def test_txn_assistance_get_or_create(): """Test TransactionFABS.get_or_create_2 method.""" agency1 = mommy.make("references.Agency") awd1 = mommy.make("awards.Award", awarding_agency=agency1) txn1 = mommy.make( "awards.TransactionNormalized", award=awd1, modification_number="1", awarding_agency=agency1, last_modified_date=date(2012, 3, 1), ) mommy.make( "awards.TransactionFABS", transaction=txn1, business_funds_indicator="a", record_type=1, total_funding_amount=1000.00, ) assert TransactionFABS.objects.all().count() == 1 # an updated transaction should also update existing TranactionAssistance # data, not insert a new record ta_dict = { "business_funds_indicator": "c", "record_type": 2, "total_funding_amount": 2000 } ta2 = TransactionFABS.get_or_create_2(txn1, **ta_dict) ta2.save() assert TransactionFABS.objects.all().count() == 1 t = TransactionNormalized.objects.get(id=txn1.id) assert t.assistance_data.business_funds_indicator == "c" assert t.assistance_data.record_type == 2 assert t.assistance_data.total_funding_amount == "2000" # a new transaction gets a new TransactionFABS record ta_dict = { "business_funds_indicator": "z", "record_type": 5, "total_funding_amount": 8000 } ta3 = TransactionFABS.get_or_create_2( mommy.make("awards.TransactionNormalized"), **ta_dict) ta3.save() assert TransactionFABS.objects.all().count() == 2
def test_txn_assistance_get_or_create(): """Test TransactionFABS.get_or_create_2 method.""" agency1 = mommy.make('references.Agency') awd1 = mommy.make('awards.Award', awarding_agency=agency1) txn1 = mommy.make( 'awards.TransactionNormalized', award=awd1, modification_number='1', awarding_agency=agency1, last_modified_date=date(2012, 3, 1), ) mommy.make('awards.TransactionFABS', transaction=txn1, business_funds_indicator='a', record_type=1, total_funding_amount=1000.00) assert TransactionFABS.objects.all().count() == 1 # an updated transaction should also update existing TranactionAssistance # data, not insert a new record ta_dict = { 'business_funds_indicator': 'c', 'record_type': 2, 'total_funding_amount': 2000, } ta2 = TransactionFABS.get_or_create_2(txn1, **ta_dict) ta2.save() assert TransactionFABS.objects.all().count() == 1 t = TransactionNormalized.objects.get(id=txn1.id) assert t.assistance_data.business_funds_indicator == 'c' assert t.assistance_data.record_type == 2 assert t.assistance_data.total_funding_amount == '2000' # a new transaction gets a new TransactionFABS record ta_dict = { 'business_funds_indicator': 'z', 'record_type': 5, 'total_funding_amount': 8000, } ta3 = TransactionFABS.get_or_create_2( mommy.make('awards.TransactionNormalized'), **ta_dict) ta3.save() assert TransactionFABS.objects.all().count() == 2
def load_transaction_fabs(self, fabs_broker_data, total_rows): logger.info('Starting bulk loading for FABS data') start_time = datetime.now() for index, row in enumerate(fabs_broker_data, 1): if not (index % 10000): logger.info('Transaction FABS: Loading row {} of {} ({})'.format(str(index), str(total_rows), datetime.now() - start_time)) fabs_instance_data = load_data_into_model( TransactionFABS(), # thrown away row, as_dict=True) fabs_instance = TransactionFABS(**fabs_instance_data) fabs_instance.transaction = transaction_normalized_bulk[index - 1] fabs_bulk.append(fabs_instance) logger.info('Bulk creating Transaction FABS (batch_size: {})...'.format(BATCH_SIZE)) TransactionFABS.objects.bulk_create(fabs_bulk, batch_size=BATCH_SIZE)
def load_transaction_fabs(self, fabs_broker_data, total_rows): logger.info('Starting bulk loading for FABS data') start_time = datetime.now() for index, row in enumerate(fabs_broker_data, 1): if not (index % 10000): logger.info( 'Transaction FABS: Loading row {} of {} ({})'.format( str(index), str(total_rows), datetime.now() - start_time)) fabs_instance_data = load_data_into_model( TransactionFABS(), # thrown away row, as_dict=True) fabs_instance = TransactionFABS(**fabs_instance_data) fabs_instance.transaction = transaction_normalized_bulk[index - 1] fabs_bulk.append(fabs_instance) logger.info( 'Bulk creating Transaction FABS (batch_size: {})...'.format( BATCH_SIZE)) TransactionFABS.objects.bulk_create(fabs_bulk, batch_size=BATCH_SIZE)
def insert_new_fabs(self, to_insert): place_of_performance_field_map = { "location_country_code": "place_of_perform_country_c", "country_name": "place_of_perform_country_n", "state_code": "place_of_perfor_state_code", "state_name": "place_of_perform_state_nam", "city_name": "place_of_performance_city", "county_name": "place_of_perform_county_na", "county_code": "place_of_perform_county_co", "foreign_location_description": "place_of_performance_forei", "zip_4a": "place_of_performance_zip4a", "congressional_code": "place_of_performance_congr", "performance_code": "place_of_performance_code", "zip_last4": "place_of_perform_zip_last4", "zip5": "place_of_performance_zip5", } legal_entity_location_field_map = { "location_country_code": "legal_entity_country_code", "country_name": "legal_entity_country_name", "state_code": "legal_entity_state_code", "state_name": "legal_entity_state_name", "city_name": "legal_entity_city_name", "city_code": "legal_entity_city_code", "county_name": "legal_entity_county_name", "county_code": "legal_entity_county_code", "address_line1": "legal_entity_address_line1", "address_line2": "legal_entity_address_line2", "address_line3": "legal_entity_address_line3", "foreign_location_description": "legal_entity_foreign_descr", "congressional_code": "legal_entity_congressional", "zip_last4": "legal_entity_zip_last4", "zip5": "legal_entity_zip5", "foreign_postal_code": "legal_entity_foreign_posta", "foreign_province": "legal_entity_foreign_provi", "foreign_city_name": "legal_entity_foreign_city", } for row in to_insert: upper_case_dict_values(row) # Create new LegalEntityLocation and LegalEntity from the row data legal_entity_location = create_location( legal_entity_location_field_map, row, {"recipient_flag": True}) recipient_name = row['awardee_or_recipient_legal'] legal_entity = LegalEntity.objects.create( recipient_unique_id=row['awardee_or_recipient_uniqu'], recipient_name=recipient_name if recipient_name is not None else "", parent_recipient_unique_id=row['ultimate_parent_unique_ide'], ) legal_entity_value_map = { "location": legal_entity_location, "business_categories": get_business_categories(row=row, data_type='fabs'), "business_types_description": row['business_types_desc'], } legal_entity = load_data_into_model( legal_entity, row, value_map=legal_entity_value_map, save=True) # Create the place of performance location pop_location = create_location(place_of_performance_field_map, row, {"place_of_performance_flag": True}) # Find the toptier awards from the subtier awards awarding_agency = Agency.get_by_subtier_only( row["awarding_sub_tier_agency_c"]) funding_agency = Agency.get_by_subtier_only( row["funding_sub_tier_agency_co"]) # Generate the unique Award ID # "ASST_AW_" + awarding_sub_tier_agency_c + fain + uri # this will raise an exception if the cast to an int fails, that's ok since we don't want to process # non-numeric record type values record_type_int = int(row['record_type']) if record_type_int == 1: uri = row['uri'] if row['uri'] else '-NONE-' fain = '-NONE-' elif record_type_int in (2, 3): uri = '-NONE-' fain = row['fain'] if row['fain'] else '-NONE-' else: msg = "Invalid record type encountered for the following afa_generated_unique record: {}" raise Exception(msg.format(row['afa_generated_unique'])) astac = row["awarding_sub_tier_agency_c"] if row[ "awarding_sub_tier_agency_c"] else "-NONE-" generated_unique_id = "ASST_AW_{}_{}_{}".format(astac, fain, uri) # Create the summary Award (created, award) = Award.get_or_create_summary_award( generated_unique_award_id=generated_unique_id, fain=row['fain'], uri=row['uri'], record_type=row['record_type'], ) award.save() # Append row to list of Awards updated AWARD_UPDATE_ID_LIST.append(award.id) try: last_mod_date = datetime.strptime(str( row['modified_at']), "%Y-%m-%d %H:%M:%S.