Migration Guide

You: I have this SQL transformation that calculates customer lifetime value:

SELECT 
  customer_id,
  SUM(order_total) as total_spend,
  COUNT(*) as order_count,
  MIN(order_date) as first_order,
  MAX(order_date) as last_order
FROM orders
GROUP BY customer_id

How should I model this in DataChonk?

Chonk: Great question! This looks like a customer dimension with aggregated order metrics. 
I'd recommend:

1. Source chonk: 'orders' - reference your raw orders table
2. Staging chonk: 'stg_orders' - clean and standardize the raw data
3. Entity chonk: 'dim_customers' - this is where your aggregation lives

Would you like me to generate these chonks for you?

# Scan warehouse and create sources
datachonk scan --connection snowflake://...

# Generate source chonks for specific tables
datachonk generate source --tables orders,customers,products

# Or use the web UI to visually select tables

# Import from local path
datachonk import --path ./my-dbt-project

# Or from GitHub
datachonk import --repo https://github.com/org/dbt-project --branch main

# DataChonk analyzes:
# - Model structure and dependencies
# - Naming conventions
# - Materialization patterns
# - Test coverage
# - Documentation style

-- DataChonk detected your patterns:
-- - Naming: stg_<source>__<table>
-- - Materialization: view for staging
-- - CTEs: source → renamed → final
-- - Tests: unique, not_null on keys

{{ config(materialized='view') }}

with source as (
    select * from {{ source('stripe', 'payments') }}
),

renamed as (
    select
        id as payment_id,
        customer_id,
        amount_cents / 100.0 as amount,
        created_at as payment_at
    from source
)

select * from renamed

You: I have an Informatica mapping that implements SCD Type 2 for customers. 
It tracks changes to name, email, and address. How do I replicate this in dbt?

Chonk: For SCD Type 2 in dbt, I recommend using the dbt_utils.snapshot approach 
or building a custom incremental model. Here's what I'd generate:

1. A snapshot for tracking changes:
{% snapshot snp_customers %}
{{ config(
    target_schema='snapshots',
    unique_key='customer_id',
    strategy='check',
    check_cols=['name', 'email', 'address']
) }}
select * from {{ source('crm', 'customers') }}
{% endsnapshot %}

2. A dimension model that joins current and historical records:
[generates full dim_customers model with SCD2 support]

Would you like me to create these as chonks?