def get(self, remotePath, localPath):
try:
fileU = FileUtil()
fileU.mkdirForFile(localPath)
self.__sftpClient.get(remotePath, localPath)
return True
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
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 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 __rebuildCache(self,
targetUrl,
mapNameL,
outDirPath,
rawDirPath,
fmt="pickle",
useCache=True):
"""Fetch the UniProt selected id mapping resource file and extract
UniProt Acc to 'mapIndex' mapping. Serialize the mapping as required.
Args:
targetUrl (str): source URL of the remote index file
mapNameL (list): list of key mapping names to extract from the index
outDirPath (str): directory path for raw and processed mapping files
fmt (str, optional): output format (pickle|json) . Defaults to "pickle".
useCache (bool, optional): use cached files. Defaults to True.
Returns:
dict: od[uniprotId] = mapped value
idmapping_selected.tab
1. UniProtKB-AC
2. UniProtKB-ID
3. GeneID (EntrezGene)
4. RefSeq
5. GI
6. PDB
7. GO
8. UniRef100
9. UniRef90
10. UniRef50
11. UniParc
12. PIR
13. NCBI-taxon
14. MIM
15. UniGene
16. PubMed
17. EMBL
18. EMBL-CDS
19. Ensembl
20. Ensembl_TRS
21. Ensembl_PRO
22. Additional PubMed
"""
startTime = time.time()
nL = mapNameL
oD = {}
try:
fileU = FileUtil()
fExt = "pic" if fmt == "pickle" else "json"
fExt = "tdd" if fmt == "tdd" else fExt
fN, _ = os.path.splitext(fileU.getFileName(targetUrl))
mapFileName = fN + "-map." + fExt
idMapPath = os.path.join(outDirPath, mapFileName)
mU = MarshalUtil()
if useCache and mU.exists(idMapPath):
logger.info("Reading cached serialized file %r", idMapPath)
if fmt in ["pickle", "json"]:
tD = mU.doImport(idMapPath, fmt=fmt)
nL = list(set(tD["idNameList"]))
oD = tD["uniprotMapD"]
logger.info("keys %r", list(oD.keys())[:10])
logger.info("nL %r", nL)
ok = True
elif fmt == "tdd":
ioU = IoUtil()
it = ioU.deserializeCsvIter(idMapPath,
delimiter="\t",
rowFormat="list",
encodingErrors="ignore")
tL = next(it, [])
nL = tL[1:]
if len(nL) == 1:
for row in it:
oD[row[0]] = row[1]
else:
for row in it:
oD[row[0]] = row[1:]
ok = True
else:
idPath = os.path.join(rawDirPath, fileU.getFileName(targetUrl))
if not fileU.exists(idPath):
logger.info(
"Fetching selected UniProt idmapping data from %r in %r",
targetUrl, outDirPath)
ok = fileU.get(targetUrl, idPath)
if not ok:
logger.error("Failed to downlowd %r", targetUrl)
return oD
else:
logger.info("Using cached mapping file %r", idPath)
# ---
ioU = IoUtil()
if fmt in ["pickle", "json"]:
if len(mapNameL) == 1:
for row in ioU.deserializeCsvIter(
idPath,
delimiter="\t",
rowFormat="list",
encodingErrors="ignore"):
oD[row[0]] = str(
row[self.__mapRecordD[mapNameL[0]] - 1])
else:
for row in ioU.deserializeCsvIter(
idPath,
delimiter="\t",
rowFormat="list",
encodingErrors="ignore"):
for mapName in mapNameL:
oD.setdefault(row[0], []).append(
str(row[self.__mapRecordD[mapName] - 1]))
logger.info("Writing serialized mapping file %r",
idMapPath)
ok = mU.doExport(idMapPath, {
"idNameList": mapNameL,
"uniprotMapD": oD
},
fmt=fmt)
elif fmt == "tdd":
#
logger.info("Writing serialized mapping file %r",
idMapPath)
fU = FileUtil()
fU.mkdirForFile(idMapPath)
colNameL = []
colNameL.append("UniProtId")
colNameL.extend(mapNameL)
with open(idMapPath, "w", encoding="utf-8") as ofh:
ofh.write("%s\n" % "\t".join(colNameL))
if len(mapNameL) == 1:
idx = self.__mapRecordD[mapNameL[0]] - 1
for row in ioU.deserializeCsvIter(
idPath,
delimiter="\t",
rowFormat="list",
encodingErrors="ignore"):
ofh.write("%s\t%s\n" % (row[0], row[idx]))
else:
idxL = [0]
idxL.extend([
self.__mapRecordD[mapName] - 1
for mapName in mapNameL
])
for row in ioU.deserializeCsvIter(
idPath,
delimiter="\t",
rowFormat="list",
encodingErrors="ignore"):
ofh.write(
"%s\n" %
"\t".join([str(row[idx]) for idx in idxL]))
#
nL, oD = self.__rebuildCache(targetUrl,
mapNameL,
outDirPath,
rawDirPath,
fmt=fmt,
useCache=True)
ok = True if nL and oD else False
logger.info("Completed reload (%r) at %s (%.4f seconds)", ok,
time.strftime("%Y %m %d %H:%M:%S", time.localtime()),
time.time() - startTime)
except Exception as e:
logger.exception("Failing with %s", str(e))
#
return nL, oD
