def testRightVersion(self):
repo = datarepo.SqlDataRepository(self._repoPath)
repo.open(datarepo.MODE_WRITE)
repo.initialise()
anotherRepo = datarepo.SqlDataRepository(self._repoPath)
anotherRepo.open(datarepo.MODE_READ)
self.assertEquals(anotherRepo._schemaVersion, str(repo.version))
def testWrongVersion(self):
repo = datarepo.SqlDataRepository(self._repoPath)
repo.version = datarepo.SqlDataRepository.SchemaVersion(
"wrong.version")
repo.open(datarepo.MODE_WRITE)
repo.initialise()
anotherRepo = datarepo.SqlDataRepository(self._repoPath)
with self.assertRaises(exceptions.RepoSchemaVersionMismatchException):
anotherRepo.open(datarepo.MODE_READ)
def __init__(self, inputDirectory, outputDirectory):
"""
Converts human readable dataset from compliance repository,
and translates it into a reference-server readable filesystem
with binary files.
:param inputDirectory: location of
the human readable compliance dataset
:param outputDirectory: location of
the file hierarchy suitable for deploying on the reference server
"""
self.inputDirectory = inputDirectory
self.outputDirectory = outputDirectory
self.repoPath = os.path.join(outputDirectory, "repo.db")
self.tempdir = None
# If no input directory is specified download from GitHub
if inputDirectory is None:
utils.log("Downloading test data...")
self.tempdir = tempfile.mkdtemp()
assert(os.path.exists(self.tempdir))
url = "https://github.com/ga4gh/compliance/archive/master.zip"
filePath = os.path.join(self.tempdir, 'compliance-master.zip')
downloader = utils.HttpFileDownloader(url, filePath)
downloader.download()
utils.log("Extracting test data...")
with zipfile.ZipFile(filePath, "r") as z:
z.extractall(self.tempdir)
self.inputDirectory = os.path.join(
self.tempdir, 'compliance-master', 'test-data')
repo = datarepo.SqlDataRepository(self.repoPath)
self.repo = repo
def setUp(self):
self._maxDiff = None
repoPath = paths.testDataRepo
self._dataUrl = "file://{}".format(repoPath)
dataRepository = datarepo.SqlDataRepository(repoPath)
dataRepository.open(datarepo.MODE_READ)
self._backend = backend.Backend(dataRepository)
self._client = client.LocalClient(self._backend)
def __init__(self, featureSetLocalName, dataPath):
"""
:param localId: Name of the GFF3 resource corresponding to a pair
of files, .db and .gff3
:param dataPath: string representing full path to the .db file
:return:
"""
self._dataset = datasets.Dataset(_datasetName)
self._repo = datarepo.SqlDataRepository(paths.testDataRepo)
self._repo.open(datarepo.MODE_READ)
self._ontology = self._repo.getOntologyByName(paths.ontologyName)
self._referenceSet = references.AbstractReferenceSet("test_rs")
featureSetLocalName = featureSetLocalName[:-3] # remove '.db'
self._testData = _testDataForFeatureSetName[featureSetLocalName]
super(FeatureSetTests, self).__init__(featureSetLocalName, dataPath)
def _createVariantAnnotationSet(self, vcfDir):
"""
Creates a VariantAnnotationSet from the specified directory of
VCF files.
"""
self._variantSetName = "testVariantSet"
self._repo = datarepo.SqlDataRepository(paths.testDataRepo)
self._repo.open(datarepo.MODE_READ)
self._dataset = datasets.Dataset("testDs")
self._variantSet = variants.HtslibVariantSet(
self._dataset, self._variantSetName)
self._variantSet.populateFromDirectory(vcfDir)
self._variantAnnotationSet = variants.HtslibVariantAnnotationSet(
self._variantSet, "testVAs")
self._variantAnnotationSet.setOntology(
self._repo.getOntologyByName(paths.ontologyName))
def __init__(self, inputDirectory, outputDirectory, force):
"""
Converts human readable dataset from compliance repository,
and translates it into a reference-server readable filesystem
with binary files.
