Seeds for static reference data in dbt - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

Output of a dbt seed CSV import

Given a seed CSV file named countries.csv with the following content:

country_code,country_name
US,United States
CA,Canada
MX,Mexico

and a dbt project configured to load this seed, what will be the output of the SQL query select * from {{ ref('countries') }}?

A [{'country_code': 'US', 'country_name': 'United States'}, {'country_code': 'CA', 'country_name': 'Canada'}, {'country_code': 'MX', 'country_name': 'Mexico'}]

B [{'country_code': 'US', 'country_name': 'United States'}, {'country_code': 'CA', 'country_name': 'Canada'}]

C []

D SyntaxError: table 'countries' does not exist

💡 Hint

Think about what dbt seeds do with CSV files and how they become tables.

dbt seeds load CSV files as tables in the data warehouse. The query selects all rows from the seeded table, so all three rows appear.

❓ data_output

intermediate

Number of rows in a seeded table after dbt seed

If you have a seed CSV file with 10 rows (excluding header) and run dbt seed, how many rows will the resulting table contain?

A 9

B 11

C 10

D 0

💡 Hint

Remember that the header row is not counted as data.

dbt seeds create tables with the exact number of data rows from the CSV, excluding the header.

🔧 Debug

advanced

Why does dbt seed fail to load a CSV with inconsistent columns?

You have a seed CSV file where some rows have fewer columns than the header row. Running dbt seed results in an error. What is the most likely cause?

A dbt seed requires the CSV file to be sorted alphabetically.

B The CSV file has inconsistent number of columns per row causing parsing errors.

C The CSV file must have no header row for dbt seed to work.

D dbt seed only supports CSV files with exactly 5 columns.

💡 Hint

Think about how CSV parsers handle rows with missing columns.

dbt seed uses CSV parsing that expects each row to have the same number of columns as the header. Missing columns cause errors.

🚀 Application

advanced

Using seeds for static lookup tables in dbt models

You want to use a static lookup table for country codes in your dbt models. Which approach correctly uses a seed for this purpose?

A Create a CSV file with country codes, add it to the seeds folder, run <code>dbt seed</code>, then reference it in models using <code>{{ ref('filename_without_extension') }}</code>.

B Write the country codes directly inside the model SQL using a VALUES clause.

C Create a separate model with hardcoded country codes and join it in other models.

D Use an external API call inside the dbt model to fetch country codes dynamically.

💡 Hint

Seeds are designed to load static CSV data as tables for easy reference.

Seeds allow you to maintain static reference data in CSV files and use them as tables in your dbt models via refs.

🧠 Conceptual

expert

Best practice for updating static reference data with dbt seeds

You have a seed CSV file used for static reference data in production. What is the best practice to update this data safely?

A Rename the seed CSV file and add a new seed to avoid overwriting existing data.

B Manually delete the seed table in the warehouse, then run <code>dbt seed</code>.

C Edit the seed table directly in the warehouse using SQL UPDATE statements.

D Update the CSV file, run <code>dbt seed --full-refresh</code> to reload the table with new data.

💡 Hint

Think about how dbt manages seed tables and refreshing data.

Using dbt seed --full-refresh reloads the seed table fully with updated CSV data, ensuring consistency.