Commit eef43def authored by Margrenzo Gunawan's avatar Margrenzo Gunawan 💬

Production

Dags dari production
parent f16c0353
File added
import json
from datetime import date
import calendar
from airflow import DAG
from airflow.operators.python import PythonOperator, BranchPythonOperator
from airflow.operators.bash import BashOperator
from datetime import datetime, timedelta
from airflow.providers.sftp.operators.sftp import SFTPOperator
from airflow.models import Variable
from airflow.providers.postgres.operators.postgres import PostgresOperator
from airflow.operators.trigger_dagrun import TriggerDagRunOperator
from airflow.utils.trigger_rule import TriggerRule
from airflow.models.xcom import XCom
from airflow.decorators import task
from airflow.hooks.postgres_hook import PostgresHook
from airflow import XComArg
yesterday_nodash = (datetime.now() - timedelta(1)).strftime('%Y%m%d') if Variable.get(
"DATE_OF_DATA") == 'today' else Variable.get("DATE_OF_DATA")
yesterday_strip = datetime.strptime(yesterday_nodash, '%Y%m%d').strftime('%Y-%m-%d')
POSTGRES_CONN_ID = Variable.get("DS_DB")
POSTGRES_SCHEMA = 'ds_ebanking' + Variable.get("ENV_T24")
DS_CONN_ID = Variable.get("DS_EBANKING")
DS_FOLDER = 'ebanking'
DS_DB = 'ebanking'
DS_SCHEMA = 'ebanking'
DS_CREATE_TABLE = ''
def create_ddl(schema,column):
return f"""select string_agg(x,'') from (select
'create table {POSTGRES_SCHEMA}.'||table_name || ' ( ' || string_agg(column_name, ' text,') || ' text'|| ' ); ' as x
from information_schema.columns where table_schema = '{schema}'
and table_name in ({column})
group by table_schema, table_name) as x ;"""
def _start():
print("Start :: Extractor ")
def ds_list_extractor():
sql_stmt = f"""select * from ds_conf.ds_extractor_list_extractor('ebanking', '{yesterday_strip}');"""
pg_hook = PostgresHook(
postgres_conn_id=POSTGRES_CONN_ID,
)
pg_conn = pg_hook.get_conn()
cursor = pg_conn.cursor()
cursor.execute(sql_stmt)
files = cursor.fetchall()
return files
def ds_get_syntax(ti):
iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
if not iris:
raise Exception('No data.')
return [{"op_kwargs": {
"copy_sql": f"""COPY {DS_SCHEMA}.{table['file_id'].replace(".", "_")} to STDOUT delimiter '|' CSV header""",
"file_id": table['file_id']}}
for
table in json.loads(iris[0][0][0])]
def ds_get_csv(ti, copy_sql, file_id):
pg_hook = PostgresHook.get_hook(DS_CONN_ID)
pg_hook.copy_expert(copy_sql, filename=f"""{Variable.get("LOCAL_PATH")}{DS_FOLDER}/""" + (
f"""{yesterday_nodash}/{yesterday_nodash}.""" if Variable.get(
"WITH_DATE") == 'Y' else '') + f"""{file_id}.csv""")
#
# def ds_drop_syntax(ti):
# iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
# if not iris:
# raise Exception('No data.')
# return [{
# "sql": f"""drop table ds_ebanking.{table['file_id'].replace(".", "_")} CASCADE;"""}
# for
# table in json.loads(iris[0][0][0])]
def ds_get_ddl(ti):
global DS_CREATE_TABLE
iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
column = []
for table in json.loads(iris[0][0][0]):
column.append(f"""'{table['file_id']}'""")
print(create_ddl(DS_SCHEMA, ','.join(column)))
sql_stmt = create_ddl(DS_SCHEMA, ','.join(column))
pg_hook = PostgresHook(
postgres_conn_id=DS_CONN_ID,
)
pg_conn = pg_hook.get_conn()
cursor = pg_conn.cursor()
cursor.execute(sql_stmt)
return cursor.fetchall()[0][0]
def pg_ddl_syntax(ti):
print(ti.xcom_pull(task_ids=['ds_get_ddl'])[0])
return [{"sql": ti.xcom_pull(task_ids=['ds_get_ddl'])[0]}]
def pg_push_syntax(ti):
iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
if not iris:
raise Exception('No data.')
return [{"op_kwargs": {
"copy_sql": f"""COPY {POSTGRES_SCHEMA}.{table['file_id'].replace(".", "_")} from STDOUT delimiter '|' CSV header""",
"file_id": table['file_id']}}
for
table in json.loads(iris[0][0][0])]
def pg_push_csv(ti, copy_sql, file_id):
pg_hook = PostgresHook.get_hook(POSTGRES_CONN_ID)
pg_hook.copy_expert(copy_sql, filename=f"""{Variable.get("LOCAL_PATH")}{DS_FOLDER}/""" + (
f"""{yesterday_nodash}/{yesterday_nodash}.""" if Variable.get(
"WITH_DATE") == 'Y' else '') + f"""{file_id}.csv""")
with DAG("APJ_1_ebanking",
start_date=datetime(2021, 1, 1),
schedule_interval='0 * * * *',
catchup=False,
concurrency=3) as dag:
begin = PythonOperator(
task_id=f"Begin",
python_callable=_start,
op_kwargs={
}
)
create_folder = BashOperator(
task_id='create_folder',
bash_command=f"""mkdir -p {Variable.get("LOCAL_PATH")}{DS_FOLDER}/{yesterday_nodash}"""
)
ds_list_extractor = PythonOperator(
task_id='ds_list_extractor',
python_callable=ds_list_extractor,
do_xcom_push=True
)
ds_get_syntax = PythonOperator(
task_id='ds_syntax_get',
python_callable=ds_get_syntax
)
ds_table_to_csv = PythonOperator.partial(
task_id="ds_table_to_csv",
python_callable=ds_get_csv,
dag=dag
).expand_kwargs(
XComArg(ds_get_syntax),
)
ds_get_ddl = PythonOperator(
task_id='ds_get_ddl',
python_callable=ds_get_ddl,
do_xcom_push=True
)
# ds_drop_syntax = PythonOperator(
# task_id='ds_drop_syntax',
# python_callable=ds_drop_syntax
# )
pg_drop_schema = PostgresOperator(
sql= f"""drop schema IF EXISTS {POSTGRES_SCHEMA} CASCADE;""",
task_id="pg_drop_schema",
postgres_conn_id=POSTGRES_CONN_ID,
)
pg_create_schema = PostgresOperator(
sql= f"""create schema IF NOT EXISTS {POSTGRES_SCHEMA};""",
task_id="pg_create_schema",
postgres_conn_id=POSTGRES_CONN_ID,
)
pg_ddl_syntax = PythonOperator(
task_id='pg_ddl_syntax',
python_callable=pg_ddl_syntax
)
pg_create_table = PostgresOperator.partial(
task_id="pg_create_table",
postgres_conn_id=POSTGRES_CONN_ID,
).expand_kwargs(
XComArg(pg_ddl_syntax),
)
pg_push_syntax = PythonOperator(
task_id='pg_syntax_push',
python_callable=pg_push_syntax
)
pg_csv_to_table = PythonOperator.partial(
task_id="pg_csv_to_table",
python_callable=pg_push_csv,
dag=dag
).expand_kwargs(
XComArg(pg_push_syntax),
)
set_access_schemma = PostgresOperator(
sql=f"""GRANT USAGE ON SCHEMA {POSTGRES_SCHEMA} TO readaccess;""",
task_id="set_access_schemma",
postgres_conn_id=POSTGRES_CONN_ID,
)
set_access_all_table = PostgresOperator(
sql=f"""GRANT SELECT ON ALL TABLES IN SCHEMA {POSTGRES_SCHEMA} TO readaccess;""",
task_id="set_access_all_table",
postgres_conn_id=POSTGRES_CONN_ID,
)
begin >> create_folder >> ds_list_extractor >> ds_get_syntax >> ds_table_to_csv >> ds_get_ddl >> pg_drop_schema >> pg_create_schema >> pg_ddl_syntax >> pg_create_table >> pg_push_syntax >> pg_csv_to_table >> set_access_schemma >> set_access_all_table
\ No newline at end of file
import json
from datetime import date
import calendar
from airflow import DAG
from airflow.operators.python import PythonOperator, BranchPythonOperator
from airflow.operators.bash import BashOperator
from datetime import datetime, timedelta
from airflow.providers.sftp.operators.sftp import SFTPOperator
from airflow.models import Variable
from airflow.providers.postgres.operators.postgres import PostgresOperator
from airflow.operators.trigger_dagrun import TriggerDagRunOperator
from airflow.utils.trigger_rule import TriggerRule
from airflow.models.xcom import XCom
from airflow.decorators import task
from airflow.hooks.postgres_hook import PostgresHook
from airflow import XComArg
yesterday_nodash = (datetime.now() - timedelta(1)).strftime('%Y%m%d') if Variable.get(
"DATE_OF_DATA") == 'today' else Variable.get("DATE_OF_DATA")
yesterday_strip = datetime.strptime(yesterday_nodash, '%Y%m%d').strftime('%Y-%m-%d')
POSTGRES_CONN_ID = Variable.get("DS_DB")
POSTGRES_SCHEMA = 'ds_bib' + Variable.get("ENV_T24")
DS_CONN_ID = Variable.get("DS_IBB")
DS_FOLDER = 'bib'
DS_DB = 'bib'
DS_SCHEMA = 'bib'
DS_CREATE_TABLE = ''
def create_ddl(schema,column):
return f"""select string_agg(x,'') from (select
'create table {POSTGRES_SCHEMA}.'||table_name || ' ( ' || string_agg(column_name, ' text,') || ' text'|| ' ); ' as x
from information_schema.columns where table_schema = '{schema}'
and table_name in ({column})
group by table_schema, table_name) as x ;"""
def _start():
print("Start :: Extractor ")
def ds_list_extractor():
sql_stmt = f"""select * from ds_conf.ds_extractor_list_extractor('ibb', '{yesterday_strip}');"""
pg_hook = PostgresHook(
postgres_conn_id=POSTGRES_CONN_ID,
)
pg_conn = pg_hook.get_conn()
cursor = pg_conn.cursor()
cursor.execute(sql_stmt)
files = cursor.fetchall()
return files
def ds_get_syntax(ti):
iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
if not iris:
raise Exception('No data.')
