jaffle-shop-clickzetta

jaffle-shop adapted for ClickZetta Lakehouse. Use this project to get started with dbt + ClickZetta in minutes.

What's inside

A coffee shop order dataset with a full dbt project:

• 6 CSV seed files → raw tables (153,864 rows total)
• 6 staging views (data cleaning layer)
• 7 mart tables (business-ready layer)
• 27 data quality tests + 3 unit tests
• 3 example analyses (store revenue, customer segments, top products)

Prerequisites

• Python 3.10+
• A ClickZetta Lakehouse instance (workspace, vcluster, username, password)
• dbt-clickzetta 1.7.0+

Quickstart

1. Clone and set up environment

git clone https://github.com/clickzetta/jaffle-shop-clickzetta.git
cd jaffle-shop-clickzetta

python3 -m venv .venv
source .venv/bin/activate  # Windows: .venv\Scripts\activate
pip install dbt-clickzetta

2. Configure connection

cp profiles.yml.example profiles.yml

Edit profiles.yml with your instance details:

jaffle_shop:
  target: dev
  outputs:
    dev:
      type: clickzetta
      service: <your-service-endpoint>
      instance: <your-instance-id>
      workspace: <your-workspace>
      username: <your-username>
      password: <your-password>
      schema: dbt_jaffle
      vcluster: default_ap

Set the profiles directory so you don't need --profiles-dir . on every command:

export DBT_PROFILES_DIR=.

Verify the connection:

dbt debug

3. Run the project

dbt deps
dbt seed    # ~60s — loads 153,864 rows via COPY INTO
dbt run     # ~5s  — builds 13 models
dbt test    # ~10s — runs 27 data quality tests + 3 unit tests

Expected output:

dbt seed  → Done. PASS=6  WARN=0 ERROR=0 SKIP=0 TOTAL=6
dbt run   → Done. PASS=13 WARN=0 ERROR=0 SKIP=0 TOTAL=13
dbt test  → Done. PASS=30 WARN=0 ERROR=0 SKIP=0 TOTAL=30

4. Explore the results

Query the mart tables directly in ClickZetta Studio or via cz-cli. Three ready-to-run example queries are in analyses/:

File	Query
analyses/store_revenue.sql	Revenue and order count by store location
analyses/customer_segments.sql	Customer count and avg spend by segment
analyses/top_products.sql	Top 10 products by order frequency

Quick samples:

-- Store revenue ranking
SELECT
    l.location_name,
    COUNT(DISTINCT o.order_id)        AS total_orders,
    ROUND(SUM(o.subtotal) / 100.0, 2) AS revenue_usd
FROM dbt_jaffle.orders o
JOIN dbt_jaffle.locations l ON o.location_id = l.location_id
GROUP BY l.location_name
ORDER BY revenue_usd DESC;

-- Customer value segments
SELECT
    customer_type,
    COUNT(*)                               AS customers,
    ROUND(AVG(lifetime_spend) / 100.0, 2) AS avg_spend_usd
FROM dbt_jaffle.customers
GROUP BY customer_type
ORDER BY avg_spend_usd DESC;

-- Top 10 products by order count
SELECT
    product_name,
    product_type,
    COUNT(*) AS times_ordered
FROM dbt_jaffle.order_items
GROUP BY product_name, product_type
ORDER BY times_ordered DESC
LIMIT 10;

5. View data lineage (optional)

dbt docs generate
dbt docs serve

Open http://localhost:8080 to explore the data lineage graph.

ClickZetta-specific adaptations

File	Change
profiles.yml.example	ClickZetta connection template
dbt_project.yml	column_types: string for all seed columns — required for COPY INTO compatibility (dbt-clickzetta 1.7.0+)
macros/generate_schema_name.sql	Preserves custom schema names in non-default targets

Project structure

models/
  staging/    ← views: clean raw data
  marts/      ← tables: business-ready
analyses/     ← example queries to run after dbt build
seeds/
  jaffle-data/  ← 6 CSV files (153,864 rows)
profiles.yml.example

Documentation

Full step-by-step guide (Chinese): dbt + ClickZetta Lakehouse 快速入门

dbt-clickzetta adapter: github.com/clickzetta/dbt-clickzetta