def setUp(self):
self.__verbose = True
#
self.__mockTopPath = os.path.join(TOPDIR, "rcsb", "mock-data")
self.__pathConfig = os.path.join(TOPDIR, "rcsb", "db", "config",
"exdb-config-example.yml")
self.__cachePath = os.path.join(TOPDIR, "CACHE")
self.__updateId = "2018_25"
self.__export = False
#
configName = "site_info_configuration"
self.__cfgOb = ConfigUtil(configPath=self.__pathConfig,
defaultSectionName=configName,
mockTopPath=self.__mockTopPath)
self.__schP = SchemaProvider(self.__cfgOb,
self.__cachePath,
useCache=True)
self.__sandboxPath = self.__cfgOb.getPath("RCSB_EXCHANGE_SANDBOX_PATH",
sectionName=configName)
#
self.__mU = MarshalUtil(workPath=self.__cachePath)
self.__startTime = time.time()
logger.debug("Starting %s at %s", self.id(),
time.strftime("%Y %m %d %H:%M:%S", time.localtime()))
def setUp(self):
self.__isMac = platform.system() == "Darwin"
self.__excludeType = None if self.__isMac else "optional"
self.__verbose = True
#
fileLimit = 100
numProc = 2
self.__cachePath = os.path.join(TOPDIR, "CACHE")
self.__workPath = os.path.join(HERE, "test-output")
mockTopPath = os.path.join(TOPDIR, "rcsb", "mock-data")
configPath = os.path.join(TOPDIR, "rcsb", "db", "config",
"exdb-config-example.yml")
#
configName = "site_info_configuration"
self.__cfgOb = ConfigUtil(configPath=configPath,
defaultSectionName=configName,
mockTopPath=mockTopPath)
self.__resourceName = "MYSQL_DB"
#
self.__schP = SchemaProvider(self.__cfgOb,
self.__cachePath,
useCache=True)
self.__rpP = RepositoryProvider(cfgOb=self.__cfgOb,
numProc=numProc,
fileLimit=fileLimit,
cachePath=self.__cachePath)
#
#
self.__startTime = time.time()
logger.debug("Starting %s at %s", self.id(),
time.strftime("%Y %m %d %H:%M:%S", time.localtime()))
def __init__(self,
cfgOb,
cachePath,
useCache=True,
numProc=2,
chunkSize=10,
readBackCheck=False,
documentLimit=None,
doValidate=False,
verbose=False):
self.__cfgOb = cfgOb
self.__cachePath = cachePath
self.__useCache = useCache
self.__readBackCheck = readBackCheck
self.__numProc = numProc
self.__chunkSize = chunkSize
self.__documentLimit = documentLimit
#
self.__resourceName = "MONGO_DB"
self.__verbose = verbose
self.__statusList = []
self.__schP = SchemaProvider(self.__cfgOb,
self.__cachePath,
useCache=self.__useCache)
self.__docHelper = DocumentDefinitionHelper(cfgOb=self.__cfgOb)
self.__valInst = None
self.__doValidate = doValidate
def setUp(self):
self.__verbose = True
self.__numProc = 2
self.__fileLimit = 100
self.__chunkSize = 0
self.__workPath = os.path.join(HERE, "test-output")
self.__mockTopPath = os.path.join(TOPDIR, "rcsb", "mock-data")
configPath = os.path.join(TOPDIR, "rcsb", "db", "config",
"exdb-config-example.yml")
configName = "site_info_configuration"
self.__cfgOb = ConfigUtil(configPath=configPath,
defaultSectionName=configName)
self.__resourceName = "CRATE_DB"
self.__schP = SchemaProvider(self.__cfgOb,
self.__workPath,
useCache=True)
self.__rpP = RepositoryProvider(cfgOb=self.__cfgOb,
numProc=self.__numProc,
fileLimit=self.__fileLimit,
cachePath=self.__workPath)
#
#
self.__tableIdSkipD = {
"ATOM_SITE": True,
"ATOM_SITE_ANISOTROP": True,
"__LOAD_STATUS__": True
}
self.__ioObj = IoAdapter(verbose=self.__verbose)
#
self.__startTime = time.time()
logger.debug("Starting %s at %s", self.id(),
time.strftime("%Y %m %d %H:%M:%S", time.localtime()))
def __init__(
self,
cfgOb,
cachePath,
resourceName="MONGO_DB",
numProc=4,
chunkSize=15,
documentLimit=None,
verbose=False,
readBackCheck=False,
maxStepLength=2000,
schemaRebuildFlag=False,
):
self.__verbose = verbose
#
# Limit the load length of each file type for testing - Set to None to remove -
self.__documentLimit = documentLimit
self.__maxStepLength = maxStepLength
#
# Controls for multiprocessing execution -
self.__numProc = numProc
self.__chunkSize = chunkSize
#
self.__cfgOb = cfgOb
self.__resourceName = resourceName
#
self.__cachePath = cachePath if cachePath else "."
self.__schP = SchemaProvider(cfgOb,
cachePath,
useCache=True,
rebuildFlag=schemaRebuildFlag)
#
self.__readBackCheck = readBackCheck
self.__mpFormat = "[%(levelname)s] %(asctime)s %(processName)s-%(module)s.%(funcName)s: %(message)s"
def setUp(self):
self.__numProc = 2
# self.__fileLimit = 200
self.__fileLimit = None
self.__mockTopPath = os.path.join(TOPDIR, "rcsb", "mock-data")
self.__cachePath = os.path.join(TOPDIR, "CACHE")
self.__configPath = os.path.join(TOPDIR, "rcsb", "db", "config", "exdb-config-example-ihm.yml")
configName = "site_info_configuration"
self.__configName = configName
self.__cfgOb = ConfigUtil(configPath=self.__configPath, defaultSectionName=configName, mockTopPath=self.__mockTopPath)
self.__mU = MarshalUtil(workPath=self.__cachePath)
#self.__schP = SchemaProvider(self.__cfgOb, self.__cachePath, useCache=False, rebuildFlag=True)
self.__schP = SchemaProvider(self.__cfgOb, self.__cachePath, useCache=True)
self.__rpP = RepositoryProvider(cfgOb=self.__cfgOb, numProc=self.__numProc, fileLimit=self.__fileLimit, cachePath=self.__cachePath)
#
self.__birdRepoPath = self.__cfgOb.getPath("BIRD_REPO_PATH", sectionName=configName)
#
self.__fTypeRow = "drop-empty-attributes|drop-empty-tables|skip-max-width|convert-iterables|normalize-enums|translateXMLCharRefs"
self.__fTypeCol = "drop-empty-tables|skip-max-width|convert-iterables|normalize-enums|translateXMLCharRefs"
self.__verbose = True
#
self.__modulePathMap = self.__cfgOb.get("DICT_METHOD_HELPER_MODULE_PATH_MAP", sectionName=configName)
self.__testDirPath = os.path.join(HERE, "test-output", "pdbx-files")
self.__testIhmDirPath = os.path.join(HERE, "test-output", "ihm-files")
self.__export = True
#
#self.__extraOpts = None
# The following for extended parent/child info -
self.__extraOpts = 'addParentRefs|addPrimaryKey'
#
self.__alldatabaseNameD = {
"ihm_dev": ["ihm_dev"],
"pdbx": ["pdbx", "pdbx_ext"],
"pdbx_core": ["pdbx_core_entity", "pdbx_core_entry", "pdbx_core_assembly", "pdbx_core_entity_instance", "pdbx_core_entity_instance_validation"],
"bird": ["bird"],
"bird_family": ["family"],
"chem_comp": ["chem_comp"],
"bird_chem_comp": ["bird_chem_comp"],
"bird_chem_comp_core": ["bird_chem_comp_core"],
}
self.__databaseNameD = {
"pdbx_core": ["pdbx_core_entity", "pdbx_core_entry", "pdbx_core_assembly", "pdbx_core_entity_instance", "pdbx_core_entity_instance_validation"],
"bird_chem_comp_core": ["bird_chem_comp_core"],
}
self.__mergeContentTypeD = {"pdbx_core": ["vrpt"]}
# self.__databaseNameD = {"chem_comp_core": ["chem_comp_core"], "bird_chem_comp_core": ["bird_chem_comp_core"]}
# self.__databaseNameD = {"ihm_dev_full": ["ihm_dev_full"]}
# self.__databaseNameD = {"pdbx_core": ["pdbx_core_entity_instance_validation"]}
# self.__databaseNameD = {"pdbx_core": ["pdbx_core_entity_monomer"]}
self.__startTime = time.time()
logger.debug("Starting %s at %s", self.id(), time.strftime("%Y %m %d %H:%M:%S", time.localtime()))
def __init__(self,
cfgOb,
objectAdapter=None,
cachePath=".",
useCache=True,
**kwargs):
self.__cfgOb = cfgOb
self.__oAdapt = objectAdapter
self.__resourceName = "MONGO_DB"
_ = kwargs
self.__statusList = []
self.__schP = SchemaProvider(self.__cfgOb,
cachePath,
useCache=useCache)
self.__valInst = None
def setUp(self):
self.__verbose = True
mockTopPath = os.path.join(TOPDIR, "rcsb", "mock-data")
pathConfig = os.path.join(TOPDIR, "rcsb", "db", "config",
"exdb-config-example.yml")
configName = "site_info_configuration"
self.__cachePath = os.path.join(TOPDIR, "CACHE")
self.__cfgOb = ConfigUtil(configPath=pathConfig,
defaultSectionName=configName,
mockTopPath=mockTopPath)
self.__schP = SchemaProvider(self.__cfgOb,
self.__cachePath,
useCache=False)
#
self.__validationLevels = self.__cfgOb.getList(
"VALIDATION_LEVELS_TEST",
sectionName="database_catalog_configuration")
self.__encodingTypes = self.__cfgOb.getList(
"ENCODING_TYPES_TEST",
sectionName="database_catalog_configuration")
#
buildAll = True
if buildAll:
self.__databaseNameList = self.__cfgOb.getList(
"DATABASE_NAMES_DEPLOYED",
sectionName="database_catalog_configuration")
self.__dataTypingList = self.__cfgOb.getList(
"DATATYPING_DEPLOYED",
sectionName="database_catalog_configuration")
#
else:
self.__databaseNameList = self.__cfgOb.getList(
"DATABASE_NAMES_TEST",
sectionName="database_catalog_configuration")
# self.__databaseNameList = ["repository_holdings"]
self.__dataTypingList = self.__cfgOb.getList(
"DATATYPING_TEST",
sectionName="database_catalog_configuration")
# self.__databaseNameList = ["sequence_clusters"]
self.__saveSchema = True
self.__compareDefSchema = False
self.__compareSchema = False
#
self.__startTime = time.time()
logger.debug("Starting %s at %s", self.id(),
time.strftime("%Y %m %d %H:%M:%S", time.localtime()))
def setUp(self):
self.__verbose = True
mockTopPath = os.path.join(TOPDIR, "rcsb", "mock-data")
self.__cachePath = os.path.join(TOPDIR, "CACHE")
pathConfig = os.path.join(TOPDIR, "rcsb", "db", "config",
"exdb-config-example.yml")
configName = "site_info_configuration"
self.__cfgOb = ConfigUtil(configPath=pathConfig,
defaultSectionName=configName,
mockTopPath=mockTopPath)
self.__schP = SchemaProvider(self.__cfgOb,
self.__cachePath,
useCache=True,
clearPath=False)
#
self.__startTime = time.time()
logger.debug("Starting %s at %s", self.id(),
time.strftime("%Y %m %d %H:%M:%S", time.localtime()))
def setUp(self):
self.__verbose = True
#
self.__mockTopPath = os.path.join(TOPDIR, "rcsb", "mock-data")
self.__pathConfig = os.path.join(TOPDIR, "rcsb", "db", "config", "exdb-config-example.yml")
self.__cachePath = os.path.join(TOPDIR, "CACHE")
self.__updateId = "2018_25"
#
configName = "site_info_configuration"
self.__cfgOb = ConfigUtil(configPath=self.__pathConfig, defaultSectionName=configName, mockTopPath=self.__mockTopPath)
self.__schP = SchemaProvider(self.__cfgOb, self.__cachePath, useCache=True)
#
self.__sandboxPath = self.__cfgOb.getPath("RCSB_EXCHANGE_SANDBOX_PATH", sectionName=configName)
#
self.__dataSetId = "2018_23"
self.__pathClusterData = self.__cfgOb.getPath("RCSB_SEQUENCE_CLUSTER_DATA_PATH", sectionName=configName)
# self.__levels = ['100', '95', '90', '70', '50', '30']
self.__levels = ["100"]
#
self.__startTime = time.time()
logger.debug("Starting %s at %s", self.id(), time.strftime("%Y %m %d %H:%M:%S", time.localtime()))
class SchemaDefLoaderDbTests(unittest.TestCase):
def __init__(self, methodName="runTest"):
super(SchemaDefLoaderDbTests, self).__init__(methodName)
self.__verbose = True
def setUp(self):
self.__isMac = platform.system() == "Darwin"
self.__excludeType = None if self.__isMac else "optional"
self.__verbose = True
#
fileLimit = 100
numProc = 2
self.__cachePath = os.path.join(TOPDIR, "CACHE")
self.__workPath = os.path.join(HERE, "test-output")
mockTopPath = os.path.join(TOPDIR, "rcsb", "mock-data")
configPath = os.path.join(TOPDIR, "rcsb", "db", "config",
"exdb-config-example.yml")
#
configName = "site_info_configuration"
self.__cfgOb = ConfigUtil(configPath=configPath,
defaultSectionName=configName,
mockTopPath=mockTopPath)
self.__resourceName = "MYSQL_DB"
#
self.__schP = SchemaProvider(self.__cfgOb,
self.__cachePath,
useCache=True)
self.__rpP = RepositoryProvider(cfgOb=self.__cfgOb,
numProc=numProc,
fileLimit=fileLimit,
cachePath=self.__cachePath)
#
#
self.__startTime = time.time()
logger.debug("Starting %s at %s", self.id(),
time.strftime("%Y %m %d %H:%M:%S", time.localtime()))
def tearDown(self):
endTime = time.time()
logger.debug("Completed %s at %s (%.4f seconds)", self.id(),
time.strftime("%Y %m %d %H:%M:%S", time.localtime()),
endTime - self.__startTime)
def __schemaCreate(self, schemaDefObj):
"""Create table schema using schema definition"""
try:
tableIdList = schemaDefObj.getSchemaIdList()
sqlGen = SqlGenAdmin(self.__verbose)
sqlL = sqlGen.createDatabaseSQL(schemaDefObj.getDatabaseName())
for tableId in tableIdList:
tableDefObj = schemaDefObj.getSchemaObject(tableId)
sqlL.extend(
sqlGen.createTableSQL(
databaseName=schemaDefObj.getDatabaseName(),
tableDefObj=tableDefObj))
logger.debug("Schema creation SQL string\n %s\n\n",
"\n".join(sqlL))
with Connection(cfgOb=self.__cfgOb,
resourceName=self.__resourceName) as client:
myQ = MyDbQuery(dbcon=client, verbose=self.__verbose)
#
# Permit warnings to support "drop table if exists" for missing tables.
#
myQ.setWarning("ignore")
ret = myQ.sqlCommand(sqlCommandList=sqlL)
logger.debug("\n\n+INFO mysql server returns %r\n", ret)
self.assertTrue(ret)
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
# ------------- - -------------------------------------------------------------------------------------------
def testSchemaCreate(self):
"""Create table schema for BIRD, chemical component, and PDBx data."""
