def collectionsOverview():
# get username from session
username = session['username']
# get objects
PIDs = jobs.getSelPIDs()
object_package = {}
# get collection objects
'''
This can be improved
'''
riquery = fedora_handle.risearch.get_subjects(predicate="info:fedora/fedora-system:def/relations-external#hasContentModel", object="info:fedora/CM:Collection")
collections = list(riquery)
# assemble sizes
object_package['coll_size_dict'] = {}
for collection in collections:
print "Working on",collection
results = solr_handle.search(**{ "q":"rels_isMemberOfCollection:"+collection.replace(":","\:"), "stats":"true", "stats.field":"obj_size_i", "rows":0 })
print results.stats
if results != None and results.total_results > 0 and results.stats['obj_size_i'] != None:
collection_obj_sum = results.stats['obj_size_i']['sum']
object_package['coll_size_dict'][collection] = (collection_obj_sum,utilities.sizeof_fmt(collection_obj_sum),results.total_results)
# print object_package['coll_size_dict']
object_package['coll_size_dict'] = json.dumps(object_package['coll_size_dict'])
return render_template("collectionsOverview.html", object_package=object_package)
def search(self):
solr_params = {
"q": self.q,
"rows": self.rows,
"start": self.start,
"sort": self.sort,
"fl": self.fl
}
return solr_handle.search(**solr_params)
def search(self):
solr_params = {
"q":self.q,
"rows":self.rows,
"start":self.start,
"sort":self.sort,
"fl":self.fl
}
return solr_handle.search(**solr_params)
def getLink(self, doc_id):
try:
self.id = doc_id
solr_params = {
"q": "id:{doc_id}".format(doc_id=doc_id),
"start": 0,
"rows": 1
}
search_results = solr_handle.search(**solr_params)
doc = search_results.documents[0]
self.doc = doc
self.date_parsed = dateutil.parser.parse(self.doc['last_modified'])
except:
return False
def getLink(self, doc_id):
try:
self.id = doc_id
solr_params = {
"q":"id:{doc_id}".format(doc_id=doc_id),
"start":0,
"rows":1
}
search_results = solr_handle.search(**solr_params)
doc = search_results.documents[0]
self.doc = doc
self.date_parsed = dateutil.parser.parse(self.doc['last_modified'])
except:
return False
def total_record_count(self):
# get total record count from Solr DB
return solr_handle.search(q="*:*", rows=0).total_results
def PIDSolr():
'''
Current Approach: If POST, send results as large array to template, save as JS variable
- works great so far at 800+ items, but what about 100,000+?
- documentation says ~ 50,000 is the limit
- will need to think of a server-side option
'''
# get username from session
username = session['username']
# get form
form = forms.solrSearch(request.form)
# collection selection
coll_query = {'q':"rels_hasContentModel:*Collection", 'fl':["id","dc_title"], 'rows':1000}
coll_results = solr_handle.search(**coll_query)
coll_docs = coll_results.documents
# check for title, give generic if not present
for each in coll_docs:
if 'dc_title' not in each:
each['dc_title'] = [ 'Unknown Collection Title' + each['id'].encode('ascii','ignore') ]
form.collection_object.choices = [(each['id'].encode('ascii','ignore'), each['dc_title'][0].encode('ascii','ignore')) for each in coll_docs]
form.collection_object.choices.insert(0,("","All Collections"))
# content model
cm_query = {'q':'*', 'facet' : 'true', 'facet.field' : 'rels_hasContentModel'}
cm_results = solr_handle.search(**cm_query)
form.content_model.choices = [(each, each.split(":")[-1]) for each in cm_results.facets['facet_fields']['rels_hasContentModel']]
form.content_model.choices.insert(0,("","All Content Types"))
# perform search
if request.method == 'POST':
# build base with native Solr queries
query = {'q':form.q.data, 'fq':[form.fq.data], 'fl':[form.fl.data], 'rows':100000}
# Fedora RELS-EXT
# collection selection
if form.collection_object.data:
print "Collection refinement:",form.collection_object.data
escaped_coll = form.collection_object.data.replace(":","\:")
query['fq'].append("rels_isMemberOfCollection:info\:fedora/"+escaped_coll)
# content model / type selection
if form.content_model.data:
print "Content Model refinement:",form.content_model.data
escaped_cm = form.content_model.data.replace(":","\:")
query['fq'].append("rels_hasContentModel:"+escaped_cm)
# issue query
print query
stime = time.time()
q_results = solr_handle.search(**query)
etime = time.time()
ttime = (etime - stime) * 1000
print "Solr Query took:",ttime,"ms"
output_dict = {}
data = []
stime = time.time()
for each in q_results.documents:
try:
PID = each['id'].encode('ascii','ignore')
dc_title = each['dc_title'][0].encode('ascii','ignore')
data.append([PID,dc_title])
except:
print "Could not render:",each['id'] #unicdoe solr id
etime = time.time()
ttime = (etime - stime) * 1000
print "Solr Munging for DataTables took::",ttime,"ms"
output_dict['data'] = data
json_output = json.dumps(data)
return render_template("PIDSolr.html",username=username, form=form, q_results=q_results, json_output=json_output, coll_docs=coll_docs,APP_HOST=localConfig.APP_HOST)
# pass the current PIDs to page as list
return render_template("PIDSolr.html",username=username, form=form, coll_docs=coll_docs,APP_HOST=localConfig.APP_HOST)
def objPreview(PIDnum):
object_package = {}
# GET CURRENT OBJECTS
PIDlet = jobs.genPIDlet(int(PIDnum))
if PIDlet == False:
return utilities.applicationError("PIDnum is out of range or invalid. Object-at-a-Glance is displeased.")