return [{"op_kwargs": {
"copy_sql": f"""COPY {DS_SCHEMA}.{table['file_id'].replace(".", "_")} to STDOUT delimiter '|' CSV header""",
"file_id": table['file_id']}}
for
table in json.loads(iris[0][0][0])]
def ds_get_csv(ti, copy_sql, file_id):
pg_hook = PostgresHook.get_hook(DS_CONN_ID)
pg_hook.copy_expert(copy_sql, filename=f"""{Variable.get("LOCAL_PATH")}{DS_FOLDER}/""" + (
f"""{yesterday_nodash}/{yesterday_nodash}.""" if Variable.get(
"WITH_DATE") == 'Y' else '') + f"""{file_id}.csv""")
#
# def ds_drop_syntax(ti):
# iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
# if not iris:
# raise Exception('No data.')
# return [{
# "sql": f"""drop table ds_ebanking.{table['file_id'].replace(".", "_")} CASCADE;"""}
# for
# table in json.loads(iris[0][0][0])]
def ds_get_ddl(ti):
global DS_CREATE_TABLE
iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
column = []
for table in json.loads(iris[0][0][0]):
column.append(f"""'{table['file_id']}'""")
print(create_ddl(DS_SCHEMA, ','.join(column)))
sql_stmt = create_ddl(DS_SCHEMA, ','.join(column))
pg_hook = PostgresHook(
postgres_conn_id=DS_CONN_ID,
)
pg_conn = pg_hook.get_conn()
cursor = pg_conn.cursor()
cursor.execute(sql_stmt)
return cursor.fetchall()[0][0]
def pg_ddl_syntax(ti):
print(ti.xcom_pull(task_ids=['ds_get_ddl'])[0])
return [{"sql": ti.xcom_pull(task_ids=['ds_get_ddl'])[0]}]
def pg_push_syntax(ti):
iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
if not iris:
raise Exception('No data.')
return [{"op_kwargs": {
"copy_sql": f"""COPY {POSTGRES_SCHEMA}.{table['file_id'].replace(".", "_")} from STDOUT delimiter '|' CSV header""",
"file_id": table['file_id']}}
for
table in json.loads(iris[0][0][0])]
def pg_push_csv(ti, copy_sql, file_id):
pg_hook = PostgresHook.get_hook(POSTGRES_CONN_ID)
pg_hook.copy_expert(copy_sql, filename=f"""{Variable.get("LOCAL_PATH")}{DS_FOLDER}/""" + (
f"""{yesterday_nodash}/{yesterday_nodash}.""" if Variable.get(
"WITH_DATE") == 'Y' else '') + f"""{file_id}.csv""")
with DAG("APJ_1_ibb",
start_date=datetime(2021, 1, 1),
schedule_interval=None,
catchup=False,
concurrency=3) as dag:
begin = PythonOperator(
task_id=f"Begin",
python_callable=_start,
op_kwargs={
}
)
create_folder = BashOperator(
task_id='create_folder',
bash_command=f"""mkdir -p {Variable.get("LOCAL_PATH")}{DS_FOLDER}/{yesterday_nodash}"""
)
ds_list_extractor = PythonOperator(
task_id='ds_list_extractor',
python_callable=ds_list_extractor,
do_xcom_push=True
)
ds_get_syntax = PythonOperator(
task_id='ds_syntax_get',
python_callable=ds_get_syntax
)
ds_table_to_csv = PythonOperator.partial(
task_id="ds_table_to_csv",
python_callable=ds_get_csv,
dag=dag
).expand_kwargs(
XComArg(ds_get_syntax),
)
ds_get_ddl = PythonOperator(
task_id='ds_get_ddl',
python_callable=ds_get_ddl,
do_xcom_push=True
)
# ds_drop_syntax = PythonOperator(
# task_id='ds_drop_syntax',
# python_callable=ds_drop_syntax
# )
pg_drop_schema = PostgresOperator(
sql= f"""drop schema IF EXISTS {POSTGRES_SCHEMA} CASCADE;""",
task_id="pg_drop_schema",
postgres_conn_id=POSTGRES_CONN_ID,
)
pg_create_schema = PostgresOperator(
sql= f"""create schema IF NOT EXISTS {POSTGRES_SCHEMA};""",
task_id="pg_create_schema",
postgres_conn_id=POSTGRES_CONN_ID,
)
pg_ddl_syntax = PythonOperator(
task_id='pg_ddl_syntax',
python_callable=pg_ddl_syntax
)
pg_create_table = PostgresOperator.partial(
task_id="pg_create_table",
postgres_conn_id=POSTGRES_CONN_ID,
).expand_kwargs(
XComArg(pg_ddl_syntax),
)
pg_push_syntax = PythonOperator(
task_id='pg_syntax_push',
python_callable=pg_push_syntax
)
pg_csv_to_table = PythonOperator.partial(
task_id="pg_csv_to_table",
python_callable=pg_push_csv,
dag=dag
).expand_kwargs(
XComArg(pg_push_syntax),
)
set_access_schemma = PostgresOperator(
sql=f"""GRANT USAGE ON SCHEMA {POSTGRES_SCHEMA} TO readaccess;""",
task_id="set_access_schemma",
postgres_conn_id=POSTGRES_CONN_ID,
)
set_access_all_table = PostgresOperator(
sql=f"""GRANT SELECT ON ALL TABLES IN SCHEMA {POSTGRES_SCHEMA} TO readaccess;""",
task_id="set_access_all_table",
postgres_conn_id=POSTGRES_CONN_ID,
)
begin >> create_folder >> ds_list_extractor >> ds_get_syntax >> ds_table_to_csv >> ds_get_ddl >> pg_drop_schema >> pg_create_schema >> pg_ddl_syntax >> pg_create_table >> pg_push_syntax >> pg_csv_to_table >> set_access_schemma >> set_access_all_table
\ No newline at end of file
import json
from datetime import date
import calendar
from airflow import DAG
from airflow.operators.python import PythonOperator, BranchPythonOperator
from airflow.operators.bash import BashOperator
from datetime import datetime, timedelta
from airflow.providers.sftp.operators.sftp import SFTPOperator
from airflow.models import Variable
from airflow.providers.postgres.operators.postgres import PostgresOperator
from airflow.operators.trigger_dagrun import TriggerDagRunOperator
from airflow.utils.trigger_rule import TriggerRule
from airflow.models.xcom import XCom
from airflow.decorators import task
from airflow.hooks.postgres_hook import PostgresHook
from airflow import XComArg
yesterday_nodash = (datetime.now() - timedelta(1)).strftime('%Y%m%d') if Variable.get(
"DATE_OF_DATA") == 'today' else Variable.get("DATE_OF_DATA")
yesterday_strip = datetime.strptime(yesterday_nodash, '%Y%m%d').strftime('%Y-%m-%d')
POSTGRES_CONN_ID = Variable.get("DS_DB")
POSTGRES_SCHEMA = 'ds_igate' + Variable.get("ENV_T24")
DS_CONN_ID = Variable.get("DS_IGATE")
DS_FOLDER = 'igate'
DS_DB = 'igate'
DS_SCHEMA = 'public'
DS_CREATE_TABLE = ''
def create_ddl(schema,column):
return f"""select string_agg(x,'') from (select
'create table {POSTGRES_SCHEMA}.'||table_name || ' ( ' || string_agg(column_name, ' text,') || ' text'|| ' ); ' as x
from information_schema.columns where table_schema = '{schema}'
and table_name in ({column})
group by table_schema, table_name) as x ;"""
def _start():
print("Start :: Extractor ")
def ds_list_extractor():
sql_stmt = f"""select * from ds_conf.ds_extractor_list_extractor('igate', '{yesterday_strip}');"""
pg_hook = PostgresHook(
postgres_conn_id=POSTGRES_CONN_ID,
)
pg_conn = pg_hook.get_conn()
cursor = pg_conn.cursor()
cursor.execute(sql_stmt)
files = cursor.fetchall()
return files
def ds_get_syntax(ti):
iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
if not iris:
raise Exception('No data.')
return [{"op_kwargs": {
"copy_sql": f"""COPY ( select * from {DS_SCHEMA}.{table['file_id'].replace(".", "_")}{table['partition_table_format']} {table['condition']}) to STDOUT delimiter '{table['delimiter']}' CSV header""",
"file_id": table['file_id']}}
for
table in json.loads(iris[0][0][0])]
def ds_get_csv(ti, copy_sql, file_id):
pg_hook = PostgresHook.get_hook(DS_CONN_ID)
pg_hook.copy_expert(copy_sql, filename=f"""{Variable.get("LOCAL_PATH")}{DS_FOLDER}/""" + (
f"""{yesterday_nodash}/{yesterday_nodash}.""" if Variable.get(
"WITH_DATE") == 'Y' else '') + f"""{file_id}.csv""")
#
# def ds_drop_syntax(ti):
# iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
# if not iris:
# raise Exception('No data.')
# return [{
# "sql": f"""drop table ds_ebanking.{table['file_id'].replace(".", "_")} CASCADE;"""}
# for
# table in json.loads(iris[0][0][0])]
def ds_get_ddl(ti):
global DS_CREATE_TABLE
iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
column = []
for table in json.loads(iris[0][0][0]):
column.append(f"""'{table['file_id']}'""")
print(create_ddl(DS_SCHEMA, ','.join(column)))
sql_stmt = create_ddl(DS_SCHEMA, ','.join(column))
pg_hook = PostgresHook(
postgres_conn_id=DS_CONN_ID,
)
pg_conn = pg_hook.get_conn()
cursor = pg_conn.cursor()
cursor.execute(sql_stmt)
return cursor.fetchall()[0][0]
def pg_ddl_syntax(ti):
print(ti.xcom_pull(task_ids=['ds_get_ddl'])[0])
return [{"sql": ti.xcom_pull(task_ids=['ds_get_ddl'])[0]}]
def pg_push_syntax(ti):
iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
if not iris:
raise Exception('No data.')
