def get_next_pid(self, namespace=None, count=None):
"""
Request next available pid or pids from Fedora, optionally in a specified
namespace. Calls :meth:`ApiFacade.getNextPID`.
.. deprecated :: 0.14
Mint pids for new objects with
:func:`eulfedora.models.DigitalObject.get_default_pid`
instead, or call :meth:`ApiFacade.getNextPID` directly.
:param namespace: (optional) get the next pid in the specified pid namespace;
otherwise, Fedora will return the next pid in the configured default namespace.
:param count: (optional) get the specified number of pids; by default, returns 1 pid
:rtype: string or list of strings
"""
# this method should no longer be needed - default pid logic moved to DigitalObject
warnings.warn(
"""get_next_pid() method is deprecated; you should mint new pids via DigitalObject or ApiFacade.getNextPID() instead.""",
DeprecationWarning)
kwargs = {}
if namespace:
kwargs['namespace'] = namespace
elif self.default_pidspace:
kwargs['namespace'] = self.default_pidspace
if count:
kwargs['numPIDs'] = count
r = self.api.getNextPID(**kwargs)
nextpids = parse_xml_object(NewPids, r.content, r.url)
if count is None:
return nextpids.pids[0]
else:
return nextpids.pids
def get_next_pid(self, namespace=None, count=None):
"""
Request next available pid or pids from Fedora, optionally in a specified
namespace. Calls :meth:`ApiFacade.getNextPID`.
.. deprecated :: 0.14
Mint pids for new objects with
:func:`eulfedora.models.DigitalObject.get_default_pid`
instead, or call :meth:`ApiFacade.getNextPID` directly.
:param namespace: (optional) get the next pid in the specified pid namespace;
otherwise, Fedora will return the next pid in the configured default namespace.
:param count: (optional) get the specified number of pids; by default, returns 1 pid
:rtype: string or list of strings
"""
# this method should no longer be needed - default pid logic moved to DigitalObject
warnings.warn("""get_next_pid() method is deprecated; you should mint new pids via DigitalObject or ApiFacade.getNextPID() instead.""",
DeprecationWarning)
kwargs = {}
if namespace:
kwargs['namespace'] = namespace
elif self.default_pidspace:
kwargs['namespace'] = self.default_pidspace
if count:
kwargs['numPIDs'] = count
r = self.api.getNextPID(**kwargs)
nextpids = parse_xml_object(NewPids, r.content, r.url)
if count is None:
return nextpids.pids[0]
else:
return nextpids.pids
def datastream_lastmodified(request, pid, dsid, repo=None,
as_of_date=None, *args, **kwargs):
'''Method suitable for use as a a last-modified function with
:class:`django.views.decorators.http.condition`. Takes basically
the same arguments as :meth:`~eulfedora.views.raw_datastream`.
'''
try:
if repo is None:
repo = Repository()
resp = repo.api.getDatastream(pid, dsid, asOfDateTime=as_of_date)
dsprofile = parse_xml_object(DatastreamProfile, resp.content, resp.url)
if dsprofile:
return dsprofile.created
except RequestFailed:
pass
def datastream_lastmodified(request,
pid,
dsid,
repo=None,
as_of_date=None,
*args,
**kwargs):
'''Method suitable for use as a a last-modified function with
:class:`django.views.decorators.http.condition`. Takes basically
the same arguments as :meth:`~eulfedora.views.raw_datastream`.
'''
try:
if repo is None:
repo = Repository()
resp = repo.api.getDatastream(pid, dsid, asOfDateTime=as_of_date)
dsprofile = parse_xml_object(DatastreamProfile, resp.content, resp.url)
if dsprofile:
return dsprofile.created
except RequestFailed:
pass
def datastream_etag(request, pid, dsid, repo=None, as_of_date=None, **kwargs):
'''Method suitable for use as an etag function with
:class:`django.views.decorators.http.condition`. Takes the same
arguments as :meth:`~eulfedora.views.raw_datastream`.
'''
# if a range is requested and it is not for the entire file,
# do *NOT* return an etag
# NOTE: using api directly here instead of object/ds models
# to avoid making unneeded api calls
try:
if repo is None:
repo = Repository()
resp = repo.api.getDatastream(pid, dsid, asOfDateTime=as_of_date)
dsprofile = parse_xml_object(DatastreamProfile, resp.content, resp.url)
if dsprofile and dsprofile.checksum_type != 'DISABLED':
return dsprofile.checksum
except RequestFailed:
pass
return None
def datastream_etag(request, pid, dsid, repo=None,
as_of_date=None, **kwargs):
'''Method suitable for use as an etag function with
:class:`django.views.decorators.http.condition`. Takes the same
arguments as :meth:`~eulfedora.views.raw_datastream`.
'''
# if a range is requested and it is not for the entire file,
# do *NOT* return an etag
# NOTE: using api directly here instead of object/ds models
# to avoid making unneeded api calls
try:
if repo is None:
repo = Repository()
resp = repo.api.getDatastream(pid, dsid, asOfDateTime=as_of_date)
dsprofile = parse_xml_object(DatastreamProfile, resp.content, resp.url)
if dsprofile and dsprofile.checksum_type != 'DISABLED':
return dsprofile.checksum
except RequestFailed:
pass
return None
def find_objects(self, terms=None, type=None, chunksize=None, **kwargs):
"""
Find objects in Fedora. Find query should be generated via keyword
args, based on the fields in Fedora documentation. By default, the
query uses a contains (~) search for all search terms. Calls
:meth:`ApiFacade.findObjects`. Results seem to return consistently
in ascending PID order.