:param inputDirectory: location of
the human readable compliance dataset
:param outputDirectory: location of
the file hierarchy suitable for deploying on the reference server
"""
self.inputDirectory = inputDirectory
self.outputDirectory = outputDirectory
self.repoPath = os.path.abspath(
os.path.join(outputDirectory, "registry.db"))
self.tempdir = None
if os.path.exists(self.outputDirectory):
if force:
utils.log("Removing existing output directory at '{}'".format(
self.outputDirectory))
shutil.rmtree(self.outputDirectory)
else:
utils.log("Output directory '{}' already exists".format(
self.outputDirectory))
utils.log("Please specify an output path that does not exist")
utils.log("Exiting...")
exit(1)
# If no input directory is specified download from GitHub
if inputDirectory is None:
utils.log("Downloading test data...")
self.tempdir = tempfile.mkdtemp()
assert (os.path.exists(self.tempdir))
url = "https://github.com/ga4gh/compliance/archive/master.zip"
filePath = os.path.join(self.tempdir, 'compliance-master.zip')
downloader = utils.HttpFileDownloader(url, filePath)
downloader.download()
utils.log("Extracting test data...")
with zipfile.ZipFile(filePath, "r") as z:
z.extractall(self.tempdir)
self.inputDirectory = os.path.join(self.tempdir,
'compliance-master',
'test-data')
repo = datarepo.SqlDataRepository(self.repoPath)
self.repo = repo
def createRepo(self):
"""
Creates the repository for all the data we've just downloaded.
"""
repo = datarepo.SqlDataRepository(self.repoPath)
repo.open("w")
repo.initialise()
referenceSet = references.HtslibReferenceSet("GRCh37-subset")
referenceSet.populateFromFile(self.fastaFilePath)
referenceSet.setDescription("Subset of GRCh37 used for demonstration")
referenceSet.setNcbiTaxonId(9606)
for reference in referenceSet.getReferences():
reference.setNcbiTaxonId(9606)
reference.setSourceAccessions(
self.accessions[reference.getName()] + ".subset")
repo.insertReferenceSet(referenceSet)
dataset = datasets.Dataset("1kg-p3-subset")
dataset.setDescription("Sample data from 1000 Genomes phase 3")
repo.insertDataset(dataset)
variantSet = variants.HtslibVariantSet(dataset, "mvncall")
variantSet.setReferenceSet(referenceSet)
dataUrls = [vcfFile for vcfFile, _ in self.vcfFilePaths]
indexFiles = [indexFile for _, indexFile in self.vcfFilePaths]
variantSet.populateFromFile(dataUrls, indexFiles)
variantSet.checkConsistency()
repo.insertVariantSet(variantSet)
for sample, (bamFile, indexFile) in zip(self.samples,
self.bamFilePaths):
readGroupSet = reads.HtslibReadGroupSet(dataset, sample)
readGroupSet.populateFromFile(bamFile, indexFile)
readGroupSet.setReferenceSet(referenceSet)
repo.insertReadGroupSet(readGroupSet)
repo.commit()
repo.close()
self.log("Finished creating the repository; summary:\n")
repo.open("r")
repo.printSummary()
def __init__(self, registryDb):
repo = datarepo.SqlDataRepository(registryDb)
repo.open(datarepo.MODE_READ)
super(CpuProfilerBackend, self).__init__(repo)
self.profiler = cProfile.Profile()
def __init__(self, registryDb):
