def location_mapper_vendor(row):
loc = {
"city_name": row.get("city", ""),
"congressional_code": row.get(
"vendor_cd", "").zfill(2), # Need to add leading zeroes here
"location_country_code":
row.get("vendorcountrycode",
""), # Never actually a country code, just the string name
"location_zip": row.get("zipcode", "").replace("-", ""),
"state_code": h.up2colon(row.get("vendor_state_code", "")),
"address_line1": row.get("streetaddress", ""),
"address_line2": row.get("streetaddress2", ""),
"address_line3": row.get("streetaddress3", "")
}
return loc
def location_mapper_vendor(row):
loc = {
"city_name": row.get("city", ""),
"congressional_code": row.get(
"vendor_cd", "").zfill(2), # Need to add leading zeroes here
"location_country_code":
row.get("vendorcountrycode",
""), # Never actually a country code, just the string name
"location_zip": row.get("zipcode", "").replace("-", ""),
"state_code": h.up2colon(row.get("vendor_state_code", "")),
"address_line1": row.get("streetaddress", ""),
"address_line2": row.get("streetaddress2", ""),
"address_line3": row.get("streetaddress3", "")
}
return loc
def location_mapper_place_of_performance(row):
loc = {
"city_name": row.get("placeofperformancecity", ""),
"congressional_code":
row.get("placeofperformancecongressionaldistrict",
"")[2:], # Need to strip the state off the front
"location_country_code": row.get("placeofperformancecountrycode", ""),
"location_zip": row.get("placeofperformancezipcode", "").replace(
"-", ""), # Either ZIP5, or ZIP5+4, sometimes with hypens
"state_code": h.up2colon(
row.get("pop_state_code", "")
) # Format is VA: VIRGINIA, so we need to grab the first bit
}
return loc
def location_mapper_place_of_performance(row):
loc = {
"city_name": row.get("placeofperformancecity", ""),
"congressional_code":
row.get("placeofperformancecongressionaldistrict",
"")[2:], # Need to strip the state off the front
"location_country_code": row.get("placeofperformancecountrycode", ""),
"location_zip": row.get("placeofperformancezipcode", "").replace(
"-", ""), # Either ZIP5, or ZIP5+4, sometimes with hypens
"state_code": h.up2colon(
row.get("pop_state_code", "")
) # Format is VA: VIRGINIA, so we need to grab the first bit
}
return loc
def evaluate_contract_award_type(row):
first_element = h.up2colon(row['contractactiontype'].split()[0])
if len(first_element) == 1:
return first_element
else:
cat = row['contractactiontype'].lower()
# Not using DAIMS enumeration . . .
if 'bpa' in cat:
return 'A'
elif 'purchase' in cat:
return 'B'
elif 'delivery' in cat:
return 'C'
elif 'definitive' in cat:
return 'D'
else:
return None
def evaluate_contract_award_type(row):
first_element = h.up2colon(row['contractactiontype'].split()[0])
if len(first_element) == 1:
return first_element
else:
cat = row['contractactiontype'].lower()
# Not using DAIMS enumeration . . .
if 'bpa' in cat:
return 'A'
elif 'purchase' in cat:
return 'B'
elif 'delivery' in cat:
return 'C'
elif 'definitive' in cat:
return 'D'
else:
return None
def handle(self, *args, **options):
h.clear_caches()
csv_file = options['file'][0]
self.logger.info("Starting load for file {}".format(csv_file))
# Create the csv reader
reader = CsvDataReader(csv_file)
# Create a new submission attributes object for this timestamp
subattr = SubmissionAttributes()
subattr.usaspending_update = datetime.now()
subattr.save()
# Create lists to hold model instances for bulk insert
txn_list = []
txn_assistance_list = []
for idx, row in enumerate(reader):
if len(reader) % 1000 == 0:
self.logger.info("Read row {}".format(len(reader)))
row = h.cleanse_values(row)
awarding_agency = self.get_awarding_agency(
row) # todo: use agency dict?