cD = self.__schP.makeSchemaDef("bird",
dataTyping="SQL",
saveSchema=True)
sd = SchemaDefAccess(cD)
self.__schemaCreate(sd)
#
cD = self.__schP.makeSchemaDef("chem_comp",
dataTyping="SQL",
saveSchema=True)
sd = SchemaDefAccess(cD)
self.__schemaCreate(sd)
#
# cD = self.__schP.makeSchemaDef("pdbx", dataTyping="SQL", saveSchema=True)
# sd = SchemaDefAccess(cD)
self.__schemaCreate(sd)
def testLoadBirdReference(self):
try:
cD = self.__schP.makeSchemaDef("bird",
dataTyping="SQL",
saveSchema=True)
sd = SchemaDefAccess(cD)
self.__schemaCreate(sd)
inputPathList = self.__rpP.getLocatorObjList(contentType="bird")
inputPathList.extend(
self.__rpP.getLocatorObjList(contentType="bird_family"))
#
with Connection(cfgOb=self.__cfgOb,
resourceName=self.__resourceName) as client:
sdl = SchemaDefLoader(
self.__cfgOb,
schemaDefObj=sd,
dbCon=client,
cachePath=self.__cachePath,
workPath=self.__workPath,
cleanUp=False,
warnings="error",
verbose=self.__verbose,
restoreUseStash=False,
restoreUseGit=True,
providerTypeExclude=self.__excludeType,
)
ok = sdl.load(inputPathList=inputPathList,
loadType="batch-file")
self.assertTrue(ok)
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def testReLoadBirdReference(self):
try:
cD = self.__schP.makeSchemaDef("bird",
dataTyping="SQL",
saveSchema=True)
sd = SchemaDefAccess(cD)
self.__schemaCreate(sd)
inputPathList = self.__rpP.getLocatorObjList(contentType="bird")
inputPathList.extend(
self.__rpP.getLocatorObjList(contentType="bird_family"))
#
with Connection(cfgOb=self.__cfgOb,
resourceName=self.__resourceName) as client:
sdl = SchemaDefLoader(
self.__cfgOb,
schemaDefObj=sd,
dbCon=client,
cachePath=self.__cachePath,
workPath=self.__workPath,
cleanUp=False,
warnings="error",
verbose=self.__verbose,
restoreUseStash=False,
restoreUseGit=True,
providerTypeExclude=self.__excludeType,
)
sdl.load(inputPathList=inputPathList, loadType="batch-file")
#
logger.debug(
"INFO BATCH FILE RELOAD TEST --------------------------------------------\n"
)
ok = sdl.load(inputPathList=inputPathList,
loadType="batch-file",
deleteOpt="all")
self.assertTrue(ok)
#
logger.debug(
"\n\n\n+INFO BATCH INSERT RELOAD TEST --------------------------------------------\n"
)
ok = sdl.load(inputPathList=inputPathList,
loadType="batch-file",
deleteOpt="selected")
self.assertTrue(ok)
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def testLoadChemCompReference(self):
try:
cD = self.__schP.makeSchemaDef("chem_comp",
dataTyping="SQL",
saveSchema=True)
sd = SchemaDefAccess(cD)
self.__schemaCreate(sd)
inputPathList = self.__rpP.getLocatorObjList(
contentType="chem_comp")
with Connection(cfgOb=self.__cfgOb,
resourceName=self.__resourceName) as client:
sdl = SchemaDefLoader(
self.__cfgOb,
schemaDefObj=sd,
dbCon=client,
cachePath=self.__cachePath,
workPath=self.__workPath,
cleanUp=False,
warnings="error",
verbose=self.__verbose,
restoreUseStash=False,
restoreUseGit=True,
providerTypeExclude=self.__excludeType,
)
ok = sdl.load(inputPathList=inputPathList,
loadType="batch-file")
self.assertTrue(ok)
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
@unittest.skip("Disable test - schema not optimized for mysql limitations")
def testLoadPdbxFiles(self):
try:
cD = self.__schP.makeSchemaDef("pdbx",
dataTyping="SQL",
saveSchema=True)
sd = SchemaDefAccess(cD)
self.__schemaCreate(sd)
inputPathList = self.__rpP.getLocatorObjList(contentType="pdbx")
logger.debug("Input path list %r", inputPathList)
with Connection(cfgOb=self.__cfgOb,
resourceName=self.__resourceName) as client:
sdl = SchemaDefLoader(
self.__cfgOb,
schemaDefObj=sd,
dbCon=client,
cachePath=self.__cachePath,
workPath=self.__workPath,
cleanUp=False,
warnings="error",
verbose=self.__verbose,
restoreUseStash=False,
restoreUseGit=True,
providerTypeExclude=self.__excludeType,
)
ok = sdl.load(inputPathList=inputPathList,
loadType="batch-insert",
deleteOpt="all")
self.assertTrue(ok)
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
class ObjectValidator(object):
"""Utilities to extract and update object from the document object server with validation."""
def __init__(self,
cfgOb,
objectAdapter=None,
cachePath=".",
useCache=True,
**kwargs):
self.__cfgOb = cfgOb
self.__oAdapt = objectAdapter
self.__resourceName = "MONGO_DB"
_ = kwargs
self.__statusList = []
self.__schP = SchemaProvider(self.__cfgOb,
cachePath,
useCache=useCache)
self.__valInst = None
def __getValidator(self, databaseName, collectionName, schemaLevel="full"):
_ = self.__schP.makeSchemaDef(databaseName,
dataTyping="ANY",
saveSchema=True)
cD = self.__schP.makeSchema(databaseName,
collectionName,
encodingType="JSON",
level=schemaLevel,
saveSchema=True)
# Raises exceptions for schema compliance.
Draft4Validator.check_schema(cD)
valInst = Draft4Validator(cD, format_checker=FormatChecker())
return valInst
def __validateObj(self, databaseName, collectionName, rObj, label=""):
try:
eCount = 0
tId = rObj["rcsb_id"] if rObj and "rcsb_id" in rObj else "anonymous"
for error in sorted(self.__valInst.iter_errors(rObj), key=str):
logger.info(
"Database %s collection %s (%s %r) path %s error: %s",
databaseName, collectionName, label, tId, error.path,
error.message)
logger.debug(">>> Failing object is %r", rObj)
eCount += 1
except Exception as e:
logger.exception("Validation failing %s", str(e))
return eCount
def doTransform(self, **kwargs):
desp = DataExchangeStatus()
statusStartTimestamp = desp.setStartTime()
#
databaseName = kwargs.get("databaseName", "pdbx_core")
collectionName = kwargs.get("collectionName", "pdbx_core_entry")
selectionQueryD = kwargs.get("selectionQuery", {})
fetchLimit = kwargs.get("fetchLimit", None)
#
#
tU = TimeUtil()
updateId = kwargs.get("updateId", tU.getCurrentWeekSignature())
#
docSelectList = self.__selectObjectIds(databaseName, collectionName,
selectionQueryD)
docSelectList = docSelectList[:fetchLimit] if fetchLimit else docSelectList
ok = self.__transform(databaseName, collectionName, docSelectList)
#
if updateId:
okS = self.__updateStatus(updateId, databaseName, collectionName,
ok, statusStartTimestamp)
return ok and okS
def __selectObjectIds(self, databaseName, collectionName, selectionQueryD):
"""Return a list of object identifiers for the input selection query."""
try:
with Connection(cfgOb=self.__cfgOb,
resourceName=self.__resourceName) as client:
mg = MongoDbUtil(client)
if mg.collectionExists(databaseName, collectionName):
logger.info("%s %s document count is %d", databaseName,
collectionName,
mg.count(databaseName, collectionName))
qD = {}
if selectionQueryD:
qD.update(selectionQueryD)
selectL = ["_id"]
dL = mg.fetch(databaseName,
collectionName,
selectL,
queryD=qD)
logger.info("Selection %r fetch result count %d", selectL,
len(dL))
except Exception as e:
logger.exception("Failing with %s", str(e))
return dL
#
def __transform(self,
databaseName,
collectionName,
docSelectList,
logIncrement=100):
"""Return a list of object identifiers for the input selection query."""
#
ok = True
try:
self.__valInst = self.__getValidator(databaseName,
collectionName,
schemaLevel="full")
with Connection(cfgOb=self.__cfgOb,
resourceName=self.__resourceName) as client:
mg = MongoDbUtil(client)
if mg.collectionExists(databaseName, collectionName):
numDoc = len(docSelectList)
for ii, dD in enumerate(docSelectList, 1):
if "_id" not in dD:
continue
rObj = mg.fetchOne(databaseName, collectionName, "_id",
dD["_id"])
del rObj["_id"]
#
fOk = True
if self.__oAdapt:
self.__validateObj(databaseName,
collectionName,
rObj,
label="Original")
fOk, rObj = self.__oAdapt.filter(rObj)
self.__validateObj(databaseName,
collectionName,
rObj,
label="Updated")
if fOk:
rOk = mg.replace(databaseName, collectionName,
rObj, dD)
if rOk is None:
tId = rObj[
"rcsb_id"] if rObj and "rcsb_id" in rObj else "anonymous"
logger.error("%r %r (%r) failing",
databaseName, collectionName, tId)
# logger.info("rObj.keys() %r", list(rObj.keys()))
# logger.info("rObj.items() %s", rObj.items())
rOk = False
ok = ok and rOk
#
if ii % logIncrement == 0 or ii == numDoc:
logger.info("Replace status %r object (%d of %d)",
ok, ii, numDoc)
#
except Exception as e:
logger.exception("Failing with %s", str(e))
return ok
def getLoadStatus(self):
return self.__statusList
def __updateStatus(self, updateId, databaseName, collectionName, status,
startTimestamp):
try:
sFlag = "Y" if status else "N"
desp = DataExchangeStatus()
desp.setStartTime(tS=startTimestamp)
desp.setObject(databaseName, collectionName)
desp.setStatus(updateId=updateId, successFlag=sFlag)
desp.setEndTime()
self.__statusList.append(desp.getStatus())
return True
except Exception as e:
logger.exception("Failing with %s", str(e))
return False
class SqlGenTests(unittest.TestCase):
def setUp(self):
self.__verbose = True
#
mockTopPath = os.path.join(TOPDIR, "rcsb", "mock-data")
pathConfig = os.path.join(TOPDIR, "rcsb", "db", "config",
"exdb-config-example.yml")
self.__cachePath = os.path.join(TOPDIR, "CACHE")
#
configName = "site_info_configuration"
self.__cfgOb = ConfigUtil(configPath=pathConfig,
defaultSectionName=configName,
mockTopPath=mockTopPath)
self.__sdu = SchemaProvider(self.__cfgOb,
self.__cachePath,
useCache=True)
#
self.__startTime = time.time()
logger.debug("Starting %s at %s", self.id(),
time.strftime("%Y %m %d %H:%M:%S", time.localtime()))
def testSQLMethods(self):
schemaNames = ["pdbx_core"]
dataTyping = "SQL"
for schemaName in schemaNames:
dD = self.__sdu.makeSchemaDef(schemaName,
dataTyping=dataTyping,
saveSchema=False)
sD = SchemaDefAccess(dD)
self.__testSchemaCreate(sD)
self.__testImportExport(sD)
self.__testSelectionAndConditions(sD)
#
def __getHelper(self, modulePath, **kwargs):
aMod = __import__(modulePath, globals(), locals(), [""])
sys.modules[modulePath] = aMod
#
# Strip off any leading path to the module before we instaniate the object.
mpL = modulePath.split(".")
moduleName = mpL[-1]
#
aObj = getattr(aMod, moduleName)(**kwargs)
return aObj
def __testSchemaCreate(self, sD):
"""Test case - create table schema using input schema definition as an example
"""
try:
tableIdList = sD.getSchemaIdList()
myAd = SqlGenAdmin(self.__verbose)
sqlL = []
for tableId in tableIdList:
tableDefObj = sD.getSchemaObject(tableId)
sqlL.extend(
myAd.createTableSQL(databaseName=sD.getDatabaseName(),
tableDefObj=tableDefObj))
logger.debug(
"\n\n+SqlGenTests table creation SQL string\n %s\n\n",
"\n".join(sqlL))
self.assertGreaterEqual(len(sqlL), 10)
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def __testImportExport(self, sD):
"""Test case - import and export commands --
"""
try:
databaseName = sD.getDatabaseName()
tableIdList = sD.getSchemaIdList()
myAd = SqlGenAdmin(self.__verbose)
for tableId in tableIdList:
tableDefObj = sD.getSchemaObject(tableId)
exportPath = os.path.join(HERE, "test-output",
tableDefObj.getName() + ".tdd")
sqlExport = myAd.exportTable(databaseName,
tableDefObj,
exportPath=exportPath)
logger.debug(
"\n\n+SqlGenTests table export SQL string\n %s\n\n",
sqlExport)
sqlImport = myAd.importTable(databaseName,
tableDefObj,
importPath=exportPath)
logger.debug(
"\n\n+SqlGenTests table import SQL string\n %s\n\n",
sqlImport)
self.assertGreaterEqual(len(sqlImport), 100)
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def __testSelectionAndConditions(self, sD):
"""Test case - selection everything for a simple condition-
"""
try:
# get delete attribute -
#
tableIdList = sD.getSchemaIdList()
logger.debug("TableIdList %r", tableIdList)
sqlGen = SqlGenQuery(schemaDefObj=sD, verbose=self.__verbose)
for tableId in tableIdList:
tableDefObj = sD.getSchemaObject(tableId)
dAtId = tableDefObj.getDeleteAttributeId()
if dAtId:
sqlCondition = SqlGenCondition(schemaDefObj=sD,
verbose=self.__verbose)
sqlCondition.addValueCondition((tableId, dAtId), "EQ",
("D000001", "CHAR"))
aIdList = sD.getAttributeIdList(tableId)
for aId in aIdList:
sqlGen.addSelectAttributeId(attributeTuple=(tableId,
aId))
sqlGen.setCondition(sqlCondition)
sqlGen.addOrderByAttributeId(attributeTuple=(tableId,
dAtId))
sqlS = sqlGen.getSql()
logger.debug(
"\n\n+SqlGenTests table creation SQL string\n %s\n\n",
sqlS)
self.assertGreaterEqual(len(sqlS), 50)
sqlGen.clear()
else:
logger.debug("Missing delete atttribe for table %r",
tableId)
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
class CockroachDbLoaderCockroachDbTests(unittest.TestCase):
def __init__(self, methodName="runTest"):
super(CockroachDbLoaderCockroachDbTests, self).__init__(methodName)
self.__verbose = True
self.__createFlag = False
def setUp(self):
self.__verbose = True
self.__numProc = 2
self.__fileLimit = 100
self.__workPath = os.path.join(HERE, "test-output")
self.__mockTopPath = os.path.join(TOPDIR, "rcsb", "mock-data")
configPath = os.path.join(TOPDIR, "rcsb", "db", "config", "exdb-config-example.yml")
configName = "site_info_configuration"
self.__cfgOb = ConfigUtil(configPath=configPath, defaultSectionName=configName)
self.__resourceName = "COCKROACH_DB"
self.__schP = SchemaProvider(self.__cfgOb, self.__workPath, useCache=True)
self.__rpP = RepositoryProvider(cfgOb=self.__cfgOb, numProc=self.__numProc, fileLimit=self.__fileLimit, cachePath=self.__workPath)
#
self.__tableIdSkipD = {"ATOM_SITE": True, "ATOM_SITE_ANISOTROP": True}
self.__ioObj = IoAdapter(verbose=self.__verbose)
#
self.__startTime = time.time()
logger.debug("Starting %s at %s", self.id(), time.strftime("%Y %m %d %H:%M:%S", time.localtime()))
def tearDown(self):
endTime = time.time()
logger.debug("Completed %s at %s (%.4f seconds)\n", self.id(), time.strftime("%Y %m %d %H:%M:%S", time.localtime()), endTime - self.__startTime)
def testConnection(self):
try:
with Connection(cfgOb=self.__cfgOb, resourceName=self.__resourceName) as client:
self.assertNotEqual(client, None)
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def testSchemaCreate(self):
"""Create table schema (live) for BIRD, chemical component, and PDBx data."""
try:
sd, _, _, _ = self.__schP.getSchemaInfo("bird")
ret = self.__schemaCreate(schemaDefObj=sd)
self.assertEqual(ret, True)
#
sd, _, _, _ = self.__schP.getSchemaInfo("chem_comp")
ret = self.__schemaCreate(schemaDefObj=sd)
self.assertEqual(ret, True)
#
sd, _, _, _ = self.__schP.getSchemaInfo("pdbx")
ret = self.__schemaCreate(schemaDefObj=sd)
self.assertEqual(ret, True)
#
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def testSchemaRemove(self):
"""Remove table schema (live) for BIRD, chemical component, and PDBx data."""
