def _cache_parsed_xml(self):
""" Caches every Journal and Publisher found in the xml """
if not self.parsed:
return
outcome = self.xml['header']['outcome'] is 'notFound'
if outcome is 'notFound' or outcome is 'failed':
return
if self.xml['header']['outcome'] == 'singleJournal' \
or self.xml['header']['outcome'] == 'uniqueZetoc':
journal = self.xml['journals']['journal']
cache.set("journal:" + journal['jtitle'].lower(), journal)
if self.xml['header']['outcome'] != 'uniqueZetoc':
# if the publisher has been indexed by RoMEO
publisher = self.xml['publishers']['publisher']
# Associate a Journal with a Publisher key in cache
cache.set("journal-publisher:" + journal['jtitle'].lower(),
"publisher:" + publisher['name'].lower())
elif self.xml['journals'] is not None:
for journal in self.xml['journals']['journal']:
cache.set("journal:" + journal['jtitle'].lower(), journal)
if self.xml['header']['numhits'] == '1' \
and self.xml['header']['outcome'] != 'uniqueZetoc':
publisher = self.xml['publishers']['publisher']
cache.set("publisher:" + publisher['name'].lower(), publisher)
elif self.xml['publishers'] is not None:
for publisher in self.xml['publishers']['publisher']:
cache.set("publisher:" + publisher['name'].lower(), publisher)
def set_single_item(self, journal=None, publisher=None):
"""
Used to initialize the parser with items retrieved from cache
Note: if both a journal and a publisher are defined
the publisher is associated with the journal
"""
self.xml = dict()
self.xml['header'] = dict()
self.xml['header']['outcome'] = ''
if journal is not None:
self.xml['journals'] = dict()
self.xml['journals']['journal'] = journal
self.xml['header']['numhits'] = '1'
self.parsed = True
self.single_item = True
if publisher is not None:
# Associate a Journal with a Publisher key in cache
self.xml['header']['outcome'] = 'singleJournal'
cache.set("journal-publisher:" + journal['jtitle'].lower(),
"publisher:" + publisher['name'].lower())
elif publisher is not None:
self.xml['header']['outcome'] = 'publisherFound'
self.xml['header']['numhits'] = '1'
self.xml['publishers'] = dict()
self.xml['publishers']['publisher'] = publisher
self.single_item = True
self.parsed = True
def curate():
"""
Index page with uploader and list of existing depositions
"""
action = request.values.get('action')
usercollection_id = request.values.get('collection')
recid = request.values.get('recid', 0, type=int)
# Allowed actions
if action not in ['accept', 'reject', 'remove']:
abort(400)
# Check recid
if not recid:
abort(400)
recid = int(recid)
# Does collection exists
u = UserCollection.query.filter_by(id=usercollection_id).first()
if not u:
abort(400)
# Check permission to perform action on this record
# - Accept and reject is done by community owner
# - Remove is done by record owner
if action in ['accept', 'reject', ]:
if u.id_user != current_user.get_id():
abort(403)
elif action == 'remove':
try:
email = get_fieldvalues(recid, '8560_f')[0]
if email != current_user['email']:
abort(403)
# User not allowed to remove from the zenodo user collection
if u.id == 'zenodo':
abort(403)
except (IndexError, KeyError):
abort(403)
# Prevent double requests (i.e. give bibupload a chance to make the change)
key = "usercoll_curate:%s_%s" % (usercollection_id, recid)
cache_action = cache.get(key)
if cache_action == action or cache_action in ['reject', 'remove']:
return jsonify({'status': 'success', 'cache': 1})
elif cache_action:
# Operation under way, but the same action
return jsonify({'status': 'failure', 'cache': 1})
if action == "accept":
res = u.accept_record(recid)
elif action == "reject" or action == "remove":
res = u.reject_record(recid)
if res:
# Set 5 min cache to allow bibupload/webcoll to finish
cache.set(key, action, timeout=5*60)
return jsonify({'status': 'success', 'cache': 0})
else:
return jsonify({'status': 'failure', 'cache': 0})
def save(self):
"""
Saves modified data pernamently for logged users.
"""
if not self.is_guest and self.modified:
cache.set(self.get_key(),
dict(self.info),
timeout=CFG_WEBSESSION_EXPIRY_LIMIT_DEFAULT * 3600)
def preingest_form_data(user_id,
form_data,
uuid=None,
append=False,
cached_data=False):
"""Used to insert form data to the workflow before running it
Creates an identical json structure to the draft json.