PIDlet['pURL'] = "/objPreview/"+str(int(PIDnum)-1)
PIDlet['nURL'] = "/objPreview/"+str(int(PIDnum)+1)
# WSUDOR handle
obj_handle = WSUDOR_ContentTypes.WSUDOR_Object(PIDlet['cPID'])
# General Metadata
solr_params = {'q':utilities.escapeSolrArg(PIDlet['cPID']), 'rows':1}
solr_results = solr_handle.search(**solr_params)
solr_package = solr_results.documents[0]
object_package['solr_package'] = solr_package
# COMPONENTS
object_package['components_package'] = []
riquery = fedora_handle.risearch.spo_search(subject=None, predicate="info:fedora/fedora-system:def/relations-external#isMemberOf", object="info:fedora/"+PIDlet['cPID'])
for s,p,o in riquery:
object_package['components_package'].append(s.encode('utf-8'))
if len(object_package['components_package']) == 0:
object_package.pop('components_package')
# RDF RELATIONSHIPS
riquery = fedora_handle.risearch.spo_search(subject="info:fedora/"+PIDlet['cPID'], predicate=None, object=None)
# parse
riquery_filtered = []
for s,p,o in riquery:
riquery_filtered.append((p,o))
riquery_filtered.sort()
object_package['rdf_package'] = riquery_filtered
# DATASTREAMS
ds_list = obj_handle.ohandle.ds_list
object_package['datastream_package'] = ds_list
# Object size of datastreams
size_dict = obj_handle.objSizeDict
object_package['size_dict'] = size_dict
object_package['size_dict_json'] = json.dumps(size_dict)
# OAI
OAI_dict = {}
#identifer
try:
riquery = fedora_handle.risearch.spo_search(subject="info:fedora/"+PIDlet['cPID'], predicate="http://www.openarchives.org/OAI/2.0/itemID", object=None)
OAI_ID = riquery.objects().next().encode('utf-8')
OAI_dict['ID'] = OAI_ID
except:
print "No OAI Identifier found."
# sets
OAI_dict['sets'] = []
try:
riquery = fedora_handle.risearch.spo_search(subject="info:fedora/"+PIDlet['cPID'], predicate="http://digital.library.wayne.edu/fedora/objects/wayne:WSUDOR-Fedora-Relations/datastreams/RELATIONS/content/isMemberOfOAISet", object=None)
for each in riquery.objects():
OAI_dict['sets'].append(each)
except:
print "No OAI sets found."
object_package['OAI_package'] = OAI_dict
print object_package['OAI_package']
# RENDER
return render_template("objPreview.html",PIDnum=(int(PIDnum)+1),PIDlet=PIDlet,object_package=object_package,localConfig=localConfig)
def PIDSolr():
'''
Current Approach: If POST, send results as large array to template, save as JS variable
- works great so far at 800+ items, but what about 100,000+?
- documentation says ~ 50,000 is the limit
- will need to think of a server-side option
'''
# get username from session
username = session['username']
# get form
form = forms.solrSearch(request.form)
# dynamically update fields
# collection selection
coll_query = {
'q': "rels_hasContentModel:*Collection",
'fl': ["id", "dc_title"],
'rows': 1000
}
coll_results = solr_handle.search(**coll_query)
coll_docs = coll_results.documents
form.collection_object.choices = [
(each['id'].encode('ascii', 'ignore'),
each['dc_title'][0].encode('ascii', 'ignore')) for each in coll_docs
]
form.collection_object.choices.insert(0, ("", "All Collections"))
# content model
cm_query = {
'q': '*',
'facet': 'true',
'facet.field': 'rels_hasContentModel'
}
cm_results = solr_handle.search(**cm_query)
form.content_model.choices = [
(each, each.split(":")[-1])
for each in cm_results.facets['facet_fields']['rels_hasContentModel']
]
form.content_model.choices.insert(0, ("", "All Content Types"))
# perform search
if request.method == 'POST':
# build base with native Solr queries
query = {
'q': form.q.data,
'fq': [form.fq.data],
'fl': [form.fl.data],
'rows': 100000
}
# Fedora RELS-EXT
# collection selection
if form.collection_object.data:
print "Collection refinement:", form.collection_object.data
escaped_coll = form.collection_object.data.replace(":", "\:")
query['fq'].append("rels_isMemberOfCollection:info\:fedora/" +
escaped_coll)
# content model / type selection
if form.content_model.data:
print "Content Model refinement:", form.content_model.data
escaped_cm = form.content_model.data.replace(":", "\:")
query['fq'].append("rels_hasContentModel:" + escaped_cm)
# issue query
print query
stime = time.time()
q_results = solr_handle.search(**query)
etime = time.time()
ttime = (etime - stime) * 1000
print "Solr Query took:", ttime, "ms"
output_dict = {}
data = []
stime = time.time()
for each in q_results.documents:
try:
PID = each['id'].encode('ascii', 'ignore')
dc_title = each['dc_title'][0].encode('ascii', 'ignore')
data.append([PID, dc_title])
except:
print "Could not render:", each['id'] #unicdoe solr id
etime = time.time()
ttime = (etime - stime) * 1000
print "Solr Munging for DataTables took::", ttime, "ms"
output_dict['data'] = data
json_output = json.dumps(data)
return render_template("PIDSolr.html",
username=username,
form=form,
q_results=q_results,
json_output=json_output,
coll_docs=coll_docs)
# pass the current PIDs to page as list
return render_template("PIDSolr.html",
username=username,
form=form,
coll_docs=coll_docs)
def total_record_count(self): # get total record count from Solr DB return solr_handle.search(q="*:*",rows=0).total_results
