aboutsummaryrefslogtreecommitdiffstats
path: root/src/transform_lambda
diff options
context:
space:
mode:
authorT-Aji <tolujbd2@gmail.com>2024-08-28 17:52:45 +0100
committerT-Aji <tolujbd2@gmail.com>2024-08-28 17:52:45 +0100
commit6235a2bb04b60d57a41196b07bbf0296920c6980 (patch)
treea9c768be6a4503df12b8547eb8f21757a37ff3f5 /src/transform_lambda
parentd064b2ec2c7393f8de50560a7edfe100851bfea3 (diff)
downloadde-project-bentley-6235a2bb04b60d57a41196b07bbf0296920c6980.tar.gz
de-project-bentley-6235a2bb04b60d57a41196b07bbf0296920c6980.zip
wip commit
Diffstat (limited to 'src/transform_lambda')
-rw-r--r--src/transform_lambda/dataframes.py8
-rw-r--r--src/transform_lambda/transform_lambda.py2
2 files changed, 6 insertions, 4 deletions
diff --git a/src/transform_lambda/dataframes.py b/src/transform_lambda/dataframes.py
index bf0556b..e89a6b2 100644
--- a/src/transform_lambda/dataframes.py
+++ b/src/transform_lambda/dataframes.py
@@ -18,7 +18,7 @@ import requests
# no test, same as fact_payment
def create_fact_sales_order(dict_of_df):
- df_sales = dict_of_df["sales_order"]
+ df_sales = dict_of_df["sales_order"].rename(columns={"staff_id": "sales_staff_id"})
df_sales.index.name = "sales_record_id"
df_sales["created_date"] = df_sales["created_at"].astype("datetime64[ns]").dt.date
@@ -44,7 +44,7 @@ def create_fact_sales_order(dict_of_df):
"created_time",
"last_updated_date",
"last_updated_time",
- "staff_id",
+ "sales_staff_id",
"counterparty_id",
"units_sold",
"unit_price",
@@ -55,7 +55,7 @@ def create_fact_sales_order(dict_of_df):
"agreed_delivery_location_id"
],
]
- fact_sales.rename(columns={"staff_id": "sales_staff_id"}).reset_index(inplace=True)
+ fact_sales.reset_index(inplace=True)
return fact_sales
@@ -253,6 +253,8 @@ def create_dim_currency(dict_of_df, names=scrape_currency_names()):
df_cur, names, left_on="currency_code", right_on="currency_code", how="left"
)
dim_currency.drop_duplicates(inplace=True)
+ dim_currency.astype({"currency_name": "string", "currency_code": "string"})
+ print(dim_currency.dtypes, "<<<<<<<<<Dtype")
return dim_currency
diff --git a/src/transform_lambda/transform_lambda.py b/src/transform_lambda/transform_lambda.py
index 1453c6c..0b5748b 100644
--- a/src/transform_lambda/transform_lambda.py
+++ b/src/transform_lambda/transform_lambda.py
@@ -5,7 +5,7 @@ import logging
import pandas as pd
import pyarrow as pa
import pyarrow.parquet as pq
-from dataframes import *
+from src.transform_lambda.dataframes import *
from botocore.exceptions import ClientError
from pg8000.native import Connection, InterfaceError
from datetime import datetime
git.ajschof.me — hosted by ajschofield — powered by cgit