def fetch_metadata(self, session, id, mType):
""" Open irodsFile and get metadata from it. """
if (self.idNormalizer is not None):
id = self.idNormalizer.process_string(session, id)
elif type(id) == unicode:
id = id.encode('utf-8')
else:
id = str(id)
self._open(session)
upwards = 0
if id.find('/') > -1 and self.allowStoreSubDirs:
idp = id.split('/')
id = idp.pop()
while idp:
dn = idp.pop(0)
if not dn in self.coll.getSubCollections():
for x in range(upwards):
self.coll.upCollection()
raise ObjectDoesNotExistException(id)
self.coll.openCollection(dn)
upwards += 1
else:
id = id.replace('/', '--')
collPath = self.coll.getCollName()
# This is much more efficient than getting the file as it's simply
# interacting with iCAT
umd = irods.getFileUserMetadata(self.cxn,
'{0}/{1}'.format(collPath, id)
)
# if self.resource:
# f = self.coll.open(id, rescName=self.resource)
# else:
# f = self.coll.open(id)
#
# if not f:
# for x in range(upwards):
# self.coll.upCollection()
# return None
# umd = f.getUserMetadata()
# f.close()
val = None
for x in umd:
if x[0] == mType:
val = icatValToPy(x[1], x[2])
break
for x in range(upwards):
self.coll.upCollection()
return val
def find_documents(self, session, cache=0):
# Given a location in irods, go there and descend looking for files
c = self.coll
files = c.getObjects()
files.sort()
fList = [x[0] for x in files]
for i, f in enumerate(self._processFiles(session, fList)):
md = {}
irodsFilePath = '{0}/{1}'.format(c.getCollName(), f.filename)
for x in irods.getFileUserMetadata(self.cxn, irodsFilePath):
md[x[0]] = icatValToPy(x[1], x[2])
if len(md):
f.metadata['iRODS'] = md
# Delete the file on its resource
c.delete(files[i][0], files[i][1])
yield f
dirs = c.getSubCollections()
while dirs:
d = dirs.pop(0)
upColls = 0
for dx in d.split('/'):
c.openCollection(dx)
upColls += 1
files = c.getObjects()
files.sort()
fList = [x[0] for x in files]
for i, f in enumerate(self._processFiles(session, fList)):
md = {}
irodsFilePath = '{0}/{1}'.format(c.getCollName(), f.filename)
for x in irods.getFileUserMetadata(self.cxn, irodsFilePath):
md[x[0]] = icatValToPy(x[1], x[2])
if len(md):
f.metadata['iRODS'] = md
# Delete the file on its resource
c.delete(files[i][0], files[i][1])
yield f
ndirs = c.getSubCollections()
dirs.extend(["%s/%s" % (d, x) for x in ndirs])
for x in range(upColls):
c.upCollection()
def find_documents(self, session, cache=0):
# Given a location in irods, go there and descend looking for files
c = self.coll
files = c.getObjects()
files.sort()
fList = [x[0] for x in files]
for i, f in enumerate(self._processFiles(session, fList)):
md = {}
irodsFilePath = '{0}/{1}'.format(c.getCollName(), f.filename)
for x in irods.getFileUserMetadata(self.cxn, irodsFilePath):
md[x[0]] = icatValToPy(x[1], x[2])
if len(md):
f.metadata['iRODS'] = md
# Delete the file on its resource
c.delete(files[i][0], files[i][1])
yield f
dirs = c.getSubCollections()
while dirs:
d = dirs.pop(0)
upColls = 0
for dx in d.split('/'):
c.openCollection(dx)
upColls += 1
files = c.getObjects()
files.sort()
fList = [x[0] for x in files]
for i, f in enumerate(self._processFiles(session, fList)):
md = {}
irodsFilePath = '{0}/{1}'.format(c.getCollName(), f.filename)
for x in irods.getFileUserMetadata(self.cxn, irodsFilePath):
md[x[0]] = icatValToPy(x[1], x[2])
if len(md):
f.metadata['iRODS'] = md
# Delete the file on its resource
c.delete(files[i][0], files[i][1])
yield f
ndirs = c.getSubCollections()
dirs.extend(["%s/%s" % (d, x) for x in ndirs])
for x in range(upColls):
c.upCollection()
def find_documents(self, session, cache=0):
# given a location in irods, go there and descend looking for files
c = self.coll
files = c.getObjects()
files = [x[0] for x in files]
files.sort()
for f in self._processFiles(session, files):
md = {}
for x in irods.getFileUserMetadata(self.cxn, '{0}/{1}'.format(c.getCollName(), f.filename)):
md[x[0]] = icatValToPy(x[1], x[2])
if len(md):
f.metadata['iRODS'] = md
yield f
dirs = c.getSubCollections()
while dirs:
d = dirs.pop(0)
upColls = 0
for dx in d.split('/'):
c.openCollection(dx)
upColls += 1
files = c.getObjects()
files = [x[0] for x in files]
files.sort()
for f in self._processFiles(session, files):
md = {}
for x in irods.getFileUserMetadata(self.cxn, '{0}/{1}'.format(c.getCollName(), f.filename)):
md[x[0]] = icatValToPy(x[1], x[2])
if len(md):
f.metadata['iRODS'] = md
yield f
ndirs = c.getSubCollections()
dirs.extend(["%s/%s" % (d, x) for x in ndirs])
for x in range(upColls):
c.upCollection()
def find_documents(self, session, cache=0):
# read in single file
doc = StringDocument(self.stream.read(), filename=self.stream.getName())
# attach any iRODS metadata
umd = self.stream.getUserMetadata()
self.stream.close()
self.cxn.disconnect()
md = {}
for x in umd:
md[x[0]] = icatValToPy(x[1], x[2])
if len(md):
doc.metadata['iRODS'] = md
if cache == 0:
yield doc
elif cache == 2:
self.documents = [doc]
def find_documents(self, session, cache=0):
# Read in single file
doc = StringDocument(self.stream.read(),
filename=self.stream.getName()
)
# Attach any iRODS metadata
umd = self.stream.getUserMetadata()
self.stream.close()
self.cxn.disconnect()
md = {}
for x in umd:
md[x[0]] = icatValToPy(x[1], x[2])
if len(md):
doc.metadata['iRODS'] = md
if cache == 0:
yield doc
elif cache == 2:
self.documents = [doc]
def _open(self, session):
if self.cxn == None:
# connect to iRODS
myEnv, status = irods.getRodsEnv()
host = self.host if self.host else myEnv.getRodsHost()
port = self.port if self.port else myEnv.getRodsPort()
user = self.user if self.user else myEnv.getRodsUserName()
zone = self.zone if self.zone else myEnv.getRodsZone()
conn, errMsg = irods.rcConnect(host, port, user, zone)
if self.passwd:
status = irods.clientLoginWithPassword(conn, self.passwd)
else:
status = irods.clientLogin(conn)
if status:
raise ConfigFileException("Cannot connect to iRODS: (%s) %s" % (status, errMsg))
self.cxn = conn
self.env = myEnv
resources = irods.getResources(self.cxn)
self.resourceHash = {}
for r in resources:
self.resourceHash[r.getName()] = r
if self.coll != None:
# already open, just skip
return None
c = irods.irodsCollection(self.cxn, self.env.getRodsHome())
self.coll = c
# move into cheshire3 section
path = self.get_path(session, 'irodsCollection', 'cheshire3')
dirs = c.getSubCollections()
if not path in dirs:
c.createCollection(path)
c.openCollection(path)
if self.get_setting(session, 'createSubDir', 1):
# now look for object's storage area
# maybe move into database collection
if (isinstance(self.parent, Database)):
sc = self.parent.id
dirs = c.getSubCollections()
if not sc in dirs:
c.createCollection(sc)
c.openCollection(sc)
# move into store collection
dirs = c.getSubCollections()
if not self.id in dirs:
c.createCollection(self.id)
c.openCollection(self.id)
# Fetch user metadata
myMetadata = self.get_metadataTypes(session)
umd = c.getUserMetadata()
umdHash = {}
for u in umd:
umdHash[u[0]] = icatValToPy(*u[1:])
for md in myMetadata:
try:
setattr(self, md, umdHash[md])
except KeyError:
# hasn't been set yet
pass
if self.totalItems != 0:
self.meanWordCount = self.totalWordCount / self.totalItems
self.meanByteCount = self.totalByteCount / self.totalItems
else:
self.meanWordCount = 1
self.meanByteCount = 1
