class RepositoryProviderTests(unittest.TestCase):
def setUp(self):
#
#
mockTopPath = os.path.join(TOPDIR, "rcsb", "mock-data")
configPath = os.path.join(TOPDIR, "rcsb", "db", "config",
"exdb-config-example.yml")
self.__configName = "site_info_configuration"
self.__cfgOb = ConfigUtil(configPath=configPath,
defaultSectionName=self.__configName,
mockTopPath=mockTopPath)
self.__cachePath = os.path.join(TOPDIR, "CACHE")
self.__numProc = 2
self.__chunkSize = 20
self.__fileLimit = None
#
self.__rpP = RepositoryProvider(cfgOb=self.__cfgOb,
numProc=self.__numProc,
fileLimit=self.__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)\n", self.id(),
time.strftime("%Y %m %d %H:%M:%S", time.localtime()),
endTime - self.__startTime)
def testRepoUtils(self):
""" Test case - repository locator path utilities
"""
for contentType in ["bird_chem_comp_core", "pdbx_core", "ihm_dev"]:
mergeContentTypes = None
if contentType in ["pdbx_core"]:
mergeContentTypes = ["vrpt"]
#
locatorObjList = self.__rpP.getLocatorObjList(
contentType=contentType, mergeContentTypes=mergeContentTypes)
pathList = self.__rpP.getLocatorPaths(locatorObjList)
locatorObjList2 = self.__rpP.getLocatorsFromPaths(
locatorObjList, pathList)
logger.info("%s pathList length %d", contentType, len(pathList))
self.assertEqual(len(locatorObjList), len(pathList))
self.assertEqual(len(locatorObjList), len(locatorObjList2))
class ScanRepoUtil(object):
"""Tools for for scanning repositories and collecting coverage and type data information.
"""
def __init__(self,
cfgOb,
attributeDataTypeD=None,
numProc=4,
chunkSize=15,
fileLimit=None,
maxStepLength=2000,
workPath=None):
"""
Args:
cfgOb (object): Configuration object (ConfigUtil)
attributeDataTypeD
dictPath (str): Path to supporting data dictionary
numProc (int, optional): Number of parallel worker processes used.
chunkSize (int, optional): Size of files processed in a single multi-proc process
fileLimit (int, optional): maximum file scanned or None for no limit
mockTopPath (str, optional): Path to directory containing mock repositories or None
maxStepLength (int, optional): maximum number of multi-proc runs to perform
"""
#
self.__attributeDataTypeD = attributeDataTypeD if attributeDataTypeD else {}
# Limit the load length of each file type for testing - Set to None to remove -
self.__fileLimit = fileLimit
self.__maxStepLength = maxStepLength
#
# Controls for multiprocessing execution -
self.__numProc = numProc
self.__chunkSize = chunkSize
#
self.__cfgOb = cfgOb
#
self.__mpFormat = "[%(levelname)s] %(asctime)s %(processName)s-%(module)s.%(funcName)s: %(message)s"
self.__workPath = workPath
self.__mU = MarshalUtil(workPath=self.__workPath)
self.__rpP = RepositoryProvider(self.__cfgOb,
numProc=self.__numProc,
fileLimit=self.__fileLimit,
cachePath=self.__workPath)
def scanContentType(self,
contentType,
mergeContentTypes=None,
scanType="full",
inputPathList=None,
scanDataFilePath=None,
failedFilePath=None,
saveInputFileListPath=None):
"""Driver method for repository scan operation
Args:
contentType (str): one of 'bird','bird_family','bird_chem_comp', chem_comp','pdbx'
scanType (str, optional): 'full' [or 'incr' to be supported]
inputPathList (list, optional): list of input file paths to scan
scanDataFilePath (str, optional): file path for serialized scan data (Pickle format)
failedFilePath (str, optional): file path for list of files that fail scanning operation
saveInputFileListPath str, optional): Path to store file path list that is scanned
Returns:
bool: True for success or False otherwise
"""
try:
startTime = self.__begin(message="scanning operation")
#
locatorObjList = self.__rpP.getLocatorObjList(
contentType=contentType,
inputPathList=inputPathList,
mergeContentTypes=mergeContentTypes)
#
if saveInputFileListPath:
self.__mU.doExport(saveInputFileListPath,
self.__rpP.getLocatorPaths(locatorObjList),
fmt="list")
logger.debug("Saving %d paths in %s", len(locatorObjList),
saveInputFileListPath)
#
optD = {}
optD["contentType"] = contentType
optD["logSize"] = True
optD["scanType"] = scanType
