This lesson's interactive features are locked, please to keep using them
Handling Column Name Conflicts
When two DataFrames share a column name that isn't the merge key, Pandas needs to distinguish them in the result. By default, it appends _x and _y suffixes, which technically works... but is frankly just slop of the non-ai variety.
sales = pd.DataFrame(
{
"product_id": ["CHL-482", "TST-731"],
"price": [10, 20], # Sale price
}
)
products = pd.DataFrame(
{
"product_id": ["CHL-482", "TST-731"],
"price": [12, 25], # List price
}
)
result = pd.merge(sales, products, on="product_id", how="left")
print(result)
# product_id price_x price_y
# 0 CHL-482 10 12
# 1 TST-731 20 25
Instead, use the suffixes parameter to replace _x and _y with something descriptive:
result = pd.merge(
sales, products, on="product_id", how="left", suffixes=("_sale", "_list")
)
print(result)
# product_id price_sale price_list
# 0 CHL-482 10 12
# 1 TST-731 20 25
Rename Before Merging
When you know ahead of time that there are going to be column name conflicts, it's even better to rename them before doing the merge:
products_renamed = products.rename(columns={"price": "list_price"})
# No conflict: 'price' and 'list_price' are already distinct
result = pd.merge(sales, products_renamed, on="product_id", how="left")
print(result)
# product_id price list_price
# 0 CHL-482 10 12
# 1 TST-731 20 25
The merge call stays clean, and the resulting columns have unambiguous, self-documenting names.
Assignment
SnackStack's fulfillment team needs to track toaster orders and shipments. Both tables have a status column, so the merged report needs clear column names. Orders that haven't shipped yet should also have an explicit shipment status instead of a missing value.
Complete the merge_with_suffixes function. It accepts two DataFrames: orders and shipments. It should: