def __reload(self, dirPath, useCache, fmt):
startTime = time.time()
fD = {}
ok = False
cofactorPath = self.__getCofactorDataPath(fmt=fmt)
#
logger.info("useCache %r featurePath %r", useCache, cofactorPath)
if useCache and self.__mU.exists(cofactorPath):
fD = self.__mU.doImport(cofactorPath, fmt=fmt)
else:
fU = FileUtil()
fU.mkdir(dirPath)
# ---
logger.info("Completed reload (%r) at %s (%.4f seconds)", ok,
time.strftime("%Y %m %d %H:%M:%S", time.localtime()),
time.time() - startTime)
return fD
def __fetchUrl(self, urlTarget, dirPath, useCache=False):
fU = FileUtil()
fn = fU.getFileName(urlTarget)
filePath = os.path.join(dirPath, fn)
if not (useCache and fU.exists(filePath)):
startTime = time.time()
ok2 = fU.get(urlTarget, filePath)
endTime = time.time()
if ok2:
logger.info(
"Fetched %s for resource file %s (status = %r) (%.4f seconds)",
urlTarget, filePath, ok2, endTime - startTime)
else:
logger.error(
"Failing fetch for %s for resource file %s (status = %r) (%.4f seconds)",
urlTarget, filePath, ok2, endTime - startTime)
#
return filePath
def __reload(self, dirPath, useCache):
startTime = time.time()
aD = {}
fU = FileUtil()
fU.mkdir(dirPath)
targetMechanismFilePath = self.getTargetMechanismDataPath()
#
if useCache and fU.exists(targetMechanismFilePath):
logger.info("useCache %r using %r", useCache,
targetMechanismFilePath)
qD = self.__mU.doImport(targetMechanismFilePath, fmt="json")
aD = qD["mechanism"] if "mechanism" in qD else {}
#
logger.info("Completed reload of (%d) at %s (%.4f seconds)", len(aD),
time.strftime("%Y %m %d %H:%M:%S", time.localtime()),
time.time() - startTime)
#
return aD
def __reload(self, dirPath, useCache):
startTime = time.time()
fD = {}
ok = False
mappingPath = self.__getMappingDataPath()
#
logger.info("useCache %r mappingPath %r", useCache, mappingPath)
if useCache and self.__mU.exists(mappingPath):
fD = self.__mU.doImport(mappingPath, fmt="json")
ok = True
else:
fU = FileUtil()
fU.mkdir(dirPath)
# ---
logger.info("Completed reload with status (%r) at %s (%.4f seconds)",
ok, time.strftime("%Y %m %d %H:%M:%S", time.localtime()),
time.time() - startTime)
return fD
def put(self, localPath, remotePath):
"""Put a local file on a remote FTP server.
Arguments:
localPath (str): local file path
remotePath (str): remote file path
Returns:
bool: True for success or false otherwise
"""
try:
# First, make sure the provided localPath represents a file, not a directory
if not os.path.isfile(localPath):
logger.error(
"put failing for localPath %s - path must be to a specific file, not a directory.",
localPath)
return False
fileU = FileUtil()
remotePathDir = fileU.getFilePath(remotePath)
self.mkdir(remotePathDir)
# If provided remotePath already exists and is a directory, put the file on the remote server using the local filename
# to avoid unintentionally overwriting an entire remote directory with a single file
if (os.path.exists(remotePath) and os.path.isdir(remotePath)):
localFileName = FileUtil().getFileName(localPath)
remoteFilePath = os.path.join(remotePath, localFileName)
else:
remoteFilePath = remotePath
with open(localPath, 'rb') as lFP:
self.__ftpClient.storbinary('STOR %s' % remoteFilePath, lFP)
if remoteFilePath in self.listdir(remotePathDir):
return True
else:
logger.error("put failing for localPath %s remoteFilePath %s",
localPath, remoteFilePath)
return False
except Exception as e:
if self.__raiseExceptions:
raise e
else:
logger.error(
"put failing for localPath %s remotePath %s with %s",
localPath, remotePath, str(e))
return False
def __pharosFixture(self):
try:
ok = False
fU = FileUtil()
srcPath = os.path.join(self.__dataPath, "Pharos")
dstPath = os.path.join(self.__cachePath, "Pharos-targets")
for fn in ["drug_activity", "cmpd_activity", "protein"]:
inpPath = os.path.join(srcPath, fn + ".tdd.gz")
outPath = os.path.join(dstPath, fn + ".tdd.gz")
fU.get(inpPath, outPath)
fU.uncompress(outPath, outputDir=dstPath)
fU.remove(outPath)
fU.put(os.path.join(srcPath, "pharos-readme.txt"),
os.path.join(dstPath, "pharos-readme.txt"))
ok = True
except Exception as e:
logger.exception("Failing with %s", str(e))
ok = False
return ok
def __reload(self, dirPath, baseVersion, useCache, **kwargs):
startTime = time.time()
mU = MarshalUtil(workPath=dirPath)
chemblDbUrl = kwargs.get(
"ChEMBLDbUrl",
"ftp://ftp.ebi.ac.uk/pub/databases/chembl/ChEMBLdb/latest/")
ok = False
fU = FileUtil()
fU.mkdir(dirPath)
#
# ChEMBL current version <baseVersion>,...
# template: chembl_<baseVersion>.fa.gz
#
targetFileName = "chembl_" + str(baseVersion) + ".fa.gz"
mappingFileName = "chembl_uniprot_mapping.txt"
#
chemblTargetPath = os.path.join(dirPath, targetFileName)
chemblMappingPath = os.path.join(dirPath, mappingFileName)
mappingFilePath = os.path.join(dirPath, "chembl_uniprot_mapping.json")
#
mapD = {}
if useCache and fU.exists(mappingFilePath):
logger.info("useCache %r using %r and %r and %r", useCache,
chemblTargetPath, chemblMappingPath, mappingFilePath)
mapD = mU.doImport(mappingFilePath, fmt="json")
else:
# Get the ChEMBL UniProt mapping file
url = os.path.join(chemblDbUrl, mappingFileName)
ok = fU.get(url, chemblMappingPath)
logger.info("Fetched %r url %s path %s", ok, url,
chemblMappingPath)
logger.info("Reading ChEMBL mapping file path %s", mappingFilePath)
rowL = mU.doImport(chemblMappingPath, fmt="tdd", rowFormat="list")
for row in rowL:
mapD[row[0]] = (row[1], row[2], row[3])
ok = mU.doExport(mappingFilePath, mapD, fmt="json")
logger.info("Processed mapping path %s (%d) %r", mappingFilePath,
len(mapD), ok)
#
# Get the target FASTA files --
for vers in range(baseVersion, baseVersion + 10):
logger.info("Now fetching version %r", vers)
self.__version = vers
targetFileName = "chembl_" + str(vers) + ".fa.gz"
chemblTargetPath = os.path.join(dirPath,
"chembl_targets_raw.fa.gz")
url = os.path.join(chemblDbUrl, targetFileName)
ok = fU.get(url, chemblTargetPath)
logger.info("Fetched %r url %s path %s", ok, url,
chemblTargetPath)
if ok:
break
#
logger.info("Completed reload at %s (%.4f seconds)",
time.strftime("%Y %m %d %H:%M:%S", time.localtime()),
time.time() - startTime)
#
return mapD
def __processAppendedSections(self,
appendConfigOption,
cachePath,
useCache=True):
"""Fetch and append configuration assets assigned to input configuration option.
Args:
appendConfigOption (str): reserved configuration option to hold a list of configuration asset locators
cachePath (str): path to store cached copies configuration assets
useCache (bool, optional): use existing cached configuration assets. Defaults to True.
Returns:
bool: True for success of False otherwise
"""
try:
ret = True
appendLocL = self.getList(appendConfigOption,
sectionName=self.__defaultSectionName)
logger.debug("appendLocL is %r", appendLocL)
if appendLocL:
cP = os.path.join(cachePath, "config")
fU = FileUtil(workPath=cP)
logger.debug("Fetching append sections from %r", appendLocL)
for appendLoc in appendLocL:
fn = fU.getFileName(appendLoc)
fp = os.path.join(cP, fn)
okF = True
if not (useCache and fU.exists(fp)):
# get a fresh copy from source
okF = fU.get(appendLoc, fp)
logger.debug("Fetched %r to %r", appendLoc, fp)
ok = self.appendConfig(fp)
ret = ret and ok and okF
except Exception as e:
logger.exception("Failing for option %r cachePath %r with %s",
appendConfigOption, cachePath, str(e))
ret = False
#
if not ret:
logger.error("Fetching appended sections failing %r", appendLocL)
return ret
def __reload(self, urlTarget, urlFallbackTarget, dirPath, useCache=True):
invD = {}
fU = FileUtil()
fn = fU.getFileName(urlTarget)
fp = os.path.join(dirPath, fn)
self.__mU.mkdir(dirPath)
#
if useCache and self.__mU.exists(fp):
invD = self.__mU.doImport(fp, fmt="json")
logger.debug("Reading cached inventory (%d)", len(invD))
else:
logger.info("Fetch inventory from %s", urlTarget)
ok = fU.get(urlTarget, fp)
if not ok:
ok = fU.get(urlFallbackTarget, fp)
#
if ok:
invD = self.__mU.doImport(fp, fmt="json")
#
return invD
def setUp(self):
self.__verbose = True
self.__pathPdbxDictionaryFile = os.path.join(TOPDIR, "rcsb",
"mock-data",
"dictionaries",
"mmcif_pdbx_v5_next.dic")
self.__pathTaxonomyFile = os.path.join(TOPDIR, "rcsb", "mock-data",
"NCBI", "names.dmp.gz")
self.__zipFileUrl = "https://inventory.data.gov/dataset/794cd3d7-4d28-4408-8f7d-84b820dbf7f2/resource/6b78ec0c-4980-4ad8-9cbd-2d6eb9eda8e7/download/myfoodapediadata.zip"
#
self.__ftpFileUrl = "ftp://ftp.wwpdb.org/pub/pdb/data/component-models/complete/chem_comp_model.cif.gz"
#
self.__workPath = os.path.join(HERE, "test-output")
self.__inpDirPath = os.path.join(HERE, "test-data")
self.__fileU = FileUtil()
self.__startTime = time.time()
logger.debug("Running tests on version %s", __version__)
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, **kwargs):
"""Utilities to access and update provenance details.