%f").date() except ValueError: last_mod_date = datetime.strptime(str(row['modified_at']), "%Y-%m-%d %H:%M:%S").date() parent_txn_value_map = { "award": award, "awarding_agency": awarding_agency, "funding_agency": funding_agency, "recipient": legal_entity, "place_of_performance": pop_location, "period_of_performance_start_date": format_date(row['period_of_performance_star']), "period_of_performance_current_end_date": format_date(row['period_of_performance_curr']), "action_date": format_date(row['action_date']), "last_modified_date": last_mod_date, "type_description": row['assistance_type_desc'], "transaction_unique_id": row['afa_generated_unique'], "generated_unique_award_id": generated_unique_id, } fad_field_map = { "type": "assistance_type", "description": "award_description", "funding_amount": "total_funding_amount", } transaction_normalized_dict = load_data_into_model( TransactionNormalized(), # thrown away row, field_map=fad_field_map, value_map=parent_txn_value_map, as_dict=True, ) financial_assistance_data = load_data_into_model( TransactionFABS(), row, as_dict=True) # thrown away afa_generated_unique = financial_assistance_data[ 'afa_generated_unique'] unique_fabs = TransactionFABS.objects.filter( afa_generated_unique=afa_generated_unique) if unique_fabs.first(): transaction_normalized_dict["update_date"] = datetime.now( timezone.utc) transaction_normalized_dict["fiscal_year"] = fy( transaction_normalized_dict["action_date"]) # Update TransactionNormalized TransactionNormalized.objects.filter( id=unique_fabs.first().transaction.id).update( **transaction_normalized_dict) # Update TransactionFABS unique_fabs.update(**financial_assistance_data) else: # Create TransactionNormalized transaction = TransactionNormalized( **transaction_normalized_dict) transaction.save() # Create TransactionFABS transaction_fabs = TransactionFABS(transaction=transaction, **financial_assistance_data) transaction_fabs.save() # Update legal entity to map back to transaction legal_entity.transaction_unique_id = afa_generated_unique legal_entity.save()
def update_transaction_assistance(db_cursor, fiscal_year=None, page=1, limit=500000): # logger.info("Getting IDs for what's currently in the DB...") # current_ids = TransactionFABS.objects # # if fiscal_year: # current_ids = current_ids.filter(action_date__fy=fiscal_year) # # current_ids = current_ids.values_list('published_award_financial_assistance_id', flat=True) query = "SELECT * FROM published_award_financial_assistance" arguments = [] fy_begin = '10/01/' + str(fiscal_year - 1) fy_end = '09/30/' + str(fiscal_year) if fiscal_year: if arguments: query += " AND" else: query += " WHERE" query += ' action_date::Date BETWEEN %s AND %s' arguments += [fy_begin] arguments += [fy_end] query += ' ORDER BY published_award_financial_assistance_id LIMIT %s OFFSET %s' arguments += [limit, (page - 1) * limit] logger.info("Executing query on Broker DB => " + query % (arguments[0], arguments[1], arguments[2], arguments[3])) db_cursor.execute(query, arguments) logger.info("Running dictfetchall on db_cursor") award_financial_assistance_data = dictfetchall(db_cursor) legal_entity_location_field_map = { "address_line1": "legal_entity_address_line1", "address_line2": "legal_entity_address_line2", "address_line3": "legal_entity_address_line3", "city_name": "legal_entity_city_name", "congressional_code": "legal_entity_congressional", "county_code": "legal_entity_county_code", "county_name": "legal_entity_county_name", "foreign_city_name": "legal_entity_foreign_city", "foreign_postal_code": "legal_entity_foreign_posta", "foreign_province": "legal_entity_foreign_provi", "state_code": "legal_entity_state_code", "state_name": "legal_entity_state_name", "zip5": "legal_entity_zip5", "zip_last4": "legal_entity_zip_last4", "location_country_code": "legal_entity_country_code" } place_of_performance_field_map = { "city_name": "place_of_performance_city", "performance_code": "place_of_performance_code", "congressional_code": "place_of_performance_congr", "county_name": "place_of_perform_county_na", "foreign_location_description": "place_of_performance_forei", "state_name": "place_of_perform_state_nam", "zip4": "place_of_performance_zip4a", "location_country_code": "place_of_perform_country_c" } fad_field_map = { "type": "assistance_type", "description": "award_description", } logger.info("Getting total rows") # rows_loaded = len(current_ids) total_rows = len(award_financial_assistance_data) # - rows_loaded logger.info("Processing " + str(total_rows) + " rows of assistance data") # skip_count = 0 # ROW ITERATION STARTS HERE lel_bulk = [] pop_bulk = [] legal_entity_bulk = [] award_bulk = [] transaction_assistance_bulk = [] transaction_normalized_bulk = [] logger.info( 'Getting legal entity location objects for {} rows...'.format( len(award_financial_assistance_data))) for index, row in enumerate(award_financial_assistance_data, 1): # Recipient flag is true for LeL legal_entity_location = get_or_create_location( legal_entity_location_field_map, row, {"recipient_flag": True}, save=False) lel_bulk.append(legal_entity_location) logger.info('Bulk creating {} legal entity location rows...'.format( len(lel_bulk))) try: Location.objects.bulk_create(lel_bulk) except IntegrityError: logger.info('!!! DUPLICATES FOUND. Continuing... ') logger.info( 'Getting place of performance objects for {} rows...'.format( len(award_financial_assistance_data))) for index, row in enumerate(award_financial_assistance_data, 1): # Place of Performance flag is true for PoP pop_location = get_or_create_location( place_of_performance_field_map, row, {"place_of_performance_flag": True}, save=False) pop_bulk.append(pop_location) logger.info('Bulk creating {} place of performance rows...'.format( len(pop_bulk))) try: Location.objects.bulk_create(pop_bulk) except IntegrityError: logger.info('!!! DUPLICATES FOUND. Continuing... ') logger.info('Getting legal entity objects for {} rows...'.format( len(award_financial_assistance_data))) for index, row in enumerate(award_financial_assistance_data, 1): recipient_name = row.get('awardee_or_recipient_legal', '') legal_entity = LegalEntity.objects.filter( recipient_unique_id=row['awardee_or_recipient_uniqu'], recipient_name=recipient_name).first() if legal_entity is None: legal_entity = LegalEntity( recipient_unique_id=row['awardee_or_recipient_uniqu'], recipient_name=recipient_name) legal_entity_value_map = { "location": lel_bulk[index - 1], } legal_entity = load_data_into_model( legal_entity, row, value_map=legal_entity_value_map, save=False) legal_entity_bulk.append(legal_entity) logger.info('Bulk creating {} legal entity rows...'