return [{"op_kwargs": {
"copy_sql": f"""COPY {POSTGRES_SCHEMA}.{table['file_id'].replace(".", "_")} from STDOUT delimiter '|' CSV header""",
"file_id": table['file_id']}}
for
table in json.loads(iris[0][0][0])]
def pg_push_csv(ti, copy_sql, file_id):
pg_hook = PostgresHook.get_hook(POSTGRES_CONN_ID)
pg_hook.copy_expert(copy_sql, filename=f"""{Variable.get("LOCAL_PATH")}{DS_FOLDER}/""" + (
f"""{yesterday_nodash}/{yesterday_nodash}.""" if Variable.get(
"WITH_DATE") == 'Y' else '') + f"""{file_id}.csv""")
with DAG("APJ_1_igate",
start_date=datetime(2021, 1, 1),
schedule_interval='0 * * * *',
catchup=False,
concurrency=3) as dag:
begin = PythonOperator(
task_id=f"Begin",
python_callable=_start,
op_kwargs={
}
)
create_folder = BashOperator(
task_id='create_folder',
bash_command=f"""mkdir -p {Variable.get("LOCAL_PATH")}{DS_FOLDER}/{yesterday_nodash}"""
)
ds_list_extractor = PythonOperator(
task_id='ds_list_extractor',
python_callable=ds_list_extractor,
do_xcom_push=True
)
ds_get_syntax = PythonOperator(
task_id='ds_syntax_get',
python_callable=ds_get_syntax
)
ds_table_to_csv = PythonOperator.partial(
task_id="ds_table_to_csv",
python_callable=ds_get_csv,
dag=dag
).expand_kwargs(
XComArg(ds_get_syntax),
)
ds_get_ddl = PythonOperator(
task_id='ds_get_ddl',
python_callable=ds_get_ddl,
do_xcom_push=True
)
# ds_drop_syntax = PythonOperator(
# task_id='ds_drop_syntax',
# python_callable=ds_drop_syntax
# )
pg_drop_schema = PostgresOperator(
sql= f"""drop schema IF EXISTS {POSTGRES_SCHEMA} CASCADE;""",
task_id="pg_drop_schema",
postgres_conn_id=POSTGRES_CONN_ID,
)
pg_create_schema = PostgresOperator(
sql= f"""create schema IF NOT EXISTS {POSTGRES_SCHEMA};""",
task_id="pg_create_schema",
postgres_conn_id=POSTGRES_CONN_ID,
)
pg_ddl_syntax = PythonOperator(
task_id='pg_ddl_syntax',
python_callable=pg_ddl_syntax
)
pg_create_table = PostgresOperator.partial(
task_id="pg_create_table",
postgres_conn_id=POSTGRES_CONN_ID,
).expand_kwargs(
XComArg(pg_ddl_syntax),
)
pg_push_syntax = PythonOperator(
task_id='pg_syntax_push',
python_callable=pg_push_syntax
)
pg_csv_to_table = PythonOperator.partial(
task_id="pg_csv_to_table",
python_callable=pg_push_csv,
dag=dag
).expand_kwargs(
XComArg(pg_push_syntax),
)
set_access_schemma = PostgresOperator(
sql=f"""GRANT USAGE ON SCHEMA {POSTGRES_SCHEMA} TO readaccess;""",
task_id="set_access_schemma",
postgres_conn_id=POSTGRES_CONN_ID,
)
set_access_all_table = PostgresOperator(
sql=f"""GRANT SELECT ON ALL TABLES IN SCHEMA {POSTGRES_SCHEMA} TO readaccess;""",
task_id="set_access_all_table",
postgres_conn_id=POSTGRES_CONN_ID,
)
begin >> create_folder >> ds_list_extractor >> ds_get_syntax >> ds_table_to_csv >> ds_get_ddl >> pg_drop_schema >> pg_create_schema >> pg_ddl_syntax >> pg_create_table >> pg_push_syntax >> pg_csv_to_table >> set_access_schemma >> set_access_all_table
\ No newline at end of file
import json
from datetime import date
import calendar
from airflow import DAG
from airflow.operators.python import PythonOperator, BranchPythonOperator
from airflow.operators.bash import BashOperator
from datetime import datetime, timedelta
from airflow.providers.sftp.operators.sftp import SFTPOperator
from airflow.models import Variable
from airflow.providers.postgres.operators.postgres import PostgresOperator
from airflow.operators.trigger_dagrun import TriggerDagRunOperator
from airflow.utils.trigger_rule import TriggerRule
from airflow.models.xcom import XCom
from airflow.decorators import task
from airflow.hooks.postgres_hook import PostgresHook
from airflow import XComArg
yesterday_nodash = (datetime.now() - timedelta(1)).strftime('%Y%m%d') if Variable.get(
"DATE_OF_DATA") == 'today' else Variable.get("DATE_OF_DATA")
yesterday_strip = datetime.strptime(yesterday_nodash, '%Y%m%d').strftime('%Y-%m-%d')
POSTGRES_CONN_ID = Variable.get("DS_DB")
POSTGRES_SCHEMA = 'ds_los_fintech' + Variable.get("ENV_T24")
DS_CONN_ID = Variable.get("DS_LOS_FINTECH")
DS_FOLDER = 'los'
DS_DB = 'los-fintech'
DS_SCHEMA = 'public'
DS_CREATE_TABLE = ''
def create_ddl(schema,column):
return f"""select string_agg(x,'') from (select
'create table {POSTGRES_SCHEMA}.'||table_name || ' ( ' || string_agg(column_name, ' text,') || ' text'|| ' ); ' as x
from information_schema.columns where table_schema = '{schema}'
and table_name in ({column})
group by table_schema, table_name) as x ;"""
def _start():
print("Start :: Extractor ")
def ds_list_extractor():
sql_stmt = f"""select * from ds_conf.ds_extractor_list_extractor('los-fintech', '{yesterday_strip}');"""
pg_hook = PostgresHook(
postgres_conn_id=POSTGRES_CONN_ID,
)
pg_conn = pg_hook.get_conn()
cursor = pg_conn.cursor()
cursor.execute(sql_stmt)
files = cursor.fetchall()
return files
def ds_get_syntax(ti):
iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
if not iris:
raise Exception('No data.')
return [{"op_kwargs": {
"copy_sql": f"""COPY {DS_SCHEMA}.{table['file_id'].replace(".", "_")} to STDOUT delimiter '|' CSV header""",
"file_id": table['file_id']}}
for
table in json.loads(iris[0][0][0])]
def ds_get_csv(ti, copy_sql, file_id):
pg_hook = PostgresHook.get_hook(DS_CONN_ID)
pg_hook.copy_expert(copy_sql, filename=f"""{Variable.get("LOCAL_PATH")}{DS_FOLDER}/""" + (
f"""{yesterday_nodash}/{yesterday_nodash}.""" if Variable.get(
"WITH_DATE") == 'Y' else '') + f"""{file_id}.csv""")
#
# def ds_drop_syntax(ti):
# iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
# if not iris:
# raise Exception('No data.')
# return [{
# "sql": f"""drop table ds_ebanking.{table['file_id'].replace(".", "_")} CASCADE;"""}
# for
# table in json.loads(iris[0][0][0])]
def ds_get_ddl(ti):
global DS_CREATE_TABLE
iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
column = []
for table in json.loads(iris[0][0][0]):
column.append(f"""'{table['file_id']}'""")
print(create_ddl(DS_SCHEMA, ','.join(column)))
sql_stmt = create_ddl(DS_SCHEMA, ','.join(column))
pg_hook = PostgresHook(
postgres_conn_id=DS_CONN_ID,
)
pg_conn = pg_hook.get_conn()
cursor = pg_conn.cursor()
cursor.execute(sql_stmt)
return cursor.fetchall()[0][0]
def pg_ddl_syntax(ti):
print(ti.xcom_pull(task_ids=['ds_get_ddl'])[0])
return [{"sql": ti.xcom_pull(task_ids=['ds_get_ddl'])[0]}]
def pg_push_syntax(ti):
iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
if not iris:
raise Exception('No data.')
return [{"op_kwargs": {
"copy_sql": f"""COPY {POSTGRES_SCHEMA}.{table['file_id'].replace(".", "_")} from STDOUT delimiter '|' CSV header""",
"file_id": table['file_id']}}
for
table in json.loads(iris[0][0][0])]
def pg_push_csv(ti, copy_sql, file_id):
pg_hook = PostgresHook.get_hook(POSTGRES_CONN_ID)
pg_hook.copy_expert(copy_sql, filename=f"""{Variable.get("LOCAL_PATH")}{DS_FOLDER}/""" + (
f"""{yesterday_nodash}/{yesterday_nodash}.""" if Variable.get(
"WITH_DATE") == 'Y' else '') + f"""{file_id}.csv""")
with DAG("APJ_1_los_fintech",
start_date=datetime(2021, 1, 1),
schedule_interval=None,
catchup=False,
concurrency=3) as dag:
begin = PythonOperator(
task_id=f"Begin",
python_callable=_start,
op_kwargs={
}
)
create_folder = BashOperator(
task_id='create_folder',
bash_command=f"""mkdir -p {Variable.get("LOCAL_PATH")}{DS_FOLDER}/{yesterday_nodash}"""
)
ds_list_extractor = PythonOperator(
task_id='ds_list_extractor',
python_callable=ds_list_extractor,
do_xcom_push=True
)
ds_get_syntax = PythonOperator(
task_id='ds_syntax_get',
python_callable=ds_get_syntax
)
ds_table_to_csv = PythonOperator.partial(
task_id="ds_table_to_csv",
python_callable=ds_get_csv,
dag=dag
).expand_kwargs(
XComArg(ds_get_syntax),
)
ds_get_ddl = PythonOperator(
task_id='ds_get_ddl',
python_callable=ds_get_ddl,
do_xcom_push=True
)
# ds_drop_syntax = PythonOperator(
# task_id='ds_drop_syntax',
# python_callable=ds_drop_syntax
# )
pg_drop_schema = PostgresOperator(
sql= f"""drop schema IF EXISTS {POSTGRES_SCHEMA} CASCADE;""",
task_id="pg_drop_schema",
postgres_conn_id=POSTGRES_CONN_ID,
)
pg_create_schema = PostgresOperator(
sql= f"""create schema IF NOT EXISTS {POSTGRES_SCHEMA};""",
task_id="pg_create_schema",
postgres_conn_id=POSTGRES_CONN_ID,
)
pg_ddl_syntax = PythonOperator(
task_id='pg_ddl_syntax',
python_callable=pg_ddl_syntax
)
pg_create_table = PostgresOperator.partial(
task_id="pg_create_table",
postgres_conn_id=POSTGRES_CONN_ID,
).expand_kwargs(
XComArg(pg_ddl_syntax),
)
pg_push_syntax = PythonOperator(
task_id='pg_syntax_push',
python_callable=pg_push_syntax
)
pg_csv_to_table = PythonOperator.partial(
task_id="pg_csv_to_table",
python_callable=pg_push_csv,
dag=dag
).expand_kwargs(
XComArg(pg_push_syntax),
)
set_access_schemma = PostgresOperator(
sql=f"""GRANT USAGE ON SCHEMA {POSTGRES_SCHEMA} TO readaccess;""",
task_id="set_access_schemma",
postgres_conn_id=POSTGRES_CONN_ID,
)
set_access_all_table = PostgresOperator(
sql=f"""GRANT SELECT ON ALL TABLES IN SCHEMA {POSTGRES_SCHEMA} TO readaccess;""",
task_id="set_access_all_table",
postgres_conn_id=POSTGRES_CONN_ID,
)
begin >> create_folder >> ds_list_extractor >> ds_get_syntax >> ds_table_to_csv >> ds_get_ddl >> pg_drop_schema >> pg_create_schema >> pg_ddl_syntax >> pg_create_table >> pg_push_syntax >> pg_csv_to_table >> set_access_schemma >> set_access_all_table
\ No newline at end of file
import json
from datetime import date
import calendar
from airflow import DAG
from airflow.operators.python import PythonOperator, BranchPythonOperator
from airflow.operators.bash import BashOperator
from datetime import datetime, timedelta
from airflow.providers.sftp.operators.sftp import SFTPOperator
from airflow.models import Variable
from airflow.providers.postgres.operators.postgres import PostgresOperator
from airflow.operators.trigger_dagrun import TriggerDagRunOperator
from airflow.utils.trigger_rule import TriggerRule
from airflow.models.xcom import XCom
from airflow.decorators import task
from airflow.hooks.postgres_hook import PostgresHook
from airflow import XComArg
yesterday_nodash = (datetime.now() - timedelta(1)).strftime('%Y%m%d') if Variable.get(
"DATE_OF_DATA") == 'today' else Variable.get("DATE_OF_DATA")
yesterday_strip = datetime.strptime(yesterday_nodash, '%Y%m%d').strftime('%Y-%m-%d')
POSTGRES_CONN_ID = Variable.get("DS_DB")
POSTGRES_SCHEMA = 'ds_los_kpt' + Variable.get("ENV_T24")
DS_CONN_ID = Variable.get("DS_LOS_KPT")
DS_FOLDER = 'los'
DS_DB = 'los-kpt'
DS_SCHEMA = 'public'
DS_CREATE_TABLE = ''
def create_ddl(schema,column):
return f"""select string_agg(x,'') from (select
'create table {POSTGRES_SCHEMA}.'||table_name || ' ( ' || string_agg(column_name, ' text,') || ' text'|| ' ); ' as x
from information_schema.columns where table_schema = '{schema}'
and table_name in ({column})
group by table_schema, table_name) as x ;"""
def _start():
print("Start :: Extractor ")
def ds_list_extractor():
sql_stmt = f"""select * from ds_conf.ds_extractor_list_extractor('los-kpt', '{yesterday_strip}');"""
pg_hook = PostgresHook(
postgres_conn_id=POSTGRES_CONN_ID,
)
pg_conn = pg_hook.get_conn()
cursor = pg_conn.cursor()
cursor.execute(sql_stmt)
files = cursor.fetchall()
return files
def ds_get_syntax(ti):
iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
if not iris:
raise Exception('No data.')