repo = datarepo.SqlDataRepository(registryDb)
repo.open(datarepo.MODE_READ)
super(HeapProfilerBackend, self).__init__(repo)
self.profiler = guppy.hpy()
""")
args = parser.parse_args()
registryDb = "ga4gh-example-data/registry.db"
if args.profile == 'heap':
backendClass = HeapProfilerBackend
backend = backendClass(registryDb)
args.repeatLimit = 1
args.pageLimit = 1
elif args.profile == 'cpu':
backendClass = CpuProfilerBackend
backend = backendClass(registryDb)
else:
repo = datarepo.SqlDataRepository(registryDb)
repo.open(datarepo.MODE_READ)
backend = backend.Backend(repo)
# Get our list of callSetids
callSetIds = args.callSetIds
if callSetIds != []:
callSetIds = None
if args.callSetIds != "*":
callSetIds = args.callSetIds.split(",")
minTime = benchmarkOneQuery(_heavyQuery(args.variantSetId, callSetIds),
args.repeatLimit, args.pageLimit)
print(minTime)
if args.profile == 'cpu':
stats = pstats.Stats(backend.profiler)
def __init__(self, args):
self._args = args
self._registryPath = args.registryPath
self._repo = datarepo.SqlDataRepository(self._registryPath)
def data_repo(path):
dataRepository = datarepo.SqlDataRepository(path)
dataRepository.open(datarepo.MODE_READ)
return dataRepository
def testNonexistantFile(self):
repo = datarepo.SqlDataRepository("aFilePathThatDoesNotExist")
with self.assertRaises(exceptions.RepoNotFoundException):
repo.open(datarepo.MODE_READ)
def testDirectory(self):
repoPath = makeTempDir()
repo = datarepo.SqlDataRepository(repoPath)
with self.assertRaises(exceptions.RepoInvalidDatabaseException):
repo.open(datarepo.MODE_READ)
def testTextFile(self):
with open(self._repoPath, 'w') as textFile:
textFile.write('This is now a text file')
repo = datarepo.SqlDataRepository(self._repoPath)
with self.assertRaises(exceptions.RepoInvalidDatabaseException):
repo.open(datarepo.MODE_READ)
def testDbFileWithoutTables(self):
repo = datarepo.SqlDataRepository(self._repoPath)
with self.assertRaises(exceptions.RepoInvalidDatabaseException):
repo.open(datarepo.MODE_READ)
def readRepo(self):
repo = datarepo.SqlDataRepository(self._repoPath)
repo.open(datarepo.MODE_READ)
return repo
def setUp(self):
self._dataRepo = datarepo.SqlDataRepository(paths.testDataRepo)
self._dataRepo.open(datarepo.MODE_READ)
def configure(configFile=None, baseConfig="ProductionConfig",
port=8000, extraConfig={}):
"""
TODO Document this critical function! What does it do? What does
it assume?
"""
file_handler = StreamHandler()
file_handler.setLevel(logging.WARNING)
app.logger.addHandler(file_handler)
configStr = 'ga4gh.serverconfig:{0}'.format(baseConfig)
app.config.from_object(configStr)
if os.environ.get('GA4GH_CONFIGURATION') is not None:
app.config.from_envvar('GA4GH_CONFIGURATION')
if configFile is not None:
app.config.from_pyfile(configFile)
app.config.update(extraConfig.items())
# Setup file handle cache max size
datamodel.fileHandleCache.setMaxCacheSize(
app.config["FILE_HANDLE_CACHE_MAX_SIZE"])
# Setup CORS
cors.CORS(app, allow_headers='Content-Type')
app.serverStatus = ServerStatus()