.format( len(legal_entity_bulk))) try: LegalEntity.objects.bulk_create(legal_entity_bulk) except IntegrityError: logger.info('!!! DUPLICATES FOUND. Continuing... ') awarding_agency_list = [] funding_agency_list = [] logger.info('Getting award objects for {} rows...'.format( len(award_financial_assistance_data))) for index, row in enumerate(award_financial_assistance_data, 1): # If awarding toptier agency code (aka CGAC) is not supplied on the D2 record, # use the sub tier code to look it up. This code assumes that all incoming # records will supply an awarding subtier agency code if row['awarding_agency_code'] is None or len( row['awarding_agency_code'].strip()) < 1: awarding_subtier_agency_id = subtier_agency_map[ row["awarding_sub_tier_agency_c"]] awarding_toptier_agency_id = subtier_to_agency_map[ awarding_subtier_agency_id]['toptier_agency_id'] awarding_cgac_code = toptier_agency_map[ awarding_toptier_agency_id] row['awarding_agency_code'] = awarding_cgac_code # If funding toptier agency code (aka CGAC) is empty, try using the sub # tier funding code to look it up. Unlike the awarding agency, we can't # assume that the funding agency subtier code will always be present. if row['funding_agency_code'] is None or len( row['funding_agency_code'].strip()) < 1: funding_subtier_agency_id = subtier_agency_map.get( row["funding_sub_tier_agency_co"]) if funding_subtier_agency_id is not None: funding_toptier_agency_id = \ subtier_to_agency_map[funding_subtier_agency_id]['toptier_agency_id'] funding_cgac_code = toptier_agency_map[ funding_toptier_agency_id] else: funding_cgac_code = None row['funding_agency_code'] = funding_cgac_code # Find the award that this award transaction belongs to. If it doesn't exist, create it. awarding_agency = Agency.get_by_toptier_subtier( row['awarding_agency_code'], row["awarding_sub_tier_agency_c"]) funding_agency = Agency.get_by_toptier_subtier( row['funding_agency_code'], row["funding_sub_tier_agency_co"]) awarding_agency_list.append(awarding_agency) funding_agency_list.append(funding_agency) # award.save() is called in Award.get_or_create_summary_award by default created, award = Award.get_or_create_summary_award( awarding_agency=awarding_agency, fain=row.get('fain'), uri=row.get('uri'), save=False) award_bulk.append(award) award_update_id_list.append(award.id) logger.info('Bulk creating {} award rows...'.format(len(award_bulk))) try: Award.objects.bulk_create(award_bulk) except IntegrityError: logger.info('!!! DUPLICATES FOUND. Continuing... ') logger.info('Getting transaction_normalized for {} rows...'.format( len(award_financial_assistance_data))) for index, row in enumerate(award_financial_assistance_data, 1): parent_txn_value_map = { "award": award_bulk[index - 1], "awarding_agency": awarding_agency_list[index - 1], "funding_agency": funding_agency_list[index - 1], "recipient": legal_entity_bulk[index - 1], "place_of_performance": pop_bulk[index - 1], "period_of_performance_start_date": format_date(row['period_of_performance_star']), "period_of_performance_current_end_date": format_date(row['period_of_performance_curr']), "action_date": format_date(row['action_date']), } transaction_dict = load_data_into_model( TransactionNormalized(), # thrown away row, field_map=fad_field_map, value_map=parent_txn_value_map, as_dict=True) transaction_normalized = TransactionNormalized.get_or_create_transaction( **transaction_dict) transaction_normalized.fiscal_year = fy( transaction_normalized.action_date) transaction_normalized_bulk.append(transaction_normalized) logger.info('Bulk creating {} TransactionNormalized rows...'.format( len(transaction_normalized_bulk))) try: TransactionNormalized.objects.bulk_create( transaction_normalized_bulk) except IntegrityError: logger.info( 'Tried and failed to insert duplicate transaction_normalized row. Continuing... ' ) for index, row in enumerate(award_financial_assistance_data, 1): financial_assistance_data = load_data_into_model( TransactionFABS(), # thrown away row, as_dict=True) transaction_assistance = TransactionFABS( transaction=transaction_normalized_bulk[index - 1], **financial_assistance_data) transaction_assistance_bulk.append(transaction_assistance) logger.info('Bulk creating TransactionFABS rows...') try: TransactionFABS.objects.bulk_create(transaction_assistance_bulk) except IntegrityError: logger.info('!!! DUPLICATES FOUND. Continuing... ')
def insert_new_fabs(to_insert): fabs_normalized_field_map = { "type": "assistance_type", "description": "award_description", "funding_amount": "total_funding_amount", } fabs_field_map = { "officer_1_name": "high_comp_officer1_full_na", "officer_1_amount": "high_comp_officer1_amount", "officer_2_name": "high_comp_officer2_full_na", "officer_2_amount": "high_comp_officer2_amount", "officer_3_name": "high_comp_officer3_full_na", "officer_3_amount": "high_comp_officer3_amount", "officer_4_name": "high_comp_officer4_full_na", "officer_4_amount": "high_comp_officer4_amount", "officer_5_name": "high_comp_officer5_full_na", "officer_5_amount": "high_comp_officer5_amount", } update_award_ids = [] for row in to_insert: upper_case_dict_values(row) # Find the toptier awards from the subtier awards awarding_agency = Agency.get_by_subtier_only( row["awarding_sub_tier_agency_c"]) funding_agency = Agency.get_by_subtier_only( row["funding_sub_tier_agency_co"]) # Create the summary Award (created, award) = Award.get_or_create_summary_award( generated_unique_award_id=row["unique_award_key"], fain=row["fain"], uri=row["uri"], record_type=row["record_type"], ) award.save() # Append row to list of Awards updated update_award_ids.append(award.id) try: last_mod_date = datetime.strptime(str(row["modified_at"]), "%Y-%m-%d %H:%M:%S.%f").date() except ValueError: last_mod_date = datetime.strptime(str(row["modified_at"]), "%Y-%m-%d %H:%M:%S").date() parent_txn_value_map = { "award": award, "awarding_agency": awarding_agency, "funding_agency": funding_agency, "period_of_performance_start_date": format_date(row["period_of_performance_star"]), "period_of_performance_current_end_date": format_date(row["period_of_performance_curr"]), "action_date": format_date(row["action_date"]), "last_modified_date": last_mod_date, "type_description": row["assistance_type_desc"], "transaction_unique_id": row["afa_generated_unique"], "business_categories": get_business_categories(row=row, data_type="fabs"), } transaction_normalized_dict = load_data_into_model( TransactionNormalized(), # thrown away row, field_map=fabs_normalized_field_map, value_map=parent_txn_value_map, as_dict=True, ) financial_assistance_data = load_data_into_model( TransactionFABS(), row, field_map=fabs_field_map, as_dict=True # thrown away ) # Hack to cut back on the number of warnings dumped to the log. financial_assistance_data["updated_at"] = cast_datetime_to_utc( financial_assistance_data["updated_at"]) financial_assistance_data["created_at"] = cast_datetime_to_utc( financial_assistance_data["created_at"]) financial_assistance_data["modified_at"] = cast_datetime_to_utc( financial_assistance_data["modified_at"]) afa_generated_unique = financial_assistance_data[ "afa_generated_unique"] unique_fabs = TransactionFABS.objects.filter( afa_generated_unique=afa_generated_unique) if unique_fabs.first(): transaction_normalized_dict["update_date"] = datetime.now( timezone.utc) transaction_normalized_dict["fiscal_year"] = fy( transaction_normalized_dict["action_date"]) # Update TransactionNormalized TransactionNormalized.objects.filter( id=unique_fabs.first().transaction.id).update( **transaction_normalized_dict) # Update TransactionFABS unique_fabs.update(**financial_assistance_data) else: # Create TransactionNormalized transaction_normalized = TransactionNormalized( **transaction_normalized_dict) transaction_normalized.save() # Create TransactionFABS transaction_fabs = TransactionFABS( transaction=transaction_normalized, **financial_assistance_data) transaction_fabs.save() return update_award_ids