Example usage - search for all objects where the owner contains 'jdoe'::
repository.find_objects(ownerId='jdoe')
Supports all search operators provided by Fedora findObjects query (exact,
gt, gte, lt, lte, and contains). To specify the type of query for
a particular search term, call find_objects like this::
repository.find_objects(ownerId__exact='lskywalker')
repository.find_objects(date__gt='20010302')
:param type: type of objects to return; defaults to :class:`DigitalObject`
:param chunksize: number of objects to return at a time
:rtype: generator for list of objects
"""
type = type or self.default_object_type
find_opts = {'chunksize': chunksize}
search_operators = {
'exact': '=',
'gt': '>',
'gte': '>=',
'lt': '<',
'lte': '<=',
'contains': '~'
}
if terms is not None:
find_opts['terms'] = terms
else:
conditions = []
for field, value in six.iteritems(kwargs):
if '__' in field:
field, filtr = field.split('__')
if filtr not in search_operators:
raise Exception("Unsupported search filter '%s'" %
filtr)
op = search_operators[filtr]
else:
op = search_operators['contains'] # default search mode
if field in self.search_fields_aliases:
field = self.search_fields_aliases[field]
if field not in self.search_fields:
raise Exception("Error generating Fedora findObjects query: unknown search field '%s'" \
% field)
if ' ' in value:
# if value contains whitespace, it must be delimited with single quotes
value = "'%s'" % value
conditions.append("%s%s%s" % (field, op, value))
query = ' '.join(conditions)
find_opts['query'] = query
r = self.api.findObjects(**find_opts)
chunk = parse_xml_object(SearchResults, r.content, r.url)
while True:
for result in chunk.results:
yield type(self.api, result.pid)
if chunk.session_token:
r = self.api.findObjects(session_token=chunk.session_token,
**find_opts)
chunk = parse_xml_object(SearchResults, r.content, r.url)
else:
break
def find_objects(self, terms=None, type=None, chunksize=None, **kwargs):
"""
Find objects in Fedora. Find query should be generated via keyword
args, based on the fields in Fedora documentation. By default, the
query uses a contains (~) search for all search terms. Calls
:meth:`ApiFacade.findObjects`. Results seem to return consistently
in ascending PID order.
Example usage - search for all objects where the owner contains 'jdoe'::
repository.find_objects(ownerId='jdoe')
Supports all search operators provided by Fedora findObjects query (exact,
gt, gte, lt, lte, and contains). To specify the type of query for
a particular search term, call find_objects like this::
repository.find_objects(ownerId__exact='lskywalker')
repository.find_objects(date__gt='20010302')
:param type: type of objects to return; defaults to :class:`DigitalObject`
:param chunksize: number of objects to return at a time
:rtype: generator for list of objects
"""
type = type or self.default_object_type
find_opts = {'chunksize' : chunksize}
search_operators = {
'exact': '=',
'gt': '>',
'gte': '>=',
'lt': '<',
'lte': '<=',
'contains': '~'
}
if terms is not None:
find_opts['terms'] = terms
else:
conditions = []
for field, value in six.iteritems(kwargs):
if '__' in field:
field, filter = field.split('__')
if filter not in search_operators:
raise Exception("Unsupported search filter '%s'" % filter)
op = search_operators[filter]
else:
op = search_operators['contains'] # default search mode
if field in self.search_fields_aliases:
field = self.search_fields_aliases[field]
if field not in self.search_fields:
raise Exception("Error generating Fedora findObjects query: unknown search field '%s'" \
% field)
if ' ' in value:
# if value contains whitespace, it must be delimited with single quotes
value = "'%s'" % value
conditions.append("%s%s%s" % (field, op, value))
query = ' '.join(conditions)
find_opts['query'] = query
r = self.api.findObjects(**find_opts)
chunk = parse_xml_object(SearchResults, r.content, r.url)
while True:
for result in chunk.results:
yield type(self.api, result.pid)
if chunk.session_token:
r = self.api.findObjects(session_token=chunk.session_token, **find_opts)
chunk = parse_xml_object(SearchResults, r.content, r.url)
else:
break
def fs_members(self, *els, **kwargs):
"""Get all the members of a "container" element (i.e. an object treated
as a directory by fuse).
:param els: list of path elements
:params writable_only: if True, only return members that are capable of
being written
:rtype: list
"""
if "writable_only" in kwargs:
writable_only = kwargs["writable_only"]
else:
writable_only = False
members = []
if len(els) == 0: # top level
# no elements - top-level view of the object itself
# find all members of this object that should be listed in the directory
# - should be strings, not unicode
if not writable_only:
members.append(".info")
if len(self.history):
members.append(".versions")
# add all datastreams - using datastream id for directory listing
members.extend([str(dsid) for dsid in self.ds_list.keys()])
# relations to other objects - use short-hand rel names for directory names
members.extend(self.rel_shortnames)
if not writable_only:
# add methods - using method name for directory listing
# NOTE: it's possible to have duplicate method names defined
# on different service objects... current implementation does
# *not* take this into account
self.method_names = {}
for sdef, method_list in self.methods.iteritems():
# FIXME: any way to exclude methods that require parameters?
for m in method_list:
self.method_names[m] = sdef
members.extend(self.method_names)
elif els[0] == ".versions":
# list revisions of this object
# (not completely implemented...)
if len(els) == 1:
members = [str(dt) for dt in self.history]
else:
# TODO: get a *versioned* list of datastreams
# (don't list ones that didn't exist at the specified time!)
data, url = self.api.listDatastreams(self.pid)
dsobj = parse_xml_object(ObjectDatastreams, data, url)
return [str(ds.dsid) for ds in dsobj.datastreams]
# related objects
elif els[0] in self.rel_shortnames:
# top level: directory with objects related by the specified short-name relation
if len(els) == 1:
rel_uri = self.rel_shortnames[els[0]]
members = self.related_objects[rel_uri]
return members