try:
sd, _, _, _ = self.__schP.getSchemaInfo("bird")
ret = self.__schemaRemove(schemaDefObj=sd)
self.assertEqual(ret, True)
#
sd, _, _, _ = self.__schP.getSchemaInfo("chem_comp")
ret = self.__schemaRemove(schemaDefObj=sd)
self.assertEqual(ret, True)
#
sd, _, _, _ = self.__schP.getSchemaInfo("pdbx")
ret = self.__schemaRemove(schemaDefObj=sd)
self.assertEqual(ret, True)
#
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def testLoadInsertBirdReference(self):
try:
sd, _, _, _ = self.__schP.getSchemaInfo("bird")
if self.__createFlag:
self.__schemaCreate(schemaDefObj=sd)
inputPathList = self.__rpP.getLocatorObjList("bird")
inputPathList.extend(self.__rpP.getLocatorObjList("bird_family"))
with Connection(cfgOb=self.__cfgOb, resourceName=self.__resourceName) as client:
sdl = CockroachDbLoader(schemaDefObj=sd, ioObj=self.__ioObj, dbCon=client, workPath=self.__workPath, cleanUp=False, warnings="default", verbose=self.__verbose)
ret = sdl.load(inputPathList=inputPathList, loadType="cockroach-insert", deleteOpt="selected")
self.assertEqual(ret, True)
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def testLoadInsertManyBirdReference(self):
try:
sd, _, _, _ = self.__schP.getSchemaInfo("bird")
if self.__createFlag:
self.__schemaCreate(schemaDefObj=sd)
inputPathList = self.__rpP.getLocatorObjList("bird")
inputPathList.extend(self.__rpP.getLocatorObjList("bird_family"))
with Connection(cfgOb=self.__cfgOb, resourceName=self.__resourceName) as client:
sdl = CockroachDbLoader(schemaDefObj=sd, ioObj=self.__ioObj, dbCon=client, workPath=self.__workPath, cleanUp=False, warnings="default", verbose=self.__verbose)
ret = sdl.load(inputPathList=inputPathList, loadType="cockroach-insert-many", deleteOpt="selected")
self.assertEqual(ret, True)
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def testLoadInsertChemCompReference(self):
try:
sd, _, _, _ = self.__schP.getSchemaInfo("chem_comp")
if self.__createFlag:
self.__schemaCreate(schemaDefObj=sd)
inputPathList = self.__rpP.getLocatorObjList("chem_comp")
with Connection(cfgOb=self.__cfgOb, resourceName=self.__resourceName) as client:
sdl = CockroachDbLoader(schemaDefObj=sd, ioObj=self.__ioObj, dbCon=client, workPath=self.__workPath, cleanUp=False, warnings="default", verbose=self.__verbose)
ret = sdl.load(inputPathList=inputPathList, loadType="cockroach-insert", deleteOpt="selected")
self.assertEqual(ret, True)
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def testLoadInsertManyChemCompReference(self):
try:
sd, _, _, _ = self.__schP.getSchemaInfo("chem_comp")
if self.__createFlag:
self.__schemaCreate(schemaDefObj=sd)
inputPathList = self.__rpP.getLocatorObjList("chem_comp")
with Connection(cfgOb=self.__cfgOb, resourceName=self.__resourceName) as client:
sdl = CockroachDbLoader(schemaDefObj=sd, ioObj=self.__ioObj, dbCon=client, workPath=self.__workPath, cleanUp=False, warnings="default", verbose=self.__verbose)
ret = sdl.load(inputPathList=inputPathList, loadType="cockroach-insert-many", deleteOpt="selected")
self.assertEqual(ret, True)
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def testLoadInsertPdbxExampleFiles(self):
try:
sd, _, _, _ = self.__schP.getSchemaInfo("pdbx")
if self.__createFlag:
self.__schemaCreate(schemaDefObj=sd)
inputPathList = self.__rpP.getLocatorObjList("pdbx")
with Connection(cfgOb=self.__cfgOb, resourceName=self.__resourceName) as client:
sdl = CockroachDbLoader(schemaDefObj=sd, ioObj=self.__ioObj, dbCon=client, workPath=self.__workPath, cleanUp=False, warnings="default", verbose=self.__verbose)
ret = sdl.load(inputPathList=inputPathList, loadType="cockroach-insert", deleteOpt="selected", tableIdSkipD=self.__tableIdSkipD)
self.assertEqual(ret, True)
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def testLoadInsertManyPdbxExampleFiles(self):
try:
sd, _, _, _ = self.__schP.getSchemaInfo("pdbx")
if self.__createFlag:
self.__schemaCreate(schemaDefObj=sd)
inputPathList = self.__rpP.getLocatorObjList("pdbx")
with Connection(cfgOb=self.__cfgOb, resourceName=self.__resourceName) as client:
sdl = CockroachDbLoader(schemaDefObj=sd, ioObj=self.__ioObj, dbCon=client, workPath=self.__workPath, cleanUp=False, warnings="default", verbose=self.__verbose)
ret = sdl.load(inputPathList=inputPathList, loadType="cockroach-insert-many", deleteOpt="selected", tableIdSkipD=self.__tableIdSkipD)
self.assertEqual(ret, True)
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def __schemaCreateSQL(self, schemaDefObj):
"""Test case - create table schema using schema definition"""
sqlL = []
try:
tableIdList = schemaDefObj.getTableIdList()
sqlGen = SqlGenAdmin(self.__verbose, serverType="CockroachDb")
dbName = schemaDefObj.getVersionedDatabaseName()
sqlL = sqlGen.createDatabaseSQL(dbName)
for tableId in tableIdList:
tableDefObj = schemaDefObj.getTable(tableId)
sqlL.extend(sqlGen.createTableSQL(databaseName=schemaDefObj.getVersionedDatabaseName(), tableDefObj=tableDefObj))
logger.debug("\nSchema creation SQL string\n %s\n\n", "\n".join(sqlL))
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
return sqlL
def __schemaCreate(self, schemaDefObj):
"""Test case - create table schema using schema definition"""
ret = 0
try:
tableIdList = schemaDefObj.getTableIdList()
sqlGen = SqlGenAdmin(self.__verbose, serverType="CockroachDb")
dbName = schemaDefObj.getVersionedDatabaseName()
sqlL = sqlGen.createDatabaseSQL(dbName)
for tableId in tableIdList:
tableDefObj = schemaDefObj.getTable(tableId)
sqlL.extend(sqlGen.createTableSQL(databaseName=schemaDefObj.getVersionedDatabaseName(), tableDefObj=tableDefObj))
logger.debug("\nSchema creation SQL string\n %s\n\n", "\n".join(sqlL))
logger.info("Creating schema using database %s", schemaDefObj.getVersionedDatabaseName())
#
with Connection(cfgOb=self.__cfgOb, resourceName=self.__resourceName) as client:
crQ = CockroachDbQuery(dbcon=client, verbose=self.__verbose)
ret = crQ.sqlCommandList(sqlCommandList=sqlL)
# ret = crQ.sqlCommand(' '.join(sqlL))
logger.info("Schema create command returns %r\n", ret)
return ret
#
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def __schemaRemove(self, schemaDefObj):
"""Test case - remove table schema using schema definition"""
ret = 0
try:
dbName = schemaDefObj.getVersionedDatabaseName()
sqlGen = SqlGenAdmin(self.__verbose, serverType="CockroachDb")
sqlL = sqlGen.removeDatabaseSQL(dbName)
logger.debug("Schema Remove SQL string\n %s", "\n".join(sqlL))
with Connection(cfgOb=self.__cfgOb, resourceName=self.__resourceName) as client:
crQ = CockroachDbQuery(dbcon=client, verbose=self.__verbose)
ret = crQ.sqlCommandList(sqlCommandList=sqlL)
# ret = crQ.sqlCommand(' '.join(sqlL))
logger.debug("Schema remove command returns %r\n", ret)
return ret
#
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def main():
parser = argparse.ArgumentParser()
#
defaultConfigName = "site_info_configuration"
#
parser.add_argument(
"--update_chem_comp_ref",
default=False,
action="store_true",
help="Update schema for Chemical Component reference definitions")
parser.add_argument(
"--update_chem_comp_core_ref",
default=False,
action="store_true",
help="Update core schema for Chemical Component reference definitions")
parser.add_argument(
"--update_bird_chem_comp_ref",
default=False,
action="store_true",
help="Update schema for Bird Chemical Component reference definitions")
parser.add_argument(
"--update_bird_chem_comp_core_ref",
default=False,
action="store_true",
help=
"Update core schema for Bird Chemical Component reference definitions")
parser.add_argument("--update_bird_ref",
default=False,
action="store_true",
help="Update schema for Bird reference definitions")
parser.add_argument(
"--update_bird_family_ref",
default=False,
action="store_true",
help="Update schema for Bird Family reference definitions")
parser.add_argument("--update_pdbx",
default=False,
action="store_true",
help="Update schema for PDBx entry data")
parser.add_argument("--update_pdbx_core",
default=False,
action="store_true",
help="Update schema for PDBx core entry/entity data")
parser.add_argument(
"--update_pdbx_comp_model_core",
default=False,
action="store_true",
help="Update schema for PDBx computational model core entry/entity data"
)
#
parser.add_argument("--update_repository_holdings",
default=False,
action="store_true",
help="Update schema for repository holdings")
parser.add_argument("--update_entity_sequence_clusters",
default=False,
action="store_true",
help="Update schema for entity sequence clusters")
parser.add_argument("--update_data_exchange",
default=False,
action="store_true",
help="Update schema for data exchange status")
parser.add_argument("--update_ihm_dev",
default=False,
action="store_true",
help="Update schema for I/HM dev entry data")
parser.add_argument("--update_drugbank_core",
default=False,
action="store_true",
help="Update DrugBank schema")
#
parser.add_argument(
"--update_config_all",
default=False,
action="store_true",
help="Update using configuration settings (e.g. DATABASE_NAMES_ALL)")
parser.add_argument(
"--update_config_deployed",
default=False,
action="store_true",
help=
"Update using configuration settings (e.g. DATABASE_NAMES_DEPLOYED)")
parser.add_argument(
"--update_config_test",
default=False,
action="store_true",
help="Update using configuration settings (e.g. DATABASE_NAMES_TEST)")
#
parser.add_argument("--config_path",
default=None,
help="Path to configuration options file")
parser.add_argument("--config_name",
default=defaultConfigName,
help="Configuration section name")
#
parser.add_argument("--cache_path",
default=None,
help="Schema cache directory path")
parser.add_argument(
"--encoding_types",
default=None,
help="Schema encoding (rcsb|json|bson) (comma separated)")
parser.add_argument(
"--validation_levels",
default=None,
help="Schema validation level (full|min) (comma separated)")
parser.add_argument("--compare_only",
default=False,
action="store_true",
help="Perform comparison with cached schema")
#
parser.add_argument("--debug",
default=False,
action="store_true",
help="Turn on verbose logging")
parser.add_argument(
"--mock",
default=False,
action="store_true",
help="Use MOCK repository configuration for dependencies and testing")
# parser.add_argument("--working_path", default=None, help="Working/alternative path for temporary and schema files")
args = parser.parse_args()
#
debugFlag = args.debug
if debugFlag:
logger.setLevel(logging.DEBUG)
# ----------------------- - ----------------------- - ----------------------- - ----------------------- - ----------------------- -
# Configuration Details
configPath = args.config_path
configName = args.config_name
cachePath = args.cache_path
compareOnly = args.compare_only
#
encodingTypes = args.encoding_types.split(
",") if args.encoding_types else []
validationLevels = args.validation_levels.split(
",") if args.validation_levels else []
dataTypingList = ["ANY", "SQL"]
if not configPath:
configPath = os.getenv("DBLOAD_CONFIG_PATH", None)
try:
if os.access(configPath, os.R_OK):
os.environ["DBLOAD_CONFIG_PATH"] = configPath
logger.info("Using configuation path %s (%s)", configPath,
configName)
else:
logger.error("Missing or access issue with config file %r",
configPath)
exit(1)
mockTopPath = os.path.join(TOPDIR, "rcsb",
"mock-data") if args.mock else None
cfgOb = ConfigUtil(configPath=configPath,
defaultSectionName=defaultConfigName,
mockTopPath=mockTopPath)
if configName != defaultConfigName:
cfgOb.replaceSectionName(defaultConfigName, configName)
except Exception as e:
logger.error("Missing or access issue with config file %r with %s",
configPath, str(e))
exit(1)
#
databaseNameList = []
if args.update_chem_comp_ref:
databaseNameList.append("chem_comp")
if args.update_bird_chem_comp_ref:
databaseNameList.append("bird_chem_comp")
if args.update_chem_comp_core_ref:
databaseNameList.append("chem_comp_core")
if args.update_bird_chem_comp_core_ref:
databaseNameList.append("bird_chem_comp_core")
if args.update_bird_ref:
databaseNameList.append("bird")
if args.update_bird_family_ref:
databaseNameList.append("bird_family")
if args.update_pdbx:
databaseNameList.append("pdbx")
if args.update_pdbx_core:
databaseNameList.append("pdbx_core")
if args.update_pdbx_comp_model_core:
databaseNameList.append("pdbx_comp_model_core")
if args.update_repository_holdings:
databaseNameList.append("repository_holdings")
if args.update_entity_sequence_clusters:
databaseNameList.append("sequence_clusters")
if args.update_data_exchange:
databaseNameList.append("data_exchange")
if args.update_ihm_dev:
databaseNameList.append("ihm_dev")
if args.update_drugbank_core:
databaseNameList.append("drugbank_core")
if args.update_config_deployed:
databaseNameList = cfgOb.getList(
"DATABASE_NAMES_DEPLOYED",
sectionName="database_catalog_configuration")
dataTypingList = cfgOb.getList(
"DATATYPING_DEPLOYED",
sectionName="database_catalog_configuration")
validationLevels = cfgOb.getList(
"VALIDATION_LEVELS_DEPLOYED",
sectionName="database_catalog_configuration")
encodingTypes = cfgOb.getList(
"ENCODING_TYPES_DEPLOYED",
sectionName="database_catalog_configuration")
if args.update_config_all:
databaseNameList = cfgOb.getList(
"DATABASE_NAMES_ALL", sectionName="database_catalog_configuration")
dataTypingList = cfgOb.getList(
"DATATYPING_ALL", sectionName="database_catalog_configuration")
validationLevels = cfgOb.getList(
"VALIDATION_LEVELS_ALL",
sectionName="database_catalog_configuration")
encodingTypes = cfgOb.getList(
"ENCODING_TYPES_ALL", sectionName="database_catalog_configuration")
if args.update_config_test:
databaseNameList = cfgOb.getList(
"DATABASE_NAMES_TEST",
sectionName="database_catalog_configuration")
dataTypingList = cfgOb.getList(
"DATATYPING_TEST", sectionName="database_catalog_configuration")
validationLevels = cfgOb.getList(
"VALIDATION_LEVELS_TEST",
sectionName="database_catalog_configuration")
encodingTypes = cfgOb.getList(
"ENCODING_TYPES_TEST",
sectionName="database_catalog_configuration")
#
scnD = cfgOb.get("document_collection_names",
sectionName="document_helper_configuration")
#
databaseNameList = list(set(databaseNameList))
logger.debug("Collections %s", list(scnD.items()))
logger.debug("databaseNameList %s", databaseNameList)
if compareOnly:
schP = SchemaProvider(cfgOb, cachePath, useCache=True)
difPathList = []
for databaseName in databaseNameList:
for dataTyping in dataTypingList:
logger.debug("Building schema %s with types %s", databaseName,
dataTyping)
pth = schP.schemaDefCompare(databaseName, dataTyping)
if pth:
difPathList.append(pth)
if difPathList:
logger.info("Schema definition difference path list %r",
difPathList)
difPathList = []
for databaseName in databaseNameList:
dD = schP.makeSchemaDef(databaseName,
dataTyping="ANY",
saveSchema=False)
sD = SchemaDefAccess(dD)
for cd in sD.getCollectionInfo():
collectionName = cd["NAME"]
for encodingType in encodingTypes:
if encodingType.lower() != "json":
continue
for level in validationLevels:
pth = schP.jsonSchemaCompare(databaseName,
collectionName,
encodingType, level)
if pth:
difPathList.append(pth)
if difPathList:
logger.info("JSON schema difference path list %r", difPathList)
else:
schP = SchemaProvider(cfgOb, cachePath, useCache=False)
for databaseName in databaseNameList:
for encodingType in encodingTypes:
if encodingType == "rcsb":
for dataTyping in dataTypingList:
logger.info(
"Creating schema definition for content type %s data typing %s",
databaseName, dataTyping)
schP.makeSchemaDef(databaseName,
dataTyping=dataTyping,
saveSchema=True)
else:
if databaseName in scnD:
for dD in scnD[databaseName]:
collectionName = dD["NAME"]
for validationLevel in validationLevels:
logger.info(
"Creating %r schema for content type %s collection %s",
encodingType, databaseName, collectionName)
schP.makeSchema(databaseName,
collectionName,
encodingType=encodingType,
level=validationLevel,
saveSchema=True)
class SchemaDefAccessTests(unittest.TestCase):
def setUp(self):
self.__verbose = True
mockTopPath = os.path.join(TOPDIR, "rcsb", "mock-data")
self.__cachePath = os.path.join(TOPDIR, "CACHE")
pathConfig = os.path.join(TOPDIR, "rcsb", "db", "config",
"exdb-config-example.yml")
configName = "site_info_configuration"
self.__cfgOb = ConfigUtil(configPath=pathConfig,
defaultSectionName=configName,
mockTopPath=mockTopPath)
self.__schP = SchemaProvider(self.__cfgOb,
self.__cachePath,
useCache=True,
clearPath=False)
#
self.__startTime = time.time()
logger.debug("Starting %s at %s", self.id(),
time.strftime("%Y %m %d %H:%M:%S", time.localtime()))
def tearDown(self):
endTime = time.time()
logger.debug("Completed %s at %s (%.4f seconds)", self.id(),
time.strftime("%Y %m %d %H:%M:%S", time.localtime()),
endTime - self.__startTime)
def testAccess(self):
databaseNames = ["pdbx_core", "bird_chem_comp_core"]
dataTypingList = ["ANY", "SQL"]
for databaseName in databaseNames:
for dataTyping in dataTypingList:
self.__testAccess(databaseName, dataTyping)
def __testAccess(self, databaseName, dataTyping):
try:
sD = self.__schP.makeSchemaDef(databaseName,
dataTyping=dataTyping,
saveSchema=False)
ok = self.__testAccessors(sD)
self.assertTrue(ok)
#
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
return {}
def __testAccessors(self, schemaDef):
"""Verify data and accessor mapping -"""
sd = SchemaDefAccess(schemaDef)
logger.debug("Schema name %s", sd.getName())
logger.debug("Schema name %s", sd.getAppName())
logger.debug("Database name %s", sd.getDatabaseName())
logger.debug("Versioned database name %s",
sd.getVersionedDatabaseName())
logger.debug("Collection info %r", sd.getCollectionInfo())
for dS in sd.getDataSelectorNames():
logger.debug("Selector %s %r", dS, sd.getDataSelectors(dS))
collectionInfoL = sd.getCollectionInfo()
for dD in collectionInfoL:
collectionName = dD["NAME"]
logger.debug("Collection excluded %r",
sd.getCollectionExcluded(collectionName))
logger.debug("Collection included %r",
sd.getCollectionSelected(collectionName))
logger.debug("Collection document key attribute names %r",
sd.getDocumentKeyAttributeNames(collectionName))
schemaIdList = sd.getSchemaIdList()
for schemaId in schemaIdList:
#
aIdL = sd.getAttributeIdList(schemaId)
tObj = sd.getSchemaObject(schemaId)
attributeIdList = tObj.getAttributeIdList()
self.assertEqual(len(aIdL), len(attributeIdList))
attributeNameList = tObj.getAttributeNameList()
logger.debug("Ordered attribute Id list %s",
str(attributeIdList))
logger.debug("Ordered attribute name list %s",
str(attributeNameList))
#
mAL = tObj.getMapAttributeNameList()
logger.debug("Ordered mapped attribute name list %s", str(mAL))
mAL = tObj.getMapAttributeIdList()
logger.debug("Ordered mapped attribute id list %s", str(mAL))
cL = tObj.getMapInstanceCategoryList()
logger.debug("Mapped category list %s", str(cL))
for cV in cL:
aL = tObj.getMapInstanceAttributeList(cV)
logger.debug("Mapped attribute list in %s : %s", cV, str(aL))
return True
class ChemRefEtlWorker(object):
"""Prepare and load chemical reference data collections."""