def insert_new_fabs(to_insert): place_of_performance_field_map = { "location_country_code": "place_of_perform_country_c", "country_name": "place_of_perform_country_n", "state_code": "place_of_perfor_state_code", "state_name": "place_of_perform_state_nam", "city_name": "place_of_performance_city", "county_name": "place_of_perform_county_na", "county_code": "place_of_perform_county_co", "foreign_location_description": "place_of_performance_forei", "zip_4a": "place_of_performance_zip4a", "congressional_code": "place_of_performance_congr", "performance_code": "place_of_performance_code", "zip_last4": "place_of_perform_zip_last4", "zip5": "place_of_performance_zip5", } legal_entity_location_field_map = { "location_country_code": "legal_entity_country_code", "country_name": "legal_entity_country_name", "state_code": "legal_entity_state_code", "state_name": "legal_entity_state_name", "city_name": "legal_entity_city_name", "city_code": "legal_entity_city_code", "county_name": "legal_entity_county_name", "county_code": "legal_entity_county_code", "address_line1": "legal_entity_address_line1", "address_line2": "legal_entity_address_line2", "address_line3": "legal_entity_address_line3", "foreign_location_description": "legal_entity_foreign_descr", "congressional_code": "legal_entity_congressional", "zip_last4": "legal_entity_zip_last4", "zip5": "legal_entity_zip5", "foreign_postal_code": "legal_entity_foreign_posta", "foreign_province": "legal_entity_foreign_provi", "foreign_city_name": "legal_entity_foreign_city", } fabs_normalized_field_map = { "type": "assistance_type", "description": "award_description", "funding_amount": "total_funding_amount", } fabs_field_map = { "officer_1_name": "high_comp_officer1_full_na", "officer_1_amount": "high_comp_officer1_amount", "officer_2_name": "high_comp_officer2_full_na", "officer_2_amount": "high_comp_officer2_amount", "officer_3_name": "high_comp_officer3_full_na", "officer_3_amount": "high_comp_officer3_amount", "officer_4_name": "high_comp_officer4_full_na", "officer_4_amount": "high_comp_officer4_amount", "officer_5_name": "high_comp_officer5_full_na", "officer_5_amount": "high_comp_officer5_amount", } update_award_ids = [] for row in to_insert: upper_case_dict_values(row) # Create new LegalEntityLocation and LegalEntity from the row data legal_entity_location = create_location( legal_entity_location_field_map, row, {"recipient_flag": True}) recipient_name = row['awardee_or_recipient_legal'] legal_entity = LegalEntity.objects.create( recipient_unique_id=row['awardee_or_recipient_uniqu'], recipient_name=recipient_name if recipient_name is not None else "", parent_recipient_unique_id=row['ultimate_parent_unique_ide'], ) legal_entity_value_map = { "location": legal_entity_location, "business_categories": get_business_categories(row=row, data_type='fabs'), "business_types_description": row['business_types_desc'], } legal_entity = load_data_into_model(legal_entity, row, value_map=legal_entity_value_map, save=True) # Create the place of performance location pop_location = create_location(place_of_performance_field_map, row, {"place_of_performance_flag": True}) # Find the toptier awards from the subtier awards awarding_agency = Agency.get_by_subtier_only( row["awarding_sub_tier_agency_c"]) funding_agency = Agency.get_by_subtier_only( row["funding_sub_tier_agency_co"]) # Create the summary Award (created, award) = Award.get_or_create_summary_award( generated_unique_award_id=row['unique_award_key'], fain=row['fain'], uri=row['uri'], record_type=row['record_type'], ) award.save() # Append row to list of Awards updated update_award_ids.append(award.id) try: last_mod_date = datetime.strptime(str(row['modified_at']), "%Y-%m-%d %H:%M:%S.%f").date() except ValueError: last_mod_date = datetime.strptime(str(row['modified_at']), "%Y-%m-%d %H:%M:%S").date() parent_txn_value_map = { "award": award, "awarding_agency": awarding_agency, "funding_agency": funding_agency, "recipient": legal_entity, "place_of_performance": pop_location, "period_of_performance_start_date": format_date(row['period_of_performance_star']), "period_of_performance_current_end_date": format_date(row['period_of_performance_curr']), "action_date": format_date(row['action_date']), "last_modified_date": last_mod_date, "type_description": row['assistance_type_desc'], "transaction_unique_id": row['afa_generated_unique'], } transaction_normalized_dict = load_data_into_model( TransactionNormalized(), # thrown away row, field_map=fabs_normalized_field_map, value_map=parent_txn_value_map, as_dict=True, ) financial_assistance_data = load_data_into_model( TransactionFABS(), # thrown away row, field_map=fabs_field_map, as_dict=True) # Hack to cut back on the number of warnings dumped to the log. financial_assistance_data['updated_at'] = cast_datetime_to_utc( financial_assistance_data['updated_at']) financial_assistance_data['created_at'] = cast_datetime_to_utc( financial_assistance_data['created_at']) financial_assistance_data['modified_at'] = cast_datetime_to_utc( financial_assistance_data['modified_at']) afa_generated_unique = financial_assistance_data[ 'afa_generated_unique'] unique_fabs = TransactionFABS.objects.filter( afa_generated_unique=afa_generated_unique) if unique_fabs.first(): transaction_normalized_dict["update_date"] = datetime.now( timezone.utc) transaction_normalized_dict["fiscal_year"] = fy( transaction_normalized_dict["action_date"]) # Update TransactionNormalized TransactionNormalized.objects.filter( id=unique_fabs.first().transaction.id).update( **transaction_normalized_dict) # Update TransactionFABS unique_fabs.update(**financial_assistance_data) else: # Create TransactionNormalized transaction_normalized = TransactionNormalized( **transaction_normalized_dict) transaction_normalized.save() # Create TransactionFABS transaction_fabs = TransactionFABS( transaction=transaction_normalized, **financial_assistance_data) transaction_fabs.save() # Update legal entity to map back to transaction legal_entity.transaction_unique_id = afa_generated_unique legal_entity.save() return update_award_ids