return [{"op_kwargs": {
"copy_sql": f"""COPY {DS_SCHEMA}.{table['file_id'].replace(".", "_")} to STDOUT delimiter '|' CSV header""",
"file_id": table['file_id']}}
for
table in json.loads(iris[0][0][0])]
def ds_get_csv(ti, copy_sql, file_id):
pg_hook = PostgresHook.get_hook(DS_CONN_ID)
pg_hook.copy_expert(copy_sql, filename=f"""{Variable.get("LOCAL_PATH")}{DS_FOLDER}/""" + (
f"""{yesterday_nodash}/{yesterday_nodash}.""" if Variable.get(
"WITH_DATE") == 'Y' else '') + f"""{file_id}.csv""")
#
# def ds_drop_syntax(ti):
# iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
# if not iris:
# raise Exception('No data.')
# return [{
# "sql": f"""drop table ds_ebanking.{table['file_id'].replace(".", "_")} CASCADE;"""}
# for
# table in json.loads(iris[0][0][0])]
def ds_get_ddl(ti):
global DS_CREATE_TABLE
iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
column = []
for table in json.loads(iris[0][0][0]):
column.append(f"""'{table['file_id']}'""")
print(create_ddl(DS_SCHEMA, ','.join(column)))
sql_stmt = create_ddl(DS_SCHEMA, ','.join(column))
pg_hook = PostgresHook(
postgres_conn_id=DS_CONN_ID,
)
pg_conn = pg_hook.get_conn()
cursor = pg_conn.cursor()
cursor.execute(sql_stmt)
return cursor.fetchall()[0][0]
def pg_ddl_syntax(ti):
print(ti.xcom_pull(task_ids=['ds_get_ddl'])[0])
return [{"sql": ti.xcom_pull(task_ids=['ds_get_ddl'])[0]}]
def pg_push_syntax(ti):
iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
if not iris:
raise Exception('No data.')
return [{"op_kwargs": {
"copy_sql": f"""COPY {POSTGRES_SCHEMA}.{table['file_id'].replace(".", "_")} from STDOUT delimiter '|' CSV header""",
"file_id": table['file_id']}}
for
table in json.loads(iris[0][0][0])]
def pg_push_csv(ti, copy_sql, file_id):
pg_hook = PostgresHook.get_hook(POSTGRES_CONN_ID)
pg_hook.copy_expert(copy_sql, filename=f"""{Variable.get("LOCAL_PATH")}{DS_FOLDER}/""" + (
f"""{yesterday_nodash}/{yesterday_nodash}.""" if Variable.get(
"WITH_DATE") == 'Y' else '') + f"""{file_id}.csv""")
with DAG("APJ_1_los_kpt",
start_date=datetime(2021, 1, 1),
schedule_interval=None,
catchup=False,
concurrency=3) as dag:
begin = PythonOperator(
task_id=f"Begin",
python_callable=_start,
op_kwargs={
}
)
create_folder = BashOperator(
task_id='create_folder',
bash_command=f"""mkdir -p {Variable.get("LOCAL_PATH")}{DS_FOLDER}/{yesterday_nodash}"""
)
ds_list_extractor = PythonOperator(
task_id='ds_list_extractor',
python_callable=ds_list_extractor,
do_xcom_push=True
)
ds_get_syntax = PythonOperator(
task_id='ds_syntax_get',
python_callable=ds_get_syntax
)
ds_table_to_csv = PythonOperator.partial(
task_id="ds_table_to_csv",
python_callable=ds_get_csv,
dag=dag
).expand_kwargs(
XComArg(ds_get_syntax),
)
ds_get_ddl = PythonOperator(
task_id='ds_get_ddl',
python_callable=ds_get_ddl,
do_xcom_push=True
)
# ds_drop_syntax = PythonOperator(
# task_id='ds_drop_syntax',
# python_callable=ds_drop_syntax
# )
pg_drop_schema = PostgresOperator(
sql= f"""drop schema IF EXISTS {POSTGRES_SCHEMA} CASCADE;""",
task_id="pg_drop_schema",
postgres_conn_id=POSTGRES_CONN_ID,
)
pg_create_schema = PostgresOperator(
sql= f"""create schema IF NOT EXISTS {POSTGRES_SCHEMA};""",
task_id="pg_create_schema",
postgres_conn_id=POSTGRES_CONN_ID,
)
pg_ddl_syntax = PythonOperator(
task_id='pg_ddl_syntax',
python_callable=pg_ddl_syntax
)
pg_create_table = PostgresOperator.partial(
task_id="pg_create_table",
postgres_conn_id=POSTGRES_CONN_ID,
).expand_kwargs(
XComArg(pg_ddl_syntax),
)
pg_push_syntax = PythonOperator(
task_id='pg_syntax_push',
python_callable=pg_push_syntax
)
pg_csv_to_table = PythonOperator.partial(
task_id="pg_csv_to_table",
python_callable=pg_push_csv,
dag=dag
).expand_kwargs(
XComArg(pg_push_syntax),
)
set_access_schemma = PostgresOperator(
sql=f"""GRANT USAGE ON SCHEMA {POSTGRES_SCHEMA} TO readaccess;""",
task_id="set_access_schemma",
postgres_conn_id=POSTGRES_CONN_ID,
)
set_access_all_table = PostgresOperator(
sql=f"""GRANT SELECT ON ALL TABLES IN SCHEMA {POSTGRES_SCHEMA} TO readaccess;""",
task_id="set_access_all_table",
postgres_conn_id=POSTGRES_CONN_ID,
)
begin >> create_folder >> ds_list_extractor >> ds_get_syntax >> ds_table_to_csv >> ds_get_ddl >> pg_drop_schema >> pg_create_schema >> pg_ddl_syntax >> pg_create_table >> pg_push_syntax >> pg_csv_to_table >> set_access_schemma >> set_access_all_table
\ No newline at end of file
import json, os
from datetime import date
import calendar
from airflow import DAG
from airflow.operators.python import PythonOperator, BranchPythonOperator, ShortCircuitOperator
from airflow.operators.bash import BashOperator
from datetime import datetime, timedelta
from airflow.providers.sftp.operators.sftp import SFTPOperator
from airflow.models import Variable
from airflow.providers.postgres.operators.postgres import PostgresOperator
from airflow.operators.trigger_dagrun import TriggerDagRunOperator
from airflow.utils.trigger_rule import TriggerRule
from airflow.models.xcom import XCom
from airflow.decorators import task
from airflow.hooks.postgres_hook import PostgresHook
from airflow import XComArg
yesterday_nodash = (datetime.now() - timedelta(1)).strftime('%Y%m%d') if Variable.get(
"DATE_OF_DATA") == 'today' else Variable.get("DATE_OF_DATA")
yesterday_strip = datetime.strptime(yesterday_nodash, '%Y%m%d').strftime('%Y-%m-%d')
d = f"""{datetime.now().strftime('%Y-%m-%d %H:%M:%S')}"""
today = d[:]
POSTGRES_CONN_ID = Variable.get("DS_DB")
POSTGRES_SCHEMA = 'ds_mgate' + Variable.get("ENV_T24")
DS_CONN_ID = Variable.get("DS_MGATE")
DS_FOLDER = 'mgate'
DS_DB = 'mgate'
DS_SCHEMA = 'mgate'
DS_CREATE_TABLE = ''
def create_ddl(schema,column):
return f"""select string_agg(x,'') from (select
'create table {POSTGRES_SCHEMA}.'||table_name || ' ( ' || string_agg(column_name, ' text,' order by ordinal_position) || ' text'|| ' ); ' as x
from information_schema.columns where table_schema = '{schema}'
and table_name in ({column})
group by table_schema, table_name) as x ;"""
def _start():
print("Start :: Extractor ")
def check_done_ext(**kwargs):
from random import randint
# number = randint(0, 10)
if os.path.isfile(f'''{Variable.get("LOCAL_PATH")}{DS_FOLDER}/{yesterday_nodash}/done_ext.csv'''):
print("A")
return False
else:
print("B")
# STOP DAG
return True
def ds_list_extractor():
sql_stmt = f"""select * from ds_conf.ds_extractor_list_extractor('mgate', '{yesterday_strip}');"""
pg_hook = PostgresHook(
postgres_conn_id=POSTGRES_CONN_ID,
)
pg_conn = pg_hook.get_conn()
cursor = pg_conn.cursor()
cursor.execute(sql_stmt)
files = cursor.fetchall()
return files
def ds_get_syntax(ti):
iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
if not iris:
raise Exception('No data.')