If cached_data is enabled, the data will be used by the next workflow
initiated by the user, so the uuid can be ommited in this case.
@param user_id: the user id
@param uuid: the id of the workflow
@param form_data: a json with field_name -> value structure
@param append: set to True if you want to append the values to the existing
ones
@param cached_data: set to True if you want to cache the data.
"""
def preingest_data(form_data, append):
def preingest(json):
if 'pop_obj' not in json:
json['pop_obj'] = {}
for field, value in form_data.items():
if append:
try:
if isinstance(json['pop_obj'][field], list):
json['pop_obj'][field].append(value)
else:
new_values_list = [json['pop_obj'][field]]
new_values_list.append(value)
json['pop_obj'][field] = new_values_list
except KeyError:
json['pop_obj'][field] = [value]
else:
json['pop_obj'][field] = value
json['pop_obj']['timestamp'] = str(datetime.now())
return preingest
if cached_data:
cache.set(str(user_id) + ':cached_form_data', form_data)
else:
Workflow.set_extra_data(user_id=user_id,
uuid=uuid,
setter=preingest_data(form_data, append))
# Ingest the data in the forms, in case there are any
if append:
for field_name, value in form_data.items():
draft_field_list_add(user_id, uuid, field_name, value)
else:
for field_name, value in form_data.items():
draft_field_set(user_id, uuid, field_name, value)
def parse_url(self, url):
self.url = url
#example
#url = 'http://www.sherpa.ac.uk/romeo/api29.php?jtitle=Annals%20of%20Physics'
found_journal = url.find("jtitle=")
found_publisher = url.find("pub=")
if found_journal != -1:
self.search_type = "journal_search"
self.query = url[found_journal + 7:(len(url) - 15)]
elif found_publisher != -1:
self.search_type = "publisher_search"
found_publisher = url.find("pub=")
self.query = url[found_publisher + 4:len(url)]
else:
self.search_type = "issn_search"
found_publisher = url.find("issn=")
self.query = url[found_publisher + 4:len(url)]
cached_xml = cache.get(self.search_type + ":" + self.query.lower())
if cached_xml is None:
try:
self.data = urllib2.urlopen(url).read()
except urllib2.HTTPError:
self.error = True
return
try:
root = ElementTree.XML(self.data)
except ParseError:
self.error = True
return
self.xml = XmlDictConfig(root)
outcome = self.xml['header']['outcome']
if outcome != 'failed' and outcome != 'notFound':
cache.set(self.search_type + ":" + self.query.lower(),
self.xml, 999999999999)
else:
self.xml = cached_xml
#self.data = cached_xml
#root = ElementTree.XML(self.data)
#self.xml = XmlDictConfig(root)
if self.xml['header']['outcome'] == 'failed':
self.error = True
self.error_message = self.xml['header']['message']
self.parsed = True
self._cache_parsed_xml()
class ZenodoExtension(Extension):
"""
Temporary extension (let's see how long it will stay ;-).
This extension is made until a pull-request has been integrated in the
main Flask-Cache branch, so that generated cache keys are stable and
predictable instead of based on filename and line numbers.
"""
tags = set(['zenodocache'])
def parse(self, parser):
lineno = parser.stream.next().lineno
# Parse timeout
args = [parser.parse_expression()]
# Parse fragment name
parser.stream.skip_if('comma')
args.append(parser.parse_expression())
# Parse vary_on parameters
vary_on = []
while parser.stream.skip_if('comma'):
vary_on.append(parser.parse_expression())
if vary_on:
args.append(nodes.List(vary_on))
else:
args.append(nodes.Const([]))
body = parser.parse_statements(['name:endcache'], drop_needle=True)
return nodes.CallBlock(self.call_method('_cache', args),
[], [], body).set_lineno(lineno)
def _cache(self, timeout, fragment_name, vary_on, caller):
try:
cache = getattr(self.environment, JINJA_CACHE_ATTR_NAME)
except AttributeError, e:
raise e
key = make_template_fragment_key(fragment_name, vary_on=vary_on)
# Delete key if timeout is 'del'
if timeout == "del":
cache.delete(key)
return caller()
rv = cache.get(key)
if rv is None:
rv = caller()
cache.set(key, rv, timeout)
return rv
def _precache(self, info, force=False):
"""
Calculate prermitions for user actions.
FIXME: compatibility layer only !!!