def load_file_d2(submission_attributes, award_financial_assistance_data, db_cursor, quick, row_preprocessor=no_preprocessing): """ Process and load file D2 broker data (financial assistance award txns). """ d_start_time = time.time() if quick: setup_broker_fdw() parameters = { 'broker_submission_id': submission_attributes.broker_submission_id } run_sql_file('usaspending_api/etl/management/load_file_d2.sql', parameters) logger.info('\n\n\n\nFile D2 time elapsed: {}'.format(time.time() - d_start_time)) return legal_entity_location_field_map = { "address_line1": "legal_entity_address_line1", "address_line2": "legal_entity_address_line2", "address_line3": "legal_entity_address_line3", "city_code": "legal_entity_city_code", "city_name": "legal_entity_city_name", "congressional_code": "legal_entity_congressional", "county_code": "legal_entity_county_code", "county_name": "legal_entity_county_name", "foreign_city_name": "legal_entity_foreign_city", "foreign_postal_code": "legal_entity_foreign_posta", "foreign_province": "legal_entity_foreign_provi", "state_code": "legal_entity_state_code", "state_name": "legal_entity_state_name", "zip5": "legal_entity_zip5", "zip_last4": "legal_entity_zip_last4", "location_country_code": "legal_entity_country_code" } place_of_performance_field_map = { "city_name": "place_of_performance_city", "performance_code": "place_of_performance_code", "congressional_code": "place_of_performance_congr", "county_name": "place_of_perform_county_na", "foreign_location_description": "place_of_performance_forei", "state_name": "place_of_perform_state_nam", "zip4": "place_of_performance_zip4a", "location_country_code": "place_of_perform_country_c" } legal_entity_location_value_map = {"recipient_flag": True} place_of_performance_value_map = {"place_of_performance_flag": True} fad_field_map = { "type": "assistance_type", "description": "award_description", } total_rows = len(award_financial_assistance_data) start_time = datetime.now() for index, row in enumerate(award_financial_assistance_data, 1): if not (index % 100): logger.info('D2 File Load: Loading row {} of {} ({})'.format( str(index), str(total_rows), datetime.now() - start_time)) row = row_preprocessor(row) legal_entity_location, created = get_or_create_location( legal_entity_location_field_map, row, legal_entity_location_value_map) recipient_name = row['awardee_or_recipient_legal'] if recipient_name is None: recipient_name = "" # Create the legal entity if it doesn't exist legal_entity, created = LegalEntity.objects.get_or_create( recipient_unique_id=row['awardee_or_recipient_uniqu'], recipient_name=recipient_name) if created: legal_entity_value_map = { "location": legal_entity_location, } legal_entity = load_data_into_model( legal_entity, row, value_map=legal_entity_value_map, save=True) # Create the place of performance location pop_location, created = get_or_create_location( place_of_performance_field_map, row, place_of_performance_value_map) # If awarding toptier agency code (aka CGAC) is not supplied on the D2 record, # use the sub tier code to look it up. This code assumes that all incoming # records will supply an awarding subtier agency code if row['awarding_agency_code'] is None or len( row['awarding_agency_code'].strip()) < 1: row['awarding_agency_code'] = Agency.get_by_subtier( row["awarding_sub_tier_agency_c"]).toptier_agency.cgac_code # If funding toptier agency code (aka CGAC) is empty, try using the sub # tier funding code to look it up. Unlike the awarding agency, we can't # assume that the funding agency subtier code will always be present. if row['funding_agency_code'] is None or len( row['funding_agency_code'].strip()) < 1: funding_agency = Agency.get_by_subtier( row["funding_sub_tier_agency_co"]) row['funding_agency_code'] = ( funding_agency.toptier_agency.cgac_code if funding_agency is not None else None) # Find the award that this award transaction belongs to. If it doesn't exist, create it. awarding_agency = Agency.get_by_toptier_subtier( row['awarding_agency_code'], row["awarding_sub_tier_agency_c"]) created, award = Award.get_or_create_summary_award( awarding_agency=awarding_agency, piid=row.get('piid'), fain=row.get('fain'), uri=row.get('uri')) award.save() award_update_id_list.append(award.id) parent_txn_value_map = { "award": award, "awarding_agency": awarding_agency, "funding_agency": Agency.get_by_toptier_subtier(row['funding_agency_code'], row["funding_sub_tier_agency_co"]), "recipient": legal_entity, "place_of_performance": pop_location, 'submission': submission_attributes, "period_of_performance_start_date": format_date(row['period_of_performance_star']), "period_of_performance_current_end_date": format_date(row['period_of_performance_curr']), "action_date": format_date(row['action_date']), } transaction_dict = load_data_into_model( TransactionNormalized(), # thrown away row, field_map=fad_field_map, value_map=parent_txn_value_map, as_dict=True) transaction = TransactionNormalized.get_or_create_transaction( **transaction_dict) transaction.save() fad_value_map = { "submission": submission_attributes, "cfda": Cfda.objects.filter(program_number=row['cfda_number']).first(), 'reporting_period_start': submission_attributes.reporting_period_start, 'reporting_period_end': submission_attributes.reporting_period_end, "period_of_performance_start_date": format_date(row['period_of_performance_star']), "period_of_performance_current_end_date": format_date(row['period_of_performance_curr']), } financial_assistance_data = load_data_into_model( TransactionFABS(), # thrown away row, field_map=fad_field_map, value_map=fad_value_map, as_dict=True) transaction_assistance = TransactionFABS.get_or_create_2( transaction=transaction, **financial_assistance_data) transaction_assistance.save() logger.info('\n\n\n\nFile D2 time elapsed: {}'.format(time.time() - d_start_time))
def insert_new_fabs(self, to_insert, total_rows): logger.info('Starting insertion of new FABS data') place_of_performance_field_map = { "location_country_code": "place_of_perform_country_c", "country_name": "place_of_perform_country_n", "state_code": "place_of_perfor_state_code", "state_name": "place_of_perform_state_nam", "city_name": "place_of_performance_city", "county_name": "place_of_perform_county_na", "county_code": "place_of_perform_county_co", "foreign_location_description": "place_of_performance_forei", "zip_4a": "place_of_performance_zip4a", "congressional_code": "place_of_performance_congr", "performance_code": "place_of_performance_code", "zip_last4": "place_of_perform_zip_last4", "zip5": "place_of_performance_zip5" } legal_entity_location_field_map = { "location_country_code": "legal_entity_country_code", "country_name": "legal_entity_country_name", "state_code": "legal_entity_state_code", "state_name": "legal_entity_state_name", "city_name": "legal_entity_city_name", "city_code": "legal_entity_city_code", "county_name": "legal_entity_county_name", "county_code": "legal_entity_county_code", "address_line1": "legal_entity_address_line1", "address_line2": "legal_entity_address_line2", "address_line3": "legal_entity_address_line3", "foreign_location_description": "legal_entity_foreign_descr", "congressional_code": "legal_entity_congressional", "zip_last4": "legal_entity_zip_last4", "zip5": "legal_entity_zip5", "foreign_postal_code": "legal_entity_foreign_posta", "foreign_province": "legal_entity_foreign_provi", "foreign_city_name": "legal_entity_foreign_city" } start_time = datetime.now() for index, row in enumerate(to_insert, 1): if not (index % 1000): logger.info('Inserting Stale FABS: Inserting row {} of {} ({})'.format(str(index), str(total_rows), datetime.now() - start_time)) for key in row: if isinstance(row[key], str): row[key] = row[key].upper() # Create new LegalEntityLocation and LegalEntity from the row data legal_entity_location = create_location(legal_entity_location_field_map, row, {"recipient_flag": True}) recipient_name = row['awardee_or_recipient_legal'] legal_entity = LegalEntity.objects.create( recipient_unique_id=row['awardee_or_recipient_uniqu'], recipient_name=recipient_name if recipient_name is not None else "" ) legal_entity_value_map = { "location": legal_entity_location, "business_categories": get_business_categories(row=row, data_type='fabs'), "business_types_description": get_business_type_description(row['business_types']) } legal_entity = load_data_into_model(legal_entity, row, value_map=legal_entity_value_map, save=True) # Create the place of performance location pop_location = create_location(place_of_performance_field_map, row, {"place_of_performance_flag": True}) # Find the toptier awards from the subtier awards awarding_agency = Agency.get_by_subtier_only(row["awarding_sub_tier_agency_c"]) funding_agency = Agency.get_by_subtier_only(row["funding_sub_tier_agency_co"]) # Generate the unique Award ID # "ASST_AW_" + awarding_sub_tier_agency_c + fain + uri # this will raise an exception if the cast to an int fails, that's ok since we don't want to process # non-numeric record type values record_type_int = int(row['record_type']) if record_type_int == 1: uri = row['uri'] if row['uri'] else '-NONE-' fain = '-NONE-' elif record_type_int == 2: uri = '-NONE-' fain = row['fain'] if row['fain'] else '-NONE-' else: raise Exception('Invalid record type encountered for the following afa_generated_unique record: %s' % row['afa_generated_unique']) generated_unique_id = 'ASST_AW_' +\ (row['awarding_sub_tier_agency_c'] if row['awarding_sub_tier_agency_c'] else '-NONE-') + '_' + \ fain + '_' + uri # Create the summary Award (created, award) = Award.get_or_create_summary_award(generated_unique_award_id=generated_unique_id, fain=row['fain'], uri=row['uri'], record_type=row['record_type']) award.save() # Append row to list of Awards updated award_update_id_list.append(award.id) try: last_mod_date = datetime.strptime(str(row['modified_at']), "%Y-%m-%d %H:%M:%S.%f").date() except ValueError: last_mod_date = datetime.strptime(str(row['modified_at']), "%Y-%m-%d %H:%M:%S").date() parent_txn_value_map = { "award": award, "awarding_agency": awarding_agency, "funding_agency": funding_agency, "recipient": legal_entity, "place_of_performance": pop_location, "period_of_performance_start_date": format_date(row['period_of_performance_star']), "period_of_performance_current_end_date": format_date(row['period_of_performance_curr']), "action_date": format_date(row['action_date']), "last_modified_date": last_mod_date, "type_description": get_assistance_type_description(row['assistance_type']), "transaction_unique_id": row['afa_generated_unique'], "generated_unique_award_id": generated_unique_id } fad_field_map = { "type": "assistance_type", "description": "award_description", } transaction_normalized_dict = load_data_into_model( TransactionNormalized(), # thrown away row, field_map=fad_field_map, value_map=parent_txn_value_map, as_dict=True) financial_assistance_data = load_data_into_model( TransactionFABS(), # thrown away row, as_dict=True) afa_generated_unique = financial_assistance_data['afa_generated_unique'] unique_fabs = TransactionFABS.objects.filter(afa_generated_unique=afa_generated_unique) if unique_fabs.first(): transaction_normalized_dict["update_date"] = datetime.utcnow() transaction_normalized_dict["fiscal_year"] = fy(transaction_normalized_dict["action_date"]) # Update TransactionNormalized TransactionNormalized.objects.filter(id=unique_fabs.first().transaction.id).\ update(**transaction_normalized_dict) # Update TransactionFABS unique_fabs.update(**financial_assistance_data) else: # Create TransactionNormalized transaction = TransactionNormalized(**transaction_normalized_dict) transaction.save() # Create TransactionFABS transaction_fabs = TransactionFABS(transaction=transaction, **financial_assistance_data) transaction_fabs.save()
def insert_new_fabs(to_insert): place_of_performance_field_map = { "location_country_code": "place_of_perform_country_c", "country_name": "place_of_perform_country_n", "state_code": "place_of_perfor_state_code", "state_name": "place_of_perform_state_nam", "city_name": "place_of_performance_city", "county_name": "place_of_perform_county_na", "county_code": "place_of_perform_county_co", "foreign_location_description": "place_of_performance_forei", "zip_4a": "place_of_performance_zip4a", "congressional_code": "place_of_performance_congr", "performance_code": "place_of_performance_code", "zip_last4": "place_of_perform_zip_last4", "zip5": "place_of_performance_zip5", } legal_entity_location_field_map = { "location_country_code": "legal_entity_country_code", "country_name": "legal_entity_country_name", "state_code": "legal_entity_state_code", "state_name": "legal_entity_state_name", "city_name": "legal_entity_city_name", "city_code": "legal_entity_city_code", "county_name": "legal_entity_county_name", "county_code": "legal_entity_county_code", "address_line1": "legal_entity_address_line1", "address_line2": "legal_entity_address_line2", "address_line3": "legal_entity_address_line3", "foreign_location_description": "legal_entity_foreign_descr", "congressional_code": "legal_entity_congressional", "zip_last4": "legal_entity_zip_last4", "zip5": "legal_entity_zip5", "foreign_postal_code": "legal_entity_foreign_posta", "foreign_province": "legal_entity_foreign_provi", "foreign_city_name": "legal_entity_foreign_city", } update_award_ids = [] for row in to_insert: upper_case_dict_values(row) # Create new LegalEntityLocation and LegalEntity from the row data legal_entity_location = create_location(legal_entity_location_field_map, row, {"recipient_flag": True}) recipient_name = row['awardee_or_recipient_legal'] legal_entity = LegalEntity.objects.create( recipient_unique_id=row['awardee_or_recipient_uniqu'], recipient_name=recipient_name if recipient_name is not None else "", parent_recipient_unique_id=row['ultimate_parent_unique_ide'], ) legal_entity_value_map = { "location": legal_entity_location, "business_categories": get_business_categories(row=row, data_type='fabs'), "business_types_description": row['business_types_desc'], } legal_entity = load_data_into_model(legal_entity, row, value_map=legal_entity_value_map, save=True) # Create the place of performance location pop_location = create_location(place_of_performance_field_map, row, {"place_of_performance_flag": True}) # Find the toptier awards from the subtier awards awarding_agency = Agency.get_by_subtier_only(row["awarding_sub_tier_agency_c"]) funding_agency = Agency.get_by_subtier_only(row["funding_sub_tier_agency_co"]) # Generate the unique Award ID # "ASST_AW_" + awarding_sub_tier_agency_c + fain + uri # this will raise an exception if the cast to an int fails, that's ok since we don't want to process # non-numeric record type values record_type_int = int(row['record_type']) if record_type_int == 1: uri = row['uri'] if row['uri'] else '-NONE-' fain = '-NONE-' elif record_type_int in (2, 3): uri = '-NONE-' fain = row['fain'] if row['fain'] else '-NONE-' else: msg = "Invalid record type encountered for the following afa_generated_unique record: {}" raise Exception(msg.format(row['afa_generated_unique'])) astac = row["awarding_sub_tier_agency_c"] if row["awarding_sub_tier_agency_c"] else "-NONE-" generated_unique_id = "ASST_AW_{}_{}_{}".format(astac, fain, uri) # Create the summary Award (created, award) = Award.get_or_create_summary_award( generated_unique_award_id=generated_unique_id, fain=row['fain'], uri=row['uri'], record_type=row['record_type'], ) award.save() # Append row to list of Awards updated update_award_ids.append(award.id) try: last_mod_date = datetime.strptime(str(row['modified_at']), "%Y-%m-%d %H:%M:%S.