fixed a few logging bugs

default_config/transform_ecmc.yaml

@@ -22,8 +22,6 @@ production_columns_to_keep:
   - oil_prod
   - water_prod
 completions_columns_to_keep:
-  - name
-  - operator_num
   - facility_name
   - facility_num
   - well_name

oip_ecmc_app/convert_access_to_parquet.py

@@ -106,8 +106,7 @@ def main() -> None:
         with parquet_metadata_path.open('w') as f:
             json.dump(utils.to_json(parquet_metadata, logger=logger), f)
 
-        driver = get_access_driver(
-            config.Config.microsoft_access_driver, logger)
+        driver = get_access_driver(config.Config.microsoft_access_driver)
         data = mdb_import(access_db_metadata, logger, driver=driver)
         write_parquet(parquet_path, data, logger)
 
@@ -124,7 +123,8 @@ def read_odbc_table(
 ) -> pl.DataFrame:
     logger.info(f'loading data from {table} in {connection[ODBCKey.dbq]}')
     query = f'SELECT * FROM \"{table}\"'
-    return pl.read_database(query, connection=odbc_connection_str(connection))
+    return pl.read_database(
+        query, connection=odbc_connection_str(connection, logger))
 
 
 def get_parquet_metadata(

oip_ecmc_app/scrape_from_ecmc.py

@@ -100,7 +100,6 @@ def main() -> None:
                 utils.to_json(
                     get_db_metadata(access_db_path, zip_metadata, logger)),
                 f,
-                logger=logger,
             )
 
 

oip_ecmc_app/transform_ecmc.py

@@ -55,8 +55,6 @@ class Config:
         'water_prod',
     ]
     completions_columns_to_keep: List[str] = [
-        'name',
-        'operator_num',
         'facility_name',
         'facility_num',
         'well_name',
@@ -165,7 +163,7 @@ def write_output_data(
             data['completions'][max(data['completions'])],
             on='API_num',
             how='outer',
-        ).select(pl.exclude('^.*_right$'))
+        )
         if remove_co2_wells:
             df_out = df_out.filter(pl.col('Prod_days') != 0)
         df_out.write_csv(output_path / f'{year}.csv')