"""
# get autorization key
acc_key = self.get_acc_key()
acc = cache.get(acc_key)
if not force and acc_key is not None and acc is not None:
return acc
#FIXME: acc_authorize_action should use flask request directly
user_info = info
user_info.update(self.req)
from invenio.webuser import isUserSubmitter, isUserReferee, \
isUserAdmin, isUserSuperAdmin
from invenio.access_control_engine import acc_authorize_action
from invenio.access_control_admin import acc_get_role_id, \
acc_is_user_in_role
from invenio.search_engine import get_permitted_restricted_collections
data = {}
data['precached_permitted_restricted_collections'] = \
get_permitted_restricted_collections(user_info)
data['precached_usebaskets'] = acc_authorize_action(
user_info, 'usebaskets')[0] == 0
data['precached_useloans'] = acc_authorize_action(
user_info, 'useloans')[0] == 0
data['precached_usegroups'] = acc_authorize_action(
user_info, 'usegroups')[0] == 0
data['precached_usealerts'] = acc_authorize_action(
user_info, 'usealerts')[0] == 0
data['precached_usemessages'] = acc_authorize_action(
user_info, 'usemessages')[0] == 0
data['precached_usestats'] = acc_authorize_action(
user_info, 'runwebstatadmin')[0] == 0
data['precached_viewsubmissions'] = isUserSubmitter(user_info)
data['precached_useapprove'] = isUserReferee(user_info)
data['precached_useadmin'] = isUserAdmin(user_info)
data['precached_usesuperadmin'] = isUserSuperAdmin(user_info)
data['precached_canseehiddenmarctags'] = acc_authorize_action(
user_info, 'runbibedit')[0] == 0
usepaperclaim = False
usepaperattribution = False
viewclaimlink = False
if (CFG_BIBAUTHORID_ENABLED and acc_is_user_in_role(
user_info, acc_get_role_id("paperclaimviewers"))):
usepaperclaim = True
if (CFG_BIBAUTHORID_ENABLED and acc_is_user_in_role(
user_info, acc_get_role_id("paperattributionviewers"))):
usepaperattribution = True
viewlink = False
try:
viewlink = session['personinfo']['claim_in_process']
except (KeyError, TypeError):
pass
if (CFG_BIBAUTHORID_ENABLED and usepaperattribution and viewlink):
viewclaimlink = True
# if (CFG_BIBAUTHORID_ENABLED
# and ((usepaperclaim or usepaperattribution)
# and acc_is_user_in_role(data, acc_get_role_id("paperattributionlinkviewers")))):
# viewclaimlink = True
data['precached_viewclaimlink'] = viewclaimlink
data['precached_usepaperclaim'] = usepaperclaim
data['precached_usepaperattribution'] = usepaperattribution
cache.set(acc_key,
data,
timeout=CFG_WEBSESSION_EXPIRY_LIMIT_DEFAULT * 3600)
return data
def curate():
"""
Index page with uploader and list of existing depositions
"""
action = request.values.get('action')
usercollection_id = request.values.get('collection')
recid = request.values.get('recid', 0, type=int)
# Allowed actions
if action not in ['accept', 'reject', 'remove']:
abort(400)
# Check recid
if not recid:
abort(400)
recid = int(recid)
# Does collection exists
u = UserCollection.query.filter_by(id=usercollection_id).first()
if not u:
abort(400)
# Check permission to perform action on this record
# - Accept and reject is done by community owner
# - Remove is done by record owner
if action in [
'accept',
'reject',
]:
if u.id_user != current_user.get_id():
abort(403)
elif action == 'remove':
try:
email = get_fieldvalues(recid, '8560_f')[0]
if email != current_user['email']:
abort(403)
# User not allowed to remove from the zenodo user collection
if u.id == 'zenodo':
abort(403)
except (IndexError, KeyError):
abort(403)
# Prevent double requests (i.e. give bibupload a chance to make the change)
key = "usercoll_curate:%s_%s" % (usercollection_id, recid)
cache_action = cache.get(key)
if cache_action == action or cache_action in ['reject', 'remove']:
return jsonify({'status': 'success', 'cache': 1})
elif cache_action:
# Operation under way, but the same action
return jsonify({'status': 'failure', 'cache': 1})
if action == "accept":
res = u.accept_record(recid)
elif action == "reject" or action == "remove":
res = u.reject_record(recid)
if res:
# Set 5 min cache to allow bibupload/webcoll to finish
cache.set(key, action, timeout=5 * 60)
return jsonify({'status': 'success', 'cache': 0})
else:
return jsonify({'status': 'failure', 'cache': 0})
def set(self, name, value, timeout=None):
cache.set(CFG_CACHE_KEY_PREFIX_SESSION + name, value, timeout=3600)