def build_elasticsearch_result(self, info_buckets: List[dict]) -> List[dict]:
results = []
for bucket in info_buckets:
info = json.loads(bucket.get("key"))
# Build a list of hash IDs to handle multiple levels
recipient_hash = info.get("hash")
recipient_levels = sorted(info.get("levels") or [])
if recipient_hash and recipient_levels:
recipient_hash_list = [f"{recipient_hash}-{level}" for level in recipient_levels]
else:
recipient_hash_list = None
results.append(
{
"id": recipient_hash_list,
"code": info["unique_id"] or "DUNS Number not provided",
"description": info["name"] or None,
"award_count": int(bucket.get("doc_count", 0)),
**{
column: get_summed_value_as_float(bucket, self.sum_column_mapping[column])
for column in self.sum_column_mapping
},
}
)
return results
def build_elasticsearch_result(self, info_buckets: List[dict]) -> List[dict]:
results = []
cfda_prefetch_pks = [bucket.get("key") for bucket in info_buckets]
prefetched_cfdas = {
cfda["program_number"]: cfda for cfda in Cfda.objects.filter(program_number__in=cfda_prefetch_pks).values()
}
for bucket in info_buckets:
cfda_number = bucket.get("key")
cfda = prefetched_cfdas.get(cfda_number, {})
results.append(
{
"id": cfda.get("id"),
"code": cfda.get("program_number"),
"description": cfda.get("program_title"),
"award_count": int(bucket.get("doc_count", 0)),
"resource_link": cfda.get("url") if cfda.get("url") != "None;" else None,
"cfda_federal_agency": cfda.get("federal_agency"),
"cfda_objectives": cfda.get("objectives"),
"cfda_website": cfda.get("website_address"),
"applicant_eligibility": cfda.get("applicant_eligibility"),
"beneficiary_eligibility": cfda.get("beneficiary_eligibility"),
**{
column: get_summed_value_as_float(bucket, self.sum_column_mapping[column])
for column in self.sum_column_mapping
},
}
)
return results
def build_elasticsearch_result(self,
info_buckets: List[dict]) -> List[dict]:
results = []
cfda_prefetch_pks = [
json_str_to_dict(bucket.get("key")).get("code")
for bucket in info_buckets
]
prefetched_cfdas = Cfda.objects.filter(
program_number__in=cfda_prefetch_pks)
for bucket in info_buckets:
info = json_str_to_dict(bucket.get("key"))
toAdd = {
"id": int(info.get("id")) if info.get("id") else None,
"code": info.get("code") or None,
"description": info.get("description") or None,
"award_count": int(bucket.get("doc_count", 0)),
"resource_link": info.get("url") or None,
**{
column: get_summed_value_as_float(
bucket, self.sum_column_mapping[column])
for column in self.sum_column_mapping
},
}
cfda = [
cfda for cfda in prefetched_cfdas
if cfda.program_number == info.get("code")
]
if cfda:
toAdd.update({
"cfda_federal_agency":
cfda[0].federal_agency,
"cfda_objectives":
cfda[0].objectives,
"cfda_website":
cfda[0].website_address,
"applicant_eligibility":
cfda[0].applicant_eligibility,
"beneficiary_eligibility":
cfda[0].beneficiary_eligibility,
})
else:
toAdd.update({
"cfda_federal_agency": None,
"cfda_objectives": None,
"cfda_website": None,
"applicant_eligibility": None,
"beneficiary_eligibility": None,
})
results.append(toAdd)
return results
def _build_json_result(self, bucket: dict):
info = json_str_to_dict(bucket.get("key"))
return {
"id": int(info["id"]),
"code": info["code"],
"description": info["name"],
# the count of distinct awards contributing to the totals
"award_count": int(bucket.get("doc_count", 0)),
**{
column: get_summed_value_as_float(bucket, self.sum_column_mapping[column])
for column in self.sum_column_mapping
},
}
def build_totals(self, response: List[dict]) -> dict:
# Need to use a Postgres in this case since we only look at the first 10k results for Elasticsearch.
# Since the endpoint is performing aggregations on the entire matview with no grouping or joins
# the query takes minimal time to complete.
if self.agg_key == settings.ES_ROUTING_FIELD:
annotations = {
"cast_def_codes": Cast("def_codes", ArrayField(TextField()))
}
filters = [
Q(cast_def_codes__overlap=self.def_codes),
self.has_award_of_provided_type(should_join_awards=False),
]
aggregations = {
"face_value_of_loan": Sum("total_loan_value"),
"obligation": Sum("obligation"),
"outlay": Sum("outlay"),
}
aggregations = {
col: aggregations[col]
for col in self.sum_column_mapping.keys()
}
aggregations["award_count"] = Count("award_id")
if self.filters.get("query"):
filters.append(
Q(recipient_name__icontains=self.filters["query"]["text"]))
totals = (CovidFinancialAccountMatview.objects.annotate(
**annotations).filter(*filters).values().aggregate(
**aggregations))
return totals
totals = {key: 0 for key in self.sum_column_mapping.keys()}
award_count = 0
for bucket in response:
for key in totals.keys():
totals[key] += get_summed_value_as_float(
bucket, self.sum_column_mapping[key])
award_count += int(bucket.get("doc_count", 0))
totals["award_count"] = award_count
return totals