# Create the transaction object for this row
txn_dict = {
"submission":
subattr,
"action_date":
h.convert_date(row['obligation_action_date']),
"action_type":
h.up2colon(row['action_type']),
"award":
self.get_or_create_award(row, awarding_agency=awarding_agency),
"awarding_agency":
awarding_agency,
"description":
row["project_description"], # ?? account_title is anther contender?
"data_source":
"USA",
"federal_action_obligation":
row["fed_funding_amount"],
"last_modified_date":
h.convert_date(row['last_modified_date']),
"modification_number":
row["federal_award_mod"], # ??
"period_of_performance_start_date":
h.convert_date(row['starting_date']),
"period_of_performance_current_end_date":
h.convert_date(row['ending_date']),
"place_of_performance":
h.get_or_create_location(
row, location_mapper_fin_assistance_principal_place),
"recipient":
self.get_or_create_recipient(row),
"type":
h.up2colon(row['assistance_type']),
"usaspending_unique_transaction_id":
row["unique_transaction_id"],
# ??"funding_agency_id":
# ?? "certified date":
}
txn = Transaction(**txn_dict)
txn.fiscal_year = fy(txn.action_date)
txn_list.append(txn)
# Create the transaction contract object for this row
txn_assistance_dict = {
"submission":
subattr,
"fain":
row["federal_award_id"],
"uri":
row["uri"],
"cfda":
Cfda.objects.filter(
program_number=row["cfda_program_num"]).first(),
"correction_late_delete_indicator":
h.up2colon(row['correction_late_ind']),
"face_value_loan_guarantee":
row["face_loan_guran"],
"fiscal_year_and_quarter_correction":
row["fyq_correction"],
"non_federal_funding_amount":
row["non_fed_funding_amount"],
"original_loan_subsidy_cost":
row["orig_sub_guran"], # ??
"record_type":
int(h.up2colon(row['record_type'])),
"sai_number":
row["sai_number"],
"submitted_type":
"C", # ?? For CSV?
}
# ?? business_funds_indicator
# ?? reporting period start/end??
txn_assistance = TransactionAssistance(**txn_assistance_dict)
txn_assistance_list.append(txn_assistance)
# Bulk insert transaction rows
self.logger.info(
"Starting Transaction bulk insert ({} records)".format(
len(txn_list)))
Transaction.objects.bulk_create(txn_list)
self.logger.info("Completed Transaction bulk insert")
# Update txn assistance list with newly-inserted transactions
award_id_list = [] # we'll need this when updating the awards later on
for idx, t in enumerate(txn_assistance_list):
t.transaction = txn_list[idx]
award_id_list.append(txn_list[idx].award_id)
# Bulk insert transaction assistance rows
self.logger.info(
"Starting TransactionAssistance bulk insert ({} records)".format(
len(txn_assistance_list)))
TransactionAssistance.objects.bulk_create(txn_assistance_list)
self.logger.info("Completed TransactionAssistance bulk insert")
# Update awards to reflect latest transaction information
# (note that this can't be done via signals or a save()
# override in the model itself, because those aren't
# triggered by a bulk update
self.logger.info("Starting Awards update")
count = update_awards(tuple(award_id_list))
update_contract_awards(tuple(award_id_list))
update_model_description_fields()
self.logger.info("Completed Awards update ({} records)".format(count))
def get_awarding_agency(self, row):
toptier_code = h.up2colon(row['maj_agency_cat'])
subtier_code = h.up2colon(row['agency_code'])
return self.get_agency(toptier_code, subtier_code)
def handle(self, *args, **options):
csv_file = options['file'][0]
self.logger.info("Starting load for file {}".format(csv_file))
# Create the csv reader
reader = CsvDataReader(csv_file)
# Create a new submission attributes object for this timestamp
subattr = SubmissionAttributes()
subattr.usaspending_update = datetime.now()
subattr.save()
# Create lists to hold model instances for bulk insert
txn_list = []