return [{"op_kwargs": {
"copy_sql": f"""COPY ( select * from {DS_SCHEMA}.{table['file_id'].replace(".", "_")}{table['partition_table_format']} {table['condition']}) to STDOUT delimiter '{table['delimiter']}' CSV header""",
"file_id": table['file_id']}}
for
table in json.loads(iris[0][0][0])]
def ds_get_csv(ti, copy_sql, file_id):
pg_hook = PostgresHook.get_hook(DS_CONN_ID)
pg_hook.copy_expert(copy_sql, filename=f"""{Variable.get("LOCAL_PATH")}{DS_FOLDER}/""" + (
f"""{yesterday_nodash}/{yesterday_nodash}.""" if Variable.get(
"WITH_DATE") == 'Y' else '') + f"""{file_id}.csv""")
#
# def ds_drop_syntax(ti):
# iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
# if not iris:
# raise Exception('No data.')
# return [{
# "sql": f"""drop table ds_ebanking.{table['file_id'].replace(".", "_")} CASCADE;"""}
# for
# table in json.loads(iris[0][0][0])]
def ds_get_ddl(ti):
global DS_CREATE_TABLE
iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
column = []
for table in json.loads(iris[0][0][0]):
column.append(f"""'{table['file_id']}'""")
print(create_ddl(DS_SCHEMA, ','.join(column)))
sql_stmt = create_ddl(DS_SCHEMA, ','.join(column))
pg_hook = PostgresHook(
postgres_conn_id=DS_CONN_ID,
)
pg_conn = pg_hook.get_conn()
cursor = pg_conn.cursor()
cursor.execute(sql_stmt)
return cursor.fetchall()[0][0]
def pg_ddl_syntax(ti):
print(ti.xcom_pull(task_ids=['ds_get_ddl'])[0])
return [{"sql": ti.xcom_pull(task_ids=['ds_get_ddl'])[0]}]
def pg_push_syntax(ti):
iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
if not iris:
raise Exception('No data.')
return [{"op_kwargs": {
"copy_sql": f"""COPY {POSTGRES_SCHEMA}.{table['file_id'].replace(".", "_")} from STDOUT delimiter '|' CSV header""",
"file_id": table['file_id']}}
for
table in json.loads(iris[0][0][0])]
def pg_push_csv(ti, copy_sql, file_id):
pg_hook = PostgresHook.get_hook(POSTGRES_CONN_ID)
pg_hook.copy_expert(copy_sql, filename=f"""{Variable.get("LOCAL_PATH")}{DS_FOLDER}/""" + (
f"""{yesterday_nodash}/{yesterday_nodash}.""" if Variable.get(
"WITH_DATE") == 'Y' else '') + f"""{file_id}.csv""")
with DAG("APJ_1_mgate",
start_date=datetime(2021, 1, 1),
schedule_interval='10 0 * * *',
catchup=False,
concurrency=3) as dag:
begin = PythonOperator(
task_id=f"Begin",
python_callable=_start,
op_kwargs={
}
)
check_done_ext = ShortCircuitOperator(
task_id="check_done_ext",
provide_context=True,
python_callable=check_done_ext,
op_kwargs={},
)
history_start = PostgresOperator(
sql=f"""INSERT INTO ds_conf."Datasource_history"
(source, start_time, finish_time, status, env, date_of_data) VALUES
('{POSTGRES_SCHEMA}'::varchar(255), '{today}'::timestamp with time zone, '{datetime.now().strftime('%Y-%m-%d %H:%M:%S')}'::timestamp with time zone, 'ONPROCESS'::varchar(100), '', '{yesterday_nodash}');""",
task_id="history_start",
postgres_conn_id=POSTGRES_CONN_ID,
)
create_folder = BashOperator(
task_id='create_folder',
bash_command=f"""mkdir -p {Variable.get("LOCAL_PATH")}{DS_FOLDER}/{yesterday_nodash}"""
)
ds_list_extractor = PythonOperator(
task_id='ds_list_extractor',
python_callable=ds_list_extractor,
do_xcom_push=True
)
ds_get_syntax = PythonOperator(
task_id='ds_syntax_get',
python_callable=ds_get_syntax
)
ds_table_to_csv = PythonOperator.partial(
task_id="ds_table_to_csv",
python_callable=ds_get_csv,
dag=dag
).expand_kwargs(
XComArg(ds_get_syntax),
)
ds_get_ddl = PythonOperator(
task_id='ds_get_ddl',
python_callable=ds_get_ddl,
do_xcom_push=True
)
# ds_drop_syntax = PythonOperator(
# task_id='ds_drop_syntax',
# python_callable=ds_drop_syntax
# )
pg_drop_schema = PostgresOperator(
sql= f"""drop schema IF EXISTS {POSTGRES_SCHEMA} CASCADE;""",
task_id="pg_drop_schema",
postgres_conn_id=POSTGRES_CONN_ID,
)
pg_create_schema = PostgresOperator(
sql= f"""create schema IF NOT EXISTS {POSTGRES_SCHEMA};""",
task_id="pg_create_schema",
postgres_conn_id=POSTGRES_CONN_ID,
)
pg_ddl_syntax = PythonOperator(
task_id='pg_ddl_syntax',
python_callable=pg_ddl_syntax
)
pg_create_table = PostgresOperator.partial(
task_id="pg_create_table",
postgres_conn_id=POSTGRES_CONN_ID,
).expand_kwargs(
XComArg(pg_ddl_syntax),
)
pg_push_syntax = PythonOperator(
task_id='pg_syntax_push',
python_callable=pg_push_syntax
)
pg_csv_to_table = PythonOperator.partial(
task_id="pg_csv_to_table",
python_callable=pg_push_csv,
dag=dag
).expand_kwargs(
XComArg(pg_push_syntax),
)
set_access_schemma = PostgresOperator(
sql=f"""GRANT USAGE ON SCHEMA {POSTGRES_SCHEMA} TO readaccess;""",
task_id="set_access_schemma",
postgres_conn_id=POSTGRES_CONN_ID,
)
set_access_all_table = PostgresOperator(
sql=f"""GRANT SELECT ON ALL TABLES IN SCHEMA {POSTGRES_SCHEMA} TO readaccess;""",
task_id="set_access_all_table",
postgres_conn_id=POSTGRES_CONN_ID,
)
ds_ext_done = BashOperator(
task_id="ds_ext_done",
bash_command=f"""
touch /opt/airflow/dags/DFE/{DS_FOLDER}/{yesterday_nodash}/done_ext.csv;
""",
)
history_finish = PostgresOperator(
sql=f"""
UPDATE ds_conf."Datasource_history"
SET status = 'DONE', finish_time = '{datetime.now().strftime('%Y-%m-%d %H:%M:%S')}'::timestamp with time zone
WHERE source = '{POSTGRES_SCHEMA}' and status = 'ONPROCESS';
""",
task_id="history_finish",
postgres_conn_id=POSTGRES_CONN_ID,
)
begin >> check_done_ext >> history_start >> create_folder >> ds_list_extractor >> ds_get_syntax >> ds_table_to_csv >> ds_get_ddl >> pg_drop_schema >> pg_create_schema >> pg_ddl_syntax >> pg_create_table >> pg_push_syntax >> pg_csv_to_table >> set_access_schemma >> set_access_all_table >> ds_ext_done >> history_finish
\ No newline at end of file
import json, os
from datetime import date
import calendar
from airflow import DAG
from airflow.operators.python import PythonOperator, BranchPythonOperator, ShortCircuitOperator
from airflow.operators.bash import BashOperator
from datetime import datetime, timedelta
from airflow.providers.sftp.operators.sftp import SFTPOperator
from airflow.models import Variable
from airflow.providers.postgres.operators.postgres import PostgresOperator
from airflow.operators.trigger_dagrun import TriggerDagRunOperator
from airflow.utils.trigger_rule import TriggerRule
from airflow.models.xcom import XCom
from airflow.decorators import task
from airflow.hooks.postgres_hook import PostgresHook
from airflow import XComArg
yesterday_nodash = (datetime.now() - timedelta(1)).strftime('%Y%m%d') if Variable.get(
"DATE_OF_DATA") == 'today' else Variable.get("DATE_OF_DATA")
yesterday_strip = datetime.strptime(yesterday_nodash, '%Y%m%d').strftime('%Y-%m-%d')
d = f"""{datetime.now().strftime('%Y-%m-%d %H:%M:%S')}"""
today = d[:]
POSTGRES_CONN_ID = Variable.get("DS_DB")
POSTGRES_SCHEMA = 'ds_switching' + Variable.get("ENV_T24")
DS_CONN_ID = Variable.get("DS_SWITCHING")
DS_FOLDER = 'switching'
DS_DB = 'switching'
DS_SCHEMA = 'public'
DS_CREATE_TABLE = ''
def create_ddl(schema,column):
return f"""select string_agg(x,'') from (select
'create table {POSTGRES_SCHEMA}.'||table_name || ' ( ' || string_agg(column_name, ' text,' order by ordinal_position) || ' text'|| ' ); ' as x
from information_schema.columns where table_schema = '{schema}'
and table_name in ({column})
group by table_schema, table_name) as x ;"""
def _start():
print("Start :: Extractor ")
def check_done_ext(**kwargs):
from random import randint
# number = randint(0, 10)
if os.path.isfile(f'''{Variable.get("LOCAL_PATH")}{DS_FOLDER}/{yesterday_nodash}/done_ext.csv'''):
print("A")
return False
else:
print("B")
# STOP DAG
return True
def ds_list_extractor():
sql_stmt = f"""select * from ds_conf.ds_extractor_list_extractor('switching', '{yesterday_strip}');"""
pg_hook = PostgresHook(
postgres_conn_id=POSTGRES_CONN_ID,
)
pg_conn = pg_hook.get_conn()
cursor = pg_conn.cursor()
cursor.execute(sql_stmt)
files = cursor.fetchall()
return files
def ds_get_syntax(ti):
iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
if not iris:
raise Exception('No data.')