# ---------------- - ---------------- - ---------------- - ---------------- - ---------------- -
#
numProc = self.__numProc
chunkSize = self.__chunkSize if locatorObjList and self.__chunkSize < len(
locatorObjList) else 0
#
# ---------------- - ---------------- - ---------------- - ---------------- - ---------------- -
numPaths = len(locatorObjList)
logger.debug("Processing %d total paths", numPaths)
numProc = min(numProc, numPaths)
maxStepLength = self.__maxStepLength
if numPaths > maxStepLength:
numLists = int(numPaths / maxStepLength)
subLists = [
locatorObjList[i::numLists] for i in range(numLists)
]
else:
subLists = [locatorObjList]
#
if subLists:
logger.debug(
"Starting with numProc %d outer subtask count %d subtask length ~ %d",
numProc, len(subLists), len(subLists[0]))
#
numResults = 1
failList = []
retLists = [[] for ii in range(numResults)]
diagList = []
for ii, subList in enumerate(subLists):
logger.debug("Running outer subtask %d or %d length %d",
ii + 1, len(subLists), len(subList))
#
mpu = MultiProcUtil(verbose=True)
mpu.setOptions(optionsD=optD)
mpu.set(workerObj=self, workerMethod="scanWorker")
ok, failListT, retListsT, diagListT = mpu.runMulti(
dataList=subList,
numProc=numProc,
numResults=numResults,
chunkSize=chunkSize)
failList.extend(failListT)
# retLists is a list of lists -
for _ in range(numResults):
retLists[ii].extend(retListsT[ii])
diagList.extend(diagListT)
logger.debug("Scan failed path list %r", failList)
logger.debug(
"Scan path list success length %d load list failed length %d",
len(locatorObjList), len(failList))
logger.debug("Returned metadata length %r", len(retLists[0]))
#
if failedFilePath and failList:
wOk = self.__mU.doExport(failedFilePath,
self.__rpP.getLocatorPaths(failList),
fmt="list")
logger.debug("Writing scan failure path list to %s status %r",
failedFilePath, wOk)
#
if scanType == "incr":
scanDataD = self.__mU.doImport(scanDataFilePath,
fmt="pickle",
default=None)
logger.debug("Imported scan data with keys %r",
list(scanDataD.keys()))
else:
scanDataD = {}
#
if scanDataFilePath and retLists[0]:
for ssTup in retLists[0]:
cId = ssTup.containerId
if scanType == "full" and cId in scanDataD:
logger.error("Duplicate container id %s in %r and %r",
cId, ssTup.fromPath,
scanDataD[cId].fromPath)
#
scanDataD[cId] = ssTup
ok = self.__mU.doExport(scanDataFilePath,
scanDataD,
fmt="pickle")
tscanDataD = self.__mU.doImport(scanDataFilePath, fmt="pickle")
ok = tscanDataD == scanDataD
self.__end(startTime, "scanning operation with status " + str(ok))
#
return ok
except Exception as e:
logger.exception("Failing with %s", str(e))
return False
def evalScan(self,
scanDataFilePath,
evalJsonFilePath,
evalType="data_type"):
scanDataD = self.__mU.doImport(scanDataFilePath, fmt="pickle")
if evalType in ["data_type"]:
rD = self.__evalScanDataType(scanDataD)
elif evalType in ["data_coverage"]:
rD = self.__evalScanDataCoverage(scanDataD)
else:
logger.debug("Unknown evalType %r", evalType)
ok = self.__mU.doExport(evalJsonFilePath, rD, fmt="json")
return ok
def __evalScanDataType(self, scanDataD):
"""
ScanValue = collections.namedtuple('ScanValue', 'containerId, catName, atName, minWidth, maxWidth, minPrec, maxPrec')
ScanSummary = collections.namedtuple('ScanSummary', 'containerId, fromPath, scanDate, scanCategoryDict')
"""
# for populated sD[category] -> d[atName]->{minWidth: , maxWidth:, minPrec:, maxPrec: , count}
sD = {}
for cId in scanDataD:
ssTup = scanDataD[cId]
dD = ssTup.scanCategoryDict
for catName in dD:
if catName not in sD:
sD[catName] = {}
for svTup in dD[catName]:
if svTup.atName not in sD[catName]:
sD[catName][svTup.atName] = {
"minWidth": svTup.minWidth,
"maxWidth": svTup.maxWidth,
"minPrec": svTup.minPrec,
"maxPrec": svTup.maxPrec,
"count": 1
}
continue
sD[catName][svTup.atName]["minWidth"] = min(
sD[catName][svTup.atName]["minWidth"], svTup.minWidth)
sD[catName][svTup.atName]["maxWidth"] = max(
sD[catName][svTup.atName]["maxWidth"], svTup.maxWidth)
sD[catName][svTup.atName]["minPrec"] = min(
sD[catName][svTup.atName]["minPrec"], svTup.minPrec)
sD[catName][svTup.atName]["maxPrec"] = max(
sD[catName][svTup.atName]["maxPrec"], svTup.maxPrec)