txn_contract_list = []
subtier_agency_dict = h.get_subtier_agency_dict()
# Store some additional support data needed for the laod
award_type_dict = {a[0]: a[1] for a in AWARD_TYPES}
contract_pricing_dict = {c[0]: c[1] for c in CONTRACT_PRICING_TYPES}
for idx, row in enumerate(reader):
if len(reader) % 1000 == 0:
self.logger.info("Read row {}".format(len(reader)))
row = h.cleanse_values(row)
awarding_agency_id = self.get_agency_id(row["contractingofficeagencyid"], subtier_agency_dict)
# Create the transaction object for this row
txn_dict = {
"action_date": h.convert_date(row['signeddate']),
"award": self.get_or_create_award(row, awarding_agency_id),
"awarding_agency_id": awarding_agency_id,
"data_source": "USA",
"description": row["descriptionofcontractrequirement"],
"federal_action_obligation": row["dollarsobligated"],
"funding_agency_id": self.get_agency_id(row["fundingrequestingagencyid"], subtier_agency_dict),
"last_modified_date": h.convert_date(row['last_modified_date']),
"modification_number": row["modnumber"],
"place_of_performance": h.get_or_create_location(
row, mapper=location_mapper_place_of_performance),
"period_of_performance_current_end_date": h.convert_date(row['currentcompletiondate']),
"period_of_performance_start_date": h.convert_date(row['effectivedate']),
"recipient": self.get_or_create_recipient(row),
"submission": subattr,
"type": evaluate_contract_award_type(row),
"type_description": award_type_dict.get(evaluate_contract_award_type(row)),
"usaspending_unique_transaction_id": row["unique_transaction_id"]
}
txn = Transaction(**txn_dict)
txn_list.append(txn)
# Create the transaction contract object for this row
txn_contract_dict = {
"submission": subattr,
"piid": row['piid'],
"parent_award_id": row['idvpiid'],
"current_total_value_award": h.parse_numeric_value(row["baseandexercisedoptionsvalue"]),
"period_of_performance_potential_end_date": h.convert_date(row['ultimatecompletiondate']),
"potential_total_value_of_award": h.parse_numeric_value(row["baseandalloptionsvalue"]),
"epa_designated_product": self.parse_first_character(row['useofepadesignatedproducts']),
"gfe_gfp": h.up2colon(row['gfe_gfp']),
"cost_or_pricing_data": h.up2colon(row['costorpricingdata']),
"type_of_contract_pricing": h.up2colon(row['typeofcontractpricing']),
"type_of_contract_pricing_description": contract_pricing_dict.get(h.up2colon(row['typeofcontractpricing'])),
"multiple_or_single_award_idv": h.up2colon(row['multipleorsingleawardidc']),
"naics": h.up2colon(row['nationalinterestactioncode']),
"dod_claimant_program_code": h.up2colon(row['claimantprogramcode']),
"commercial_item_acquisition_procedures": h.up2colon(
row['commercialitemacquisitionprocedures']),
"commercial_item_test_program": h.up2colon(row['commercialitemtestprogram']),
"consolidated_contract": h.up2colon(row['consolidatedcontract']),
"contingency_humanitarian_or_peacekeeping_operation": h.up2colon(
row['contingencyhumanitarianpeacekeepingoperation']),
"contract_bundling": h.up2colon(row['contractbundling']),
"contract_financing": h.up2colon(row['contractfinancing']),
"contracting_officers_determination_of_business_size": h.up2colon(
row['contractingofficerbusinesssizedetermination']),
"country_of_product_or_service_origin": h.up2colon(row['countryoforigin']),
"davis_bacon_act": h.up2colon(row['davisbaconact']),
"evaluated_preference": h.up2colon(row['evaluatedpreference']),
"extent_competed": h.up2colon(row['extentcompeted']),
"information_technology_commercial_item_category": h.up2colon(
row['informationtechnologycommercialitemcategory']),
"interagency_contracting_authority": h.up2colon(row['interagencycontractingauthority']),
"local_area_set_aside": h.up2colon(row['localareasetaside']),
