def transient(self, data):
data['$uri'] = url_for('documents_api.view',
document_id=data.get('id'))
data['$ui'] = document_url(data.get('id'))
collection_id = data.get('collection_id')
data['$writeable'] = request.authz.can_write(collection_id)
return data
def document_links(self, data, pk, schemata):
links = {
'self': url_for('documents_api.view', document_id=pk),
'tags': url_for('entities_api.tags', id=pk),
'ui': document_url(pk)
}
if data.get('content_hash'):
links['file'] = url_for('documents_api.file', document_id=pk)
if schemata.intersection([Document.SCHEMA_PDF]):
links['pdf'] = url_for('documents_api.pdf', document_id=pk)
if schemata.intersection([Document.SCHEMA_PDF, Document.SCHEMA_TABLE]):
links['records'] = url_for('documents_api.records', document_id=pk)
if schemata.intersection([Document.SCHEMA_FOLDER]):
query = (('filter:parent.id', pk), )
links['children'] = url_for('documents_api.index', _query=query)
return links
def format_results(alert, results):
output = []
for result in results.get('hits', []):
document = unpack_result(result)
# generate document URL:
document['url'] = document_url(document['id'], dq=alert.query_text)
collection = Collection.by_id(document.pop('collection_id'))
if not collection:
continue
document['collection'] = collection
# preview snippets:
result['snippets'] = []
for field, snippets in result.get('highlight', {}).items():
result['snippets'].extend(snippets)
output.append(document)
return output
def get_results(query, limit):
collections = {}
for i, row in enumerate(query.scan()):
if i >= limit:
return
source = row.get('_source')
collection_id = source.pop('collection_id')
if collection_id not in collections:
obj = Collection.by_id(collection_id)
if obj is None:
collections[collection_id] = obj.label
data = {
'collection': collections.get(collection_id, '[Missing]'),
'url': document_url(row.get('_id'))
}
for name, value in source.items():
if name not in FIELDS:
continue
if isinstance(value, (list, tuple, set)):
value = ', '.join(value)
data[name] = value
yield data