def __init__(self,
cfgOb,
cachePath,
useCache=True,
numProc=2,
chunkSize=10,
readBackCheck=False,
documentLimit=None,
verbose=False):
self.__cfgOb = cfgOb
self.__cachePath = cachePath
self.__useCache = useCache
self.__readBackCheck = readBackCheck
self.__numProc = numProc
self.__chunkSize = chunkSize
self.__documentLimit = documentLimit
#
self.__resourceName = "MONGO_DB"
self.__verbose = verbose
self.__statusList = []
self.__schP = SchemaProvider(self.__cfgOb,
self.__cachePath,
useCache=self.__useCache)
#
def __updateStatus(self, updateId, databaseName, collectionName, status,
startTimestamp):
try:
sFlag = "Y" if status else "N"
desp = DataExchangeStatus()
desp.setStartTime(tS=startTimestamp)
desp.setObject(databaseName, collectionName)
desp.setStatus(updateId=updateId, successFlag=sFlag)
desp.setEndTime()
self.__statusList.append(desp.getStatus())
return True
except Exception as e:
logger.exception("Failing with %s", str(e))
return False
def load(self, updateId, extResource, loadType="full"):
"""Load chemical reference integrated data for the input external resource-"""
try:
self.__statusList = []
desp = DataExchangeStatus()
statusStartTimestamp = desp.setStartTime()
#
if extResource == "DrugBank":
databaseName = "drugbank_core"
configName = self.__cfgOb.getDefaultSectionName()
user = self.__cfgOb.get("_DRUGBANK_AUTH_USERNAME",
sectionName=configName)
pw = self.__cfgOb.get("_DRUGBANK_AUTH_PASSWORD",
sectionName=configName)
#
dbP = DrugBankProvider(cachePath=self.__cachePath,
useCache=self.__useCache,
username=user,
password=pw)
#
crExt = ChemRefExtractor(self.__cfgOb)
idD = crExt.getChemCompAccessionMapping(extResource)
dList = dbP.getDocuments(mapD=idD)
#
logger.info("Resource %r extracted mapped document length %d",
extResource, len(dList))
logger.debug("Objects %r", dList[:2])
sD, _, collectionList, _ = self.__schP.getSchemaInfo(
databaseName)
collectionName = collectionList[
0] if collectionList else "unassigned"
indexL = sD.getDocumentIndex(collectionName, "primary")
logger.info("Database %r collection %r index attributes %r",
databaseName, collectionName, indexL)
#
collectionVersion = sD.getCollectionVersion(collectionName)
addValues = {"_schema_version": collectionVersion}
#
addValues = {}
#
dl = DocumentLoader(
self.__cfgOb,
self.__cachePath,
self.__resourceName,
numProc=self.__numProc,
chunkSize=self.__chunkSize,
documentLimit=self.__documentLimit,
verbose=self.__verbose,
readBackCheck=self.__readBackCheck,
)
#
ok = dl.load(databaseName,
collectionName,
loadType=loadType,
documentList=dList,
indexAttributeList=indexL,
keyNames=None,
addValues=addValues)
self.__updateStatus(updateId, databaseName, collectionName, ok,
statusStartTimestamp)
return True
except Exception as e:
logger.exception("Failing with %s", str(e))
return False
def getLoadStatus(self):
return self.__statusList
class SchemaDataPrepValidateTests(unittest.TestCase):
def setUp(self):
self.__numProc = 2
# self.__fileLimit = 200
self.__fileLimit = None
self.__mockTopPath = os.path.join(TOPDIR, "rcsb", "mock-data")
self.__cachePath = os.path.join(TOPDIR, "CACHE")
self.__configPath = os.path.join(TOPDIR, "rcsb", "db", "config", "exdb-config-example-ihm.yml")
configName = "site_info_configuration"
self.__configName = configName
self.__cfgOb = ConfigUtil(configPath=self.__configPath, defaultSectionName=configName, mockTopPath=self.__mockTopPath)
self.__mU = MarshalUtil(workPath=self.__cachePath)
#self.__schP = SchemaProvider(self.__cfgOb, self.__cachePath, useCache=False, rebuildFlag=True)
self.__schP = SchemaProvider(self.__cfgOb, self.__cachePath, useCache=True)
self.__rpP = RepositoryProvider(cfgOb=self.__cfgOb, numProc=self.__numProc, fileLimit=self.__fileLimit, cachePath=self.__cachePath)
#
self.__birdRepoPath = self.__cfgOb.getPath("BIRD_REPO_PATH", sectionName=configName)
#
self.__fTypeRow = "drop-empty-attributes|drop-empty-tables|skip-max-width|convert-iterables|normalize-enums|translateXMLCharRefs"
self.__fTypeCol = "drop-empty-tables|skip-max-width|convert-iterables|normalize-enums|translateXMLCharRefs"
self.__verbose = True
#
self.__modulePathMap = self.__cfgOb.get("DICT_METHOD_HELPER_MODULE_PATH_MAP", sectionName=configName)
self.__testDirPath = os.path.join(HERE, "test-output", "pdbx-files")
self.__testIhmDirPath = os.path.join(HERE, "test-output", "ihm-files")
self.__export = True
#
#self.__extraOpts = None
# The following for extended parent/child info -
self.__extraOpts = 'addParentRefs|addPrimaryKey'
#
self.__alldatabaseNameD = {
"ihm_dev": ["ihm_dev"],
"pdbx": ["pdbx", "pdbx_ext"],
"pdbx_core": ["pdbx_core_entity", "pdbx_core_entry", "pdbx_core_assembly", "pdbx_core_entity_instance", "pdbx_core_entity_instance_validation"],
"bird": ["bird"],
"bird_family": ["family"],
"chem_comp": ["chem_comp"],
"bird_chem_comp": ["bird_chem_comp"],
"bird_chem_comp_core": ["bird_chem_comp_core"],
}
self.__databaseNameD = {
"pdbx_core": ["pdbx_core_entity", "pdbx_core_entry", "pdbx_core_assembly", "pdbx_core_entity_instance", "pdbx_core_entity_instance_validation"],
"bird_chem_comp_core": ["bird_chem_comp_core"],
}
self.__mergeContentTypeD = {"pdbx_core": ["vrpt"]}
# self.__databaseNameD = {"chem_comp_core": ["chem_comp_core"], "bird_chem_comp_core": ["bird_chem_comp_core"]}
# self.__databaseNameD = {"ihm_dev_full": ["ihm_dev_full"]}
# self.__databaseNameD = {"pdbx_core": ["pdbx_core_entity_instance_validation"]}
# self.__databaseNameD = {"pdbx_core": ["pdbx_core_entity_monomer"]}
self.__startTime = time.time()
logger.debug("Starting %s at %s", self.id(), time.strftime("%Y %m %d %H:%M:%S", time.localtime()))
def tearDown(self):
endTime = time.time()
logger.debug("Completed %s at %s (%.4f seconds)", self.id(), time.strftime("%Y %m %d %H:%M:%S", time.localtime()), endTime - self.__startTime)
def testValidateOptsRepo(self):
# schemaLevel = "min"
schemaLevel = "full"
inputPathList = None
eCount = self.__testValidateOpts(databaseNameD=self.__databaseNameD, inputPathList=inputPathList, schemaLevel=schemaLevel, mergeContentTypeD=self.__mergeContentTypeD)
logger.info("Total validation errors schema level %s : %d", schemaLevel, eCount)
self.assertLessEqual(eCount, 1)
@unittest.skip("Disable troubleshooting test")
def testValidateOptsList(self):
schemaLevel = "min"
inputPathList = self.__mU.doImport(os.path.join(HERE, "test-output", "failed-path.list"), "list")
# inputPathList = glob.glob(self.__testDirPath + "/*.cif")
if not inputPathList:
return True
databaseNameD = {"pdbx_core": ["pdbx_core_entity", "pdbx_core_entry", "pdbx_core_entity_instance", "pdbx_core_entity_instance_validation"]}
for ii, subList in enumerate(chunkList(inputPathList[::-1], 40)):
if ii < 5:
continue
eCount = self.__testValidateOpts(databaseNameD=databaseNameD, inputPathList=subList, schemaLevel=schemaLevel, mergeContentTypeD=self.__mergeContentTypeD)
logger.info("Chunk %d total validation errors schema level %s : %d", ii, schemaLevel, eCount)
# self.assertGreaterEqual(eCount, 20)
#@unittest.skip("Disable IHM troubleshooting test")
def testValidateOptsIhmRepo(self):
schemaLevel = "min"
inputPathList = None
self.__export = True
databaseNameD = {"ihm_dev_full": ["ihm_dev_full"]}
databaseNameD = {"ihm_dev": ["ihm_dev"]}
eCount = self.__testValidateOpts(databaseNameD=databaseNameD, inputPathList=inputPathList, schemaLevel=schemaLevel, mergeContentTypeD=self.__mergeContentTypeD)
logger.info("Total validation errors schema level %s : %d", schemaLevel, eCount)
# self.assertGreaterEqual(eCount, 20)
#
#@unittest.skip("Disable IHM troubleshooting test")
def testValidateOptsIhmList(self):
#schemaLevel = "full"
schemaLevel = "min"
inputPathList = glob.glob(self.__testIhmDirPath + "/*.cif")
if not inputPathList:
return True
#databaseNameD = {"ihm_dev_full": ["ihm_dev_full"]}
databaseNameD = {"ihm_dev": ["ihm_dev"]}
eCount = self.__testValidateOpts(databaseNameD=databaseNameD, inputPathList=inputPathList, schemaLevel=schemaLevel, mergeContentTypeD=self.__mergeContentTypeD)
logger.info("Total validation errors schema level %s : %d", schemaLevel, eCount)
# self.assertGreaterEqual(eCount, 20)
#
def __testValidateOpts(self, databaseNameD, inputPathList=None, schemaLevel="full", mergeContentTypeD=None):
#
eCount = 0
for databaseName in databaseNameD:
mergeContentTypes = mergeContentTypeD[databaseName] if databaseName in mergeContentTypeD else None
_ = self.__schP.makeSchemaDef(databaseName, dataTyping="ANY", saveSchema=True)
pthList = inputPathList if inputPathList else self.__rpP.getLocatorObjList(databaseName, mergeContentTypes=mergeContentTypes)
for collectionName in databaseNameD[databaseName]:
cD = self.__schP.makeSchema(databaseName, collectionName, encodingType="JSON", level=schemaLevel, saveSchema=True, extraOpts=self.__extraOpts)
#
dL, cnL = self.__testPrepDocumentsFromContainers(
pthList, databaseName, collectionName, styleType="rowwise_by_name_with_cardinality", mergeContentTypes=mergeContentTypes
)
# Raises exceptions for schema compliance.
try:
Draft4Validator.check_schema(cD)
except Exception as e:
logger.error("%s %s schema validation fails with %s", databaseName, collectionName, str(e))
#
valInfo = Draft4Validator(cD, format_checker=FormatChecker())
logger.info("Validating %d documents from %s %s", len(dL), databaseName, collectionName)
for ii, dD in enumerate(dL):
logger.debug("Schema %s collection %s document %d", databaseName, collectionName, ii)
try:
cCount = 0
#for error in sorted(valInfo.iter_errors(dD), key=str):
# logger.info("schema %s collection %s (%s) path %s error: %s", databaseName, collectionName, cnL[ii], error.path, error.message)
# logger.debug("Failing document %d : %r", ii, list(dD.items()))
# eCount += 1
# cCount += 1
#if cCount > 0:
# logger.info("schema %s collection %s container %s error count %d", databaseName, collectionName, cnL[ii], cCount)
except Exception as e:
logger.exception("Validation processing error %s", str(e))
return eCount
def __testPrepDocumentsFromContainers(self, inputPathList, databaseName, collectionName, styleType="rowwise_by_name_with_cardinality", mergeContentTypes=None):
"""Test case - create loadable PDBx data from repository files
"""
try:
sd, _, _, _ = self.__schP.getSchemaInfo(databaseName)
#
dP = DictionaryApiProviderWrapper(self.__cfgOb, self.__cachePath, useCache=False)
dictApi = dP.getApiByName(databaseName)
rP = DictMethodResourceProvider(self.__cfgOb, configName=self.__configName, cachePath=self.__cachePath, siftsAbbreviated="TEST")
dmh = DictMethodRunner(dictApi, modulePathMap=self.__modulePathMap, resourceProvider=rP)
#
dtf = DataTransformFactory(schemaDefAccessObj=sd, filterType=self.__fTypeRow)
sdp = SchemaDefDataPrep(schemaDefAccessObj=sd, dtObj=dtf, workPath=self.__cachePath, verbose=self.__verbose)
containerList = self.__rpP.getContainerList(inputPathList)
for container in containerList:
cName = container.getName()
logger.debug("Processing container %s", cName)
dmh.apply(container)
if self.__export:
savePath = os.path.join(HERE, "test-output", cName + "-with-method.cif")
#self.__mU.doExport(savePath, [container], fmt="mmcif")
#
tableIdExcludeList = sd.getCollectionExcluded(collectionName)
tableIdIncludeList = sd.getCollectionSelected(collectionName)
sliceFilter = sd.getCollectionSliceFilter(collectionName)
sdp.setSchemaIdExcludeList(tableIdExcludeList)
sdp.setSchemaIdIncludeList(tableIdIncludeList)
#
docList, containerNameList, _ = sdp.processDocuments(
containerList, styleType=styleType, filterType=self.__fTypeRow, dataSelectors=["PUBLIC_RELEASE"], sliceFilter=sliceFilter, collectionName=collectionName
)
docList = sdp.addDocumentPrivateAttributes(docList, collectionName)
docList = sdp.addDocumentSubCategoryAggregates(docList, collectionName)
#
mergeS = "-".join(mergeContentTypes) if mergeContentTypes else ""
if self.__export and docList:
# for ii, doc in enumerate(docList[:1]):
for ii, doc in enumerate(docList):
cn = containerNameList[ii]
fp = os.path.join(HERE, "test-output", "prep-%s-%s-%s-%s.json" % (cn, databaseName, collectionName, mergeS))
self.__mU.doExport(fp, [doc], fmt="json", indent=3)
logger.debug("Exported %r", fp)
#
return docList, containerNameList
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
class SchemaDefBuildTests(unittest.TestCase):
def setUp(self):
self.__verbose = True
mockTopPath = os.path.join(TOPDIR, "rcsb", "mock-data")
pathConfig = os.path.join(TOPDIR, "rcsb", "db", "config",
"exdb-config-example.yml")
configName = "site_info_configuration"
self.__cachePath = os.path.join(TOPDIR, "CACHE")
self.__cfgOb = ConfigUtil(configPath=pathConfig,
defaultSectionName=configName,
mockTopPath=mockTopPath)
self.__schP = SchemaProvider(self.__cfgOb,
self.__cachePath,
useCache=False)
#
self.__validationLevels = self.__cfgOb.getList(
"VALIDATION_LEVELS_TEST",
sectionName="database_catalog_configuration")
self.__encodingTypes = self.__cfgOb.getList(
"ENCODING_TYPES_TEST",
sectionName="database_catalog_configuration")
#
buildAll = True
if buildAll:
self.__databaseNameList = self.__cfgOb.getList(
"DATABASE_NAMES_DEPLOYED",
sectionName="database_catalog_configuration")
self.__dataTypingList = self.__cfgOb.getList(
"DATATYPING_DEPLOYED",
sectionName="database_catalog_configuration")
#
else:
self.__databaseNameList = self.__cfgOb.getList(
"DATABASE_NAMES_TEST",
sectionName="database_catalog_configuration")
# self.__databaseNameList = ["repository_holdings"]
self.__dataTypingList = self.__cfgOb.getList(
"DATATYPING_TEST",
sectionName="database_catalog_configuration")
# self.__databaseNameList = ["sequence_clusters"]
self.__saveSchema = True
self.__compareDefSchema = False
self.__compareSchema = False
#
self.__startTime = time.time()
logger.debug("Starting %s at %s", self.id(),
time.strftime("%Y %m %d %H:%M:%S", time.localtime()))
def tearDown(self):
endTime = time.time()
logger.debug("Completed %s at %s (%.4f seconds)", self.id(),
time.strftime("%Y %m %d %H:%M:%S", time.localtime()),
endTime - self.__startTime)
def testBuildSchemaDefs(self):
try:
for databaseName in self.__databaseNameList:
for dataTyping in self.__dataTypingList:
logger.debug("Building schema %s with types %s",
databaseName, dataTyping)
self.__schP.makeSchemaDef(databaseName,
dataTyping=dataTyping,
saveSchema=self.__saveSchema)
if self.__compareDefSchema:
self.__schP.schemaDefCompare(databaseName, dataTyping)
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def testBuildCollectionSchema(self):
schemaDifPathList = []
for databaseName in self.__databaseNameList:
dD = self.__schP.makeSchemaDef(databaseName,
dataTyping="ANY",
saveSchema=False)
sD = SchemaDefAccess(dD)
for cd in sD.getCollectionInfo():
collectionName = cd["NAME"]
for encodingType in self.__encodingTypes:
if encodingType.lower() == "rcsb":
continue
for level in self.__validationLevels:
self.__schP.makeSchema(databaseName,
collectionName,
encodingType=encodingType,
level=level,
saveSchema=self.__saveSchema)
if self.__compareSchema and encodingType.lower(
) == "json":
pth = self.__schP.jsonSchemaCompare(
databaseName, collectionName, encodingType,
level)
if pth:
schemaDifPathList.append(pth)
if schemaDifPathList:
logger.info("Path dif list %r", schemaDifPathList)
def testCompareSchema(self):
databaseName = "pdbx_core"
collectionName = "pdbx_core_entry"
encodingType = "json"
level = "full"
#
oldPath = os.path.join(
HERE, "test-saved-output",
"json-full-db-pdbx_core-col-pdbx_core_entry.json")
mU = MarshalUtil(workPath=os.path.join(HERE, "test-output"))
sOld = mU.doImport(oldPath, fmt="json")
sNew = self.__schP.makeSchema(databaseName,
collectionName,
encodingType=encodingType,
level=level)
numDif, difD = self.__schP.schemaCompare(sOld, sNew)
logger.debug("numDiffs %d", numDif)
self.assertGreaterEqual(numDif, 141)
self.assertGreaterEqual(len(difD["changed"]), 160)
logger.debug("difD %r", difD)
@unittest.skip("Deprecated test")
def testCompareSchemaCategories(self):
"""Compare common categories across schema definitions."""
