def get_items(args):
""" Get all the news items for the date and filters provided
For performance reasons, returns an iterator that yields an array of CHUNK_SIZE
So that aggregations can be queried while the next iteration is retrieved
"""
if not args.get('section'):
abort(400, gettext('Must provide a section for this report'))
source = {
'query':
items_query(True),
'size':
CHUNK_SIZE,
'from':
0,
'sort': [{
'versioncreated': 'asc'
}],
'_source': [
'headline', 'place', 'subject', 'service', 'versioncreated',
'anpa_take_key'
]
}
must_terms = []
if args.get('genre'):
must_terms.append(
{'terms': {
'genre.code': [genre for genre in args['genre']]
}})
args['date_to'] = args['date_from']
date_range = get_date_filters(args)
if date_range.get('gt') or date_range.get('lt'):
must_terms.append({'range': {'versioncreated': date_range}})
if len(must_terms) > 0:
source['query']['bool']['must'].append(must_terms)
# Apply the section filters
section = args['section']
get_resource_service('section_filters').apply_section_filter(
source['query'], section)
while True:
results = get_resource_service(
'{}_search'.format(section)).search(source)
items = list(results)
if not len(items):
break
source['from'] += CHUNK_SIZE
yield items
def get_genres():
"""Get the list of genres from the news items"""
query = items_query(True)
must_terms = []
source = {}
if date_range.get('gt') or date_range.get('lt'):
must_terms.append({'range': {'versioncreated': date_range}})
if len(must_terms) > 0:
query['bool']['must'].append(must_terms)
source.update({
'query': query,
'size': 0,
'aggs': {
'genres': {
'terms': {
'field': 'genre.code',
'size': 0
}
}
}
})
# Apply the section filters
section = args['section']
get_resource_service('section_filters').apply_section_filter(
source['query'], section)
results = get_resource_service(
'{}_search'.format(section)).search(source)
buckets = ((results.hits.get('aggregations') or {}).get('genres')
or {}).get('buckets') or []
return [genre['key'] for genre in buckets]