return [{"op_kwargs": {
"copy_sql": f"""COPY ( select * from {DS_SCHEMA}.{table['file_id'].replace(".", "_")}{table['partition_table_format']} {table['condition']}) to STDOUT delimiter '{table['delimiter']}' CSV header""",
"file_id": table['file_id']}}
for
table in json.loads(iris[0][0][0])]
def ds_get_csv(ti, copy_sql, file_id):
pg_hook = PostgresHook.get_hook(DS_CONN_ID)
pg_hook.copy_expert(copy_sql, filename=f"""{Variable.get("LOCAL_PATH")}{DS_FOLDER}/""" + (
f"""{yesterday_nodash}/{yesterday_nodash}.""" if Variable.get(
"WITH_DATE") == 'Y' else '') + f"""{file_id}.csv""")
#
# def ds_drop_syntax(ti):
# iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
# if not iris:
# raise Exception('No data.')
# return [{
# "sql": f"""drop table ds_ebanking.{table['file_id'].replace(".", "_")} CASCADE;"""}
# for
# table in json.loads(iris[0][0][0])]
def ds_get_ddl(ti):
global DS_CREATE_TABLE
iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
column = []
for table in json.loads(iris[0][0][0]):
column.append(f"""'{table['file_id']}'""")
print(create_ddl(DS_SCHEMA, ','.join(column)))
sql_stmt = create_ddl(DS_SCHEMA, ','.join(column))
pg_hook = PostgresHook(
postgres_conn_id=DS_CONN_ID,
)
pg_conn = pg_hook.get_conn()
cursor = pg_conn.cursor()
cursor.execute(sql_stmt)
return cursor.fetchall()[0][0]
def pg_ddl_syntax(ti):
print(ti.xcom_pull(task_ids=['ds_get_ddl'])[0])
return [{"sql": ti.xcom_pull(task_ids=['ds_get_ddl'])[0]}]
def pg_push_syntax(ti):
iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
if not iris:
raise Exception('No data.')
return [{"op_kwargs": {
"copy_sql": f"""COPY {POSTGRES_SCHEMA}.{table['file_id'].replace(".", "_")} from STDOUT delimiter '|' CSV header""",
"file_id": table['file_id']}}
for
table in json.loads(iris[0][0][0])]
def pg_push_csv(ti, copy_sql, file_id):
pg_hook = PostgresHook.get_hook(POSTGRES_CONN_ID)
pg_hook.copy_expert(copy_sql, filename=f"""{Variable.get("LOCAL_PATH")}{DS_FOLDER}/""" + (
f"""{yesterday_nodash}/{yesterday_nodash}.""" if Variable.get(
"WITH_DATE") == 'Y' else '') + f"""{file_id}.csv""")
with DAG("APJ_1_SWITCHING_DB",
start_date=datetime(2021, 1, 1),
schedule_interval='10 0 * * *',
catchup=False,
concurrency=3) as dag:
begin = PythonOperator(
task_id=f"Begin",
python_callable=_start,
op_kwargs={
}
)
check_done_ext = ShortCircuitOperator(
task_id="check_done_ext",
provide_context=True,
python_callable=check_done_ext,
op_kwargs={},
)
history_start = PostgresOperator(
sql=f"""INSERT INTO ds_conf."Datasource_history"
(source, start_time, finish_time, status, env, date_of_data) VALUES
('{POSTGRES_SCHEMA}'::varchar(255), '{today}'::timestamp with time zone, '{datetime.now().strftime('%Y-%m-%d %H:%M:%S')}'::timestamp with time zone, 'ONPROCESS'::varchar(100), '', '{yesterday_nodash}');""",
task_id="history_start",
postgres_conn_id=POSTGRES_CONN_ID,
)
create_folder = BashOperator(
task_id='create_folder',
bash_command=f"""mkdir -p {Variable.get("LOCAL_PATH")}{DS_FOLDER}/{yesterday_nodash}"""
)
ds_list_extractor = PythonOperator(
task_id='ds_list_extractor',
python_callable=ds_list_extractor,
do_xcom_push=True
)
ds_get_syntax = PythonOperator(
task_id='ds_syntax_get',
python_callable=ds_get_syntax
)
ds_table_to_csv = PythonOperator.partial(
task_id="ds_table_to_csv",
python_callable=ds_get_csv,
dag=dag
).expand_kwargs(
XComArg(ds_get_syntax),
)
ds_get_ddl = PythonOperator(
task_id='ds_get_ddl',
python_callable=ds_get_ddl,
do_xcom_push=True
)
# ds_drop_syntax = PythonOperator(
# task_id='ds_drop_syntax',
# python_callable=ds_drop_syntax
# )
pg_drop_schema = PostgresOperator(
sql= f"""drop schema IF EXISTS {POSTGRES_SCHEMA} CASCADE;""",
task_id="pg_drop_schema",
postgres_conn_id=POSTGRES_CONN_ID,
)
pg_create_schema = PostgresOperator(
sql= f"""create schema IF NOT EXISTS {POSTGRES_SCHEMA};""",
task_id="pg_create_schema",
postgres_conn_id=POSTGRES_CONN_ID,
)
pg_ddl_syntax = PythonOperator(
task_id='pg_ddl_syntax',
python_callable=pg_ddl_syntax
)
pg_create_table = PostgresOperator.partial(
task_id="pg_create_table",
postgres_conn_id=POSTGRES_CONN_ID,
).expand_kwargs(
XComArg(pg_ddl_syntax),
)
pg_push_syntax = PythonOperator(
task_id='pg_syntax_push',
python_callable=pg_push_syntax
)
pg_csv_to_table = PythonOperator.partial(
task_id="pg_csv_to_table",
python_callable=pg_push_csv,
dag=dag
).expand_kwargs(
XComArg(pg_push_syntax),
)
set_access_schemma = PostgresOperator(
sql=f"""GRANT USAGE ON SCHEMA {POSTGRES_SCHEMA} TO readaccess;""",
task_id="set_access_schemma",
postgres_conn_id=POSTGRES_CONN_ID,
)
set_access_all_table = PostgresOperator(
sql=f"""GRANT SELECT ON ALL TABLES IN SCHEMA {POSTGRES_SCHEMA} TO readaccess;""",
task_id="set_access_all_table",
postgres_conn_id=POSTGRES_CONN_ID,
)
ds_ext_done = BashOperator(
task_id="ds_ext_done",
bash_command=f"""
touch /opt/airflow/dags/DFE/{DS_FOLDER}/{yesterday_nodash}/done_ext.csv;
""",
)
history_finish = PostgresOperator(
sql=f"""
UPDATE ds_conf."Datasource_history"
SET status = 'DONE', finish_time = '{datetime.now().strftime('%Y-%m-%d %H:%M:%S')}'::timestamp with time zone
WHERE source = '{POSTGRES_SCHEMA}' and status = 'ONPROCESS';
""",
task_id="history_finish",
postgres_conn_id=POSTGRES_CONN_ID,
)
begin >> check_done_ext >> history_start >> create_folder >> ds_list_extractor >> ds_get_syntax >> ds_table_to_csv >> ds_get_ddl >> pg_drop_schema >> pg_create_schema >> pg_ddl_syntax >> pg_create_table >> pg_push_syntax >> pg_csv_to_table >> set_access_schemma >> set_access_all_table >> ds_ext_done >> history_finish
\ No newline at end of file
import json, os
from datetime import date
import calendar
from airflow import DAG
from airflow.operators.python import PythonOperator, BranchPythonOperator, ShortCircuitOperator
from airflow.operators.bash import BashOperator
from datetime import datetime, timedelta
from airflow.providers.sftp.operators.sftp import SFTPOperator
from airflow.models import Variable
from airflow.providers.postgres.operators.postgres import PostgresOperator
from airflow.operators.trigger_dagrun import TriggerDagRunOperator
from airflow.utils.trigger_rule import TriggerRule
from airflow.models.xcom import XCom
from airflow.decorators import task
from airflow.providers.postgres.hooks.postgres import PostgresHook
from airflow import XComArg
# yesterday_nodash = (datetime.now() - timedelta(1)).strftime('%Y%m%d')
yesterday_nodash = (datetime.now() - timedelta(1)).strftime('%Y%m%d') if Variable.get(
"DATE_OF_DATA") == 'today' else Variable.get("DATE_OF_DATA")
yesterday_strip = datetime.strptime(yesterday_nodash, '%Y%m%d').strftime('%Y-%m-%d')
d = f"""{datetime.now().strftime('%Y-%m-%d %H:%M:%S')}"""
today = d[:]
POSTGRES_CONN_ID = Variable.get("DS_DB")
POSTGRES_ENV = Variable.get("ENV_T24")
POSTGRES_SCHEMA = 'ds_t24' + Variable.get("ENV_T24")
DS_FOLDER = 't24'
def _start():
print(f"""Start :: Extractor :: {yesterday_nodash}""")
print(f"""Start :: Extractor :: {yesterday_strip}""")
def check_file_done(**kwargs):
from random import randint
# number = randint(0, 10)
print(f'''{Variable.get("LOCAL_PATH")}t24/{yesterday_nodash}/DONE.2.csv''')
if os.path.isfile(f'''{Variable.get("LOCAL_PATH")}t24/{yesterday_nodash}/DONE.2.csv'''):
print("A")
return True
else:
print("B")
# STOP DAG
return False
def check_done_ext(**kwargs):
from random import randint
# number = randint(0, 10)
if os.path.isfile(f'''{Variable.get("LOCAL_PATH")}t24/{yesterday_nodash}/done_ext.csv'''):
print("A")
return False
else:
print("B")
# STOP DAG
return True
# ##############
# METADATA
# ##############
def metadata_list_extractor():
sql_stmt = f"""select * from ds_conf.ds_t24_meta_file_list();"""
pg_hook = PostgresHook(
postgres_conn_id=POSTGRES_CONN_ID
)
pg_conn = pg_hook.get_conn()
cursor = pg_conn.cursor()
cursor.execute(sql_stmt)
files = cursor.fetchall()
return files
def metadata_syntax_copy(ti):
iris = ti.xcom_pull(task_ids=['metadata_list_extractor'])
if not iris:
raise Exception('No data.')
return [{"op_kwargs": {
"copy_sql": f"""COPY ds_conf.\"Datasource_{table['file_id'].replace(".", "").replace("ST","").lower()}temp\" FROM STDOUT delimiter '{table['field_delimiter']}' CSV QUOTE E'\b'""",
"file_id": table['file_id']}}
for
table in json.loads(iris[0][0][0])]
def meta_extract(ti, copy_sql, file_id):
pg_hook = PostgresHook.get_hook(POSTGRES_CONN_ID)
pg_hook.copy_expert(copy_sql, filename=f"""{Variable.get("LOCAL_PATH")}t24_bnk/""" + (
f"""{yesterday_nodash}/{yesterday_nodash}.""" if Variable.get(
"WITH_DATE") == 'Y' else '') + f"""{file_id}.csv""")
# ##############
# DATASOURCE
# ##############
def ds_list_extractor(**context):
sql_stmt = f"""select * from ds_conf.ds_t24_list_extractor();"""
pg_hook = PostgresHook(
postgres_conn_id=POSTGRES_CONN_ID
)
pg_conn = pg_hook.get_conn()
cursor = pg_conn.cursor()
cursor.execute(sql_stmt)
files = cursor.fetchall()
return files
def ds_syntax_copy(ti):
iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
if not iris:
raise Exception('No data.')