sD[catName][svTup.atName]["count"] += 1
return sD
def __evalScanDataCoverage(self, scanDataD):
"""
ScanValue = collections.namedtuple('ScanValue', 'containerId, catName, atName, minWidth, maxWidth, minPrec, maxPrec')
ScanSummary = collections.namedtuple('ScanSummary', 'containerId, fromPath, scanDate, scanCategoryDict')
"""
# for populated sD[category] -> d[atName]->{count: #, instances: [id,id,id]}
sD = {}
for cId in scanDataD:
ssTup = scanDataD[cId]
dD = ssTup.scanCategoryDict
for catName in dD:
if catName not in sD:
sD[catName] = {}
for svTup in dD[catName]:
if svTup.atName not in sD[catName]:
sD[catName][svTup.atName] = {
"count": 0,
"instances": []
}
sD[catName][svTup.atName]["instances"].append(
svTup.containerId)
sD[catName][svTup.atName]["count"] += 1
return sD
def scanWorker(self, dataList, procName, optionsD, workingDir):
""" Multi-proc worker method for scanning repository data files-
"""
try:
_ = workingDir
startTime = self.__begin(message=procName)
# Recover common options
scanType = optionsD["scanType"]
contentType = optionsD["contentType"]
#
successList = []
retList = []
containerList = self.__getContainerList(dataList)
for container in containerList:
ret = self.__scanContainer(container)
successList.append(ret.fromPath)
retList.append(ret)
#
logger.debug(
"%s scanType %s contentType %spathlist length %d containerList length %d",
procName, scanType, contentType, len(dataList),
len(containerList))
ok = len(successList) == len(dataList)
#
self.__end(startTime, procName + " with status " + str(ok))
return successList, retList, []
except Exception as e:
logger.error("Failing with dataList %r", dataList)
logger.exception("Failing with %s", str(e))
return [], [], []
def __getContainerList(self, locatorObjList):
"""
"""
utcnow = datetime.datetime.utcnow()
ts = utcnow.strftime("%Y-%m-%d:%H:%M:%S")
cL = []
myContainerList = self.__rpP.getContainerList(locatorObjList)
for loc in locatorObjList:
myContainerList = self.__rpP.getContainerList([loc])
lPathL = self.__rpP.getLocatorPaths([loc])
for cA in myContainerList:
dc = DataCategory("rcsb_load_status",
["name", "load_date", "locator"],
[[cA.getName(), ts, lPathL[0]]])
logger.debug("data category %r", dc)
cA.append(dc)
cL.append(cA)
return cL
def __scanContainer(self, container):
""" Scan the input container for
Get the file name -
"""
cName = container.getName()
loadStatusObj = container.getObj("rcsb_load_status")
lName = loadStatusObj.getValue(attributeName="name", rowIndex=0)
lFilePath = loadStatusObj.getValue(attributeName="locator", rowIndex=0)
lDate = loadStatusObj.getValue(attributeName="load_date", rowIndex=0)
#
oD = {}
for objName in container.getObjNameList():
if objName == "rcsb_load_status":
continue
obj = container.getObj(objName)
afD = self.__attributeDataTypeD[
objName] if objName in self.__attributeDataTypeD else {}
atNameList = obj.getAttributeList()
wMin = {atName: 100000 for atName in atNameList}
wMax = {atName: -1 for atName in atNameList}
pMin = {atName: 100000 for atName in atNameList}
pMax = {atName: -1 for atName in atNameList}
for row in obj.getRowList():
for ii, val in enumerate(row):
valLen = len(val)
if (valLen == 0) or (val == "?") or (val == "."):
continue
atName = atNameList[ii]
wMin[atName] = min(wMin[atName], valLen)
wMax[atName] = max(wMax[atName], valLen)
if atName in afD and afD[atName] == "float":
vPrec = 0
try:
fields = val.split(".")
vPrec = len(fields[1])
pMin[atName] = min(pMin[atName], vPrec)
pMax[atName] = max(pMax[atName], vPrec)
except Exception as e:
logger.debug("Failed to process float %s %r %r %s",
atName, val, vPrec, str(e))
pMin[atName] = 0
pMax[atName] = 0
logger.debug("Got float for %s %r %r", atName, val,
vPrec)
else:
pMin[atName] = 0
pMax[atName] = 0
# ScanValue - containerId, catName, atName, minWidth, maxWidth, minPrec, maxPrec
oD[objName] = [
ScanValue(cName, objName, atN, wMin[atN], wMax[atN], pMin[atN],
pMax[atN]) for atN in wMax if wMax[atN] != -1
]
# ScanSummary containerId, fromPath, scanCategoryDict
#
ret = ScanSummary(lName, lFilePath, lDate, oD)
#
return ret
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)