Args:
cfgOb ([type]): ConfigInfo() instance
cachePath ([type]): path to directory containing schema
useCache (bool, optional): use cached schema. Defaults to True.
"""
self.__cfgOb = cfgOb
self.__configName = self.__cfgOb.getDefaultSectionName()
self.__cachePath = cachePath
self.__useCache = useCache
#
self.__workPath = os.path.join(self.__cachePath, "work")
self.__provenanceCachePath = os.path.join(self.__cachePath, self.__cfgOb.get("PROVENANCE_INFO_CACHE_DIR", sectionName=self.__configName))
self.__provenanceLocator = self.__cfgOb.getPath("PROVENANCE_INFO_LOCATOR", sectionName=self.__configName)
#
self.__fileU = FileUtil(workPath=self.__workPath)
self.__fileU.mkdir(self.__provenanceCachePath)
self.__kwargs = kwargs
def stashDependencies(self,
url,
dirPath,
bundleLabel="A",
userName=None,
pw=None):
"""Store a copy of the bundled search dependencies remotely -
Args:
url (str): URL string for the destination host (e.g. sftp://myserver.net or None for a local file)
dirPath (str): directory path on the remote resource
bundleLabel (str, optional): optional label preppended to the stashed dependency bundle artifact (default='A')
userName (str, optional): optional access information. Defaults to None.
password (str, optional): optional access information. Defaults to None.
Returns:
bool: True for success or False otherwise
"""
try:
ok = False
fn = self.__makeBundleFileName(self.__dependFileName,
bundleLabel=bundleLabel)
if url and url.startswith("sftp://"):
sftpU = SftpUtil()
hostName = url[7:]
ok = sftpU.connect(hostName, userName, pw=pw, port=22)
if ok:
remotePath = os.path.join("/", dirPath, fn)
ok = sftpU.put(self.__dependTarFilePath, remotePath)
elif not url:
fileU = FileUtil()
remotePath = os.path.join(dirPath, fn)
ok = fileU.put(self.__dependTarFilePath, remotePath)
else:
logger.error("Unsupported stash protocol %r", url)
return ok
except Exception as e:
logger.exception("For %r %r failing with %s", url, dirPath, str(e))
return False
def __reload(self, urlTarget, dirPath, useCache=True):
"""Reload local cache of mapping resources to support validation report reader and translator.
Args:
urlTarget (list, str): URL for schema mapping file
dirPath (str): path to the directory containing cache files
useCache (bool, optional): flag to use cached files. Defaults to True.
Returns:
(object): instance of ValidationReportReader()
"""
mapD = {}
#
mU = MarshalUtil()
fU = FileUtil()
fn = fU.getFileName(urlTarget)
mappingFilePath = os.path.join(dirPath, fn)
mU.mkdir(dirPath)
#
# if not useCache:
# for fp in [mappingFilePath]:
# try:
# os.remove(fp)
# except Exception:
# pass
# #
logger.debug("Loading validation mapping data in %s (useCache %r)", fn,
useCache)
if useCache and fU.exists(mappingFilePath):
mapD = mU.doImport(mappingFilePath, fmt="json")
else:
logger.info("Fetching url %s to resource file %s", urlTarget,
mappingFilePath)
tS = uuid.uuid4().hex
tP = os.path.join(dirPath, "._" + tS)
ok = fU.get(urlTarget, tP)
if ok:
mapD = mU.doImport(tP, fmt="json")
os.replace(tP, mappingFilePath)
return mapD
def __rebuildCache(self, urlTargetPfam, urlTargetPfamFB, dirPath,
useCache):
pfamD = {}
fmt = "json"
ext = fmt if fmt == "json" else "pic"
pfamDataPath = os.path.join(dirPath, "pfam-data.%s" % ext)
#
logger.debug("Using cache data path %s", dirPath)
self.__mU.mkdir(dirPath)
#
if useCache and self.__mU.exists(pfamDataPath):
pfamD = self.__mU.doImport(pfamDataPath, fmt=fmt)
logger.debug("Pfam data length %d", len(pfamD))
elif not useCache:
# ------
fU = FileUtil()
logger.info("Fetch data from source %s in %s", urlTargetPfam,
dirPath)
fp = os.path.join(dirPath, fU.getFileName(urlTargetPfam))
ok = fU.get(urlTargetPfam, fp)
if not ok:
fp = os.path.join(dirPath, fU.getFileName(urlTargetPfamFB))
ok = fU.get(urlTargetPfamFB, fp)
logger.info("Fetch data fallback fetch status is %r", ok)
pfamD = self.__getPfamIndex(fp)
ok = self.__mU.doExport(pfamDataPath, pfamD, fmt=fmt)
logger.info("Caching %d in %s status %r", len(pfamD), pfamDataPath,
ok)
# ------
#
return pfamD
def __reload(self, dirPath, **kwargs):
startTime = time.time()
fD = {}
useCache = kwargs.get("useCache", True)
ok = False
cofactorPath = self.__getCofactorDataPath()
#
logger.info("useCache %r cofactorPath %r", useCache, cofactorPath)
if useCache and self.__mU.exists(cofactorPath):
fD = self.__mU.doImport(cofactorPath, fmt="json")
ok = True
else:
fU = FileUtil()
fU.mkdir(dirPath)
# ---
numCofactors = len(fD["cofactors"]) if fD and "cofactors" in fD else 0
logger.info(
"Completed reload of (%d) cofactors with status (%r) at %s (%.4f seconds)",
numCofactors, ok,
time.strftime("%Y %m %d %H:%M:%S", time.localtime()),
time.time() - startTime)
return fD
def __rebuildCache(self, urlTargetInterPro, urlTargetInterProFB, urlTargetInterProParent, urlTargetInterProParentFB, dirPath, useCache):
fmt = "json"
ext = fmt if fmt == "json" else "pic"
interProDataPath = os.path.join(dirPath, "interPro-data.%s" % ext)
#
logger.debug("Using cache data path %s", dirPath)
self.__mU.mkdir(dirPath)
#
if useCache and self.__mU.exists(interProDataPath):
rD = self.__mU.doImport(interProDataPath, fmt=fmt)
interProD = rD["index"]
interProParentD = rD["parents"]
logger.debug("InterPro index length %d parent length %d", len(interProD), len(interProParentD))
else:
# ------
fU = FileUtil()
logger.info("Fetch data from source %s in %s", urlTargetInterPro, dirPath)
fp = os.path.join(dirPath, fU.getFileName(urlTargetInterPro))
ok = fU.get(urlTargetInterPro, fp)
if not ok:
fp = os.path.join(dirPath, fU.getFileName(urlTargetInterProFB))
ok = fU.get(urlTargetInterProFB, fp)
logger.info("Fetch data fallback fetch status is %r", ok)
interProD = self.__getInterProIndex(fp)
logger.info("Caching %d in %s status %r", len(interProD), interProDataPath, ok)
# ------
logger.info("Fetch data from source %s in %s", urlTargetInterProParent, dirPath)
fp = os.path.join(dirPath, fU.getFileName(urlTargetInterProParent))
ok = fU.get(urlTargetInterProParent, fp)
if not ok:
fp = os.path.join(dirPath, fU.getFileName(urlTargetInterProParentFB))
ok = fU.get(urlTargetInterProParentFB, fp)
logger.info("Fetch data fallback fetch status is %r", ok)
interProParentD = self.__getInterProParents(fp)
#
ok = self.__mU.doExport(interProDataPath, {"index": interProD, "parents": interProParentD}, fmt=fmt)
#
return interProD, interProParentD
def fetchBundle(self, localRestoreDirPath, url, remoteDirPath, remoteStashPrefix="A", userName=None, password=None):
"""Restore bundled dependencies from remote storage and unbundle these in the
current local cache directory.
Args:
localRestoreDirPath (str): local restore path
url (str): remote URL
remoteDirPath (str): remote directory path on the remote resource
remoteStashPrefix (str, optional): optional label preppended to the stashed dependency bundle artifact (default='A')
userName (str, optional): optional access information. Defaults to None.
password (str, optional): optional access information. Defaults to None.
"""
try:
ok = False
fileU = FileUtil()
fn = self.__makeBundleFileName(self.__baseBundleFileName, remoteStashPrefix=remoteStashPrefix)
if not url:
remotePath = os.path.join(remoteDirPath, fn)
ok = fileU.get(remotePath, self.__localStashTarFilePath)
elif url and (url.startswith("http://") or url.startswith("https://")):
remotePath = url + os.path.join("/", remoteDirPath, fn)
ok = fileU.get(remotePath, self.__localStashTarFilePath)
elif url and url.startswith("sftp://"):
sftpU = SftpUtil()
ok = sftpU.connect(url[7:], userName, pw=password, port=22)
if ok:
remotePath = os.path.join(remoteDirPath, fn)
ok = sftpU.get(remotePath, self.__localStashTarFilePath)
else:
logger.error("Unsupported protocol %r", url)
if ok:
ok = fileU.unbundleTarfile(self.__localStashTarFilePath, dirPath=localRestoreDirPath)
return ok
except Exception as e:
logger.exception("For %r %r Failing with %s", url, remoteDirPath, str(e))
ok = False
return ok
def __init__(self,
cfgOb,
cachePath,
useCache=True,
rebuildFlag=False,
**kwargs):
"""A collection of schema build and caching methods.
Args:
cfgOb (object): ConfigInfo() instance
cachePath (str): path to directory containing schema
useCache (bool, optional): use cached schema. Defaults to True.
rebuildFlag (bool, optional): on-the-fly rebuild and cache schema
"""
self.__cfgOb = cfgOb
self.__configName = self.__cfgOb.getDefaultSectionName()
self.__cachePath = os.path.abspath(cachePath)
self.__useCache = useCache
self.__rebuildFlag = rebuildFlag
self.__useCache = rebuildFlag if rebuildFlag else useCache
#
self.__workPath = os.path.join(self.__cachePath, "work")
self.__fileU = FileUtil(
workPath=os.path.join(self.__cachePath, "work"))
self.__schemaCachePath = os.path.join(
self.__cachePath,
self.__cfgOb.get("SCHEMA_DEFINITION_CACHE_DIR",
sectionName=self.__configName))
self.__jsonSchemaCachePath = os.path.join(
self.__cachePath,
self.__cfgOb.get("JSON_SCHEMA_DEFINITION_CACHE_DIR",
sectionName=self.__configName))
self.__fileU.mkdir(self.__schemaCachePath)
self.__fileU.mkdir(self.__jsonSchemaCachePath)
self.__kwargs = kwargs
def __reload(self, dirPath, useCache):
startTime = time.time()
aD = {}
allIdD = {}
fU = FileUtil()
fU.mkdir(dirPath)
targetActivityFilePath = self.getTargetActivityDataPath()
#
if useCache and fU.exists(targetActivityFilePath):
logger.info("useCache %r using %r", useCache, targetActivityFilePath)
qD = self.__mU.doImport(targetActivityFilePath, fmt="json")
aD = qD["activity"] if "activity" in qD else {}
idL = qD["all_ids"] if "all_ids" in qD else []
allIdD = {k: k in aD for k in idL}
#
logger.info(
"Completed reload (%d activities) (%d tried identifiers) at %s (%.4f seconds)",
len(aD),
len(allIdD),
time.strftime("%Y %m %d %H:%M:%S", time.localtime()),
time.time() - startTime,
)
#
return aD, allIdD
def buildDependenices(self, ccUrlTarget, birdUrlTarget, **kwargs):
"""Convenience method to build configuration and static dependencies for the chemical search services.
Args:
ccUrlTarget (str): path to source concatenated chemical component definition file
birdUrlTarget (str): path to the source concatenated BIRD definition file
Other options are propagated to configurations of the wrapped classes in __bootstrapConfig()
"""
try:
okT = False
ok1 = self.setConfig(ccUrlTarget=ccUrlTarget,
birdUrlTarget=birdUrlTarget,
**kwargs)
useCache = kwargs.get("useCache", False)
ok2 = self.updateChemCompIndex(useCache=useCache)
ok3 = self.updateSearchIndex(useCache=useCache)
ok4 = self.updateSearchMoleculeProvider(useCache=useCache)
okBuild = ok1 and ok2 and ok3 and ok4
if okBuild:
fileU = FileUtil()
dirPathList = [
os.path.join(self.__cachePath, subDir)
for subDir in ["chem_comp", "oe_mol", "config"]
]
okT = fileU.bundleTarfile(self.__dependTarFilePath,
dirPathList,
mode="w:gz",
recursive=True)
#
return okT and okBuild
except Exception as e:
logger.exception("Failing build with %r and %r with %s",
ccUrlTarget, birdUrlTarget, str(e))
return False
def __fetchFromSource(self, urlTarget):
"""Fetch the classification names and domain assignments from the ECOD repo."""
fU = FileUtil()
fn = fU.getFileName(urlTarget)
fp = os.path.join(self.__dirPath, fn)
if not fU.exists(fp):
fU.get(urlTarget, fp)
#
with open(fp, "r", encoding="utf-8") as ifh:
line = ifh.readline()
line = ifh.readline()
line = ifh.readline()
ff = line[:-1].split()
self.__version = ff[-1]
#
nmL = self.__mU.doImport(fp, fmt="list", uncomment=True)
fU.remove(fp)
#
return nmL
def pushBundle(self, gitRepositoryPath, accessToken, gitHost="github.com", gitBranch="master", remoteStashPrefix="A", maxSizeMB=95):
"""Push bundle to remote stash git repository.
Args:
gitRepositoryPath (str): git repository path (e.g., rcsb/py-rcsb_exdb_assets_stash)
accessToken (str): git repository access token
gitHost (str, optional): git repository host name. Defaults to github.com.
gitBranch (str, optional): git branch name. Defaults to master.
remoteStashPrefix (str, optional): optional label preppended to the stashed dependency bundle artifact (default='A')
maxSizeMB (int, optional): maximum stash bundle file size that will be committed. Defaults to 95MB.
Returns:
bool: True for success or False otherwise
"""
try:
ok = False
gU = GitUtil(token=accessToken, repositoryHost=gitHost)
fU = FileUtil()
localRepositoryPath = os.path.join(self.__localBundlePath, "stash_repository")
fn = self.__makeBundleFileName(self.__baseBundleFileName, remoteStashPrefix=remoteStashPrefix)
#
# Update existing local repository, otherwise clone a new copy
if fU.exists(localRepositoryPath):
ok = gU.pull(localRepositoryPath, branch=gitBranch)
logger.debug("After pull status %r", gU.status(localRepositoryPath))
else:
ok = gU.clone(gitRepositoryPath, localRepositoryPath, branch=gitBranch)
#
# Split all bundles
mbSize = float(fU.size(self.__localStashTarFilePath)) / 1000000.0
logger.info("Splitting bundle %r (%.3f MB/Max %d MB)", fn, mbSize, maxSizeMB)
sj = SplitJoin()
splitDirPath = os.path.join(localRepositoryPath, "stash", fn[:-7])
sj.split(self.__localStashTarFilePath, splitDirPath, maxSizeMB=maxSizeMB)
fU.remove(self.__localStashTarFilePath)
# else:
# fU.put(self.__localStashTarFilePath, os.path.join(localRepositoryPath, "stash", fn))
ok = gU.addAll(localRepositoryPath, branch=gitBranch)
ok = gU.commit(localRepositoryPath, branch=gitBranch)
logger.debug("After commit status %r", gU.status(localRepositoryPath))
#
if accessToken:
ok = gU.push(localRepositoryPath, branch=gitBranch)
logger.info("After push status %r", gU.status(localRepositoryPath))
#
return ok
except Exception as e:
logger.exception("For %r %r failing with %s", gitHost, gitRepositoryPath, str(e))
return False
def setUp(self):
self.__workPath = os.path.join(HERE, "test-output")
#
self.__testLogFileMin = os.path.join(self.__workPath,
"logfile-min.json")
self.__testLogFileDetailed = os.path.join(self.__workPath,
"logfile-detailed.json")
fU = FileUtil()
fU.remove(self.__testLogFileMin)
fU.remove(self.__testLogFileDetailed)
self.__startTime = time.time()
logger.debug("Starting %s at %s", self.id(),
time.strftime("%Y %m %d %H:%M:%S", time.localtime()))
def testAABuildDependenciesAndStash(self):
"""Test case - build, stash and restore dependencies -"""
try:
ccsw = ChemCompSearchWrapper()
ccUrlTarget = os.path.join(
self.__dataPath,
"components-abbrev.cif") if not self.__testFlagFull else None
birdUrlTarget = os.path.join(
self.__dataPath,
"prdcc-abbrev.cif") if not self.__testFlagFull else None
ok = ccsw.buildDependenices(ccUrlTarget=ccUrlTarget,
birdUrlTarget=birdUrlTarget)
self.assertTrue(ok)
#
if self.__testStash:
url = "sftp://bl-east.rcsb.org"
userName = ""
pw = ""
dirPath = "4-coastal"
ok = ccsw.stashDependencies(url,
dirPath,
userName=userName,
pw=pw)
self.assertTrue(ok)
#
fileU = FileUtil()
fileU.remove(self.__cachePath)
#
url = "http://bl-east.rcsb.org"
ok = ccsw.restoreDependencies(url, dirPath)
#
fileU.remove(self.__cachePath)
#
url = "sftp://bl-east.rcsb.org"
ok = ccsw.restoreDependencies(url,
dirPath,
userName=userName,
pw=pw)
self.assertTrue(ok)
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def __reload(self, urlTarget, dirPath, useCache=True):
""" Reload input GO OBO ontology file and return a nx graph object.