return [{"op_kwargs": {
"copy_sql": f"""COPY {POSTGRES_SCHEMA}.{table['file_id'].replace(".", "_")} FROM STDOUT delimiter '{table['field_delimiter']}' CSV QUOTE E'\b'""",
"file_id": table['file_id']}}
for
table in json.loads(iris[0][0][0])]
def ds_csv_to_table(ti, copy_sql, file_id, **context):
print(file_id)
pg_hook = PostgresHook.get_hook(POSTGRES_CONN_ID)
pg_hook.copy_expert(copy_sql, filename=f"""{Variable.get("LOCAL_PATH")}t24_bnk/""" + (
f"""{yesterday_nodash}/{yesterday_nodash}.""" if Variable.get(
"WITH_DATE") == 'Y' else '') + f"""{file_id}.csv""")
with DAG("APJ_1_T24",
start_date=datetime(2021, 1, 1),
schedule_interval='0/30 * * * *',
# schedule_interval=None,
catchup=False,
concurrency=10) as dag:
# ###########################################
# START
# ###########################################
begin = PythonOperator(
task_id=f"Begin",
python_callable=_start,
op_kwargs={
}
)
check_done_ext = ShortCircuitOperator(
task_id="check_done_ext",
provide_context=True,
python_callable=check_done_ext,
op_kwargs={},
)
sftp_xx = BashOperator(
task_id="sftp_xx",
bash_command=f"""sshpass -p {Variable.get("SFTP_T24_PASSWORD")} sftp -o StrictHostKeyChecking=no -r {Variable.get("SFTP_T24_USER")}@{Variable.get("SFTP_T24_HOST")}:DFE/{yesterday_nodash}/ {Variable.get("LOCAL_PATH")}{DS_FOLDER}/""",
)
# ds_get_t24_extractor = SFTPOperator(
# task_id="ds_get_t24_extractor",
# ssh_conn_id="ssh_default",
# local_filepath=f"""{Variable.get("LOCAL_PATH")}t24/{yesterday_nodash}/""",
# remote_filepath=f"""DFE/{yesterday_nodash}/""",
# operation="get",
# create_intermediate_dirs=True,
# dag=dag
# )
check_file_done = ShortCircuitOperator(
task_id="check_file_done",
provide_context=True,
python_callable=check_file_done,
op_kwargs={},
)
# ###########################################
# METADATA
# ###########################################
history_start = PostgresOperator(
sql=f"""INSERT INTO ds_conf."Datasource_history"
(source, start_time, finish_time, status, env, date_of_data) VALUES
('{POSTGRES_SCHEMA}'::varchar(255), '{today}'::timestamp with time zone, '{datetime.now().strftime('%Y-%m-%d %H:%M:%S')}'::timestamp with time zone, 'ONPROCESS'::varchar(100), '', '{yesterday_nodash}');""",
task_id="history_start",
postgres_conn_id=POSTGRES_CONN_ID,
)
# Copt Source dari file EXTRACTOR
file_sc_cp = BashOperator(
task_id="file_sc_cp",
bash_command=f"""
cp -r /opt/airflow/dags/DFE/t24/{yesterday_nodash} /opt/airflow/dags/DFE/t24_bnk/{yesterday_nodash};
""",
)
# Menghapus hapus temporary METADATA
metadata_truncate = PostgresOperator(
sql=f"""select ds_conf.ds_t24_truncate_metadata();""",
task_id="metadata_truncate",
postgres_conn_id=POSTGRES_CONN_ID,
)
# Menghapus ST.META.DATA dari file EXTRACTOR
metadata_sed = BashOperator(
task_id="metadata_sed",
bash_command=f"""sed -i '/ST.META.DATA/d' /opt/airflow/dags/DFE/t24_bnk/{yesterday_nodash}/{yesterday_nodash}.ST.META.DATA.csv""",
)
# list file METADATA
metadata_list_extractor = PythonOperator(
task_id='metadata_list_extractor',
python_callable=metadata_list_extractor,
do_xcom_push=True
)
# syntax copy METADATA
metadata_syntax_copy = PythonOperator(
task_id='metadata_syntax_copy',
python_callable=metadata_syntax_copy
)
# file metadata to DWHHub
metadata_import = PythonOperator.partial(
task_id="metadata_import",
python_callable=meta_extract,
dag=dag
).expand_kwargs(
XComArg(metadata_syntax_copy),
)
# file refresh metadata
metadata_refresh = PostgresOperator(
sql=f"""select ds_conf.ds_t24_create_metadata();""",
task_id="metadata_refresh",
postgres_conn_id=POSTGRES_CONN_ID,
)
# ###########################################
# DATA SOURCE
# ###########################################
ds_create_table = PostgresOperator(
sql=f"""select ds_conf.ds_t24_create_table('{POSTGRES_ENV}');""",
task_id="ds_create_table",
postgres_conn_id=POSTGRES_CONN_ID,
)
ds_list_extractor = PythonOperator(
task_id='ds_list_extractor',
python_callable=ds_list_extractor,
do_xcom_push=True
)
ds_syntax_copy = PythonOperator(
task_id='ds_syntax_copy',
python_callable=ds_syntax_copy
)
ds_csv_to_table = PythonOperator.partial(
task_id="ds_csv_to_table",
python_callable=ds_csv_to_table,
dag=dag
).expand_kwargs(
XComArg(ds_syntax_copy),
)
ds_create_table_history = PostgresOperator(
sql=f"""select ds_conf.ds_t24_create_table_history();""",
task_id="ds_create_table_history",
postgres_conn_id=POSTGRES_CONN_ID,
)
ds_create_metadata_file = PostgresOperator(
sql=f"""create table {POSTGRES_SCHEMA}.a_metadata as (select * from ds_conf.fx_metadata_file());""",
task_id="ds_create_metadata_file",
postgres_conn_id=POSTGRES_CONN_ID,
)
ds_create_metadata_file_sys = PostgresOperator(
sql=f"""create table {POSTGRES_SCHEMA}.a_metadata_sys as (select * from ds_conf.fx_metadata_file_system());""",
task_id="ds_create_metadata_file_sys",
postgres_conn_id=POSTGRES_CONN_ID,
)
ds_nominatif = TriggerDagRunOperator(
task_id=f'ds_nominatif',
trigger_dag_id="APJ_1_t24_interface",
)
ds_remove_bnk = BashOperator(
task_id="ds_remove_bnk",
bash_command=f"""
rm -r /opt/airflow/dags/DFE/t24_bnk/{yesterday_nodash};
""",
)
ds_ext_done = BashOperator(
task_id="ds_ext_done",
bash_command=f"""
touch /opt/airflow/dags/DFE/t24/{yesterday_nodash}/done_ext.csv;
""",
)
history_finish = PostgresOperator(
sql=f"""
UPDATE ds_conf."Datasource_history"
SET status = 'DONE', finish_time = '{datetime.now().strftime('%Y-%m-%d %H:%M:%S')}'::timestamp with time zone
WHERE source = '{POSTGRES_SCHEMA}' and status = 'ONPROCESS';
""",
task_id="history_finish",
postgres_conn_id=POSTGRES_CONN_ID,
)
begin >> check_done_ext >> sftp_xx >> check_file_done >> history_start >> file_sc_cp >> metadata_truncate >> metadata_sed >> metadata_list_extractor >> metadata_syntax_copy >> metadata_import >> metadata_refresh >> ds_create_table >> ds_list_extractor >> ds_syntax_copy >> ds_csv_to_table >> ds_create_table_history >> ds_create_metadata_file >> ds_create_metadata_file_sys >> ds_nominatif >> ds_remove_bnk >> ds_ext_done >> history_finish
import json, os
import csv
from datetime import date
import calendar
from airflow import DAG
from airflow.operators.python import PythonOperator, BranchPythonOperator, ShortCircuitOperator
from airflow.operators.bash import BashOperator
from datetime import datetime, timedelta
from airflow.providers.sftp.operators.sftp import SFTPOperator
from airflow.models import Variable
from airflow.providers.postgres.operators.postgres import PostgresOperator
from airflow.operators.trigger_dagrun import TriggerDagRunOperator
from airflow.utils.trigger_rule import TriggerRule
from airflow.models.xcom import XCom
from airflow.decorators import task
from airflow.hooks.postgres_hook import PostgresHook
from airflow import XComArg
# yesterday_nodash = (datetime.now() - timedelta(1)).strftime('%Y%m%d')
yesterday_nodash = (datetime.now() - timedelta(1)).strftime('%Y%m%d') if Variable.get(
"DATE_OF_DATA") == 'today' else Variable.get("DATE_OF_DATA")
yesterday_strip = datetime.strptime(yesterday_nodash, '%Y%m%d').strftime('%Y-%m-%d')
yesterday_lusa = (datetime.strptime(yesterday_nodash, '%Y%m%d') - timedelta(1)).strftime('%Y%m%d')
POSTGRES_CONN_ID = Variable.get("DS_DB")
POSTGRES_ENV = Variable.get("ENV_T24")
POSTGRES_SCHEMA = 'ds_t24' + Variable.get("ENV_T24")
DS_CONN_ID = 'ds_t24'
DS_FOLDER = 't24_interface'
DS_DB = 't24_interface'
DS_SCHEMA = 't24_interface'
DS_CREATE_TABLE = ''
def _start():
print("Start :: Extractor ")
def ds_list_extractor():
sql_stmt = f"""select * from ds_conf.ds_extractor_list_extractor('t24_interface', '{yesterday_strip}');"""
pg_hook = PostgresHook(
postgres_conn_id=POSTGRES_CONN_ID,
)
pg_conn = pg_hook.get_conn()
cursor = pg_conn.cursor()
cursor.execute(sql_stmt)
files = cursor.fetchall()
return files
def ds_push_syntax(ti):
iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
if not iris:
raise Exception('No data.')
