def run(input_directory):
#create logger
today = datetime.datetime.now().strftime('%Y%m%d')
logger = logging.getLogger('dwetl')
file_handler = logging.FileHandler(f'logs/dwetl.log.{today}')
formatter = logging.Formatter('%(asctime)s %(levelname)s %(message)s')
file_handler.setFormatter(formatter)
logger.addHandler(file_handler)
logger.setLevel(logging.INFO)
time_started = datetime.datetime.now()
logger.info(f'DWETL.py started')
'''
create job_info for current process
'''
with dwetl.database_session() as session:
job_info_table_class = dwetl.Base.classes['dw_prcsng_cycle']
job_info = JobInfoFactory.create_job_info_from_db(
session, job_info_table_class)
'''
load_stage_1
'''
load_stage_1.load_stage_1(job_info, input_directory, logger)
'''
load_stage_2
load 'in_' values from stg1 to stg 2 tables
load 'in_' values
'''
load_stage_2.load_stage_2(job_info, logger)
'''
stg 2 intertable processing
'''
stage_2_intertable_processing.stage_2_intertable_processing(
job_info, logger)
'''
end of job metadata writing
'''
endtime = datetime.datetime.now()
# write end time to processing cycle table
with dwetl.database_session() as session:
job_info_table_class = dwetl.Base.classes['dw_prcsng_cycle']
# get row for current id and write end time to it
max_prcsng_id = session.query(job_info_table_class).\
filter(job_info_table_class.dw_prcsng_cycle_id == job_info.prcsng_cycle_id).\
update({'dw_prcsng_cycle_exectn_end_tmstmp': endtime})
elapsed_time = endtime - time_started
print("elapsed time: ", str(elapsed_time))
def load_fact_table(job_info, logger):
print('EZProxy loading fact table...')
logger.info('Loading to the fact table.... ')
stage2_table = dwetl.Base.classes['dw_stg_2_ezp_sessns_snap']
fact_table = dwetl.Base.classes['fact_ezp_sessns_snap']
processing_cycle_id = job_info.prcsng_cycle_id
# get max value for fact key from the reporting db
with dwetl.reporting_database_session() as session2:
reporting_fact_table = dwetl.ReportingBase.classes[
'fact_ezp_sessns_snap']
max_ezp_sessns_snap_fact_key = session2.query(
func.max(reporting_fact_table.ezp_sessns_snap_fact_key)).scalar()
if max_ezp_sessns_snap_fact_key is None:
max_ezp_sessns_snap_fact_key = 1
# load etl ezp fact table
with dwetl.database_session() as session:
reader = SqlAlchemyReader(session, stage2_table,
'em_create_dw_prcsng_cycle_id',
processing_cycle_id)
writer = SqlAlchemyWriter(session, fact_table)
processor = EzproxyFactProcessor(reader, writer, job_info, logger,
max_ezp_sessns_snap_fact_key)
processor.execute()
logger.info('Finished loading to the fact table.... ')
def load_stage_1(job_info, input_directory, logger):
print('Loading stage 1...')
logger.info('Loading stage 1...')
'''
file to table mapping
'''
ALEPH_TSV_TABLE_MAPPING = {
"mai01_z00_data": "dw_stg_1_mai01_z00",
"mai39_z00_data": "dw_stg_1_mai39_z00",
"mai01_z13_data": "dw_stg_1_mai01_z13",
"mai39_z13_data": "dw_stg_1_mai39_z13",
"mai01_z13u_data": "dw_stg_1_mai01_z13u",
"mai39_z13u_data": "dw_stg_1_mai39_z13u",
# "mai60_z00_data": "dw_stg_1_mai60_z00",
# "mai60_z13_data": "dw_stg_1_mai60_z13",
# "mai60_z13u_data": "dw_stg_1_mai60_z13u",
# "mai60_z103_bib_data": "dw_stg_1_mai50_z103_bib",
# "mai50_z30_data": "dw_stg_1_mai50_z30",
# "mai50_z35_data": "dw_stg_1_mai50_z35",
# "mai50_z30_full_data": "dw_stg_1_mai50_z30_full",
# "mai50_z103_bib_full_data": "dw_stg_2_lbry_item_z103_bib_full",
}
# Z00_FIELD_TABLE_MAPPING = {
# "mai01_z00_field_data": "dw_stg_1_mai01_z00_field",
# "mai39_z00_field_data": "dw_stg_1_mai39_z00_field",
# "mai60_z00_field_data": "dw_stg_1_mai60_z00_field",
# }
#
# MPF_TABLE_MAPPING = {
# "mpf_member-library-dimension.txt": "dw_stg_1_mpf_mbr_lbry",
# "mpf_library-entity-dimension.txt": "dw_stg_1_mpf_lbry_entity",
# "mpf_library-collection-dimension.txt": "dw_stg_1_mpf_collection",
# "mpf_item-status-dimension.txt": "dw_stg_1_mpf_item_status",
# "mpf_item-process-status-dimension.txt": "dw_stg_1_mpf_item_prcs_status",
# "mpf_material-form-dimension.txt": "dw_stg_1_mpf_matrl_form"
# }
'''
load aleph tsv files minus z00_field tables
'''
for file, table in ALEPH_TSV_TABLE_MAPPING.items():
file_path = os.path.join(input_directory, file)
print(file_path)
logger.info(file_path)
with dwetl.database_session() as session:
reader = TsvFileReader(file_path)
# writer = PrintWriter()
writer = SqlAlchemyWriter(session, dwetl.Base.classes[table])
processor = LoadAlephTsv(reader, writer, job_info, logger)
processor.execute()
def load_stage_1(job_info, input_file, logger):
print('EZProxy Loading stage 1...')
logger.info('EZ Proxy Loading stage 1...')
table = 'dw_stg_1_ezp_sessns_snap'
with dwetl.database_session() as session:
reader = EzproxyReader(input_file)
# writer = PrintWriter()
writer = SqlAlchemyWriter(session, dwetl.Base.classes[table])
processor = LoadAlephTsv(reader, writer, job_info, logger)
processor.execute()
def intertable_processing(job_info, logger):
print('EZProxy transformations started...')
logger.info('EZProxy intertable processing starts...')
stage2_table = dwetl.Base.classes['dw_stg_2_ezp_sessns_snap']
processing_cycle_id = job_info.prcsng_cycle_id
with dwetl.database_session() as session:
reader = SqlAlchemyReader(session, stage2_table,
'em_create_dw_prcsng_cycle_id',
processing_cycle_id)
writer = SqlAlchemyWriter(session, stage2_table)
processor = EzproxyProcessor(reader, writer, job_info, logger)
processor.execute()
logger.info('Finished EZProxy intertable processing .... ')
def load_stage_2(job_info, logger):
print('Loading stage 2...')
logger.info('Loading stage 2...')
stage1_to_stage2_table_mappings = {
"dw_stg_1_mai39_z13": "dw_stg_2_bib_rec_z13",
'dw_stg_1_mai01_z13': "dw_stg_2_bib_rec_z13",
"dw_stg_1_mai01_z13u": "dw_stg_2_bib_rec_z13u",
"dw_stg_1_mai01_z00": "dw_stg_2_bib_rec_z00",
"dw_stg_1_mai39_z00": "dw_stg_2_bib_rec_z00",
"dw_stg_1_mai39_z13u": "dw_stg_2_bib_rec_z13u",
# "dw_stg_1_mai60_z00": "dw_stg_2_lbry_holding_z00",
# "dw_stg_1_mai60_z13": "dw_stg_2_lbry_holding_z13",
# "dw_stg_1_mai60_z13u": "dw_stg_2_lbry_holding_z13u",
# "dw_stg_1_mai50_z30": "dw_stg_2_lbry_item_z30",
# "dw_stg_1_mai50_z35": "dw_stg_2_lbry_item_event_z35",
# "dw_stg_1_mai01_z00_field": "dw_stg_2_bib_rec_z00_field",
# "dw_stg_1_mai39_z00_field": "dw_stg_2_bib_rec_z00_field",
# "dw_stg_1_mai60_z00_field": "dw_stg_2_lbry_holding_z00_field",
# "dw_stg_1_mpf_mbr_lbry": "dw_stg_2_mpf_mbr_lbry",
# "dw_stg_1_mpf_lbry_entity": "dw_stg_2_mpf_lbry_entity",
# "dw_stg_1_mpf_collection": "dw_stg_2_mpf_collection",
# "dw_stg_1_mpf_item_status": "dw_stg_2_mpf_item_status",
# "dw_stg_1_mpf_item_prcs_status": "dw_stg_2_mpf_item_prcs_status",
# "dw_stg_1_mpf_matrl_form": "dw_stg_2_mpf_matrl_form"
}
processing_cycle_id = job_info.prcsng_cycle_id
for stage1_table, stage2_table in stage1_to_stage2_table_mappings.items():
print(stage1_table)
logger.info(stage1_table)
library = aleph_library(stage1_table)
with dwetl.database_session() as session:
stage1_table_class = dwetl.Base.classes[stage1_table]
stage2_table_class = dwetl.Base.classes[stage2_table]
reader = SqlAlchemyReader(session, stage1_table_class,
'em_create_dw_prcsng_cycle_id',
processing_cycle_id)
writer = SqlAlchemyWriter(session, stage2_table_class)
processor = CopyStage1ToStage2.create(reader, writer, job_info,
logger, library)
processor.execute()
def load_stage_2(job_info, logger):
print('EZProxy Loading stage 2...')
logger.info('EZProxy Loading stage 2...')
processing_cycle_id = job_info.prcsng_cycle_id
with dwetl.database_session() as session:
stage1_table_class = dwetl.Base.classes["dw_stg_1_ezp_sessns_snap"]
stage2_table_class = dwetl.Base.classes["dw_stg_2_ezp_sessns_snap"]
reader = SqlAlchemyReader(session, stage1_table_class,
'em_create_dw_prcsng_cycle_id',
processing_cycle_id)
writer = SqlAlchemyWriter(session, stage2_table_class)
# there is no aleph library for ez proxy data, but CopyStage1ToStage2 still will work
library = ''
processor = CopyStage1ToStage2.create(reader, writer, job_info, logger,
library)
processor.execute()
logger.info('Finished EZProxy loading stage 2 .... ')
def copy_new_facts_to_reporting_db(job_info, logger):
etl_fact_table = dwetl.Base.classes['fact_ezp_sessns_snap']
processing_cycle_id = job_info.prcsng_cycle_id
# query and select records from etl fact table
# should we use the create update processing cycle ID? or the Update processing cycle id?
with dwetl.database_session() as session:
reader = SqlAlchemyReader(session, etl_fact_table,
'em_create_dw_prcsng_cycle_id',
processing_cycle_id)
session.expunge_all()
# insert records into reporting db ezp fact table
with dwetl.reporting_database_session() as session2:
reporting_fact_table = dwetl.ReportingBase.classes[
'fact_ezp_sessns_snap']
writer = SqlAlchemyWriter(session2, reporting_fact_table)
processor = EzproxyReportingFactProcessor(reader, writer, job_info,
logger)
processor.execute()
def stage_2_intertable_processing(job_info, logger):
print("Stage 2 Intertable Processing...")
logger.info("Stage 2 Intertable Processing...")
STG_2_TABLE_CONFIG_MAPPING = {
'dw_stg_2_bib_rec_z00': 'bibliographic_record_dimension',
'dw_stg_2_bib_rec_z13': 'bibliographic_record_dimension',
'dw_stg_2_bib_rec_z13u': 'bibliographic_record_dimension',
'dw_stg_2_bib_rec_z00_field': 'bibliographic_record_dimension'
}
processing_cycle_id = job_info.prcsng_cycle_id
for table, dimension in STG_2_TABLE_CONFIG_MAPPING.items():
print(table)
logger.info(table)
# get json_config for current dimension
table_config_path = os.path.join('table_config', dimension + '.json')
json_config = load_table_config(table_config_path)
with dwetl.database_session() as session:
# gets SA base class for the current table
stage2_table_class = dwetl.Base.classes[table]
# gets list of PKs for the current table
pk_list = [pk.name for pk in stage2_table_class.__table__.primary_key]
reader = SqlAlchemyReader(session, stage2_table_class, 'em_create_dw_prcsng_cycle_id', processing_cycle_id)
writer = SqlAlchemyWriter(session, stage2_table_class)
'''
Preprocessing
'''
preprocessor = Preprocess(reader, writer, job_info, logger, json_config, pk_list)
preprocessor.execute()
'''
# Data Quality Checks
# '''
data_quality_checker = DataQualityProcessor(reader, writer, job_info, logger, json_config, pk_list)
data_quality_checker.execute()
def run(input_file):
#create logger
today = datetime.datetime.now().strftime('%Y%m%d')
logger = logging.getLogger('dwetl')
file_handler = logging.FileHandler(f'logs/ezproxy.log.{today}')
formatter = logging.Formatter('%(asctime)s %(levelname)s %(message)s')
file_handler.setFormatter(formatter)
logger.addHandler(file_handler)
logger.setLevel(logging.INFO)
time_started = datetime.datetime.now()
logger.info(f'EzProxy ETL started')
'''
check current hostname environment configuration to prevent errors
'''
hostname = socket.gethostname()
configured_host = database_credentials.configured_host()
if hostname != configured_host:
# quit program if env file hostname doesn't match with the current hostname
print(
'ERROR: EzProxy ETL ended because .env contained an error. Please double check the configured host and db configuration.'
)
logger.error(
f'EzProxy ETL ended because .env contained an error. please double check the configured host and db configuration.'
)
sys.exit()
'''
create job_info for current process
'''
with dwetl.database_session() as session:
job_info_table_class = dwetl.Base.classes['dw_prcsng_cycle']
job_info = JobInfoFactory.create_job_info_from_db(
session, job_info_table_class)
'''
load ezproxy stage 1
'''
ezproxy_load.load_stage_1(job_info, input_file, logger)
'''
load ezproxy stage 2
'''
ezproxy_load.load_stage_2(job_info, logger)
'''
stg 2 intertable processing
'''
ezproxy_load.intertable_processing(job_info, logger)
'''
fact table load
'''
ezproxy_load.load_fact_table(job_info, logger)
'''
copy new ezproxy data to reporting database
'''
ezproxy_load.copy_new_facts_to_reporting_db(job_info, logger)
'''
end of job metadata writing
'''
endtime = datetime.datetime.now()
# write end time to processing cycle table
with dwetl.database_session() as session:
job_info_table_class = dwetl.Base.classes['dw_prcsng_cycle']
# get row for current id and write end time to it
max_prcsng_id = session.query(job_info_table_class).\
filter(job_info_table_class.dw_prcsng_cycle_id == job_info.prcsng_cycle_id).\
update({'dw_prcsng_cycle_exectn_end_tmstmp': endtime})
elapsed_time = endtime - time_started
print("Ezproxy ETL elapsed time: ", str(elapsed_time))
logger.info(f'EzProxy ETL elapsed time: {str(elapsed_time)}')
