def spending_by_transaction_count(request_data):
keyword = request_data["filters"]["keywords"]
response = {}
for category in INDEX_ALIASES_TO_AWARD_TYPES.keys():
total = get_total_results(keyword, category)
if total is not None:
if category == "directpayments":
category = "direct_payments"
response[category] = total
else:
return total
return response
def spending_by_transaction_count(request_data):
keyword = request_data["filters"]["keywords"]
response = {}
results = get_total_results(keyword)
for category in INDEX_ALIASES_TO_AWARD_TYPES.keys():
if results is not None:
if category == "directpayments":
response["direct_payments"] = results[category]["doc_count"]
else:
response[category] = results[category]["doc_count"]
else:
return results
return response
def get_download_ids(keyword, field, size=10000):
"""
returns a generator that
yields list of transaction ids in chunksize SIZE
Note: this only works for fields in ES of integer type.
"""
n_iter = DOWNLOAD_QUERY_SIZE // size
results = get_total_results(keyword)
if results is None:
logger.error(
"Error retrieving total results. Max number of attempts reached")
return
total = sum(results[category]["doc_count"]
for category in INDEX_ALIASES_TO_AWARD_TYPES.keys())
required_iter = (total // size) + 1
n_iter = min(max(1, required_iter), n_iter)
for i in range(n_iter):
filter_query = QueryWithFilters.generate_transactions_elasticsearch_query(
{"keyword_search": [es_minimal_sanitize(keyword)]})
search = TransactionSearch().filter(filter_query)
group_by_agg_key_values = {
"field": field,
"include": {
"partition": i,
"num_partitions": n_iter
},
"size": size,
"shard_size": size,
}
aggs = A("terms", **group_by_agg_key_values)
search.aggs.bucket("results", aggs)
response = search.handle_execute()
if response is None:
raise Exception("Breaking generator, unable to reach cluster")
results = []
for result in response["aggregations"]["results"]["buckets"]:
results.append(result["key"])
yield results