return [{"op_kwargs": {
# "copy_sql": f"""COPY {POSTGRES_SCHEMA}.{table['table_name']} from STDOUT delimiter '{table['delimiter']}' CSV HEADER quote E'\b'""",
"copy_sql": f"""COPY {POSTGRES_SCHEMA}.{table['table_name']} from STDOUT delimiter '{table['delimiter']}' CSV HEADER quote '"'""",
"file_id": f"""{yesterday_nodash}/{table['file_id']}"""}}
for
table in json.loads(iris[0][0][0])]
def pg_ddl_syntax(ti):
iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
arr = []
for table in json.loads(iris[0][0][0]):
with open(f"""{Variable.get("LOCAL_PATH")}t24/{yesterday_nodash}/{table['file_id']}""") as csvFile:
reader = csv.reader(csvFile, delimiter=f"""{table['delimiter']}""")
field_names_list = reader.__next__()
arr.append({"sql": f"""drop table if exists {POSTGRES_SCHEMA}.{table['table_name']} cascade; create table {POSTGRES_SCHEMA}.{table['table_name']} ({' text, '.join([w.replace(' ', '_').replace('.', '_').replace('/', '_').replace('(', '_').replace(')', '_').replace('+', '_').replace('___', '_').lower().replace((table['delimiter']+'limit'+table['delimiter']), (table['delimiter']+'limit_reff'+table['delimiter'])) for w in field_names_list])} text);"""})
return arr
def csv_clean_syntax(ti):
iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
arr = []
for table in json.loads(iris[0][0][0]):
with open(f"""{Variable.get("LOCAL_PATH")}t24/{yesterday_nodash}/{table['file_id']}""") as csvFile:
reader = csv.reader(csvFile, delimiter=f"""{table['delimiter']}""")
field_names_list = reader.__next__()
arr.append({"bash_command": f"""echo 'OK' """ if table['sed_command'] == 'nil' else (f"""{Variable.get("LOCAL_PATH")}t24/{yesterday_nodash}/{table['file_id']} > {Variable.get("LOCAL_PATH")}t24/{yesterday_nodash}/{table['file_id']}_bk && mv {Variable.get("LOCAL_PATH")}t24/{yesterday_nodash}/{table['file_id']}_bk {Variable.get("LOCAL_PATH")}t24/{yesterday_nodash}/{table['file_id']} && """.join(table['sed_command'].split('|;|;|')) + f""" {Variable.get("LOCAL_PATH")}t24/{yesterday_nodash}/{table['file_id']} > {Variable.get("LOCAL_PATH")}t24/{yesterday_nodash}/{table['file_id']}_bk && mv {Variable.get("LOCAL_PATH")}t24/{yesterday_nodash}/{table['file_id']}_bk {Variable.get("LOCAL_PATH")}t24/{yesterday_nodash}/{table['file_id']}""")})
return arr
def ds_push_csv(ti, copy_sql, file_id):
pg_hook = PostgresHook.get_hook(POSTGRES_CONN_ID)
pg_hook.copy_expert(copy_sql, filename=f"""/opt/airflow/dags/DFE/t24/{file_id}""")
def sql_clean_syntax(ti):
iris = ti.xcom_pull(task_ids=['ds_list_extractor'])
arr = []
for table in json.loads(iris[0][0][0]):
with open(f"""{Variable.get("LOCAL_PATH")}t24/{yesterday_nodash}/{table['file_id']}""") as csvFile:
reader = csv.reader(csvFile, delimiter=f"""{table['delimiter']}""")
field_names_list = reader.__next__()
arr.append({"sql": f"""select 'OK' """ if table['sql_command'] == 'nil' else f"""{";".join(table['sql_command'].split('|;|;|')).replace('T24_SOURCE', POSTGRES_SCHEMA)}"""})
return arr
def stop_task(**kwargs):
return True
with DAG("APJ_1_t24_interface",
start_date=datetime(2021, 1, 1),
schedule_interval=None,
catchup=False,
concurrency=3) as dag:
begin = PythonOperator(
task_id=f"Begin",
python_callable=_start,
op_kwargs={
}
)
sftp_xx = BashOperator(
task_id="sftp_xx",
bash_command=f"""sshpass -p {Variable.get("SFTP_T24_PASSWORD")} sftp -o StrictHostKeyChecking=no -r {Variable.get("SFTP_T24_USER")}@{Variable.get("SFTP_T24_HOST")}:/REPORT.BP/NOM/TREASURY/*{yesterday_nodash}* {Variable.get("LOCAL_PATH")}t24/{yesterday_nodash}/""",
)
# sftp_neraca = BashOperator(
# task_id="sftp_xx",
# bash_command=f"""sshpass -p {Variable.get("SFTP_T24_PASSWORD")} sftp -o StrictHostKeyChecking=no -r {Variable.get("SFTP_T24_USER")}@{Variable.get("SFTP_T24_HOST")}:/REPORT.BP/NERACA/*{yesterday_nodash}* {Variable.get("LOCAL_PATH")}t24_neraca/{yesterday_nodash}/""",
# )
# sftp_neraca_tel = BashOperator(
# task_id="sftp_xx",
# bash_command=f"""sshpass -p {Variable.get("SFTP_T24_PASSWORD")} sftp -o StrictHostKeyChecking=no -r {Variable.get("SFTP_T24_USER")}@{Variable.get("SFTP_T24_HOST")}:/REPORT.BP/NERACA/*{yesterday_nodash}* {Variable.get("LOCAL_PATH")}t24_neraca/{yesterday_nodash}/""",
# )
ds_list_extractor = PythonOperator(
task_id='ds_list_extractor',
python_callable=ds_list_extractor,
do_xcom_push=True
)
# ft_reve = BashOperator(
# task_id="ft_reve",
# bash_command=f"""head -n -2 {Variable.get("LOCAL_PATH")}t24/{yesterday_nodash}/FT.REVERSE.csv > {Variable.get("LOCAL_PATH")}t24/{yesterday_nodash}/FT.REVERSE.2.csv && mv {Variable.get("LOCAL_PATH")}t24/{yesterday_nodash}/FT.REVERSE.2.csv {Variable.get("LOCAL_PATH")}t24/{yesterday_nodash}/FT.REVERSE.csv""",
# )
csv_clean_syntax = PythonOperator(
task_id='csv_clean_syntax',
python_callable=csv_clean_syntax
)
clean_csv = BashOperator.partial(
task_id="clean_csv",
# bash_command=f"""head -n -2 {Variable.get("LOCAL_PATH")}t24/{yesterday_nodash}/FT.REVERSE.csv > {Variable.get("LOCAL_PATH")}t24/{yesterday_nodash}/FT.REVERSE.2.csv && mv {Variable.get("LOCAL_PATH")}t24/{yesterday_nodash}/FT.REVERSE.2.csv {Variable.get("LOCAL_PATH")}t24/{yesterday_nodash}/FT.REVERSE.csv""",
).expand_kwargs(
XComArg(csv_clean_syntax),
)
# pg_drop_schema = PostgresOperator(
# sql= f"""DROP TABLE IF EXISTS {POSTGRES_SCHEMA} CASCADE;""",
# task_id="pg_drop_schema",
# postgres_conn_id=POSTGRES_CONN_ID,
# )
#
# pg_create_schema = PostgresOperator(
# sql= f"""create schema IF NOT EXISTS {POSTGRES_SCHEMA};""",
# task_id="pg_create_schema",
# postgres_conn_id=POSTGRES_CONN_ID,
# )
stop_task = ShortCircuitOperator(
task_id="stop_task",
provide_context=True,
python_callable=stop_task,
op_kwargs={},
)
pg_ddl_syntax = PythonOperator(
task_id='pg_ddl_syntax',
python_callable=pg_ddl_syntax
)
pg_create_table = PostgresOperator.partial(
task_id="pg_create_table",
postgres_conn_id=POSTGRES_CONN_ID,
).expand_kwargs(
XComArg(pg_ddl_syntax),
)
# ds_truncate_syntax = PythonOperator(
# task_id='ds_truncate_syntax',
# python_callable=ds_truncate_syntax
# )
#
# ds_truncate = PostgresOperator.partial(
# task_id="ds_truncate",
# postgres_conn_id=POSTGRES_CONN_ID,
# ).expand_kwargs(
# XComArg(ds_truncate_syntax)
# )
#
ds_push_syntax = PythonOperator(
task_id='ds_syntax_push',
python_callable=ds_push_syntax
)
ds_csv_to_table = PythonOperator.partial(
task_id="ds_csv_to_table",
python_callable=ds_push_csv,
dag=dag
).expand_kwargs(
XComArg(ds_push_syntax),
)
sql_clean_syntax = PythonOperator(
task_id='sql_clean_syntax',
python_callable=sql_clean_syntax
)
ds_clean_data = PostgresOperator.partial(
# sql=f"""select ds_conf.ds_t24_create_table_history_wo_t24_dfe('{POSTGRES_SCHEMA}');""",
task_id="ds_clean_data",
postgres_conn_id=POSTGRES_CONN_ID,
).expand_kwargs(
XComArg(sql_clean_syntax),
)
ds_create_table_history_nominatif = PostgresOperator(
sql=f"""select ds_conf.ds_t24_create_table_history_wo_t24_dfe('{POSTGRES_SCHEMA}');""",
task_id="ds_create_table_history_nominatif",
postgres_conn_id=POSTGRES_CONN_ID,
)
ds_to_history = PostgresOperator(
sql=f"""select ds_conf.ds_t24_copy_to_history('{yesterday_strip}', '{POSTGRES_SCHEMA}');""",
task_id="ds_to_history",
postgres_conn_id=POSTGRES_CONN_ID,
)
set_access_schemma = PostgresOperator(
sql=f"""GRANT USAGE ON SCHEMA {POSTGRES_SCHEMA} TO readaccess;""",
task_id="set_access_schemma",
postgres_conn_id=POSTGRES_CONN_ID,
)
set_access_all_table = PostgresOperator(
sql=f"""GRANT SELECT ON ALL TABLES IN SCHEMA {POSTGRES_SCHEMA} TO readaccess;""",
task_id="set_access_all_table",
postgres_conn_id=POSTGRES_CONN_ID,
)
pentaho = BashOperator(
task_id='pentaho',
bash_command=f"""curl '{Variable.get("PENTAHO_HOST_PASSWORD")}/kettle/executeJob/?job=/home/apj/files/scripts/etl/D_REPORT.kjb'"""
# bash_command=f"""curl '{Variable.get("PENTAHO_HOST")}/kettle/executeJob/?rep=test-repo&job=/home/oper/files/f/Untitled'"""
)
rep_nominatif_customer = PostgresOperator(
sql=f"""select ds_conf.dm_customer_nominative('{yesterday_strip}');""",
task_id="rep_nominatif_customer",
postgres_conn_id=POSTGRES_CONN_ID,
)
zip_today = BashOperator(
task_id="zip_today",
bash_command=f"""
zip -r {yesterday_nodash}.zip /opt/airflow/dags/DFE/t24/{yesterday_nodash};
""",
)
delete_before = BashOperator(
task_id="delete_before",
bash_command=f"""
rm -r /opt/airflow/dags/DFE/t24/{yesterday_lusa};
""",
)
begin >> sftp_xx >> ds_list_extractor >> csv_clean_syntax >> clean_csv >> pg_ddl_syntax >> pg_create_table >> ds_push_syntax >> sql_clean_syntax >> ds_csv_to_table >> ds_clean_data >> stop_task >> ds_create_table_history_nominatif >> ds_to_history >> set_access_schemma >> set_access_all_table >> pentaho >> rep_nominatif_customer >> zip_today >> delete_before
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment