Article
Clean CSV files with SQL locally.
CSV is simple until it is not. Real CSV exports often contain broken dates, mixed decimal separators, blank records and inconsistent column names.
A SQL-first cleanup flow gives you something repeatable. You can rerun the same transformation whenever a new export arrives.
Example
select
trim(product_code) as product_code,
replace(price, ',', '.')::decimal(18,2) as price,
try_cast(created_at as timestamp) as created_at
from imported_csv
where product_code is not null;
Where Bendit fits
Bendit is a practical desktop layer around this workflow: import file, inspect columns, write SQL, preview result, export clean output.