"purchase_card_as_payment_method": h.up2colon(row['purchasecardaspaymentmethod']),
"multi_year_contract": h.up2colon(row['multiyearcontract']),
"national_interest_action": h.up2colon(row['nationalinterestactioncode']),
"number_of_actions": h.up2colon(row['numberofactions']),
"number_of_offers_received": h.up2colon(row['numberofoffersreceived']),
"performance_based_service_acquisition": h.up2colon(row['performancebasedservicecontract']),
"place_of_manufacture": h.up2colon(row['placeofmanufacture']),
"product_or_service_code": h.up2colon(row['productorservicecode']),
"recovered_materials_sustainability": h.up2colon(row['recoveredmaterialclauses']),
"research": h.up2colon(row['research']),
"sea_transportation": h.up2colon(row['seatransportation']),
"service_contract_act": h.up2colon(row['servicecontractact']),
"small_business_competitiveness_demonstration_program": self.parse_first_character(
row['smallbusinesscompetitivenessdemonstrationprogram']),
"solicitation_procedures": h.up2colon(row['solicitationprocedures']),
"subcontracting_plan": h.up2colon(row['subcontractplan']),
"type_set_aside": h.up2colon(row['typeofsetaside']),
"walsh_healey_act": h.up2colon(row['walshhealyact']),
"rec_flag": self.parse_first_character(h.up2colon(row['rec_flag'])),
"type_of_idc": self.parse_first_character(row['typeofidc']),
"a76_fair_act_action": self.parse_first_character(row['a76action']),
"clinger_cohen_act_planning": self.parse_first_character(row['clingercohenact']),
"cost_accounting_standards": self.parse_first_character(
row['costaccountingstandardsclause']),
"fed_biz_opps": self.parse_first_character(row['fedbizopps']),
"foreign_funding": self.parse_first_character(row['fundedbyforeignentity']),
"major_program": self.parse_first_character(row['majorprogramcode']),
"program_acronym": self.parse_first_character(row['programacronym']),
"referenced_idv_modification_number": self.parse_first_character(
row['idvmodificationnumber']),
"transaction_number": self.parse_first_character(row['transactionnumber']),
"solicitation_identifier": self.parse_first_character(row['solicitationid'])
}
txn_contract = TransactionContract(**txn_contract_dict)
txn_contract_list.append(txn_contract)
# Bulk insert transaction rows
self.logger.info("Starting Transaction bulk insert ({} records)".format(len(txn_list)))
Transaction.objects.bulk_create(txn_list)
self.logger.info("Completed Transaction bulk insert")
# Update txn contract list with newly-inserted transactions
award_id_list = [] # we'll need this when updating the awards later on
for idx, t in enumerate(txn_contract_list):
# add transaction info to this TransactionContract object
t.transaction = txn_list[idx]
# add the corresponding award id to a list we'll use when batch-updating award data
award_id_list.append(txn_list[idx].award_id)
# Bulk insert transaction contract rows
self.logger.info("Starting TransactionContract bulk insert ({} records)".format(len(txn_contract_list)))
TransactionContract.objects.bulk_create(txn_contract_list)
self.logger.info("Completed TransactionContract bulk insert")
# Update awards to reflect latest transaction information
# (note that this can't be done via signals or a save()
# override in the model itself, because those aren't
# triggered by a bulk update
self.logger.info("Starting Awards update")
count = update_awards(tuple(award_id_list))
update_contract_awards(tuple(award_id_list))
self.logger.info("Completed Awards update ({} records)".format(count))
def get_agency_id(self, agency_string, subtier_agency_dict):
agency_code = h.up2colon(agency_string)
agency_id = subtier_agency_dict.get(agency_code)
if not agency_id:
self.logger.error("Missing agency: " + agency_string)
return agency_id