%f").date() except ValueError: last_mod_date = datetime.strptime(str(row['modified_at']), "%Y-%m-%d %H:%M:%S").date() parent_txn_value_map = { "award": award, "awarding_agency": awarding_agency, "funding_agency": funding_agency, "recipient": legal_entity, "place_of_performance": pop_location, "period_of_performance_start_date": format_date(row['period_of_performance_star']), "period_of_performance_current_end_date": format_date(row['period_of_performance_curr']), "action_date": format_date(row['action_date']), "last_modified_date": last_mod_date, "type_description": row['assistance_type_desc'], "transaction_unique_id": row['afa_generated_unique'], "generated_unique_award_id": generated_unique_id, } fad_field_map = { "type": "assistance_type", "description": "award_description", "funding_amount": "total_funding_amount", } transaction_normalized_dict = load_data_into_model( TransactionNormalized(), # thrown away row, field_map=fad_field_map, value_map=parent_txn_value_map, as_dict=True, ) financial_assistance_data = load_data_into_model(TransactionFABS(), row, as_dict=True) # thrown away # Hack to cut back on the number of warnings dumped to the log. financial_assistance_data['updated_at'] = cast_datetime_to_utc(financial_assistance_data['updated_at']) financial_assistance_data['created_at'] = cast_datetime_to_utc(financial_assistance_data['created_at']) financial_assistance_data['modified_at'] = cast_datetime_to_utc(financial_assistance_data['modified_at']) afa_generated_unique = financial_assistance_data['afa_generated_unique'] unique_fabs = TransactionFABS.objects.filter(afa_generated_unique=afa_generated_unique) if unique_fabs.first(): transaction_normalized_dict["update_date"] = datetime.now(timezone.utc) transaction_normalized_dict["fiscal_year"] = fy(transaction_normalized_dict["action_date"]) # Update TransactionNormalized TransactionNormalized.objects.filter(id=unique_fabs.first().transaction.id).update( **transaction_normalized_dict ) # Update TransactionFABS unique_fabs.update(**financial_assistance_data) else: # Create TransactionNormalized transaction_normalized = TransactionNormalized(**transaction_normalized_dict) transaction_normalized.save() # Create TransactionFABS transaction_fabs = TransactionFABS(transaction=transaction_normalized, **financial_assistance_data) transaction_fabs.save() # Update legal entity to map back to transaction legal_entity.transaction_unique_id = afa_generated_unique legal_entity.save() return update_award_ids
def update_transaction_assistance(db_cursor, fiscal_year=None, page=1, limit=500000): query = "SELECT * FROM published_award_financial_assistance" arguments = [] fy_begin = "10/01/" + str(fiscal_year - 1) fy_end = "09/30/" + str(fiscal_year) if fiscal_year: if arguments: query += " AND" else: query += " WHERE" query += " action_date::Date BETWEEN %s AND %s" arguments += [fy_begin] arguments += [fy_end] query += " ORDER BY published_award_financial_assistance_id LIMIT %s OFFSET %s" arguments += [limit, (page - 1) * limit] logger.info("Executing query on Broker DB => " + query % (arguments[0], arguments[1], arguments[2], arguments[3])) db_cursor.execute(query, arguments) logger.info("Running dictfetchall on db_cursor") award_financial_assistance_data = dictfetchall(db_cursor) fabs_normalized_field_map = { "type": "assistance_type", "description": "award_description", "funding_amount": "total_funding_amount", } fabs_field_map = { "officer_1_name": "high_comp_officer1_full_na", "officer_1_amount": "high_comp_officer1_amount", "officer_2_name": "high_comp_officer2_full_na", "officer_2_amount": "high_comp_officer2_amount", "officer_3_name": "high_comp_officer3_full_na", "officer_3_amount": "high_comp_officer3_amount", "officer_4_name": "high_comp_officer4_full_na", "officer_4_amount": "high_comp_officer4_amount", "officer_5_name": "high_comp_officer5_full_na", "officer_5_amount": "high_comp_officer5_amount", } logger.info("Getting total rows") total_rows = len(award_financial_assistance_data) # - rows_loaded logger.info("Processing " + str(total_rows) + " rows of assistance data") # ROW ITERATION STARTS HERE award_bulk = [] transaction_assistance_bulk = [] transaction_normalized_bulk = [] awarding_agency_list = [] funding_agency_list = [] logger.info("Getting award objects for {} rows...".format( len(award_financial_assistance_data))) for index, row in enumerate(award_financial_assistance_data, 1): # If awarding toptier agency code (aka CGAC) is not supplied on the D2 record, # use the sub tier code to look it up. This code assumes that all incoming # records will supply an awarding subtier agency code if row["awarding_agency_code"] is None or len( row["awarding_agency_code"].strip()) < 1: awarding_subtier_agency_id = subtier_agency_map[ row["awarding_sub_tier_agency_c"]] awarding_toptier_agency_id = subtier_to_agency_map[ awarding_subtier_agency_id]["toptier_agency_id"] awarding_toptier_code = toptier_agency_map[ awarding_toptier_agency_id] row["awarding_agency_code"] = awarding_toptier_code # If funding toptier agency code (aka CGAC) is empty, try using the sub # tier funding code to look it up. Unlike the awarding agency, we can't # assume that the funding agency subtier code will always be present. if row["funding_agency_code"] is None or len( row["funding_agency_code"].strip()) < 1: funding_subtier_agency_id = subtier_agency_map.get( row["funding_sub_tier_agency_co"]) if funding_subtier_agency_id is not None: funding_toptier_agency_id = subtier_to_agency_map[ funding_subtier_agency_id]["toptier_agency_id"] funding_toptier_code = toptier_agency_map[ funding_toptier_agency_id] else: funding_toptier_code = None row["funding_agency_code"] = funding_toptier_code # Find the award that this award transaction belongs to. If it doesn't exist, create it. awarding_agency = Agency.get_by_toptier_subtier( row["awarding_agency_code"], row["awarding_sub_tier_agency_c"]) funding_agency = Agency.get_by_toptier_subtier( row["funding_agency_code"], row["funding_sub_tier_agency_co"]) awarding_agency_list.append(awarding_agency) funding_agency_list.append(funding_agency) # award.save() is called in Award.get_or_create_summary_award by default created, award = Award.get_or_create_summary_award( awarding_agency=awarding_agency, fain=row.get("fain"), uri=row.get("uri"), generated_unique_award_id=row.get("unique_award_key"), save=False, ) award_bulk.append(award) award_update_id_list.append(award.id) award_assistance_update_id_list.append(award.id) logger.info("Bulk creating {} award rows...".format(len(award_bulk))) try: Award.objects.bulk_create(award_bulk) except IntegrityError: logger.info("!!! DUPLICATES FOUND. Continuing... ") logger.info("Getting transaction_normalized for {} rows...".format( len(award_financial_assistance_data))) for index, row in enumerate(award_financial_assistance_data, 1): parent_txn_value_map = { "award": award_bulk[index - 1], "awarding_agency": awarding_agency_list[index - 1], "funding_agency": funding_agency_list[index - 1], "period_of_performance_start_date": format_date(row["period_of_performance_star"]), "period_of_performance_current_end_date": format_date(row["period_of_performance_curr"]), "action_date": format_date(row["action_date"]), } transaction_dict = load_data_into_model( TransactionNormalized(), # thrown away row, field_map=fabs_normalized_field_map, value_map=parent_txn_value_map, as_dict=True, ) transaction_normalized = TransactionNormalized.get_or_create_transaction( **transaction_dict) transaction_normalized.fiscal_year = fy( transaction_normalized.action_date) transaction_normalized_bulk.append(transaction_normalized) logger.info("Bulk creating {} TransactionNormalized rows...".format( len(transaction_normalized_bulk))) try: TransactionNormalized.objects.bulk_create( transaction_normalized_bulk) except IntegrityError: logger.info( "Tried and failed to insert duplicate transaction_normalized row. Continuing... " ) for index, row in enumerate(award_financial_assistance_data, 1): financial_assistance_data = load_data_into_model( TransactionFABS(), row, field_map=fabs_field_map, as_dict=True # thrown away ) transaction_assistance = TransactionFABS( transaction=transaction_normalized_bulk[index - 1], **financial_assistance_data) transaction_assistance_bulk.append(transaction_assistance) logger.info("Bulk creating TransactionFABS rows...") try: TransactionFABS.objects.bulk_create(transaction_assistance_bulk) except IntegrityError: logger.info("!!! DUPLICATES FOUND. Continuing... ")
def load_file_d2( submission_attributes, award_financial_assistance_data, db_cursor, quick, row_preprocessor=no_preprocessing ): """ Process and load file D2 broker data (financial assistance award txns). """ d_start_time = time.time() if quick: setup_broker_fdw() parameters = {'broker_submission_id': submission_attributes.broker_submission_id} run_sql_file('usaspending_api/etl/management/load_file_d2.sql', parameters) logger.info('\n\n\n\nFile D2 time elapsed: {}'.format(time.time() - d_start_time)) return legal_entity_location_field_map = { "address_line1": "legal_entity_address_line1", "address_line2": "legal_entity_address_line2", "address_line3": "legal_entity_address_line3", "city_code": "legal_entity_city_code", "city_name": "legal_entity_city_name", "congressional_code": "legal_entity_congressional", "county_code": "legal_entity_county_code", "county_name": "legal_entity_county_name", "foreign_city_name": "legal_entity_foreign_city", "foreign_postal_code": "legal_entity_foreign_posta", "foreign_province": "legal_entity_foreign_provi", "state_code": "legal_entity_state_code", "state_name": "legal_entity_state_name", "zip5": "legal_entity_zip5", "zip_last4": "legal_entity_zip_last4", "location_country_code": "legal_entity_country_code" } place_of_performance_field_map = { "city_name": "place_of_performance_city", "performance_code": "place_of_performance_code", "congressional_code": "place_of_performance_congr", "county_name": "place_of_perform_county_na", "foreign_location_description": "place_of_performance_forei", "state_name": "place_of_perform_state_nam", "zip4": "place_of_performance_zip4a", "location_country_code": "place_of_perform_country_c" } legal_entity_location_value_map = { "recipient_flag": True } place_of_performance_value_map = { "place_of_performance_flag": True } fad_field_map = { "type": "assistance_type", "description": "award_description", } total_rows = len(award_financial_assistance_data) start_time = datetime.now() for index, row in enumerate(award_financial_assistance_data, 1): if not (index % 100): logger.info('D2 File Load: Loading row {} of {} ({})'.format(str(index), str(total_rows), datetime.now() - start_time)) row = row_preprocessor(row) legal_entity_location, created = get_or_create_location( legal_entity_location_field_map, row, legal_entity_location_value_map ) recipient_name = row['awardee_or_recipient_legal'] if recipient_name is None: recipient_name = "" # Create the legal entity if it doesn't exist legal_entity, created = LegalEntity.objects.get_or_create( recipient_unique_id=row['awardee_or_recipient_uniqu'], recipient_name=recipient_name ) if created: legal_entity_value_map = { "location": legal_entity_location, } legal_entity = load_data_into_model(legal_entity, row, value_map=legal_entity_value_map, save=True) # Create the place of performance location pop_location, created = get_or_create_location( place_of_performance_field_map, row, place_of_performance_value_map ) # If awarding toptier agency code (aka CGAC) is not supplied on the D2 record, # use the sub tier code to look it up. This code assumes that all incoming # records will supply an awarding subtier agency code if row['awarding_agency_code'] is None or len(row['awarding_agency_code'].strip()) < 1: row['awarding_agency_code'] = Agency.get_by_subtier( row["awarding_sub_tier_agency_c"]).toptier_agency.cgac_code # If funding toptier agency code (aka CGAC) is empty, try using the sub # tier funding code to look it up. Unlike the awarding agency, we can't # assume that the funding agency subtier code will always be present. if row['funding_agency_code'] is None or len(row['funding_agency_code'].strip()) < 1: funding_agency = Agency.get_by_subtier(row["funding_sub_tier_agency_co"]) row['funding_agency_code'] = ( funding_agency.toptier_agency.cgac_code if funding_agency is not None else None) # Find the award that this award transaction belongs to. If it doesn't exist, create it. awarding_agency = Agency.get_by_toptier_subtier( row['awarding_agency_code'], row["awarding_sub_tier_agency_c"] ) created, award = Award.get_or_create_summary_award( awarding_agency=awarding_agency, piid=row.get('piid'), fain=row.get('fain'), uri=row.get('uri')) award.save() award_update_id_list.append(award.id) parent_txn_value_map = { "award": award, "awarding_agency": awarding_agency, "funding_agency": Agency.get_by_toptier_subtier(row['funding_agency_code'], row["funding_sub_tier_agency_co"]), "recipient": legal_entity, "place_of_performance": pop_location, 'submission': submission_attributes, "period_of_performance_start_date": format_date(row['period_of_performance_star']), "period_of_performance_current_end_date": format_date(row['period_of_performance_curr']), "action_date": format_date(row['action_date']), } transaction_dict = load_data_into_model( TransactionNormalized(), # thrown away row, field_map=fad_field_map, value_map=parent_txn_value_map, as_dict=True) transaction = TransactionNormalized.get_or_create_transaction(**transaction_dict) transaction.save() fad_value_map = { "submission": submission_attributes, "cfda": Cfda.objects.filter(program_number=row['cfda_number']).first(), 'reporting_period_start': submission_attributes.reporting_period_start, 'reporting_period_end': submission_attributes.reporting_period_end, "period_of_performance_start_date": format_date(row['period_of_performance_star']), "period_of_performance_current_end_date": format_date(row['period_of_performance_curr']), } financial_assistance_data = load_data_into_model( TransactionFABS(), # thrown away row, field_map=fad_field_map, value_map=fad_value_map, as_dict=True) transaction_assistance = TransactionFABS.get_or_create_2(transaction=transaction, **financial_assistance_data) transaction_assistance.save() logger.info('\n\n\n\nFile D2 time elapsed: {}'.format(time.time() - d_start_time))