def get_n_latest_records(n_latest, field="last_updated", index=None):
""" Gets latest N records from the index """
query = {"size": n_latest, "query": QueryBuilder.generate_query_string(), "sort": [{field: {"order": "desc"}}]}
query_result = es.search(index=index, doc_type=CFG_PUB_TYPE, body=query)
return query_result["hits"]["hits"]
def get_n_latest_records(n_latest, field="last_updated", index=None):
""" Gets latest N records from the index """
query = {
"size": n_latest,
"query": QueryBuilder.generate_query_string(),
"sort": [{
field: {
"order": "desc"
}
}]
}
query_result = es.search(index=index, doc_type=CFG_PUB_TYPE, body=query)
return query_result['hits']['hits']
def search(query,
index=None,
filters=list(),
size=10,
include="*",
exclude="authors",
offset=0,
sort_field=None,
sort_order='',
post_filter=None):
""" Perform a search query.
:param query: [string] query string e.g. 'higgs boson'
:param index: [string] name of the index. If None a default is used
:param filters: [list of tuples] list of filters for the query.
Currently supported: ('author', author_fullname),
('collaboration', collaboration_name), ('date', date)
:param size: [int] max number of hits that should be returned
:param offset: [int] offset for the results (used for pagination)
:param sort_by: [string] sorting field. Currently supported fields:
"title", "collaboration", "date", "relevance"
:param sort_order: [string] order of the sorting either original
(for a particular field) or reversed. Supported:
'' or 'rev'
:return: [dict] dictionary with processed results and facets
"""
# If empty query then sort by date
if query == '' and not sort_field:
sort_field = 'date'
query = HEPDataQueryParser.parse_query(query)
# Build core query
data_query = QueryBuilder.generate_query_string(query)
pub_query = QueryBuilder.generate_query_string(query)
authors_query = QueryBuilder.generate_nested_query('authors', query)
query_builder = QueryBuilder()
query_builder.add_child_parent_relation(
CFG_DATA_TYPE,
relation="child",
related_query=data_query,
other_queries=[pub_query, authors_query])
# Add additional options
query_builder.add_pagination(size=size, offset=offset)
query_builder.add_sorting(sort_field=sort_field, sort_order=sort_order)
query_builder.add_filters(filters)
query_builder.add_post_filter(post_filter)
query_builder.add_aggregations()
query_builder.add_source_filter(include, exclude)
if query:
# Randomize search among the available shard copies.
pub_result = es.search(index=index,
body=query_builder.query,
doc_type=CFG_PUB_TYPE)
else:
# Execute search only on the primary shards (to ensure no missing or duplicate results).
pub_result = es.search(index=index,
body=query_builder.query,
doc_type=CFG_PUB_TYPE,
preference="_primary")
parent_filter = {
"filtered": {
"filter": {
"terms": {
"_id": [hit["_id"] for hit in pub_result['hits']['hits']]
}
}
}
}
query_builder = QueryBuilder()
query_builder.add_child_parent_relation(CFG_PUB_TYPE,
relation="parent",
related_query=parent_filter,
must=True,
other_queries=[data_query])
query_builder.add_pagination(size=size * 50)
data_result = es.search(index=index,
body=query_builder.query,
doc_type=CFG_DATA_TYPE)
merged_results = merge_results(pub_result, data_result)
return map_result(merged_results)