try:
sdCc = SchemaDefAccess(
self.__schP.makeSchemaDef("chem_comp_core",
dataTyping="ANY",
saveSchema=False))
sdBcc = SchemaDefAccess(
self.__schP.makeSchemaDef("bird_chem_comp_core",
dataTyping="ANY",
saveSchema=False))
#
logger.info("")
for schemaId in ["CHEM_COMP", "PDBX_CHEM_COMP_AUDIT"]:
atCcL = sdCc.getAttributeIdList(schemaId)
atBCcL = sdBcc.getAttributeIdList(schemaId)
logger.debug("%s attributes (%d) %r", schemaId, len(atCcL),
atCcL)
logger.debug("%s attributes (%d) %r", schemaId, len(atBCcL),
atBCcL)
sDif = set(atCcL) - set(atBCcL)
if sDif:
logger.info("For %s attribute differences %r", schemaId,
sDif)
self.assertEqual(len(sDif), 0)
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def testBuildColSchemaWithRefs(self):
for databaseName in ["ihm_dev_full"]:
dD = self.__schP.makeSchemaDef(databaseName,
dataTyping="ANY",
saveSchema=False)
sD = SchemaDefAccess(dD)
for cd in sD.getCollectionInfo():
collectionName = cd["NAME"]
for schemaType in self.__encodingTypes:
if schemaType.lower() == "rcsb":
continue
for level in self.__validationLevels:
self.__schP.makeSchema(
databaseName,
collectionName,
encodingType=schemaType,
level=level,
saveSchema=True,
extraOpts="addParentRefs|addPrimaryKey")
class ChemRefDataPrepValidateTests(unittest.TestCase):
def setUp(self):
self.__verbose = True
#
self.__mockTopPath = os.path.join(TOPDIR, "rcsb", "mock-data")
self.__pathConfig = os.path.join(TOPDIR, "rcsb", "db", "config",
"exdb-config-example.yml")
self.__cachePath = os.path.join(TOPDIR, "CACHE")
#
self.__configName = "site_info_configuration"
self.__cfgOb = ConfigUtil(configPath=self.__pathConfig,
defaultSectionName=self.__configName,
mockTopPath=self.__mockTopPath)
self.__schP = SchemaProvider(self.__cfgOb,
self.__cachePath,
useCache=True)
#
self.__startTime = time.time()
logger.debug("Starting %s at %s", self.id(),
time.strftime("%Y %m %d %H:%M:%S", time.localtime()))
def tearDown(self):
endTime = time.time()
logger.debug("Completed %s at %s (%.4f seconds)", self.id(),
time.strftime("%Y %m %d %H:%M:%S", time.localtime()),
endTime - self.__startTime)
def testValidateFull(self):
self.__validateChemRef("DrugBank", schemaLevel="full")
def __validateChemRef(self, extResource, schemaLevel="full"):
eCount = 0
if extResource == "DrugBank":
schemaName = "drugbank_core"
collectionNames = ["drugbank_core"]
user = self.__cfgOb.get("_DRUGBANK_AUTH_USERNAME",
sectionName=self.__configName)
pw = self.__cfgOb.get("_DRUGBANK_AUTH_PASSWORD",
sectionName=self.__configName)
# cacheDir = self.__cfgOb.get("DRUGBANK_CACHE_DIR", sectionName=self.__configName)
dbP = DrugBankProvider(cachePath=self.__cachePath,
useCache=True,
username=user,
password=pw)
# idD = dbP.getMapping()
# crExt = ChemRefExtractor(self.__cfgOb)
# idD = crExt.getChemCompAccesionMapping(extResource)
dList = dbP.getDocuments()
logger.info("Validating %d Drugbank documents", len(dList))
eCount = self.__validate(schemaName,
collectionNames,
dList,
schemaLevel=schemaLevel)
return eCount
def __validate(self,
databaseName,
collectionNames,
dList,
schemaLevel="full"):
eCount = 0
for collectionName in collectionNames:
_ = self.__schP.makeSchemaDef(databaseName,
dataTyping="ANY",
saveSchema=True)
cD = self.__schP.makeSchema(databaseName,
collectionName,
encodingType="JSON",
level=schemaLevel,
saveSchema=True)
# Raises exceptions for schema compliance.
Draft4Validator.check_schema(cD)
#
valInfo = Draft4Validator(cD, format_checker=FormatChecker())
for ii, dD in enumerate(dList):
logger.debug("Database %s collection %s document %d",
databaseName, collectionName, ii)
try:
cCount = 0
for error in sorted(valInfo.iter_errors(dD), key=str):
logger.info(
"database %s collection %s path %s error: %s",
databaseName, collectionName, error.path,
error.message)
logger.info(">>> failing object is %r", dD)
eCount += 1
cCount += 1
#
logger.debug("database %s collection %s count %d",
databaseName, collectionName, cCount)
except Exception as e:
logger.exception("Validation error %s", str(e))
return eCount
class SchemaDefDataPrepTests(unittest.TestCase):
def __init__(self, methodName="runTest"):
super(SchemaDefDataPrepTests, self).__init__(methodName)
self.__loadPathList = []
self.__verbose = True
def setUp(self):
self.__isMac = platform.system() == "Darwin"
self.__excludeType = None if self.__isMac else "optional"
self.__numProc = 2
mockTopPath = os.path.join(TOPDIR, "rcsb", "mock-data")
self.__cachePath = os.path.join(TOPDIR, "CACHE")
self.__outputPath = os.path.join(HERE, "test-output")
self.__savedOutputPath = os.path.join(HERE, "test-saved-output")
configPath = os.path.join(TOPDIR, "rcsb", "db", "config",
"exdb-config-example.yml")
configName = "site_info_configuration"
self.__configName = configName
self.__cfgOb = ConfigUtil(configPath=configPath,
defaultSectionName=configName,
mockTopPath=mockTopPath)
self.__mU = MarshalUtil(workPath=self.__cachePath)
self.__discoveryMode = self.__cfgOb.get("DISCOVERY_MODE",
sectionName=configName,
default="local")
self.__fileLimit = 100 if self.__discoveryMode == "local" else 10
self.__schP = SchemaProvider(self.__cfgOb,
self.__cachePath,
useCache=True)
self.__rpP = RepositoryProvider(cfgOb=self.__cfgOb,
numProc=self.__numProc,
fileLimit=self.__fileLimit,
cachePath=self.__cachePath)
#
#
self.__fTypeRow = "drop-empty-attributes|drop-empty-tables|skip-max-width|convert-iterables|normalize-enums|translateXMLCharRefs"
self.__fTypeCol = "drop-empty-tables|skip-max-width|convert-iterables|normalize-enums|translateXMLCharRefs"
self.__chemCompMockLen = 24
self.__pdbxMockLen = 30
# removes timestamped data items to allow diffs.)
excludeExtras = ["rcsb_load_status"]
# excludeExtras = []
#
self.__verbose = True
self.__modulePathMap = self.__cfgOb.get(
"DICT_METHOD_HELPER_MODULE_PATH_MAP", sectionName=configName)
#
self.__exportFlag = True
self.__diffFlag = False
#
self.__simpleTestCaseList = [
{
"contentType": "chem_comp",
"mockLength": self.__chemCompMockLen,
"filterType": self.__fTypeRow,
"styleType": "rowwise_by_name",
"mergeContentTypes": None,
"rejectLength": 2,
},
{
"contentType": "chem_comp",
"mockLength": self.__chemCompMockLen,
"filterType": self.__fTypeRow,
"styleType": "rowwise_no_name",
"mergeContentTypes": None,
"rejectLength": 2,
},
{
"contentType": "chem_comp",
"mockLength": self.__chemCompMockLen,
"filterType": self.__fTypeCol,
"styleType": "columnwise_by_name",
"mergeContentTypes": None,
"rejectLength": 2,
},
{
"contentType": "chem_comp",
"mockLength": self.__chemCompMockLen,
"filterType": self.__fTypeRow,
"styleType": "rowwise_by_name",
"mergeContentTypes": None,
"rejectLength": 2,
},
{
"contentType": "pdbx_core",
"mockLength": self.__pdbxMockLen,
"filterType": self.__fTypeRow,
"styleType": "rowwise_by_name",
"mergeContentTypes": None,
"rejectLength": 0,
},
]
#
self.__fullTestCaseList = [
{
"contentType": "pdbx_core",
"mockLength": self.__pdbxMockLen,
"filterType": self.__fTypeRow,
"styleType": "rowwise_by_name_with_cardinality",
"mergeContentTypes": ["vrpt"],
"rejectLength": 0,
"excludeExtras": excludeExtras,
},
{
"contentType": "bird_chem_comp_core",
"mockLength": self.__chemCompMockLen,
"filterType": self.__fTypeRow,
"styleType": "rowwise_by_name_with_cardinality",
"mergeContentTypes": None,
"rejectLength": 2,
"excludeExtras": excludeExtras,
},
]
#
self.__fullTestCaseListA = [
{
"contentType": "pdbx_core",
"mockLength": self.__pdbxMockLen,
"filterType": self.__fTypeRow,
"styleType": "rowwise_by_name_with_cardinality",
"mergeContentTypes": ["vrpt"],
"rejectLength": 0,
"excludeExtras": excludeExtras,
},
]
#
self.__startTime = time.time()
logger.debug("Starting %s at %s", self.id(),
time.strftime("%Y %m %d %H:%M:%S", time.localtime()))
def tearDown(self):
unitS = "MB" if platform.system() == "Darwin" else "GB"
rusageMax = resource.getrusage(resource.RUSAGE_SELF).ru_maxrss
logger.info("Maximum resident memory size %.4f %s", rusageMax / 1.0e6,
unitS)
endTime = time.time()
logger.info("Completed %s at %s (%.4f seconds)", self.id(),
time.strftime("%Y %m %d %H:%M:%S", time.localtime()),
endTime - self.__startTime)
def __timeStep(self, msg):
endTime = time.time()
logger.info("Completed %s at %s (%.4f seconds)", msg,
time.strftime("%Y %m %d %H:%M:%S", time.localtime()),
endTime - self.__startTime)
def testSimpleSchemaDefDataPrep(self):
for tcD in self.__simpleTestCaseList:
rejectLength = 0 if self.__discoveryMode == "remote" else tcD[
"rejectLength"]
mockLength = self.__fileLimit if self.__discoveryMode == "remote" else tcD[
"mockLength"]
if tcD["contentType"] == "bird_chem_comp_core" and self.__discoveryMode == "remote":
logger.info("Skipping %r in discovery mode %r",
tcD["contentType"], self.__discoveryMode)
continue
self.__simpleSchemaDataPrep(
tcD["contentType"],
tcD["filterType"],
tcD["styleType"],
mockLength,
rejectLength=rejectLength,
mergeContentTypes=tcD["mergeContentTypes"])
def testFullSchemaDefDataPrep(self):
for tcD in self.__fullTestCaseList:
rejectLength = 0 if self.__discoveryMode == "remote" else tcD[
"rejectLength"]
mockLength = self.__fileLimit if self.__discoveryMode == "remote" else tcD[
"mockLength"]
if tcD["contentType"] == "bird_chem_comp_core" and self.__discoveryMode == "remote":
logger.info("Skipping %r in discovery mode %r",
tcD["contentType"], self.__discoveryMode)
continue
self.__fullSchemaDataPrep(
tcD["contentType"],
tcD["filterType"],
tcD["styleType"],
mockLength,
rejectLength=rejectLength,
mergeContentTypes=tcD["mergeContentTypes"],
excludeExtras=tcD["excludeExtras"],
)
def __simpleSchemaDataPrep(self,
contentType,
filterType,
styleType,
mockLength,
rejectLength=0,
dataSelectors=None,
mergeContentTypes=None):
"""Internal method for preparing file-based data NOT requiring dynamic methods, slicing, or key injection.
