results))
self.assertEquals(
1,
len(
self.ingester_platform.search(
DataEntrySearchCriteria(found_dataset_id), 1, 1).results))
result = self.ingester_platform.search(
DataEntrySearchCriteria(found_dataset_id), 2, 1)
self.assertEquals(0, len(result.results))
self.assertEquals(
0,
len(
self.ingester_platform.search(
DataEntrySearchCriteria(found_dataset_id,
end_time=timestamp -
datetime.timedelta(seconds=60)), 0,
10).results))
self.assertEquals(
0,
len(
self.ingester_platform.search(
DataEntrySearchCriteria(found_dataset_id,
start_time=timestamp +
datetime.timedelta(seconds=60)), 0,
10).results))
self.assertEquals(
2,
len(
self.ingester_platform.search(
DataEntrySearchCriteria(
found_dataset_id,
start_time=timestamp - datetime.timedelta(seconds=60),
end_time=timestamp + datetime.timedelta(seconds=60)),
0, 10).results))
work = self.ingester_platform.createUnitOfWork()
data_entry_3 = DataEntry(dataset2.id, datetime.datetime.now())
data_entry_3['file'] = FileObject(f_handle=open(
os.path.join(os.path.dirname(jcudc24ingesterapi.__file__),
"tests/test_ingest.xml")),
mime_type="text/xml")
work.post(data_entry_3)
work.commit()
self.assertIsNotNone(data_entry_3.id)
f_in = self.ingester_platform.getDataEntryStream(
dataset2.id, data_entry_3.id, "file")
self.assertIsNotNone(f_in)
data = f_in.read()
f_in.close()
self.assertLess(0, len(data), "Expected data in file")
# User enters quality assurance metadata
quality_metadata_schema = DatasetMetadataSchema()
quality_metadata_schema.addAttr(String("unit"))
quality_metadata_schema.addAttr(String("description"))
quality_metadata_schema.addAttr(Double("value"))
quality_metadata_schema = self.ingester_platform.post(
quality_metadata_schema)
entered_metadata = DatasetMetadataEntry(data_entry_1.dataset,
quality_metadata_schema.id)
entered_metadata['unit'] = "%"
entered_metadata['description'] = "Percent error"
entered_metadata['value'] = 0.98
entered_metadata = self.ingester_platform.post(entered_metadata)
# Now find that metadata
results = self.ingester_platform.search(
DatasetMetadataSearchCriteria(data_entry_1.dataset), 0, 10).results
self.assertEqual(1, len(results))
data_entry_md_schema = DataEntryMetadataSchema("test")
data_entry_md_schema.addAttr(String("description"))
data_entry_md_schema.addAttr(Double("value"))
data_entry_md_schema = self.ingester_platform.post(
data_entry_md_schema)
calibration = DataEntryMetadataEntry(metadata_schema_id=int(
data_entry_md_schema.id),
dataset_id=dataset2.id,
object_id=data_entry_3.id)
calibration["description"] = "Test"
calibration["value"] = 1.2
calibration2 = DataEntryMetadataEntry(metadata_schema_id=int(
data_entry_md_schema.id),
dataset_id=dataset2.id,
object_id=data_entry_3.id)
calibration2["description"] = "Test2"
calibration2["value"] = 2.3
calibration2 = self.ingester_platform.post(calibration2)
calibrations = self.ingester_platform.search(
DataEntryMetadataSearchCriteria(int(81), int(3648)),
offset=0,
limit=1000)
self.assertEquals(1, len(calibrations.results))
self.assertEquals(calibrations.results[0].schema_id,
data_entry_md_schema.id)
self.ingester_platform.delete(calibration2)
self.ingester_platform.delete(calibration)
self.ingester_platform.delete(data_entry_md_schema)
# User changes sampling rate
# FIXME: This test is going to be changed to be done by editing the dataset
# sampling_rate_changed = Metadata(dataset1.id, type(dataset1), SampleRateMetadataSchema())
# sampling_rate_changed.change_time = datetime.datetime.now().strftime("%Y-%m-%d %H:%M")
# sampling_rate_changed.sampling = CustomSampling("file://d:/sampling_scripts/awsome_sampling.py")
#
# try:
# sampling_rate_changed = self.ingester_platform.post(sampling_rate_changed)
# assert(sampling_rate_changed.metadata_id is None, "Sampling rate change failed")
# except:
# assert(True, "Sampling rate change failed")
# User wants some random metadata specific to their project
# FIXME: Not sure what use case this is trying to demonstrate
# random_metadata_schema = DataEntryMetadataSchema()
# random_metadata_schema.addAttr('random_field', Double())
# random_metadata = Metadata(data_entry.data_entry_id, type(data_entry), random_metadata_schema)
# random_metadata.random_field = 1.5
# try:
# random_metadata = self.ingester_platform.post(random_metadata)
# assert(random_metadata.metadata_id is None, "random_metadata failed")
# except:
# assert(True, "random_metadata failed")
# User changes the data source of the dataset
new_data_source = PullDataSource("http://test.com/new_data",
"file_handle")
dataset1.data_source = new_data_source
dataset1 = self.ingester_platform.post(dataset1)
self.assertNotEqual(None, dataset1)
# External, 3rd party searches for data
# TODO: external 3rd parties should be able to use the api to get data without authentication
# TODO: I'm not sure exactly how this should work, but the search api could be open access (need spam limitations or something?)
# Project is disabled/finished
work = self.ingester_platform.createUnitOfWork()
work.disable(dataset1.id)
work.disable(dataset2.id)
work.commit()
# Project is obsolete and data should be deleted
work = self.ingester_platform.createUnitOfWork()
work.delete(dataset1.id)
work.delete(dataset2.id)
work.commit()