diff options
| -rw-r--r-- | src/dataframes.py | 50 | ||||
| -rw-r--r-- | tests/test_dataframes.py | 13 |
2 files changed, 63 insertions, 0 deletions
diff --git a/src/dataframes.py b/src/dataframes.py index f122368..36361d2 100644 --- a/src/dataframes.py +++ b/src/dataframes.py @@ -20,6 +20,7 @@ import requests def create_fact_sales_order(dict_of_df): df_sales = dict_of_df["sales_order"] df_sales.index.name = "sales_record_id" +<<<<<<< HEAD df_sales["created_date"] = df_sales["created_at"].astype("datetime64[ns]").dt.date df_sales["created_time"] = ( df_sales["created_at"].astype("datetime64[ns]").dt.floor("s").dt.time @@ -29,6 +30,15 @@ def create_fact_sales_order(dict_of_df): ) df_sales["last_updated_time"] = ( df_sales["last_updated"].astype("datetime64[ns]").dt.floor("s").dt.time +======= + df_sales["created_date"] = pd.to_datetime(df_sales["created_at"], format="%Y-%m-%d") + df_sales["created_time"] = pd.to_datetime(df_sales["created_at"], format="%H-%M-%S") + df_sales["last_updated_date"] = pd.to_datetime( + df_sales["last_updated"], format="%Y-%m-%d" + ) + df_sales["last_updated_time"] = pd.to_datetime( + df_sales["last_updated"], format="%H-%M-%S" +>>>>>>> 5db3f61 (style: format code with Autopep8, Black and Ruff Formatter) ) df_sales["agreed_delivery_date"] = pd.to_datetime( df_sales["agreed_delivery_date"], format="%Y-%m-%d" @@ -36,7 +46,11 @@ def create_fact_sales_order(dict_of_df): df_sales["agreed_payment_date"] = pd.to_datetime( df_sales["agreed_payment_date"], format="%Y-%m-%d" ) +<<<<<<< HEAD df_sales = df_sales.drop(labels=["created_at", "last_updated"], axis=1) +======= + df_sales.drop(labels=["created_at", "last_updated"], axis=1, inplace=True) +>>>>>>> 5db3f61 (style: format code with Autopep8, Black and Ruff Formatter) df_sales.reset_index(inplace=True) return df_sales @@ -47,6 +61,7 @@ def create_fact_sales_order(dict_of_df): def create_fact_purchase_orders(dict_of_df): df_po = dict_of_df["purchase_order"] df_po.index.name = "purchase_record_id" +<<<<<<< HEAD df_po["created_date"] = df_po["created_at"].astype("datetime64[ns]").dt.date df_po["created_time"] = ( df_po["created_at"].astype("datetime64[ns]").dt.floor("s").dt.time @@ -54,6 +69,15 @@ def create_fact_purchase_orders(dict_of_df): df_po["last_updated_date"] = df_po["last_updated"].astype("datetime64[ns]").dt.date df_po["last_updated_time"] = ( df_po["last_updated"].astype("datetime64[ns]").dt.floor("s").dt.time +======= + df_po["created_date"] = pd.to_datetime(df_po["created_at"], format="%Y-%m-%d") + df_po["created_time"] = pd.to_datetime(df_po["created_at"], format="%H-%M-%S") + df_po["last_updated_date"] = pd.to_datetime( + df_po["last_updated"], format="%Y-%m-%d" + ) + df_po["last_updated_time"] = pd.to_datetime( + df_po["last_updated"], format="%H-%M-%S" +>>>>>>> 5db3f61 (style: format code with Autopep8, Black and Ruff Formatter) ) df_po["agreed_delivery_date"] = pd.to_datetime( df_po["agreed_delivery_date"], format="%Y-%m-%d" @@ -61,7 +85,11 @@ def create_fact_purchase_orders(dict_of_df): df_po["agreed_payment_date"] = pd.to_datetime( df_po["agreed_payment_date"], format="%Y-%m-%d" ) +<<<<<<< HEAD df_po = df_po.drop(labels=["created_at", "last_updated"], axis=1) +======= + df_po.drop(labels=["created_at", "last_updated"], axis=1, inplace=True) +>>>>>>> 5db3f61 (style: format code with Autopep8, Black and Ruff Formatter) df_po.reset_index(inplace=True) return df_po @@ -72,6 +100,7 @@ def create_fact_purchase_orders(dict_of_df): def create_fact_payment(dict_of_df): df_payment = dict_of_df["payment"] df_payment.index.name = "payment_record_id" +<<<<<<< HEAD df_payment["created_date"] = ( df_payment["created_at"].astype("datetime64[ns]").dt.date ) @@ -83,11 +112,28 @@ def create_fact_payment(dict_of_df): ) df_payment["last_updated_time"] = ( df_payment["last_updated"].astype("datetime64[ns]").dt.floor("s").dt.time +======= + df_payment["created_date"] = pd.to_datetime( + df_payment["created_at"], format="%Y-%m-%d" + ) + df_payment["created_time"] = pd.to_datetime( + df_payment["created_at"], format="%H-%M-%S" + ) + df_payment["last_updated_date"] = pd.to_datetime( + df_payment["last_updated"], format="%Y-%m-%d" + ) + df_payment["last_updated_time"] = pd.to_datetime( + df_payment["last_updated"], format="%H-%M-%S" +>>>>>>> 5db3f61 (style: format code with Autopep8, Black and Ruff Formatter) ) df_payment["payment_date"] = pd.to_datetime( df_payment["payment_date"], format="%Y-%m-%d" ) +<<<<<<< HEAD df_payment = df_payment.drop(labels=["created_at", "last_updated"], axis=1) +======= + df_payment.drop(labels=["created_at", "last_updated"], axis=1, inplace=True) +>>>>>>> 5db3f61 (style: format code with Autopep8, Black and Ruff Formatter) df_payment.reset_index(inplace=True) return df_payment @@ -143,7 +189,11 @@ def create_dim_date(dict_of_df): list_of_date_columns = [] for df in fact_dfs: date_col_names = [ +<<<<<<< HEAD col_name for col_name in list(df.columns) if "_date" in col_name +======= + col_name for col_name in list(df.columns) if "date" in col_name +>>>>>>> 5db3f61 (style: format code with Autopep8, Black and Ruff Formatter) ] for col in date_col_names: list_of_date_columns.append(df[col]) diff --git a/tests/test_dataframes.py b/tests/test_dataframes.py index c9ff43f..cc133fe 100644 --- a/tests/test_dataframes.py +++ b/tests/test_dataframes.py @@ -252,6 +252,7 @@ class TestCreateFactPayment: "payment": pd.DataFrame( data=[ [ +<<<<<<< HEAD dt.strptime( "2022-11-03 14:20:49.962846", "%Y-%m-%d %H:%M:%S.%f" ), @@ -261,6 +262,13 @@ class TestCreateFactPayment: 1, "SE18 9QO", "2020-07-16", +======= + dt(2020, 5, 17, 6, 15, 20), + dt(2020, 5, 20, 8, 19, 30), + 1, + "SE18 9QO", + "2020-7-16", +>>>>>>> 5db3f61 (style: format code with Autopep8, Black and Ruff Formatter) ] ], columns=[ @@ -287,5 +295,10 @@ class TestCreateFactPayment: for col in list(result.columns): assert col in expected_cols for col in expected_cols: +<<<<<<< HEAD if "_date" or "_time" in col: assert result[col].dtype == "O" +======= + if "date" in col: + assert result[col].dtype == "datetime64[ns]" +>>>>>>> 5db3f61 (style: format code with Autopep8, Black and Ruff Formatter) |