Args:
contentType (str): Content type name
filterType (str): List of data processing options (separated by '|') (e.g. "drop-empty-attributes|drop-empty-tables|skip-max-width|...)
styleType (str): organization of output document (e.g. rowise-by-name)
mockLength (int): Expected length of the test data for the input content type
rejectLength (int, optional): number of input data sets rejected by the dataselection criteria. Defaults to 0.
dataSelectors (list of str, optional): data selection criteria. Defaults to None.
mergeContentTypes (list of str, optional): list content types to merge with the input data set. Defaults to None. (e.g. ['vrpt'])
"""
try:
dataSelectors = dataSelectors if dataSelectors else [
"PUBLIC_RELEASE"
]
dD = self.__schP.makeSchemaDef(contentType,
dataTyping="ANY",
saveSchema=True)
_ = SchemaDefAccess(dD)
inputPathList = self.__rpP.getLocatorObjList(
contentType=contentType, mergeContentTypes=mergeContentTypes)
sd, _, _, _ = self.__schP.getSchemaInfo(databaseName=contentType,
dataTyping="ANY")
dtf = DataTransformFactory(schemaDefAccessObj=sd,
filterType=filterType)
sdp = SchemaDefDataPrep(schemaDefAccessObj=sd,
dtObj=dtf,
workPath=self.__cachePath,
verbose=self.__verbose)
#
logger.debug("For %s mock length %d length of path list %d\n",
contentType, mockLength, len(inputPathList))
self.assertGreaterEqual(len(inputPathList), mockLength)
tableDataDictList, containerNameList, rejectList = sdp.fetchDocuments(
inputPathList,
styleType=styleType,
filterType=filterType,
dataSelectors=dataSelectors)
logger.debug(
"For %s mock length %d reject length %d length of tddl list %d\n",
contentType, mockLength, rejectLength, len(tableDataDictList))
self.assertGreaterEqual(len(tableDataDictList),
mockLength - rejectLength)
self.assertGreaterEqual(len(containerNameList),
mockLength - rejectLength)
if rejectList:
logger.debug("For %s rejecting components %r", contentType,
rejectList)
#
self.assertEqual(len(rejectList), rejectLength)
fName = "simple-prep-%s-%s.json" % (contentType, styleType)
if self.__exportFlag:
fPath = os.path.join(self.__outputPath, fName)
self.__mU.doExport(fPath,
tableDataDictList,
fmt="json",
indent=3)
if self.__diffFlag:
fPath = os.path.join(self.__savedOutputPath, fName)
refDocList = self.__mU.doImport(fPath, fmt="json")
self.assertEqual(len(refDocList), len(tableDataDictList))
#
jD = diff(refDocList,
tableDataDictList,
syntax="explicit",
marshal=True)
if jD:
_, fn = os.path.split(fPath)
bn, _ = os.path.splitext(fn)
fPath = os.path.join(self.__outputPath, bn + "-diff.json")
logger.debug("jsondiff for %s %s = \n%s", contentType,
styleType,
pprint.pformat(jD, indent=3, width=100))
self.__mU.doExport(fPath, jD, fmt="json", indent=3)
self.assertEqual(len(jD), 0)
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def __logDocumentOrder(self, docList):
for doc in docList:
logger.debug("keys %r", list(doc.keys()))
def __filterDocuments(self, docList, excludeList=None):
excludeList = excludeList if excludeList else []
for doc in docList:
for excl in excludeList:
if excl in doc:
del doc[excl]
def __fullSchemaDataPrep(self,
contentType,
filterType,
styleType,
mockLength,
rejectLength=0,
dataSelectors=None,
mergeContentTypes=None,
excludeExtras=None):
"""Internal method for preparing file-based data requiring dynamic methods, slicing, or key injection.
Args:
contentType (str): Content type name
filterType (str): List of data processing options (separated by '|') (e.g. "drop-empty-attributes|drop-empty-tables|skip-max-width|...)
styleType (str): organization of output document (e.g. rowise-by-name)
mockLength (int): Expected length of the test data for the input content type
rejectLength (int, optional): number of input data sets rejected by the dataselection criteria. Defaults to 0.
dataSelectors (list of str, optional): data selection criteria. Defaults to None.
mergeContentTypes (list of str, optional): list content types to merge with the input data set. Defaults to None. (e.g. ['vrpt'])
"""
try:
excludeExtras = excludeExtras if excludeExtras else []
_ = mockLength
_ = rejectLength
dD = self.__schP.makeSchemaDef(contentType,
dataTyping="ANY",
saveSchema=True)
_ = SchemaDefAccess(dD)
inputPathList = self.__rpP.getLocatorObjList(
contentType=contentType, mergeContentTypes=mergeContentTypes)
sd, _, collectionNameList, _ = self.__schP.getSchemaInfo(
databaseName=contentType, dataTyping="ANY")
#
dP = DictionaryApiProviderWrapper(self.__cachePath,
cfgOb=self.__cfgOb,
configName=self.__configName,
useCache=True)
dictApi = dP.getApiByName(contentType)
#
rP = DictMethodResourceProvider(
self.__cfgOb,
configName=self.__configName,
cachePath=self.__cachePath,
restoreUseStash=False,
restoreUseGit=True,
providerTypeExclude=self.__excludeType,
)
dmh = DictMethodRunner(dictApi,
modulePathMap=self.__modulePathMap,
resourceProvider=rP)
#
dtf = DataTransformFactory(schemaDefAccessObj=sd,
filterType=filterType)
sdp = SchemaDefDataPrep(schemaDefAccessObj=sd,
dtObj=dtf,
workPath=self.__cachePath,
verbose=self.__verbose)
containerList = self.__rpP.getContainerList(inputPathList)
for container in containerList:
cName = container.getName()
logger.debug("Processing container %s", cName)
dmh.apply(container)
#
for collectionName in collectionNameList:
tableIdExcludeList = sd.getCollectionExcluded(collectionName)
tableIdIncludeList = sd.getCollectionSelected(collectionName)
sliceFilter = sd.getCollectionSliceFilter(collectionName)
sdp.setSchemaIdExcludeList(tableIdExcludeList)
sdp.setSchemaIdIncludeList(tableIdIncludeList)
#
docList, _, _ = sdp.processDocuments(
containerList,
styleType=styleType,
sliceFilter=sliceFilter,
filterType=filterType,
dataSelectors=dataSelectors,
collectionName=collectionName)
docList = sdp.addDocumentPrivateAttributes(
docList, collectionName)
docList = sdp.addDocumentSubCategoryAggregates(
docList, collectionName)
# Special exclusions for the test harness. (removes timestamped data items to allow diffs.)
self.__filterDocuments(docList, excludeExtras)
mergeS = "-".join(
mergeContentTypes) if mergeContentTypes else ""
fName = "full-prep-%s-%s-%s-%s.json" % (
contentType, collectionName, mergeS, styleType)
if self.__exportFlag:
self.__logDocumentOrder(docList)
fPath = os.path.join(self.__outputPath, fName)
self.__mU.doExport(fPath, docList, fmt="json", indent=3)
logger.debug("Exported %r", fPath)
#
if self.__diffFlag:
fPath = os.path.join(self.__savedOutputPath, fName)
refDocList = self.__mU.doImport(fPath, fmt="json")
self.assertEqual(len(refDocList), len(docList))
logger.debug("For %s %s len refDocList %d", contentType,
collectionName, len(refDocList))
logger.debug("For %s %s len docList %d", contentType,
collectionName, len(docList))
jD = diff(refDocList,
docList,
syntax="explicit",
marshal=True)
if jD:
_, fn = os.path.split(fPath)
bn, _ = os.path.splitext(fn)
fPath = os.path.join(self.__outputPath,
bn + "-diff.json")
logger.debug("jsondiff for %s %s = \n%s", contentType,
collectionName,
pprint.pformat(jD, indent=3, width=100))
self.__mU.doExport(fPath, jD, fmt="json", indent=3)
self.assertEqual(len(jD), 0)
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
class ClusterDataPrepValidateTests(unittest.TestCase):
def setUp(self):
self.__verbose = True
#
self.__mockTopPath = os.path.join(TOPDIR, "rcsb", "mock-data")
self.__pathConfig = os.path.join(TOPDIR, "rcsb", "db", "config", "exdb-config-example.yml")
self.__cachePath = os.path.join(TOPDIR, "CACHE")
self.__updateId = "2018_25"
#
configName = "site_info_configuration"
self.__cfgOb = ConfigUtil(configPath=self.__pathConfig, defaultSectionName=configName, mockTopPath=self.__mockTopPath)
self.__schP = SchemaProvider(self.__cfgOb, self.__cachePath, useCache=True)
#
self.__sandboxPath = self.__cfgOb.getPath("RCSB_EXCHANGE_SANDBOX_PATH", sectionName=configName)
#
self.__dataSetId = "2018_23"
self.__pathClusterData = self.__cfgOb.getPath("RCSB_SEQUENCE_CLUSTER_DATA_PATH", sectionName=configName)
# self.__levels = ['100', '95', '90', '70', '50', '30']
self.__levels = ["100"]
#
self.__startTime = time.time()
logger.debug("Starting %s at %s", self.id(), time.strftime("%Y %m %d %H:%M:%S", time.localtime()))
def tearDown(self):
endTime = time.time()
logger.debug("Completed %s at %s (%.4f seconds)", self.id(), time.strftime("%Y %m %d %H:%M:%S", time.localtime()), endTime - self.__startTime)
def testValidateOptsStrict(self):
updateId = self.__updateId
validationLevel = "full"
eCount = self.__testValidateOpts(updateId, validationLevel=validationLevel)
logger.info("Total validation errors validation level %s : %d", validationLevel, eCount)
self.assertTrue(eCount <= 1)
def __testValidateOpts(self, updateId, validationLevel="full"):
_ = updateId
databaseNames = ["sequence_clusters"]
collectionNames = {"sequence_clusters": ["cluster_provenance", "cluster_members", "entity_members"]}
#
eCount = 0
for databaseName in databaseNames:
for collectionName in collectionNames[databaseName]:
_ = self.__schP.makeSchemaDef(databaseName, dataTyping="ANY", saveSchema=True)
cD = self.__schP.makeSchema(databaseName, collectionName, encodingType="JSON", level=validationLevel, saveSchema=True)
#
dL = self.__getSequenceClusterData(collectionName, levels=self.__levels, dataSetId=self.__dataSetId, dataLocator=self.__pathClusterData)
# Raises exceptions for schema compliance.
Draft4Validator.check_schema(cD)
#
valInfo = Draft4Validator(cD, format_checker=FormatChecker())
for _, dD in enumerate(dL):
# logger.debug("Schema %s collection %s document %d" % (schemaName, collectionName, ii))
try:
cCount = 0
for error in sorted(valInfo.iter_errors(dD), key=str):
logger.info("schema %s collection %s path %s error: %s", databaseName, collectionName, error.path, error.message)
logger.info(">>> failing object is %r", dD)
eCount += 1
cCount += 1
#
logger.debug("schema %s collection %s count %d", databaseName, collectionName, cCount)
except Exception as e:
logger.exception("Validation error %s", str(e))
return eCount
def __fetchProvenance(self):
"""Test case for fetching a provenance dictionary content."""
try:
provKeyName = "rcsb_entity_sequence_cluster_prov"
provU = ProvenanceProvider(self.__cfgOb, self.__cachePath, useCache=True)
pD = provU.fetch()
return pD[provKeyName] if provKeyName in pD else {}
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def __getSequenceClusterData(self, collectionName, dataSetId=None, dataLocator=None, levels=None):
"""Test extraction on an example sequence cluster data set."""
try:
#
if collectionName == "cluster_provenance":
return [self.__fetchProvenance()]
#
entitySchemaName = "rcsb_entity_sequence_cluster_list"
clusterSchemaName = "rcsb_entity_sequence_cluster_identifer_list"
cdp = ClusterDataPrep(workPath=self.__cachePath, entitySchemaName=entitySchemaName, clusterSchemaName=clusterSchemaName)
cifD, docBySequenceD, docByClusterD = cdp.extract(dataSetId, clusterSetLocator=dataLocator, levels=levels, clusterType="entity")
self.assertEqual(len(cifD), 1)
self.assertEqual(len(docBySequenceD), 1)
self.assertEqual(len(docByClusterD), 1)
if collectionName == "entity_members":
return docBySequenceD[entitySchemaName]
elif collectionName == "cluster_members":
return docByClusterD[clusterSchemaName]
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
return None
class SchemaDefCompareTests(unittest.TestCase):
skipFlag = True
def setUp(self):
self.__verbose = True
mockTopPath = os.path.join(TOPDIR, "rcsb", "mock-data")
pathConfig = os.path.join(TOPDIR, "rcsb", "db", "config",
"exdb-config-example.yml")
configName = "site_info_configuration"
self.__cachePath = os.path.join(TOPDIR, "CACHE")
self.__cfgOb = ConfigUtil(configPath=pathConfig,
defaultSectionName=configName,
mockTopPath=mockTopPath)
self.__schP = SchemaProvider(self.__cfgOb,
self.__cachePath,
useCache=True)
#
self.__validationLevels = self.__cfgOb.getList(
"VALIDATION_LEVELS_TEST",
sectionName="database_catalog_configuration")
self.__encodingTypes = self.__cfgOb.getList(
"ENCODING_TYPES_TEST",
sectionName="database_catalog_configuration")
#
buildAll = True
if buildAll:
self.__databaseNameList = self.__cfgOb.getList(
"DATABASE_NAMES_DEPLOYED",
sectionName="database_catalog_configuration")
self.__dataTypingList = self.__cfgOb.getList(
"DATATYPING_DEPLOYED",
sectionName="database_catalog_configuration")
#
else:
self.__databaseNameList = self.__cfgOb.getList(
"DATABASE_NAMES_TEST",
sectionName="database_catalog_configuration")
# self.__databaseNameList = ["repository_holdings"]
self.__dataTypingList = self.__cfgOb.getList(
"DATATYPING_TEST",
sectionName="database_catalog_configuration")
#
self.__startTime = time.time()
logger.debug("Starting %s at %s", self.id(),
time.strftime("%Y %m %d %H:%M:%S", time.localtime()))
def tearDown(self):
endTime = time.time()
logger.debug("Completed %s at %s (%.4f seconds)", self.id(),
time.strftime("%Y %m %d %H:%M:%S", time.localtime()),
endTime - self.__startTime)
@unittest.skipIf(skipFlag, "Troubleshooting test")
def testCompareSchemaDefs(self):
try:
difPathList = []
for databaseName in self.__databaseNameList:
for dataTyping in self.__dataTypingList:
logger.debug("Building schema %s with types %s",
databaseName, dataTyping)
pth = self.__schP.schemaDefCompare(databaseName,
dataTyping)
if pth:
difPathList.append(pth)
if difPathList:
logger.info("Schema definition difference path list %r",
[os.path.split(pth)[1] for pth in difPathList])
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
@unittest.skipIf(skipFlag, "Troubleshooting test")
def testCompareCollectionSchema(self):
try:
difPathList = []
for databaseName in self.__databaseNameList:
dD = self.__schP.makeSchemaDef(databaseName,
dataTyping="ANY",
saveSchema=False)
sD = SchemaDefAccess(dD)
for cd in sD.getCollectionInfo():
collectionName = cd["NAME"]
for encodingType in self.__encodingTypes:
if encodingType.lower() != "json":
continue
for level in self.__validationLevels:
pth = self.__schP.jsonSchemaCompare(
databaseName, collectionName, encodingType,
level)
if pth:
difPathList.append(pth)
if difPathList:
logger.info("JSON schema difference path list %r",
[os.path.split(pth)[1] for pth in difPathList])
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
class SchemaDefLoadercrateDbMultiTests(unittest.TestCase):
def __init__(self, methodName="runTest"):
super(SchemaDefLoadercrateDbMultiTests, self).__init__(methodName)
self.__verbose = True
self.__createFlag = True
def setUp(self):
self.__verbose = True
self.__numProc = 2
self.__fileLimit = 100
self.__chunkSize = 0
self.__workPath = os.path.join(HERE, "test-output")
self.__mockTopPath = os.path.join(TOPDIR, "rcsb", "mock-data")
configPath = os.path.join(TOPDIR, "rcsb", "db", "config",
"exdb-config-example.yml")
configName = "site_info_configuration"
self.__cfgOb = ConfigUtil(configPath=configPath,
defaultSectionName=configName)
self.__resourceName = "CRATE_DB"
self.__schP = SchemaProvider(self.__cfgOb,
self.__workPath,
useCache=True)
self.__rpP = RepositoryProvider(cfgOb=self.__cfgOb,
numProc=self.__numProc,
fileLimit=self.__fileLimit,
cachePath=self.__workPath)
#
#
self.__tableIdSkipD = {
"ATOM_SITE": True,
"ATOM_SITE_ANISOTROP": True,
"__LOAD_STATUS__": True
}
self.__ioObj = IoAdapter(verbose=self.__verbose)
#
self.__startTime = time.time()
logger.debug("Starting %s at %s", self.id(),
time.strftime("%Y %m %d %H:%M:%S", time.localtime()))
def tearDown(self):
endTime = time.time()
logger.debug("Completed %s at %s (%.4f seconds)", self.id(),
time.strftime("%Y %m %d %H:%M:%S", time.localtime()),
endTime - self.__startTime)
def testConnection(self):
try:
with Connection(cfgOb=self.__cfgOb,
resourceName=self.__resourceName) as client:
self.assertNotEqual(client, None)
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def testSchemaCreate(self):
"""Create table schema (live) for BIRD, chemical component, and PDBx data."""
try:
sd, _, _, _ = self.__schP.getSchemaInfo("bird")
ret = self.__schemaCreate(schemaDefObj=sd)
self.assertEqual(ret, True)
#
sd, _, _, _ = self.__schP.getSchemaInfo("chem_comp")
ret = self.__schemaCreate(schemaDefObj=sd)
self.assertEqual(ret, True)
#
sd, _, _, _ = self.__schP.getSchemaInfo("pdbx")
ret = self.__schemaCreate(schemaDefObj=sd)
self.assertEqual(ret, True)
#
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def testSchemaRemove(self):
"""Remove table schema (live) for BIRD, chemical component, and PDBx data."""