'
Returns:
dictionary[goId] = {'name_list': ... , 'id_list': ... 'depth_list': ... }
"""
goGraph = None
#
# mU = MarshalUtil()
fU = FileUtil()
fn = fU.getFileName(urlTarget)
oboFilePath = os.path.join(dirPath, fn)
fU.mkdir(dirPath)
#
if not useCache:
for fp in [oboFilePath]:
try:
os.remove(fp)
except Exception:
pass
#
if useCache and fU.exists(oboFilePath):
goGraph = obonet.read_obo(oboFilePath)
else:
logger.info("Fetching url %s to resource file %s", urlTarget,
oboFilePath)
ok = fU.get(urlTarget, oboFilePath)
if ok:
goGraph = obonet.read_obo(oboFilePath)
if goGraph:
logger.info("Reading %d nodes and %d edges", len(goGraph),
goGraph.number_of_edges())
else:
logger.info("Go graph construction failing")
#
return goGraph
def get(self, remotePath, localPath):
"""Get a file from a remote FTP server.
Arguments:
remotePath (str): remote file path
localPath (str): local file path
Returns:
bool: True for success or false otherwise
"""
try:
fileU = FileUtil()
fileU.mkdirForFile(localPath)
# If provided localPath already exists and is a directory, retrieve the file using the name on the remote server
# to avoid unintentionally overwriting an entire local directory with a single retrieved file
if (os.path.exists(localPath) and os.path.isdir(localPath)):
remoteFileName = FileUtil().getFileName(remotePath)
localFilePath = os.path.join(localPath, remoteFileName)
else:
localFilePath = localPath
with open(localFilePath, 'wb') as lFP:
self.__ftpClient.retrbinary('RETR %s' % remotePath, lFP.write)
ok = fileU.exists(localFilePath)
if ok:
return True
else:
logger.error("get failing for remotePath %s localFilePath %s",
remotePath, localFilePath)
return False
except Exception as e:
if self.__raiseExceptions:
raise e
else:
logger.error(
"get failing for remotePath %s localPath %s with %s",
remotePath, localPath, str(e))
return False
class SchemaProvider(SingletonClass):
""" A collection of schema build and caching methods.
Static cache worflow:
<authorative source> <-- <cache dir> <- client API
Compute workflow:
<dependent resource files, config file, dictionaries> -> [schema builder] --> <schema def> --> <Json schema>
"""
def __init__(self, cfgOb, cachePath, useCache=True, rebuildFlag=False, **kwargs):
"""A collection of schema build and caching methods.
Args:
cfgOb (object): ConfigInfo() instance
cachePath (str): path to directory containing schema
useCache (bool, optional): use cached schema. Defaults to True.
rebuildFlag (bool, optional): on-the-fly rebuild and cache schema
"""
self.__cfgOb = cfgOb
self.__configName = self.__cfgOb.getDefaultSectionName()
self.__cachePath = os.path.abspath(cachePath)
self.__useCache = useCache
self.__rebuildFlag = rebuildFlag
self.__useCache = rebuildFlag if rebuildFlag else useCache
#
self.__workPath = os.path.join(self.__cachePath, "work")
self.__fileU = FileUtil(workPath=os.path.join(self.__cachePath, "work"))
self.__schemaCachePath = os.path.join(self.__cachePath, self.__cfgOb.get("SCHEMA_DEFINITION_CACHE_DIR", sectionName=self.__configName))
self.__jsonSchemaCachePath = os.path.join(self.__cachePath, self.__cfgOb.get("JSON_SCHEMA_DEFINITION_CACHE_DIR", sectionName=self.__configName))
self.__fileU.mkdir(self.__schemaCachePath)
self.__fileU.mkdir(self.__jsonSchemaCachePath)
self.__kwargs = kwargs
def getSchemaOptions(self, schemaLevel, extraOpts=None):
opts = extraOpts + "|" if extraOpts else ""
if schemaLevel == "full":
return opts + "mandatoryKeys|mandatoryAttributes|bounds|enums|rcsb"
elif schemaLevel in ["min", "minimum"]:
return opts + "mandatoryKeys|enums|rcsb"
else:
return opts
def getSchemaInfo(self, databaseName, dataTyping="ANY"):
"""Convenience method to return essential schema details for the input repository content type.
Args:
databaseName (str): schema name (e.g. pdbx, bird, chem_comp, ...)
dataTyping (str, optional): Application name for the target schema (e.g. ANY, SQL, ...)
Returns:
tuple: SchemaDefAccess(object), target database name, target collection name list, primary index attribute list
"""
sd = None
dbName = None
collectionNameList = []
docIndexD = {}
try:
mU = MarshalUtil(workPath=self.__workPath)
schemaLocator = self.__getSchemaDefLocator(databaseName, dataTyping=dataTyping)
if self.__rebuildFlag:
filePath = os.path.join(self.__schemaCachePath, self.__fileU.getFileName(schemaLocator))
self.makeSchemaDef(databaseName, dataTyping=dataTyping, saveSchema=True)
else:
filePath = self.__reload(schemaLocator, self.__schemaCachePath, useCache=self.__useCache)
if not filePath:
logger.error("Unable to recover schema %s (%s)", databaseName, dataTyping)
logger.debug("ContentType %r dataTyping %r schemaLocator %r", databaseName, dataTyping, schemaLocator)
schemaDef = mU.doImport(filePath, fmt="json")
if schemaDef:
logger.debug("Using cached schema definition for %s application %s", databaseName, dataTyping)
sd = SchemaDefAccess(schemaDef)
if sd:
dbName = sd.getDatabaseName()
collectionInfoList = sd.getCollectionInfo()
logger.debug("Schema %s database name %s collections %r", databaseName, dbName, collectionInfoList)
for cd in collectionInfoList:
collectionName = cd["NAME"]
collectionNameList.append(collectionName)
docIndexD[collectionName] = sd.getDocumentIndices(collectionName)
except Exception as e:
logger.exception("Retreiving schema %s for %s failing with %s", databaseName, dataTyping, str(e))
return sd, dbName, collectionNameList, docIndexD
def schemaDefCompare(self, databaseName, dataTyping="ANY"):
"""Compare computed schema defintion with current source/cached version.
Args:
databaseName (str): schema definition name for comparison
dataTyping (str, optional): data type conventions for the schema comparison. Defaults to "ANY".
Returns:
(str): file path for schema difference or None
"""
mU = MarshalUtil(workPath=self.__workPath)
schemaDiffPath = os.path.join(self.__cachePath, "schema_diff")
mU.mkdir(schemaDiffPath)
schemaPath = self.__getSchemaDefLocator(databaseName, dataTyping=dataTyping)
fn = self.__fileU.getFileName(schemaPath)
sD = self.makeSchemaDef(databaseName, dataTyping=dataTyping)
v2 = sD["DATABASE_VERSION"]
# ----
# tPath = os.path.join(self.__schemaCachePath, self.__fileU.getFileName(schemaPath) + "-test")
# logger.info("Exporting schema def to %s", tPath)
# mU.doExport(tPath, sD, fmt="json", indent=3)
# sD = mU.doImport(tPath, fmt="json")
# ----
cPath = os.path.join(self.__schemaCachePath, self.__fileU.getFileName(schemaPath))
sDCache = mU.doImport(cPath, fmt="json")
v1 = sDCache["DATABASE_VERSION"]
#
numDiff, difD = self.schemaCompare(sDCache, sD)
#
# jD = diff(sDCache, sD, syntax="explicit", marshal=True)
diffPath = None
if numDiff:
bn, _ = os.path.splitext(fn)
diffPath = os.path.join(schemaDiffPath, bn + "-" + v1 + "-" + v2 + "-diff.json")
# logger.info("diff for %s %s = \n%s", databaseName, dataTyping, pprint.pformat(difD, indent=3, width=100))
mU.doExport(diffPath, difD, fmt="json", indent=3)
#
return diffPath
def jsonSchemaCompare(self, databaseName, collectionName, encodingType, level, extraOpts=None):
"""Compare computed JSON schema defintion with current source/cached version.
Args:
databaseName (str): schema name
collectionName (str): collection name
encodingType (str): schema data type conventions (JSON|BSON)
level (str): metadata level (min|full)
extraOpts (str): extra schema construction options
Returns:
(str): path to the difference file or None
"""
mU = MarshalUtil(workPath=self.__workPath)
schemaDiffPath = os.path.join(self.__cachePath, "schema_diff")
mU.mkdir(schemaDiffPath)
schemaLocator = self.__getJsonSchemaLocator(databaseName, collectionName, encodingType, level)
fn = self.__fileU.getFileName(schemaLocator)
schemaPath = os.path.join(self.__jsonSchemaCachePath, fn)
#
sD = self.makeSchema(databaseName, collectionName, encodingType=encodingType, level=level, saveSchema=False, extraOpts=extraOpts)
v2 = self.__getSchemaVersion(sD)
# ----
# tPath = os.path.join(self.__jsonSchemaCachePath, self.__fileU.getFileName(schemaPath) + "-test")
# logger.info("Exporting json schema to %s", tPath)
# mU.doExport(tPath, sD, fmt="json", indent=3)
# ----
#
sDCache = mU.doImport(schemaPath, fmt="json")
v1 = self.__getSchemaVersion(sDCache)
if not v1:
logger.error("no version for %s - %s %s", schemaLocator, databaseName, collectionName)
#
numDiff, difD = self.schemaCompare(sDCache, sD)
# jD = diff(sDCache, sD, marshal=True, syntax="explicit")
diffPath = None
if numDiff:
logger.debug("diff for %s %s %s %s = \n%s", databaseName, collectionName, encodingType, level, pprint.pformat(difD, indent=3, width=100))
bn, _ = os.path.splitext(fn)
diffPath = os.path.join(schemaDiffPath, bn + "-" + v1 + "-" + v2 + "-diff.json")
mU.doExport(diffPath, difD, fmt="json", indent=3)
return diffPath
def __getSchemaVersion(self, jsonSchema):
try:
comment = jsonSchema["$comment"] if "$comment" in jsonSchema else ""
ff = comment.split(":")
version = ff[1].strip()
return version
except Exception as e:
logger.exception("Failing for with %s", str(e))
return ""
def __getSchemaDefLocator(self, databaseName, dataTyping="ANY"):
"""Internal method returning schema definition path for the input content type and application.
