def import_items(request):
raw = request.POST.getlist('item')
items = []
o={}
o['imported_docs'] = []
o['existing_docs'] = []
for item in raw:
items.append(json.loads(item,strict=False))
for doc_import in items:
if not ZoteroLink.objects.filter(zotero_url=doc_import['url']):
d = Document(creator=request.user, last_updater=request.user, description=doc_import['csl'])
d.save()
o['imported_docs'].append(d)
link = ZoteroLink(zotero_data=json.dumps(doc_import['json']), zotero_url=doc_import['url'], doc_id=d.id)
try:
doc_import['date']['year']
link.date_information = json.dumps(doc_import['date'])
except KeyError:
pass
link.save()
else:
existing_link = ZoteroLink.objects.filter(zotero_url=doc_import['url'])[0]
o['existing_docs'].append(existing_link.doc)
if doc_import['related_object'] == 'topic':
related_topic = Topic.objects.get(id=int(doc_import['related_id']))
if TopicAssignment.objects.filter(document=d, topic=related_topic):
pass
else:
new_assignment = TopicAssignment.objects.create(content_object=d, topic=related_topic, creator=request.user)
new_assignment.save()
return render_to_response(
'success.html', o, context_instance=RequestContext(request))
def import_items(request):
if request.session.get('import_complete', False):
return HttpResponse('Please only submit items once')
item_data = request.POST.getlist('data')
item_citations = request.POST.getlist('changed-citation')
user = request.user
o={}
o['created_items'] = []
item_counter = 0
for item_data_string, updated_citation in zip(item_data, item_citations):
item_counter += 1
action = request.POST.get('import-action-%s' % item_counter)
if action not in ['create', 'update']:
continue
item_data = json.loads(item_data_string, object_pairs_hook=OrderedDict)
if updated_citation:
citation = updated_citation
else:
citation = item_data['citation']
if action == "create":
d = Document(creator=user, last_updater=user, description=citation)
d.save()
elif action == "update":
update_id = request.POST.get('item-update-%s' % item_counter)
d = Document.objects.get(id=update_id)
d.last_updated = datetime.datetime.now()
d.last_updater = user
d.save()
if d.zotero_link():
d.zotero_link().delete()
link = ZoteroLink(zotero_data=item_data['json'], zotero_url=item_data['url'], doc_id=d.id)
try:
item_data['date']['year']
link.date_information = json.dumps(item_data['date'])
except KeyError:
pass
link.save()
reltopic = request.GET.get('reltopic', False)
if reltopic:
related_topic = Topic.objects.get(id=int(reltopic))
new_assignment = TopicAssignment.objects.create(
content_object=d,
topic=related_topic,
creator=user
)
new_assignment.save()
o['created_items'].append(d)
request.session['import_complete'] = True
redirect_url = request.GET.get('return_to', '/')
return HttpResponseRedirect(redirect_url)
def forwards(self, orm):
"Write your forwards methods here."
# Note: Remember to use orm['appname.ModelName'] rather than "from appname.models..."
digest_update_query = """
UPDATE main_document
SET description_digest = %s
WHERE id = %s
"""
for document in orm['main.Document'].objects.all():
db.execute(digest_update_query,
params=[Document.hash_description(document.description),
document.id])
duplicates_query = """
SELECT project_id, description_digest, COUNT(description_digest) as occurences
FROM main_document
GROUP BY project_id, description_digest
HAVING COUNT(description_digest) > 1;
"""
duplicates = db.execute(duplicates_query);
for project_id, description_digest, ct in duplicates:
qs = orm['main.Document'].objects.filter(project_id=project_id,
description_digest=description_digest)
for i, document in enumerate(qs[1:], 2):
description = document.description
last_el = (list(description.iterdescendants()) or [description])[-1]
appended_text = u' ({})'.format(i)
if last_el.tail:
last_el.tail += appended_text
else:
last_el.text += appended_text
update_query = """
UPDATE main_document
SET description = %s, description_digest = %s
WHERE id = %s
"""
db.execute(update_query, params=[etree.tostring(description),
Document.hash_description(description),
document.id])
remaining_duplicates = db.execute(duplicates_query)
assert not len(remaining_duplicates), "Still duplicates"
def __call__(self, attrs):
if self.instance is not None:
description = attrs.get('description', self.instance.description)
else:
description = attrs['description']
project = attrs['project']
qs = Document.objects.filter(
description_digest=Document.hash_description(description),
project=project)
if self.instance is not None:
qs = qs.exclude(id=self.instance.id)
if qs.exists():
raise serializers.ValidationError({'description': [self.message]})
def __call__(self, attrs):
if self.instance is not None:
description = attrs.get('description', self.instance.description)
else:
description = attrs['description']
project = attrs['project']
qs = Document.objects.filter(
description_digest=Document.hash_description(description),
project=project)
if self.instance is not None:
qs = qs.exclude(id=self.instance.id)
if qs.exists():
raise serializers.ValidationError({
'description': [self.message]
})
def validate_description(self, value):
description_stripped = Document.strip_description(value)
if not description_stripped:
raise serializers.ValidationError('Field required.')
remove_stray_brs(value)
return value
def import_from_jigsaw_root(root, user, project):
order = 1
for node in root:
id = node.findtext("docID")
text = node.findtext("docText")
#text.tag = 'div'
docs = list(Document.objects.filter(import_id=id))
if len(docs)==0:
d = Document(creator=user, last_updater=user, import_id=id, description=id, ordering=order, project=project)
d.save()
t = Transcript(creator=user, last_updater=user, document=d, content=text)
t.save()
else:
d = docs[0]
d.last_updater = user
d.description = id
d.order = order
d.language = gl.guessLanguageName(text)
order += 1
d.save()
date = node.find("docDate")
if date is not None and date.text:
try:
# pdb.set_trace()
dt = datetime.strptime(date.text, "%m/%d/%Y")
res = list(Topic.objects.filter(date=dt))
cnt = len(res)
if cnt==0:
normalized = "Date: %d/%d/%d" % (dt.year, dt.month, dt.day)
t=get_or_create_topic(user, normalized, 'EVT', project, dt)
t.date = dt
#t=Topic(creator=user, last_updater=user, preferred_name=normalized, date=dt, type='EVT',project=project)
t.save()
else:
t=res[0]
d.related_topics.create(creator=user, topic=t)
except ValueError as e:
pass
for p in node.findall("concept"):
name=p.text
t=get_or_create_topic(user, name, 'TAG',project)
if t:
d.related_topics.create(creator=user, topic=t)
else:
print "Cannot create topic(%s,type=%s)" % (t,'TAG')
for p in node.findall("person"):
name=p.text
t=get_or_create_topic(user, name, 'PER',project)
if t:
d.related_topics.create(creator=user, topic=t)
else:
print "Cannot create topic(%s,type=%s)" % (t,'PER')
for p in node.findall("location"):
name=p.text
t=get_or_create_topic(user, name, 'PLA',project)
if t:
d.related_topics.create(creator=user, topic=t)
else:
print "Cannot create topic(%s,type=%s)" % (t,'PLA')
def validate_description(self, value):
description_stripped = Document.strip_description(value)
if not description_stripped:
raise serializers.ValidationError('Field required.')
remove_stray_brs(value)
return value