try:
sd, _, _, _ = self.__schP.getSchemaInfo("bird")
ret = self.__schemaRemove(schemaDefObj=sd)
self.assertEqual(ret, True)
#
sd, _, _, _ = self.__schP.getSchemaInfo("chem_comp")
ret = self.__schemaRemove(schemaDefObj=sd)
self.assertEqual(ret, True)
#
sd, _, _, _ = self.__schP.getSchemaInfo("pdbx")
ret = self.__schemaRemove(schemaDefObj=sd)
self.assertEqual(ret, True)
#
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def testLoadChemCompMulti(self):
self.__testLoadFilesMulti("chem_comp")
def testLoadBirdMulti(self):
self.__testLoadFilesMulti("bird")
def testLoadPdbxMulti(self):
self.__testLoadFilesMulti("pdbx")
def __getPathList(self, fType):
pathList = []
if fType == "chem_comp":
pathList = self.__rpP.getLocatorObjList("chem_comp")
elif fType == "bird":
pathList = self.__rpP.getLocatorObjList("bird")
pathList.extend(self.__rpP.getLocatorObjList("bird_family"))
elif fType == "pdbx":
pathList = self.__rpP.getLocatorObjList("pdbx")
return pathList
def loadInsertMany(self, dataList, procName, optionsD, workingDir):
try:
_ = workingDir
ret = None
sd = optionsD["sd"]
skipD = optionsD["skip"]
ioObj = IoAdapter(verbose=self.__verbose)
logger.debug("%s pathlist %r", procName, dataList)
with Connection(cfgOb=self.__cfgOb,
resourceName=self.__resourceName) as client:
sdl = CrateDbLoader(schemaDefObj=sd,
ioObj=ioObj,
dbCon=client,
workPath=self.__workPath,
cleanUp=False,
warnings="default",
verbose=self.__verbose)
ret = sdl.load(inputPathList=dataList,
loadType="crate-insert-many",
deleteOpt="selected",
tableIdSkipD=skipD)
# all or nothing here
if ret:
return dataList, dataList, []
else:
return [], [], []
except Exception as e:
logger.info("Failing with dataList %r", dataList)
logger.exception("Failing with %s", str(e))
return [], [], []
def __testLoadFilesMulti(self, contentType):
"""Test case - create load w/insert-many all chemical component definition data files - (multiproc test)"""
numProc = self.__numProc
chunkSize = self.__chunkSize
try:
#
sd, _, _, _ = self.__schP.getSchemaInfo(contentType)
if self.__createFlag:
self.__schemaCreate(schemaDefObj=sd)
optD = {}
optD["sd"] = sd
if contentType == "pdbx":
optD["skip"] = self.__tableIdSkipD
else:
optD["skip"] = {}
#
pathList = self.__getPathList(fType=contentType)
logger.debug("Input path list %r", pathList)
mpu = MultiProcUtil(verbose=True)
mpu.setOptions(optionsD=optD)
mpu.set(workerObj=self, workerMethod="loadInsertMany")
ok, _, _, _ = mpu.runMulti(dataList=pathList,
numProc=numProc,
numResults=1,
chunkSize=chunkSize)
self.assertEqual(ok, True)
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def __schemaCreate(self, schemaDefObj):
"""Test case - create table schema using schema definition"""
ret = 0
try:
tableIdList = schemaDefObj.getTableIdList()
sqlGen = SqlGenAdmin(self.__verbose, serverType="cratedb")
sqlL = []
for tableId in tableIdList:
if tableId in self.__tableIdSkipD:
continue
tableDefObj = schemaDefObj.getTable(tableId)
sqlL.extend(
sqlGen.createTableSQL(
databaseName=schemaDefObj.getVersionedDatabaseName(),
tableDefObj=tableDefObj))
logger.debug("Schema creation SQL string\n %s\n\n",
"\n".join(sqlL))
logger.info("Creating schema using database %s",
schemaDefObj.getVersionedDatabaseName())
#
with Connection(cfgOb=self.__cfgOb,
resourceName=self.__resourceName) as client:
crQ = CrateDbQuery(dbcon=client, verbose=self.__verbose)
ret = crQ.sqlCommandList(sqlCommandList=sqlL)
logger.debug("Schema create command returns %r\n", ret)
return ret
#
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def __schemaRemove(self, schemaDefObj):
"""Test case - remove table schema using schema definition"""
ret = 0
try:
tableIdList = schemaDefObj.getTableIdList()
sqlGen = SqlGenAdmin(self.__verbose, serverType="cratedb")
sqlL = []
for tableId in tableIdList:
if tableId in self.__tableIdSkipD:
continue
tableDefObj = schemaDefObj.getTable(tableId)
sqlL.extend(
sqlGen.dropTableSQL(
databaseName=schemaDefObj.getVersionedDatabaseName(),
tableDefObj=tableDefObj))
sqlL.extend(
sqlGen.dropTableSQL(
databaseName=schemaDefObj.getDatabaseName(),
tableDefObj=tableDefObj))
logger.debug("Schema Remove SQL string\n %s", "\n".join(sqlL))
with Connection(cfgOb=self.__cfgOb,
resourceName=self.__resourceName) as client:
crQ = CrateDbQuery(dbcon=client, verbose=self.__verbose)
ret = crQ.sqlCommandList(sqlCommandList=sqlL)
logger.debug("Schema remove command returns %r\n", ret)
return ret
#
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
class RepoHoldingsDataPrepValidateTests(unittest.TestCase):
def setUp(self):
self.__verbose = True
#
self.__mockTopPath = os.path.join(TOPDIR, "rcsb", "mock-data")
self.__pathConfig = os.path.join(TOPDIR, "rcsb", "db", "config",
"exdb-config-example.yml")
self.__cachePath = os.path.join(TOPDIR, "CACHE")
self.__updateId = "2018_25"
self.__export = False
#
configName = "site_info_configuration"
self.__cfgOb = ConfigUtil(configPath=self.__pathConfig,
defaultSectionName=configName,
mockTopPath=self.__mockTopPath)
self.__schP = SchemaProvider(self.__cfgOb,
self.__cachePath,
useCache=True)
self.__sandboxPath = self.__cfgOb.getPath("RCSB_EXCHANGE_SANDBOX_PATH",
sectionName=configName)
#
self.__mU = MarshalUtil(workPath=self.__cachePath)
self.__startTime = time.time()
logger.debug("Starting %s at %s", self.id(),
time.strftime("%Y %m %d %H:%M:%S", time.localtime()))
def tearDown(self):
endTime = time.time()
logger.debug("Completed %s at %s (%.4f seconds)", self.id(),
time.strftime("%Y %m %d %H:%M:%S", time.localtime()),
endTime - self.__startTime)
def testValidateOptsStrict(self):
updateId = self.__updateId
schemaLevel = "full"
eCount = self.__testValidateOpts(updateId, schemaLevel=schemaLevel)
logger.info("Total validation errors schema level %s : %d",
schemaLevel, eCount)
self.assertTrue(eCount <= 1)
@unittest.skip("Troubleshooting test")
def testValidateOptsMin(self):
updateId = self.__updateId
schemaLevel = "min"
eCount = self.__testValidateOpts(updateId, schemaLevel=schemaLevel)
logger.info("Total validation errors schema level %s : %d",
schemaLevel, eCount)
self.assertTrue(eCount <= 1)
def __testValidateOpts(self, updateId, schemaLevel="full"):
schemaNames = ["repository_holdings"]
collectionNames = {
"repository_holdings": [
"repository_holdings_update_entry",
"repository_holdings_current_entry",
"repository_holdings_unreleased_entry",
"repository_holdings_removed_entry",
"repository_holdings_combined_entry",
],
"entity_sequence_clusters":
["cluster_members", "cluster_provenance", "entity_members"],
}
#
eCount = 0
for schemaName in schemaNames:
for collectionName in collectionNames[schemaName]:
_ = self.__schP.makeSchemaDef(schemaName,
dataTyping="ANY",
saveSchema=True)
cD = self.__schP.makeSchema(schemaName,
collectionName,
encodingType="JSON",
level=schemaLevel,
saveSchema=True)
dL = self.__getRepositoryHoldingsDocuments(
schemaName, collectionName, updateId)
if self.__export:
savePath = os.path.join(HERE, "test-output",
collectionName + ".json")
self.__mU.doExport(savePath, dL, fmt="json", indent=3)
# Raises exceptions for schema compliance.
Draft4Validator.check_schema(cD)
#
valInfo = Draft4Validator(cD, format_checker=FormatChecker())
for ii, dD in enumerate(dL):
logger.debug("Schema %s collection %s document %d",
schemaName, collectionName, ii)
try:
cCount = 0
for error in sorted(valInfo.iter_errors(dD), key=str):
logger.info(
"schema %s collection %s path %s error: %s",
schemaName, collectionName, error.path,
error.message)
logger.info(">>>")
logger.info(">>> failing object is %r", dD)
logger.info(">>>")
eCount += 1
cCount += 1
#
logger.debug("schema %s collection %s count %d",
schemaName, collectionName, cCount)
except Exception as e:
logger.exception("Validation error %s", str(e))
return eCount
def __getRepositoryHoldingsDocuments(self, schemaName, collectionName,
updateId):
"""Test loading and processing operations for legacy holdings and status data."""
rL = []
try:
rhdp = RepoHoldingsDataPrep(cfgOb=self.__cfgOb,
sandboxPath=self.__sandboxPath,
workPath=self.__cachePath)
if collectionName == "repository_holdings_update_entry":
rL = rhdp.getHoldingsUpdateEntry(updateId=updateId)
self.assertGreaterEqual(len(rL), 10)
logger.debug("update data length %r", len(rL))
#
elif collectionName == "repository_holdings_current_entry":
rL = rhdp.getHoldingsCurrentEntry(updateId=updateId)
self.assertGreaterEqual(len(rL), 10)
logger.debug("holdings data length %r", len(rL))
#
elif collectionName == "repository_holdings_unreleased_entry":
rL = rhdp.getHoldingsUnreleasedEntry(updateId=updateId)
self.assertGreaterEqual(len(rL), 10)
logger.debug("unreleased data length %r", len(rL))
#
elif collectionName in ["repository_holdings_removed_entry"]:
rL = rhdp.getHoldingsRemovedEntry(updateId=updateId)
if collectionName == "repository_holdings_removed":
self.assertGreaterEqual(len(rL), 10)
logger.debug("removed data length %r", len(rL))
elif collectionName == "repository_holdings_combined_entry":
rL = rhdp.getHoldingsCombinedEntry(updateId=updateId)
self.assertGreaterEqual(len(rL), 10)
logger.debug("holdings data length %r", len(rL))
#
except Exception as e:
logger.exception("%s %s failing with %s", schemaName,
collectionName, str(e))
self.fail()
return rL
class UniProtCoreEtlWorker(object):
"""Prepare and load UniProt 'core' sequence reference data collections."""
def __init__(self,
cfgOb,
cachePath,
useCache=True,
numProc=2,
chunkSize=10,
readBackCheck=False,
documentLimit=None,
doValidate=False,
verbose=False):
self.__cfgOb = cfgOb
self.__cachePath = cachePath
self.__useCache = useCache
self.__readBackCheck = readBackCheck
self.__numProc = numProc
self.__chunkSize = chunkSize
self.__documentLimit = documentLimit
#
self.__resourceName = "MONGO_DB"
self.__verbose = verbose
self.__statusList = []
self.__schP = SchemaProvider(self.__cfgOb,
self.__cachePath,
useCache=self.__useCache)
self.__docHelper = DocumentDefinitionHelper(cfgOb=self.__cfgOb)
self.__valInst = None
self.__doValidate = doValidate
#
def __updateStatus(self, updateId, databaseName, collectionName, status,
startTimestamp):
try:
sFlag = "Y" if status else "N"
desp = DataExchangeStatus()
desp.setStartTime(tS=startTimestamp)
desp.setObject(databaseName, collectionName)
desp.setStatus(updateId=updateId, successFlag=sFlag)
desp.setEndTime()
self.__statusList.append(desp.getStatus())
return True
except Exception as e:
logger.exception("Failing with %s", str(e))
return False
def __getReferenceSequenceProvider(self):
""" """
try:
rsaP = ReferenceSequenceAssignmentProvider(
self.__cfgOb,
databaseName="pdbx_core",
collectionName="pdbx_core_polymer_entity",
polymerType="Protein",
referenceDatabaseName="UniProt",
provSource="PDB",
useCache=self.__useCache,
cachePath=self.__cachePath,
fetchLimit=self.__documentLimit,
siftsAbbreviated="TEST",
)
ok = rsaP.testCache()
return ok, rsaP
except Exception as e:
logger.exception("Failing with %s", str(e))
return None
def load(self, updateId, extResource, loadType="full"):
"""Load sequence reference data"""
try:
self.__statusList = []
desp = DataExchangeStatus()
statusStartTimestamp = desp.setStartTime()
#
dList = indexL = []
databaseName = collectionName = collectionVersion = None
#
if extResource == "UniProt":
databaseName = "uniprot_core"
# configName = self.__cfgOb.getDefaultSectionName()
# dirPath = os.path.join(self.__cachePath, self.__cfgOb.get("EXDB_CACHE_DIR", self.__cfgOb.getDefaultSectionName()))
#
ok, rsP = self.__getReferenceSequenceProvider()
if not ok:
return False
#
dList = rsP.getDocuments()
logger.info("Resource %r extracted mapped document length %d",
extResource, len(dList))
logger.debug("Objects %r", dList[:2])
#
cDL = self.__docHelper.getCollectionInfo(databaseName)
collectionName = cDL[0]["NAME"]
collectionVersion = cDL[0]["VERSION"]
indexL = self.__docHelper.getDocumentIndexAttributes(
collectionName, "primary")
logger.info(
"Database %r collection %r version %r index attributes %r",
databaseName, collectionName, collectionVersion, indexL)
addValues = {}
else:
logger.error("Unsupported external resource %r", extResource)
#
if self.__doValidate:
self.__valInst = self.__getValidator(databaseName,
collectionName,
schemaLevel="full")
for dObj in dList:
self.__validateObj(databaseName,
collectionName,
dObj,
label="Original")
#
dl = DocumentLoader(
self.__cfgOb,
self.__cachePath,
self.__resourceName,
numProc=self.__numProc,
chunkSize=self.__chunkSize,
documentLimit=self.__documentLimit,
verbose=self.__verbose,
readBackCheck=self.__readBackCheck,
)
#
ok = dl.load(databaseName,
collectionName,
loadType=loadType,
documentList=dList,
indexAttributeList=indexL,
keyNames=None,
addValues=addValues)
okS = self.__updateStatus(updateId, databaseName, collectionName,
ok, statusStartTimestamp)
return ok and okS
except Exception as e:
logger.exception("Failing with %s", str(e))
return False
def getLoadStatus(self):
return self.__statusList
def __getValidator(self, databaseName, collectionName, schemaLevel="full"):
# _ = self.__schP.makeSchemaDef(databaseName, dataTyping="ANY", saveSchema=True)
# cD = self.__schP.makeSchema(databaseName, collectionName, encodingType="JSON", level=schemaLevel, saveSchema=True)
logger.info("Fetch schema for %r %r validation level %r", databaseName,
collectionName, schemaLevel)
cD = self.__schP.getJsonSchema(databaseName,
collectionName,
encodingType="JSON",
level=schemaLevel)
# Raises exceptions for schema compliance.
Draft4Validator.check_schema(cD)
valInst = Draft4Validator(cD, format_checker=FormatChecker())
return valInst
def __validateObj(self, databaseName, collectionName, rObj, label=""):
try:
eCount = 0
tId = rObj["rcsb_id"] if rObj and "rcsb_id" in rObj else "anonymous"
for error in sorted(self.__valInst.iter_errors(rObj), key=str):
logger.info(
"Database %s collection %s (%s %r) path %s error: %s",
databaseName, collectionName, label, tId, error.path,
error.message)
logger.debug(">>> Failing object is %r", rObj)
if "rcsb_uniprot_feature" in rObj:
for dd in rObj["rcsb_uniprot_feature"]:
if "feature_id" in dd:
logger.info("feature_id %r", dd["feature_id"])
else:
logger.info("no feature_id keys %r",
sorted(dd.keys()))
logger.info("description %r", dd["description"])
eCount += 1
except Exception as e:
logger.exception("Validation failing %s", str(e))
return eCount
class SchemaProviderTests(unittest.TestCase):
def setUp(self):
self.__verbose = True
mockTopPath = os.path.join(TOPDIR, "rcsb", "mock-data")
pathConfig = os.path.join(TOPDIR, "rcsb", "db", "config",
"exdb-config-example.yml")
configName = "site_info_configuration"
self.__cachePath = os.path.join(TOPDIR, "CACHE")
self.__cfgOb = ConfigUtil(configPath=pathConfig,
defaultSectionName=configName,
mockTopPath=mockTopPath)
self.__schP = SchemaProvider(self.__cfgOb,
self.__cachePath,
useCache=False)
#
self.__validationLevels = self.__cfgOb.getList(
"VALIDATION_LEVELS_TEST",
sectionName="database_catalog_configuration")
self.__encodingTypes = self.__cfgOb.getList(
"ENCODING_TYPES_TEST",
sectionName="database_catalog_configuration")
#
buildAll = True
if buildAll:
self.__databaseNameList = self.__cfgOb.getList(
"DATABASE_NAMES_DEPLOYED",
sectionName="database_catalog_configuration")
self.__dataTypingList = self.__cfgOb.getList(
"DATATYPING_DEPLOYED",
sectionName="database_catalog_configuration")
#
else:
self.__databaseNameList = self.__cfgOb.getList(
"DATABASE_NAMES_TEST",
sectionName="database_catalog_configuration")
self.__dataTypingList = self.__cfgOb.getList(
"DATATYPING_TEST",
sectionName="database_catalog_configuration")
#
self.__docHelper = DocumentDefinitionHelper(cfgOb=self.__cfgOb)
self.__startTime = time.time()
logger.debug("Starting %s at %s", self.id(),
time.strftime("%Y %m %d %H:%M:%S", time.localtime()))
def tearDown(self):
endTime = time.time()
logger.debug("Completed %s at %s (%.4f seconds)", self.id(),
time.strftime("%Y %m %d %H:%M:%S", time.localtime()),
endTime - self.__startTime)
def testSchemaAccessDefault(self):
for databaseName in self.__databaseNameList:
cDL = self.__docHelper.getCollectionInfo(databaseName)
for cD in cDL:
collectionName = cD["NAME"]
for encodingType in self.__encodingTypes:
if encodingType.lower() == "rcsb":
continue
for level in self.__validationLevels:
logger.debug("Loading ->%s %s %s %s", databaseName,
collectionName, encodingType, level)
sD = self.__schP.getJsonSchema(
databaseName,
collectionName,
encodingType=encodingType,
level=level)
self.assertTrue(sD is not None)
class DocumentLoader(object):
def __init__(
self,
cfgOb,
cachePath,
resourceName="MONGO_DB",
numProc=4,
chunkSize=15,
documentLimit=None,
verbose=False,
readBackCheck=False,
maxStepLength=2000,
schemaRebuildFlag=False,
):
self.__verbose = verbose
#
# Limit the load length of each file type for testing - Set to None to remove -
self.__documentLimit = documentLimit
self.__maxStepLength = maxStepLength
#
# Controls for multiprocessing execution -
self.__numProc = numProc
self.__chunkSize = chunkSize
#
self.__cfgOb = cfgOb
self.__resourceName = resourceName
#
self.__cachePath = cachePath if cachePath else "."
self.__schP = SchemaProvider(cfgOb,
cachePath,
useCache=True,
rebuildFlag=schemaRebuildFlag)
#
self.__readBackCheck = readBackCheck
self.__mpFormat = "[%(levelname)s] %(asctime)s %(processName)s-%(module)s.%(funcName)s: %(message)s"
#
#
def load(self,
databaseName,
collectionName,
loadType="full",
documentList=None,
indexAttributeList=None,
keyNames=None,
schemaLevel="full",
addValues=None):
"""Driver method for loading MongoDb content -
loadType: "full" or "replace"
"""
try:
startTime = self.__begin(message="loading operation")
#
#
optionsD = {}
optionsD["collectionName"] = collectionName
optionsD["databaseName"] = databaseName
optionsD["readBackCheck"] = self.__readBackCheck
optionsD["loadType"] = loadType
optionsD["keyNames"] = keyNames
# ---------------- - ---------------- - ---------------- - ---------------- - ---------------- -
#
docList = documentList[:self.
__documentLimit] if self.__documentLimit else documentList
logger.debug("Full document list length %d limit %r",
len(documentList), self.__documentLimit)
numProc = self.__numProc
chunkSize = self.__chunkSize if docList and self.__chunkSize < len(
docList) else 0
#
if addValues:
try:
for doc in docList:
for k, v in addValues.items():
doc[k] = v
except Exception as e:
logger.error("Add values %r fails with %s", addValues,
str(e))
#
indAtList = indexAttributeList if indexAttributeList else []
bsonSchema = None
if schemaLevel and schemaLevel in ["min", "full"]:
bsonSchema = self.__schP.getJsonSchema(databaseName,
collectionName,
encodingType="BSON",
level=schemaLevel)
logger.debug("Using schema validation for %r %r %r",
databaseName, collectionName, schemaLevel)
if loadType == "full":
self.__removeCollection(databaseName, collectionName)
ok = self.__createCollection(databaseName,
collectionName,
indAtList,
bsonSchema=bsonSchema)
logger.info("Collection %s create status %r", collectionName,
ok)
elif loadType == "append":
# create only if object does not exist -
ok = self.__createCollection(databaseName,
collectionName,
indexAttributeNames=indAtList,
checkExists=True,
bsonSchema=bsonSchema)
logger.debug("Collection %s create status %r", collectionName,
ok)
# ---------------- - ---------------- - ---------------- - ---------------- - ---------------- -
numDocs = len(docList)
logger.debug("Processing %d total documents", numDocs)
numProc = min(numProc, numDocs)
maxStepLength = self.__maxStepLength
if numDocs > maxStepLength:
numLists = int(numDocs / maxStepLength)
subLists = [docList[i::numLists] for i in range(numLists)]
else:
subLists = [docList]
#
if subLists:
logger.debug(
"Starting with numProc %d outer subtask count %d subtask length ~ %d",
numProc, len(subLists), len(subLists[0]))
#
failList = []
for ii, subList in enumerate(subLists):
logger.debug("Running outer subtask %d of %d length %d",
ii + 1, len(subLists), len(subList))
#
mpu = MultiProcUtil(verbose=True)
mpu.setOptions(optionsD=optionsD)
mpu.set(workerObj=self, workerMethod="loadWorker")
ok, failListT, _, _ = mpu.runMulti(dataList=subList,
numProc=numProc,
numResults=1,
chunkSize=chunkSize)
failList.extend(failListT)
logger.debug("Completed load with failing document list %r",
failList)
logger.debug("Document list length %d failed load list length %d",
len(docList), len(failList))
#
self.__end(startTime, "loading operation with status " + str(ok))
#
return ok
except Exception as e:
logger.exception("Failing with %s", str(e))
return False
def loadWorker(self, dataList, procName, optionsD, workingDir):
"""Multi-proc worker method for MongoDb document loading -"""
try:
startTime = self.__begin(message=procName)
readBackCheck = optionsD["readBackCheck"]
loadType = optionsD["loadType"]
collectionName = optionsD["collectionName"]
databaseName = optionsD["databaseName"]
keyNames = optionsD["keyNames"]
#
logger.debug("%s databaseName %s collectionName %s workingDir %s",
procName, databaseName, collectionName, workingDir)
#
if dataList:
ok, successList, failedList = self.__loadDocuments(
databaseName,
collectionName,
dataList,
loadType=loadType,
readBackCheck=readBackCheck,
keyNames=keyNames)
#
logger.debug(
"%s database %s collection %s inputList length %d successList length %d failed %d",
procName,
databaseName,
collectionName,
len(dataList),
len(successList),
len(failedList),
)
#
self.__end(startTime, procName + " with status " + str(ok))
return successList, [], []
except Exception as e:
logger.exception("Failing with %s", str(e))
return [], [], []
# -------------- -------------- -------------- -------------- -------------- -------------- --------------
# --- Supporting code follows ---
#
def __begin(self, message=""):
startTime = time.time()
ts = time.strftime("%Y %m %d %H:%M:%S", time.localtime())
logger.debug("Starting %s at %s", message, ts)
return startTime
def __end(self, startTime, message=""):
endTime = time.time()
ts = time.strftime("%Y %m %d %H:%M:%S", time.localtime())
delta = endTime - startTime
logger.debug("Completed %s at %s (%.4f seconds)", message, ts, delta)
def __createCollection(self,
dbName,
collectionName,
indexAttributeNames=None,
checkExists=False,
bsonSchema=None):
"""Create database and collection and optionally a primary index -"""
try:
logger.debug("Create database %s collection %s", dbName,
collectionName)
with Connection(cfgOb=self.__cfgOb,
resourceName=self.__resourceName) as client:
mg = MongoDbUtil(client)
if checkExists and mg.databaseExists(
dbName) and mg.collectionExists(
dbName, collectionName):
ok1 = True
else:
ok1 = mg.createCollection(dbName,
collectionName,
bsonSchema=bsonSchema)
ok2 = mg.databaseExists(dbName)
ok3 = mg.collectionExists(dbName, collectionName)
okI = True
if indexAttributeNames:
okI = mg.createIndex(dbName,
collectionName,
indexAttributeNames,
indexName="primary",
indexType="DESCENDING",
uniqueFlag=False)
return ok1 and ok2 and ok3 and okI
#
except Exception as e:
logger.exception("Failing with %s", str(e))
return False
def __removeCollection(self, dbName, collectionName):
"""Drop collection within database"""
try:
with Connection(cfgOb=self.__cfgOb,
resourceName=self.__resourceName) as client:
mg = MongoDbUtil(client)
#
logger.debug("Remove collection database %s collection %s",
dbName, collectionName)
logger.debug("Starting databases = %r", mg.getDatabaseNames())
logger.debug("Starting collections = %r",
mg.getCollectionNames(dbName))
ok = mg.dropCollection(dbName, collectionName)
logger.debug("Databases = %r", mg.getDatabaseNames())
logger.debug("Post drop collections = %r",
mg.getCollectionNames(dbName))
ok = mg.collectionExists(dbName, collectionName)
logger.debug("Post drop collections = %r",
mg.getCollectionNames(dbName))
return ok
except Exception as e:
logger.exception("Failing with %s", str(e))
return False
def __loadDocuments(self,
dbName,
collectionName,
docList,
loadType="full",
readBackCheck=False,
keyNames=None):
#
# Load database/collection with input document list -
#
failList = []
rIdL = []
successList = []
logger.debug(
"Loading dbName %s collectionName %s with document count %d keynames %r",
dbName, collectionName, len(docList), keyNames)
if keyNames:
# map the document list to some document key if this is provided
indD = {}
indL = []
try:
for ii, doc in enumerate(docList):
dIdTup = self.__getKeyValues(doc, keyNames)
indD[dIdTup] = ii
indL = list(range(len(docList)))
except Exception as e:
logger.exception("Failing ii %d d %r with %s", ii, doc, str(e))
try:
with Connection(cfgOb=self.__cfgOb,
resourceName=self.__resourceName) as client:
mg = MongoDbUtil(client)
#
if loadType == "replace" and keyNames:
dTupL = mg.deleteList(dbName, collectionName, docList,
keyNames)
logger.debug("Deleted document status %r", (dTupL, ))
#
rIdL = mg.insertList(dbName,
collectionName,
docList,
keyNames=keyNames)
logger.debug("Insert returns rIdL length %r", len(rIdL))
# ---
# If there is a failure then determine the specific successes and failures -
#
successList = docList
failList = []
if len(rIdL) != len(docList):
if keyNames:
successIndList = []
for rId in rIdL:
rObj = mg.fetchOne(dbName, collectionName, "_id",
rId)
dIdTup = self.__getKeyValues(rObj, keyNames)
successIndList.append(indD[dIdTup])
failIndList = list(set(indL) - set(successIndList))
failList = [docList[ii] for ii in failIndList]
successList = [docList[ii] for ii in successIndList]
else:
# fail the whole batch if we don't have visibility into each document
failList = docList
successList = []
#
rbStatus = True
if readBackCheck and keyNames:
#
# Note that objects in docList are mutated by the insert operation with the additional key '_id',
# hence, it is possible to compare the fetched object with the input object.
#
for ii, rId in enumerate(rIdL):
rObj = mg.fetchOne(dbName, collectionName, "_id", rId)
dIdTup = self.__getKeyValues(rObj, keyNames)
jj = indD[dIdTup]
if rObj != docList[jj]:
rbStatus = False
break
#
if readBackCheck and not rbStatus:
return False, successList, failList
#
return len(rIdL) == len(docList), successList, failList
except Exception as e:
logger.exception("Failing %r %r (len=%d) %s with %s", dbName,
collectionName, len(docList), keyNames, str(e))
return False, [], docList
def __getKeyValues(self, dct, keyNames):
"""Return the tuple of values of corresponding to the input dictionary key names expressed in dot notation.
Args:
dct (dict): source dictionary object (nested)
keyNames (list): list of dictionary keys in dot notatoin
Returns:
tuple: tuple of values corresponding to the input key names
"""
rL = []
try:
for keyName in keyNames:
rL.append(self.__getKeyValue(dct, keyName))
except Exception as e:
logger.exception("Failing for key names %r with %s", keyNames,
str(e))
return tuple(rL)
def __getKeyValue(self, dct, keyName):
"""Return the value of the corresponding key expressed in dot notation in the input dictionary object (nested)."""
try:
kys = keyName.split(".")
for key in kys:
try:
dct = dct[key]
except KeyError:
return None
return dct
except Exception as e:
logger.exception("Failing for key %r with %s", keyName, str(e))
return None
def setUp(self):
self.__isMac = platform.system() == "Darwin"
self.__excludeType = None if self.__isMac else "optional"
self.__numProc = 2
mockTopPath = os.path.join(TOPDIR, "rcsb", "mock-data")
self.__cachePath = os.path.join(TOPDIR, "CACHE")
self.__outputPath = os.path.join(HERE, "test-output")
self.__savedOutputPath = os.path.join(HERE, "test-saved-output")
configPath = os.path.join(TOPDIR, "rcsb", "db", "config",
"exdb-config-example.yml")
configName = "site_info_configuration"
self.__configName = configName
self.__cfgOb = ConfigUtil(configPath=configPath,
defaultSectionName=configName,
mockTopPath=mockTopPath)
self.__mU = MarshalUtil(workPath=self.__cachePath)
self.__discoveryMode = self.__cfgOb.get("DISCOVERY_MODE",
sectionName=configName,
default="local")
self.__fileLimit = 100 if self.__discoveryMode == "local" else 10
self.__schP = SchemaProvider(self.__cfgOb,
self.__cachePath,
useCache=True)
self.__rpP = RepositoryProvider(cfgOb=self.__cfgOb,
numProc=self.__numProc,
fileLimit=self.__fileLimit,
cachePath=self.__cachePath)
#
#
self.__fTypeRow = "drop-empty-attributes|drop-empty-tables|skip-max-width|convert-iterables|normalize-enums|translateXMLCharRefs"
self.__fTypeCol = "drop-empty-tables|skip-max-width|convert-iterables|normalize-enums|translateXMLCharRefs"
self.__chemCompMockLen = 24
self.__pdbxMockLen = 30
# removes timestamped data items to allow diffs.)
excludeExtras = ["rcsb_load_status"]
# excludeExtras = []
#
self.__verbose = True
self.__modulePathMap = self.__cfgOb.get(
"DICT_METHOD_HELPER_MODULE_PATH_MAP", sectionName=configName)
#
self.__exportFlag = True
self.__diffFlag = False
#
self.__simpleTestCaseList = [
{
"contentType": "chem_comp",
"mockLength": self.__chemCompMockLen,
"filterType": self.__fTypeRow,
"styleType": "rowwise_by_name",
"mergeContentTypes": None,
"rejectLength": 2,
},
{
"contentType": "chem_comp",
"mockLength": self.__chemCompMockLen,
"filterType": self.__fTypeRow,
"styleType": "rowwise_no_name",
"mergeContentTypes": None,
"rejectLength": 2,
},
{
"contentType": "chem_comp",
"mockLength": self.__chemCompMockLen,
"filterType": self.__fTypeCol,
"styleType": "columnwise_by_name",
"mergeContentTypes": None,
"rejectLength": 2,
},
{
"contentType": "chem_comp",
"mockLength": self.__chemCompMockLen,
"filterType": self.__fTypeRow,
"styleType": "rowwise_by_name",
"mergeContentTypes": None,
"rejectLength": 2,
},
{
"contentType": "pdbx_core",
"mockLength": self.__pdbxMockLen,
"filterType": self.__fTypeRow,
"styleType": "rowwise_by_name",
"mergeContentTypes": None,
"rejectLength": 0,
},
]
#
self.__fullTestCaseList = [
{
"contentType": "pdbx_core",
"mockLength": self.__pdbxMockLen,
"filterType": self.__fTypeRow,
"styleType": "rowwise_by_name_with_cardinality",
"mergeContentTypes": ["vrpt"],
"rejectLength": 0,
"excludeExtras": excludeExtras,
},
{
"contentType": "bird_chem_comp_core",
"mockLength": self.__chemCompMockLen,
"filterType": self.__fTypeRow,
"styleType": "rowwise_by_name_with_cardinality",
"mergeContentTypes": None,
"rejectLength": 2,
"excludeExtras": excludeExtras,
},
]
#
self.__fullTestCaseListA = [
{
"contentType": "pdbx_core",
"mockLength": self.__pdbxMockLen,
"filterType": self.__fTypeRow,
"styleType": "rowwise_by_name_with_cardinality",
"mergeContentTypes": ["vrpt"],
"rejectLength": 0,
"excludeExtras": excludeExtras,
},
]
#
self.__startTime = time.time()
logger.debug("Starting %s at %s", self.id(),
time.strftime("%Y %m %d %H:%M:%S", time.localtime()))