Defines schema definition naming convention -
Args:
databaseName (str): schema name (e.g. pdbx, bird, chem_comp, ...)
dataTyping (str, optional): Application name for the target schema (e.g. ANY, SQL, ...)
Returns:
str: schema definition file locator
"""
schemaLocator = None
try:
locPath = self.__cfgOb.get("SCHEMA_DEFINITION_LOCATOR_PATH", sectionName=self.__configName)
fn = "schema_def-%s-%s.json" % (databaseName, dataTyping.upper())
schemaLocator = os.path.join(locPath, fn)
except Exception as e:
logger.exception("Retreiving schema definition path %s for %s failing with %s", databaseName, dataTyping, str(e))
return schemaLocator
def __getJsonSchemaLocator(self, databaseName, collectionName, encodingType="BSON", level="full"):
"""Internal method returning JSON schema path for the input collection data type convention and level.
Defines the JSON/BSON schema naming convention -
Args:
databaseName (str): database name in the document store
collectionName (str): collection name in document store
encodingType (str, optional): data type convention (BSON|JSON)
level (str, optional): Completeness of the schema (e.g. min or full)
Returns:
str: schema file locator
"""
schemaLocator = None
try:
sdType = None
sLevel = None
schemaLocator = None
if encodingType.upper() in ["JSON", "BSON"]:
sdType = encodingType.lower()
if level.lower() in ["min", "minimun"]:
sLevel = "min"
elif level.lower() in ["full"]:
sLevel = level.lower()
#
if sdType and sLevel:
locPath = self.__cfgOb.get("JSON_SCHEMA_DEFINITION_LOCATOR_PATH", sectionName=self.__configName)
fn = "%s-%s-db-%s-col-%s.json" % (sdType, sLevel, databaseName, collectionName)
schemaLocator = os.path.join(locPath, fn)
else:
logger.error("Unsupported schema options: %s level %r type %r", collectionName, level, encodingType)
schemaLocator = None
except Exception as e:
logger.debug("Retreiving JSON schema definition for %s type %s failing with %s", collectionName, encodingType, str(e))
#
return schemaLocator
def __reload(self, locator, dirPath, useCache=True):
#
fn = self.__fileU.getFileName(locator)
filePath = os.path.join(dirPath, fn)
logger.debug("Target cache filePath %s", filePath)
self.__fileU.mkdir(dirPath)
if not useCache:
try:
os.remove(filePath)
except Exception:
pass
#
if useCache and self.__fileU.exists(filePath):
ok = True
else:
logger.info("Fetch data from source %s to %s", locator, filePath)
ok = self.__fileU.get(locator, filePath)
return filePath if ok else None
def getJsonSchema(self, databaseName, collectionName, encodingType="BSON", level="full", extraOpts=None):
"""Return JSON schema (w/ BSON types) object for the input collection and level.and
Args:
databaseName (str): database name
collectionName (str): collection name in document store
encodingType (str, optional): data type convention (BSON|JSON)
level (str, optional): Completeness of the schema (e.g. min or full)
Returns:
dict: Schema object
"""
sObj = None
schemaLocator = self.__getJsonSchemaLocator(databaseName, collectionName, encodingType=encodingType, level=level)
#
if self.__rebuildFlag:
filePath = os.path.join(self.__schemaCachePath, self.__fileU.getFileName(schemaLocator))
self.makeSchema(databaseName, collectionName, encodingType=encodingType, level=level, extraOpts=extraOpts)
else:
filePath = self.__reload(schemaLocator, self.__jsonSchemaCachePath, useCache=self.__useCache)
mU = MarshalUtil(workPath=self.__workPath)
if filePath and mU.exists(filePath):
mU = MarshalUtil(workPath=self.__workPath)
sObj = mU.doImport(filePath, fmt="json")
else:
logger.debug("Failed to read schema for %s %r", collectionName, level)
return sObj
def makeSchema(self, databaseName, collectionName, encodingType="BSON", level="full", saveSchema=False, extraOpts=None):
try:
smb = SchemaDefBuild(databaseName, self.__cfgOb, cachePath=self.__cachePath)
#
cD = None
stU = encodingType.upper()
cD = smb.build(collectionName, dataTyping=stU, encodingType=stU, enforceOpts=self.getSchemaOptions(level, extraOpts=extraOpts))
if cD and saveSchema:
schemaLocator = self.__getJsonSchemaLocator(databaseName, collectionName, encodingType=encodingType, level=level)
localPath = os.path.join(self.__jsonSchemaCachePath, self.__fileU.getFileName(schemaLocator))
mU = MarshalUtil(workPath=self.__workPath)
mU.doExport(localPath, cD, fmt="json", indent=3, enforceAscii=False)
except Exception as e:
logger.exception("Building schema %s collection %s failing with %s", databaseName, collectionName, str(e))
return cD
def makeSchemaDef(self, databaseName, dataTyping="ANY", saveSchema=False):
schemaDef = None
try:
smb = SchemaDefBuild(databaseName, self.__cfgOb, cachePath=self.__cachePath)
schemaDef = smb.build(dataTyping=dataTyping, encodingType="rcsb")
if schemaDef and saveSchema:
schemaLocator = self.__getSchemaDefLocator(databaseName, dataTyping=dataTyping)
localPath = os.path.join(self.__schemaCachePath, self.__fileU.getFileName(schemaLocator))
mU = MarshalUtil(workPath=self.__workPath)
mU.doExport(localPath, schemaDef, fmt="json", indent=3, enforceAscii=False)
except Exception as e:
logger.exception("Building schema %s failing with %s", databaseName, str(e))
return schemaDef
def schemaCompare(self, orgD, newD):
""" Compute the difference of nested dictionaries.
"""
fOrgD = self.__flatten(orgD)
fNewD = self.__flatten(newD)
if len(fOrgD) != len(fNewD):
logger.debug("Schema lengths differ: org %d new %d", len(fOrgD), len(fNewD))
#
addedD = {k: fNewD[k] for k in set(fNewD) - set(fOrgD)}
removedD = {k: fOrgD[k] for k in set(fOrgD) - set(fNewD)}
changedOrgD = {k: fOrgD[k] for k in set(fOrgD) & set(fNewD) if fOrgD[k] != fNewD[k]}
changedNewD = {k: fNewD[k] for k in set(fOrgD) & set(fNewD) if fOrgD[k] != fNewD[k]}
chD = {}
for ky in changedOrgD:
kyS = ".".join(ky)
vOrg = changedOrgD[ky]
vNew = changedNewD[ky]
if isinstance(vOrg, (list, tuple)) and isinstance(vNew, (list, tuple)):
# logger.info(" >> %r vOrg %r vNew %r", ky, vOrg, vNew)
dV = list(set(vNew) - set(vOrg))
if dV:
chD[kyS] = {"diff": dV}
else:
chD[kyS] = {"from": vOrg, "to": vNew}
#
nT = len(addedD) + len(removedD) + len(chD)
diffD = {"added": [".".join(kk) for kk in addedD.keys()], "removed": [".".join(kk) for kk in removedD.keys()], "changed": chD}
return nT, diffD
def __flatten(self, inpDict, prefix=None):
prefix = prefix[:] if prefix else []
outDict = {}
for key, value in inpDict.items():
if isinstance(value, dict) and value:
deeper = self.__flatten(value, prefix + [key])
outDict.update({tuple(key2): val2 for key2, val2 in deeper.items()})
elif isinstance(value, (list, tuple)) and value:
for index, sublist in enumerate(value, start=1):
if isinstance(sublist, dict) and sublist:
deeper = self.__flatten(sublist, prefix + [key] + [str(index)])
outDict.update({tuple(key2): val2 for key2, val2 in deeper.items()})
else:
outDict[tuple(prefix + [key] + [str(index)])] = value
else:
outDict[tuple(prefix + [key])] = value
return outDict
def __flattenX(self, inpDict, prefix=None):
prefix = prefix[:] if prefix else []
# separator = "."
outDict = {}
for key, value in inpDict.items():
if isinstance(value, dict) and value:
deeper = self.__flatten(value, prefix + [key])
outDict.update({tuple(key2): val2 for key2, val2 in deeper.items()})
elif isinstance(value, list) and value:
for index, sublist in enumerate(value, start=1):
if isinstance(sublist, dict) and sublist:
deeper = self.__flatten(sublist, prefix + [key] + [str(index)])
outDict.update({tuple(key2): val2 for key2, val2 in deeper.items()})
else:
outDict[tuple(prefix + [key] + [str(index)])] = value
else:
outDict[tuple(prefix + [key])] = value
return outDict
def __flattenOrg(self, inpDict, separator=".", prefix=""):
outDict = {}
for key, value in inpDict.items():
if isinstance(value, dict) and value:
deeper = self.__flattenOrg(value, separator, prefix + key + separator)
outDict.update({key2: val2 for key2, val2 in deeper.items()})
elif isinstance(value, list) and value:
for index, sublist in enumerate(value, start=1):
if isinstance(sublist, dict) and sublist:
deeper = self.__flattenOrg(sublist, separator, prefix + key + separator + str(index) + separator)
outDict.update({key2: val2 for key2, val2 in deeper.items()})
else:
outDict[prefix + key + separator + str(index)] = value
else:
outDict[prefix + key] = value
return outDict
def __dictGen(self, indict, pre=None):
pre = pre[:] if pre else []
if isinstance(indict, dict):
for key, value in indict.items():
if isinstance(value, dict):
for dD in self.__dictGen(value, pre + [key]):
yield dD
elif isinstance(value, list) or isinstance(value, tuple):
for v in value:
for dD in self.__dictGen(v, pre + [key]):
yield dD
else:
yield pre + [key, value]
else:
yield indict
def __init__(self, **kwargs):
self.__fileU = FileUtil(**kwargs)
class IoUtil(object):
def __init__(self, **kwargs):
self.__fileU = FileUtil(**kwargs)
def serialize(self, filePath, myObj, fmt="pickle", **kwargs):
"""Public method to serialize format appropriate objects
Args:
filePath (str): local file path'
myObj (object): format appropriate object to be serialized
format (str, optional): one of ['mmcif', mmcif-dict', json', 'list', 'text-dump', pickle' (default)]
**kwargs: additional keyword arguments passed to worker methods -
Returns:
bool: status of serialization operation; true for success or false otherwise
"""
ret = False
fmt = str(fmt).lower()
ret = self.__fileU.mkdirForFile(filePath)
if not ret:
return ret
if fmt in ["mmcif"]:
ret = self.__serializeMmCif(filePath, myObj, **kwargs)
elif fmt in ["json"]:
ret = self.__serializeJson(filePath, myObj, **kwargs)
elif fmt in ["pickle"]:
ret = self.__serializePickle(filePath, myObj, **kwargs)
elif fmt in ["list"]:
ret = self.__serializeList(filePath,
myObj,
enforceAscii=True,
**kwargs)
elif fmt in ["mmcif-dict"]:
ret = self.__serializeMmCifDict(filePath, myObj, **kwargs)
elif fmt in ["text-dump"]:
ret = self.__textDump(filePath, myObj, **kwargs)
elif fmt in ["fasta"]:
ret = self.__serializeFasta(filePath, myObj, **kwargs)
elif fmt in ["csv"]:
ret = self.__serializeCsv(filePath, myObj, **kwargs)
else:
pass
return ret
def deserialize(self, filePath, fmt="pickle", **kwargs):
"""Public method to deserialize objects in supported formats.
Args:
filePath (str): local file path
format (str, optional): one of ['mmcif', 'json', 'list', ..., 'pickle' (default)]
**kwargs: additional keyword arguments passed to worker methods -
Returns:
object: deserialized object data
"""
fmt = str(fmt).lower()
if fmt in ["mmcif"]:
ret = self.__deserializeMmCif(filePath, **kwargs) # type: ignore
elif fmt in ["json"]:
ret = self.__deserializeJson(filePath, **kwargs) # type: ignore
elif fmt in ["pickle"]:
ret = self.__deserializePickle(filePath, **kwargs) # type: ignore
elif fmt in ["list"]:
ret = self.__deserializeList(filePath, enforceAscii=True,
**kwargs) # type: ignore
elif fmt in ["mmcif-dict"]:
ret = self.__deserializeMmCifDict(filePath,
**kwargs) # type: ignore
elif fmt in ["fasta"]:
ret = self.__deserializeFasta(filePath, **kwargs) # type: ignore
# elif fmt in ["vrpt-xml-to-cif"]:
# ret = self.__deserializeVrptToCif(filePath, **kwargs) # type: ignore
elif fmt in ["csv", "tdd"]:
delimiter = kwargs.get("csvDelimiter",
"," if fmt == "csv" else "\t")
ret = self.__deserializeCsv(filePath,
delimiter=delimiter,
**kwargs) # type: ignore
elif fmt in ["xml"]:
ret = self.__deserializeXml(filePath, **kwargs) # type: ignore
else:
ret = None # type: ignore
return ret
def __sliceInChunks(self, myList, numChunks):
mc = min(len(myList), numChunks)
chunkSize = int(len(myList) / mc)
if len(myList) % mc:
chunkSize += 1
for i in range(0, len(myList), chunkSize):
yield myList[i:i + chunkSize]
def serializeInParts(self,
filePath,
myObj,
numParts,
fmt="json",
**kwargs):
"""Public method to serialize format appropriate (json, pickle) objects in multiple parts
Args:
filePath (str): local file path
myObj (object): format appropriate object to be serialized
numParts (int): divide the data into numParts segments
format (str, optional): one of ['json' or 'pickle']. Defaults to json
**kwargs: additional keyword arguments passed to worker methods -
Returns:
bool: True for success or False otherwise
"""
if fmt not in ["json", "pickle"]:
logger.error("Unsupported format for %s", fmt)
return False
pth, fn = os.path.split(filePath)
self.__fileU.mkdirForFile(pth)
bn, ext = os.path.splitext(fn)
ret = True
if isinstance(myObj, list):
for ii, subList in enumerate(self.__sliceInChunks(myObj,
numParts)):
fp = os.path.join(pth, bn + "_part_%d" % (ii + 1) + ext)
ok = self.serialize(fp, subList, fmt=fmt, **kwargs)
ret = ret and ok
elif isinstance(myObj, dict):
for ii, keyList in enumerate(
self.__sliceInChunks(list(myObj.keys()), numParts)):
fp = os.path.join(pth, bn + "_part_%d" % (ii + 1) + ext)
ok = self.serialize(fp,
OrderedDict([(k, myObj[k])
for k in keyList]),
fmt=fmt,
**kwargs)
ret = ret and ok
else:
logger.error("Unsupported data type for serialization in parts")
ret = False
#
return ret
def deserializeInParts(self, filePath, numParts, fmt="json", **kwargs):
"""Public method to deserialize objects in supported formats from multiple parts
Args:
filePath (str): local file path
numParts (int): reconstruct the data object from numParts segments
format (str, optional): one of ['json' or 'pickle']. Defaults to json
**kwargs: additional keyword arguments passed to worker methods -
Returns:
object: deserialized object data
"""
rObj = None
if fmt not in ["json", "pickle"]:
logger.error("Unsupported format for %s", fmt)
return rObj
#
pth, fn = os.path.split(filePath)
bn, ext = os.path.splitext(fn)
if not numParts:
fp = os.path.join(pth, bn + "_part_*" + ext)
numParts = len(glob.glob(fp))
#
for ii in range(numParts):
fp = os.path.join(pth, bn + "_part_%d" % (ii + 1) + ext)
tObj = self.deserialize(fp, fmt=fmt, **kwargs)
if isinstance(tObj, list):
if not rObj:
rObj = []
rObj.extend(tObj)
elif isinstance(tObj, dict):
if not rObj:
rObj = OrderedDict()
rObj.update(tObj)
else:
logger.error(
"Unsupported data type for deserialization in parts")
return rObj
def exists(self, filePath, mode=os.R_OK):
return self.__fileU.exists(filePath, mode=mode)
def mkdir(self, dirPath, mode=0o755):
return self.__fileU.mkdir(dirPath, mode=mode)
def remove(self, pth):
return self.__fileU.remove(pth)
def __deserializeFasta(self, filePath, **kwargs):
try:
commentStyle = kwargs.get("commentStyle", "uniprot")
fau = FastaUtil()
return fau.readFasta(filePath, commentStyle=commentStyle)
except Exception as e:
logger.error("Unable to deserialize %r %r ", filePath, str(e))
return {}
def __serializeFasta(self, filePath, myObj, **kwargs):
try:
maxLineLength = int(kwargs.get("maxLineLength", 70))
makeComment = kwargs.get("makeComment", False)
fau = FastaUtil()
ok = fau.writeFasta(filePath,
myObj,
maxLineLength=maxLineLength,
makeComment=makeComment)
return ok
except Exception as e:
logger.error("Unable to serialize FASTA file %r %r", filePath,
str(e))
return False
def __textDump(self, filePath, myObj, **kwargs):
try:
indent = kwargs.get("indent", 1)
width = kwargs.get("width", 120)
sOut = pprint.pformat(myObj, indent=indent, width=width)
with open(filePath, "w") as ofh:
ofh.write("\n%s\n" % sOut)
return True
except Exception as e:
logger.error("Unable to dump to %r %r", filePath, str(e))
return False
def __serializePickle(self, filePath, myObj, **kwargs):
try:
pickleProtocol = kwargs.get("pickleProtocol",
pickle.DEFAULT_PROTOCOL)
with open(filePath, "wb") as outfile:
pickle.dump(myObj, outfile, pickleProtocol)
return True