# Allocate the backend
# We use URLs to specify the backend. Currently we have file:// URLs (or
# URLs with no scheme) for the SqlDataRepository, and special empty:// and
# simulated:// URLs for empty or simulated data sources.
dataSource = urlparse.urlparse(app.config["DATA_SOURCE"], "file")
if dataSource.scheme == "simulated":
# Ignore the query string
randomSeed = app.config["SIMULATED_BACKEND_RANDOM_SEED"]
numCalls = app.config["SIMULATED_BACKEND_NUM_CALLS"]
variantDensity = app.config["SIMULATED_BACKEND_VARIANT_DENSITY"]
numVariantSets = app.config["SIMULATED_BACKEND_NUM_VARIANT_SETS"]
numReferenceSets = app.config[
"SIMULATED_BACKEND_NUM_REFERENCE_SETS"]
numReferencesPerReferenceSet = app.config[
"SIMULATED_BACKEND_NUM_REFERENCES_PER_REFERENCE_SET"]
numAlignmentsPerReadGroup = app.config[
"SIMULATED_BACKEND_NUM_ALIGNMENTS_PER_READ_GROUP"]
numReadGroupsPerReadGroupSet = app.config[
"SIMULATED_BACKEND_NUM_READ_GROUPS_PER_READ_GROUP_SET"]
dataRepository = datarepo.SimulatedDataRepository(
randomSeed=randomSeed, numCalls=numCalls,
variantDensity=variantDensity, numVariantSets=numVariantSets,
numReferenceSets=numReferenceSets,
numReferencesPerReferenceSet=numReferencesPerReferenceSet,
numReadGroupsPerReadGroupSet=numReadGroupsPerReadGroupSet,
numAlignments=numAlignmentsPerReadGroup)
elif dataSource.scheme == "empty":
dataRepository = datarepo.EmptyDataRepository()
elif dataSource.scheme == "file":
path = os.path.join(dataSource.netloc, dataSource.path)
dataRepository = datarepo.SqlDataRepository(path)
dataRepository.open(datarepo.MODE_READ)
else:
raise exceptions.ConfigurationException(
"Unsupported data source scheme: " + dataSource.scheme)
theBackend = backend.Backend(dataRepository)
theBackend.setRequestValidation(app.config["REQUEST_VALIDATION"])
theBackend.setResponseValidation(app.config["RESPONSE_VALIDATION"])
theBackend.setDefaultPageSize(app.config["DEFAULT_PAGE_SIZE"])
theBackend.setMaxResponseLength(app.config["MAX_RESPONSE_LENGTH"])
app.backend = theBackend
app.secret_key = os.urandom(SECRET_KEY_LENGTH)
app.oidcClient = None
app.tokenMap = None
app.myPort = port
if "OIDC_PROVIDER" in app.config:
# The oic client. If we're testing, we don't want to verify
# SSL certificates
app.oidcClient = oic.oic.Client(
verify_ssl=('TESTING' not in app.config))
app.tokenMap = {}
try:
app.oidcClient.provider_config(app.config['OIDC_PROVIDER'])
except requests.exceptions.ConnectionError:
configResponse = message.ProviderConfigurationResponse(
issuer=app.config['OIDC_PROVIDER'],
authorization_endpoint=app.config['OIDC_AUTHZ_ENDPOINT'],
token_endpoint=app.config['OIDC_TOKEN_ENDPOINT'],
revocation_endpoint=app.config['OIDC_TOKEN_REV_ENDPOINT'])
app.oidcClient.handle_provider_config(configResponse,
app.config['OIDC_PROVIDER'])
# The redirect URI comes from the configuration.
# If we are testing, then we allow the automatic creation of a
# redirect uri if none is configured
redirectUri = app.config.get('OIDC_REDIRECT_URI')
if redirectUri is None and 'TESTING' in app.config:
redirectUri = 'https://{0}:{1}/oauth2callback'.format(
socket.gethostname(), app.myPort)
app.oidcClient.redirect_uris = [redirectUri]
if redirectUri is []:
raise exceptions.ConfigurationException(
'OIDC configuration requires a redirect uri')
# We only support dynamic registration while testing.
if ('registration_endpoint' in app.oidcClient.provider_info and
'TESTING' in app.config):
app.oidcClient.register(
app.oidcClient.provider_info["registration_endpoint"],
redirect_uris=[redirectUri])
else:
response = message.RegistrationResponse(
client_id=app.config['OIDC_CLIENT_ID'],
client_secret=app.config['OIDC_CLIENT_SECRET'],
redirect_uris=[redirectUri],
verify_ssl=False)
app.oidcClient.store_registration_info(response)
def setUp(self):
self._repo = datarepo.SqlDataRepository(paths.testDataRepo)
self._repo.open(datarepo.MODE_READ)
self._backend = backend.Backend(self._repo)
self._client = client.LocalClient(self._backend)