except Exception as e:
logger.error("Unable to serialize %r %r", filePath, str(e))
return False
def __deserializePickle(self, filePath, **kwargs):
myDefault = kwargs.get("default", {})
try:
if sys.version_info[0] > 2:
encoding = kwargs.get("encoding", "ASCII")
errors = kwargs.get("errors", "strict")
with open(filePath, "rb") as outfile:
return pickle.load(outfile,
encoding=encoding,
errors=errors)
else:
with open(filePath, "rb") as outfile:
return pickle.load(outfile)
except Exception as e:
logger.warning("Unable to deserialize %r %r", filePath, str(e))
return myDefault
def __serializeJson(self, filePath, myObj, **kwargs):
"""Internal method to serialize the input object as JSON. An encoding
helper class is included to handle selected python data types (e.g., datetime)
"""
indent = kwargs.get("indent", 0)
enforceAscii = kwargs.get("enforceAscii", True)
try:
if enforceAscii:
with open(filePath, "w") as outfile:
json.dump(myObj,
outfile,
indent=indent,
cls=JsonTypeEncoder,
ensure_ascii=enforceAscii)
else:
with io.open(filePath, "w", encoding="utf-8") as outfile:
json.dump(myObj,
outfile,
indent=indent,
cls=JsonTypeEncoder,
ensure_ascii=enforceAscii)
return True
except Exception as e:
logger.error("Unable to serialize %r %r", filePath, str(e))
return False
def __deserializeJson(self, filePath, **kwargs):
myDefault = kwargs.get("default", {})
encoding = kwargs.get("encoding", "utf-8-sig")
encodingErrors = kwargs.get("encodingErrors", "ignore")
try:
if filePath[-3:] == ".gz":
if sys.version_info[0] > 2:
with gzip.open(filePath,
"rt",
encoding=encoding,
errors=encodingErrors) as inpFile:
return json.load(inpFile,
object_pairs_hook=OrderedDict)
else:
# Py2 situation non-ascii encodings is problematic
# with gzip.open(filePath, "rb") as csvFile:
# oL = self.__csvReader(csvFile, rowFormat, delimiter)
tPath = self.__fileU.uncompress(filePath, outputDir=None)
with io.open(tPath,
newline="",
encoding=encoding,
errors="ignore") as inpFile:
return json.load(inpFile,
object_pairs_hook=OrderedDict)
else:
with open(filePath, "r") as inpFile:
return json.load(inpFile, object_pairs_hook=OrderedDict)
except Exception as e:
logger.warning("Unable to deserialize %r %r", filePath, str(e))
return myDefault
def __hasMinSize(self, pth, minSize):
try:
return os.path.getsize(pth) >= minSize
except Exception:
return False
def __deserializeMmCif(self, locator, **kwargs):
""" """
try:
containerList = []
workPath = kwargs.get("workPath", None)
enforceAscii = kwargs.get("enforceAscii", True)
raiseExceptions = kwargs.get("raiseExceptions", True)
useCharRefs = kwargs.get("useCharRefs", True)
minSize = kwargs.get("minSize", 5)
#
if self.__fileU.isLocal(locator):
if minSize >= 0 and not self.__hasMinSize(locator, minSize):
logger.warning("Minimum file size not satisfied for: %r",
locator)
myIo = IoAdapter(raiseExceptions=raiseExceptions,
useCharRefs=useCharRefs)
containerList = myIo.readFile(
locator, enforceAscii=enforceAscii,
outDirPath=workPath) # type: ignore
else:
# myIo = IoAdapterPy(raiseExceptions=raiseExceptions, useCharRefs=useCharRefs)
# containerList = myIo.readFile(locator, enforceAscii=enforceAscii, outDirPath=workPath)
containerList = self.__deserializeMmCifRemote(
locator, useCharRefs, enforceAscii, workPath)
except Exception as e:
logger.error("Failing for %s with %s", locator, str(e))
return containerList
@retry((requests.exceptions.RequestException),
maxAttempts=3,
delaySeconds=1,
multiplier=2,
defaultValue=[],
logger=logger)
def __deserializeMmCifRemote(self, locator, useCharRefs, enforceAscii,
workPath):
containerList = []
try:
myIo = IoAdapterPy(raiseExceptions=True, useCharRefs=useCharRefs)
containerList = myIo.readFile(locator,
enforceAscii=enforceAscii,
outDirPath=workPath)
except Exception as e:
raise e
return containerList
def __serializeMmCif(self, filePath, containerList, **kwargs):
""" """
try:
ret = False
workPath = kwargs.get("workPath", None)
enforceAscii = kwargs.get("enforceAscii", True)
raiseExceptions = kwargs.get("raiseExceptions", True)
useCharRefs = kwargs.get("useCharRefs", True)
#
myIo = IoAdapter(raiseExceptions=raiseExceptions,
useCharRefs=useCharRefs)
if filePath.endswith(".gz") and workPath:
rfn = "".join(
random.choice(string.ascii_uppercase + string.digits)
for _ in range(10))
tPath = os.path.join(workPath, rfn)
ret = myIo.writeFile(tPath,
containerList=containerList,
enforceAscii=enforceAscii)
ret = self.__fileU.compress(tPath,
filePath,
compressType="gzip")
else:
ret = myIo.writeFile(filePath,
containerList=containerList,
enforceAscii=enforceAscii)
except Exception as e:
logger.error("Failing for %s with %s", filePath, str(e))
return ret
def __deserializeMmCifDict(self, filePath, **kwargs):
""" """
try:
containerList = []
workPath = kwargs.get("workPath", None)
enforceAscii = kwargs.get("enforceAscii", True)
raiseExceptions = kwargs.get("raiseExceptions", True)
useCharRefs = kwargs.get("useCharRefs", True)
#
myIo = IoAdapterPy(raiseExceptions=raiseExceptions,
useCharRefs=useCharRefs)
containerList = myIo.readFile(filePath,
enforceAscii=enforceAscii,
outDirPath=workPath)
except Exception as e:
logger.error("Failing for %s with %s", filePath, str(e))
return containerList
def __serializeMmCifDict(self, filePath, containerList, **kwargs):
""" """
try:
ret = False
# workPath = kwargs.get('workPath', None)
enforceAscii = kwargs.get("enforceAscii", True)
raiseExceptions = kwargs.get("raiseExceptions", True)
useCharRefs = kwargs.get("useCharRefs", True)
#
myIo = IoAdapterPy(raiseExceptions=raiseExceptions,
useCharRefs=useCharRefs)
ret = myIo.writeFile(filePath,
containerList=containerList,
enforceAscii=enforceAscii)
except Exception as e:
logger.error("Failing for %s with %s", filePath, str(e))
return ret
def __serializeList(self, filePath, aList, enforceAscii=True, **kwargs):
""" """
try:
_ = kwargs
if enforceAscii:
encoding = "ascii"
else:
encoding = "utf-8"
#
if sys.version_info[0] > 2:
with open(filePath, "w") as ofh:
if enforceAscii:
for st in aList:
ofh.write("%s\n" % st.encode(
"ascii", "xmlcharrefreplace").decode("ascii"))
else:
for st in aList:
ofh.write("%s\n" % st)
else:
if enforceAscii:
with io.open(filePath, "w", encoding=encoding) as ofh:
for st in aList:
ofh.write("%s\n" % st.encode(
"ascii", "xmlcharrefreplace").decode("ascii"))
else:
with open(filePath, "wb") as ofh:
for st in aList:
ofh.write("%s\n" % st)
return True
except Exception as e:
logger.error("Unable to serialize %r %r", filePath, str(e))
return False
def __processList(self, ifh, enforceAscii=True, **kwargs):
uncomment = kwargs.get("uncomment", True)
aList = []
for line in ifh:
if enforceAscii:
pth = line[:-1].encode("ascii",
"xmlcharrefreplace").decode("ascii")
else:
pth = line[:-1]
if not pth or (uncomment and pth.startswith("#")):
continue
aList.append(pth)
return aList
def __deserializeList(self,
filePath,
enforceAscii=True,
encodingErrors="ignore",
**kwargs):
aList = []
_ = kwargs
try:
if filePath[-3:] == ".gz":
if sys.version_info[0] > 2:
with gzip.open(filePath,
"rt",
encoding="utf-8-sig",
errors=encodingErrors) as ifh:
aList = self.__processList(ifh,
enforceAscii=enforceAscii,
**kwargs)
else:
tPath = self.__fileU.uncompress(filePath, outputDir=None)
# for py2 this commented code is problematic for non-ascii data
# with gzip.open(filePath, "rb") as ifh:
# aList = self.__processList(ifh, enforceAscii=enforceAscii)
with io.open(tPath, encoding="utf-8-sig",
errors="ignore") as ifh:
aList = self.__processList(ifh,
enforceAscii=enforceAscii)
else:
with io.open(filePath, encoding="utf-8-sig",
errors="ignore") as ifh:
aList = self.__processList(ifh,
enforceAscii=enforceAscii,
**kwargs)
except Exception as e:
logger.error("Unable to deserialize %r %s", filePath, str(e))
#
logger.debug("Reading list length %d", len(aList))
return aList
def __csvReader(self, csvFile, rowFormat, delimiter, uncomment=True):
oL = []
maxInt = sys.maxsize
csv.field_size_limit(maxInt)
if rowFormat == "dict":
if uncomment:
reader = csv.DictReader(uncommentFilter(csvFile),
delimiter=delimiter)
else:
reader = csv.DictReader(csvFile, delimiter=delimiter)
for rowD in reader:
oL.append(rowD)
elif rowFormat == "list":
if uncomment:
reader = csv.reader(uncommentFilter(csvFile),
delimiter=delimiter)
else:
reader = csv.reader(csvFile, delimiter=delimiter)
for rowL in reader:
oL.append(rowL)
return oL
def deserializeCsvIter(self,
filePath,
delimiter=",",
rowFormat="dict",
encodingErrors="ignore",
uncomment=True,
**kwargs):
"""Return an iterator to input CSV format file.
Args:
filePath (str): input file path
delimiter (str, optional): CSV delimiter. Defaults to ",".
rowFormat (str, optional): format for each process row (list or dict). Defaults to "dict".
encodingErrors (str, optional): treatment of encoding errors. Defaults to "ignore".
uncomment (bool, optional): flag to ignore leading comments. Defaults to True.
Returns:
(iterator): iterator for rowwise access to processed CSV data
"""
encoding = kwargs.get("encoding", "utf-8-sig")
maxInt = sys.maxsize
csv.field_size_limit(maxInt)
try:
if filePath[-3:] == ".gz":
with gzip.open(filePath,
"rt",
encoding=encoding,
errors=encodingErrors) as csvFile:
startIt = itertools.dropwhile(
lambda x: x.startswith("#"),
csvFile) if uncomment else csvFile
if rowFormat == "dict":
reader = csv.DictReader(startIt, delimiter=delimiter)
elif rowFormat == "list":
reader = csv.reader(startIt, delimiter=delimiter)
for row in reader:
yield row
else:
with io.open(filePath,
newline="",
encoding=encoding,
errors="ignore") as csvFile:
startIt = itertools.dropwhile(
lambda x: x.startswith("#"),
csvFile) if uncomment else csvFile
if rowFormat == "dict":
reader = csv.DictReader(startIt, delimiter=delimiter)
elif rowFormat == "list":
reader = csv.reader(startIt, delimiter=delimiter)
for row in reader:
# if uncomment and row.startswith("#"):
# continue
yield row
except Exception as e:
logger.error("Unable to deserialize %r %s", filePath, str(e))
def __deserializeCsv(self,
filePath,
delimiter=",",
rowFormat="dict",
encodingErrors="ignore",
uncomment=True,
**kwargs):
oL = []
encoding = kwargs.get("encoding", "utf-8-sig")
try:
if filePath[-3:] == ".gz":
if sys.version_info[0] > 2:
with gzip.open(filePath,
"rt",
encoding=encoding,
errors=encodingErrors) as csvFile:
oL = self.__csvReader(csvFile,
rowFormat,
delimiter,
uncomment=uncomment)
else:
# Py2 situation non-ascii encodings is problematic
# with gzip.open(filePath, "rb") as csvFile:
# oL = self.__csvReader(csvFile, rowFormat, delimiter)
tPath = self.__fileU.uncompress(filePath, outputDir=None)
with io.open(tPath,
newline="",
encoding=encoding,
errors="ignore") as csvFile:
oL = self.__csvReader(csvFile,
rowFormat,
delimiter,
uncomment=uncomment)
else:
with io.open(filePath,
newline="",
encoding=encoding,
errors="ignore") as csvFile:
oL = self.__csvReader(csvFile,
rowFormat,
delimiter,
uncomment=uncomment)
return oL
except Exception as e:
logger.error("Unable to deserialize %r %s", filePath, str(e))
#
logger.debug("Reading list length %d", len(oL))
return oL
def __serializeCsv(self, filePath, rowDictList, fieldNames=None, **kwargs):
""" """
_ = kwargs
try:
wD = {}
ret = False
fNames = fieldNames if fieldNames else list(rowDictList[0].keys())
# with io.open(filePath, 'w', newline='') as csvFile:
with open(filePath, "w") as csvFile:
writer = csv.DictWriter(csvFile, fieldnames=fNames)
writer.writeheader()
for ii, rowDict in enumerate(rowDictList):
try:
wD = {k: v for k, v in rowDict.items() if k in fNames}
writer.writerow(wD)
except Exception as e:
logger.error(
"Skipping bad CSV record %d wD %r rowDict %r with %s",
ii + 1, wD, rowDict, str(e))
continue
ret = True
except Exception as e:
logger.error("Failing for %s : %r with %s", filePath, wD, str(e))
return ret
def __csvEncoder(self,
csvData,
encoding="utf-8-sig",
encodingErrors="ignore"):
"""Handle encoding issues for gzipped data in Py2. (beware of the BOM chars)
Args:
csvData (text lines): uncompressed data from gzip open
encoding (str, optional): character encoding. Defaults to "utf-8-sig".
encodingErrors (str, optional): error treatment. Defaults to "ignore".
"""
for line in csvData:
yield line.decode("utf-8-sig", errors=encodingErrors).encode(
encoding, errors=encodingErrors)
def __deserializeXmlPrev(self, filePath, **kwargs):
"""Read the input XML file path and return an ElementTree data object instance.
Args:
filePath (sting): input XML file path
Returns:
object: instance of an ElementTree tree object
"""
_ = kwargs
tree = None
try:
logger.debug("Parsing XML path %s", filePath)
if filePath[-3:] == ".gz":
with gzip.open(filePath, mode="rb") as ifh:
tV = time.time()
tree = ET.parse(ifh)
else:
with open(filePath, mode="rb") as ifh:
tV = time.time()
tree = ET.parse(ifh)
logger.debug("Parsed %s in %.2f seconds", filePath,
time.time() - tV)
except Exception as e:
logger.error("Unable to deserialize %r %s", filePath, str(e))
#
return tree
def __testGzip(self, filePath):
ok = True
with gzip.open(filePath, "r") as fh:
try:
fh.read(1)
except gzip.BadGzipFile:
ok = False
except Exception:
ok = False
logger.debug("Gzip file check %r", ok)
return ok
def __deserializeXml(self, filePath, **kwargs):
"""Read the input XML file path and return an ElementTree data object instance.
Args:
filePath (sting): input XML file path
Returns:
object: instance of an ElementTree tree object
"""
_ = kwargs
tree = None
encoding = kwargs.get("encoding", "utf-8-sig")
encodingErrors = kwargs.get("encodingErrors", "ignore")
#
try:
logger.debug("Parsing XML path %s", filePath)
if filePath[-3:] == ".gz" and self.__testGzip(filePath):
if sys.version_info[0] > 2:
with gzip.open(filePath,
"rt",
encoding=encoding,
errors=encodingErrors) as ifh:
tV = time.time()
tree = ET.parse(ifh)
else:
tPath = self.__fileU.uncompress(filePath, outputDir=None)
with io.open(tPath,
encoding=encoding,
errors=encodingErrors) as ifh:
tV = time.time()
tree = ET.parse(ifh)
else:
with io.open(filePath,
encoding=encoding,
errors=encodingErrors) as ifh:
tV = time.time()
tree = ET.parse(ifh)
logger.debug("Parsed %s in %.2f seconds", filePath,
time.time() - tV)
except Exception as e:
logger.error("Unable to deserialize %r %s", filePath, str(e))
#
return tree
def __reload(self,
dirPath,
useCache=False,
imgtDumpUrl=None,
testList=None,
maxCount=None):
imgtD = {}
startTime = time.time()
fU = FileUtil()
fU.mkdir(dirPath)
#
imgtDataPath = os.path.join(self.__dirPath, "imgt-data.json")
#
logger.info("useCache %r imgtFeaturePath %r", useCache, imgtDataPath)
if useCache and self.__mU.exists(imgtDataPath):
imgtD = self.__mU.doImport(imgtDataPath, fmt="json")
self.__version = imgtD["version"]
else:
imgtDumpUrl = imgtDumpUrl if imgtDumpUrl else "http://www.imgt.org/download/3Dstructure-DB/IMGT3DFlatFiles.tgz"
imgtReadmeUrl = "http://www.imgt.org/download/3Dstructure-DB/RELEASE"
imgtDumpFileName = fU.getFileName(imgtDumpUrl)
imgtDumpPath = os.path.join(dirPath, imgtDumpFileName)
imgtReleasePath = os.path.join(dirPath, "IMGT-release.txt")
_, fn = os.path.split(imgtDumpUrl)
imgtFlatFilePath = os.path.join(self.__dirPath, fn[:-4])
#
logger.info("Fetching url %s path %s", imgtDumpUrl, imgtDumpPath)
ok1 = fU.get(imgtDumpUrl, imgtDumpPath)
ok2 = fU.get(imgtReadmeUrl, imgtReleasePath)
fU.unbundleTarfile(imgtDumpPath, dirPath=dirPath)
logger.info("Completed fetch (%r) at %s (%.4f seconds)", ok1
and ok2,
time.strftime("%Y %m %d %H:%M:%S", time.localtime()),
time.time() - startTime)
# ---
readmeLines = self.__mU.doImport(imgtReleasePath, fmt="list")
self.__version = readmeLines[0].strip() if readmeLines else None
logger.info("IMGT version %r", self.__version)
# ---
chainD, rawD = self.__imgtFlatFileProcessor(imgtFlatFilePath,
maxCount=maxCount,
testList=testList)
# ---
tS = datetime.datetime.now().isoformat()
# vS = datetime.datetime.now().strftime("%Y-%m-%d")
if testList:
imgtD = {
"version": self.__version,
"date": tS,
"chains": chainD,
"raw": rawD
}
else:
imgtD = {
"version": self.__version,
"date": tS,
"chains": chainD
}
ok = self.__mU.doExport(imgtDataPath, imgtD, fmt="json", indent=3)
logger.info("Completed flatfile prep (%r) at %s (%.4f seconds)",
ok, time.strftime("%Y %m %d %H:%M:%S",
time.localtime()),
time.time() - startTime)
return imgtD
