DocsConfigurationTemplates

DocsConfigurationTemplates

Configuration

Code Templates

Customize how DataChonk generates dbt code to match your team's standards and conventions.

DataChonk uses a template system for code generation. You can customize templates to match your team's coding standards, naming conventions, and patterns. Templates use Jinja-like syntax and can be stored locally or in a shared registry.

Built-in

Default templates bundled with DataChonk. Good starting point for customization.

Project

.datachonk/templates/
Project-specific templates in your repo.

User

~/.datachonk/templates/
Personal templates across all projects.

Viewing Templates

List all available templates with the CLI:

bash

# List all templates
datachonk templates

# Filter by category
datachonk templates --category staging

# Output:
# Staging Templates:
#   staging-model - Standard staging model that cleans and renames columns
#   staging-snapshot - Staging model with SCD Type 2 tracking
#
# Mart Templates:
#   dimension-model - Dimension table following Kimball methodology
#   fact-model - Fact table with foreign key joins
#   ...

Template Structure

Templates are JSON files with metadata and Jinja-like content:

my-staging-template.json

{
  "name": "my-staging-template",
  "version": "1.0.0",
  "description": "Custom staging model with audit columns",
  "category": "staging",
  "author": "Data Team",
  "tags": ["staging", "audit", "custom"],
  
  "variables": [
    {
      "name": "source_name",
      "type": "string",
      "required": true,
      "description": "Name of the source"
    },
    {
      "name": "source_table",
      "type": "string",
      "required": true,
      "description": "Source table name"
    },
    {
      "name": "columns",
      "type": "array",
      "required": false,
      "description": "Columns to select"
    },
    {
      "name": "add_audit",
      "type": "boolean",
      "required": false,
      "default": true,
      "description": "Add audit columns"
    }
  ],
  
  "content": "... (template content)"
}

Template Syntax

Templates use Jinja-like syntax for variable substitution and control flow:

Variable Substitution

-- Basic variable
select * from {{ source('{{ source_name }}', '{{ source_table }}') }}

-- With default value
{{ model_name | default: 'stg_default' }}

-- From config
{{ dbt_utils.generate_surrogate_key(['{{ primary_key }}']) }}

Conditionals

{%- if add_audit %}
    -- Audit columns
    , current_timestamp as _loaded_at
    , '{{ source_name }}' as _source
    , md5(row_to_json(source.*)::text) as _row_hash
{%- endif %}

{%- if scd_type == '2' %}
    -- SCD Type 2 columns
    , valid_from
    , valid_to
    , is_current
{%- endif %}

Loops

select
{%- for col in columns %}
    {{ col }}{% if not loop.last %},{% endif %}
{%- endfor %}
from source

-- With rename mapping
{%- for col in columns %}
    {%- if rename_columns and col in rename_columns %}
    {{ col }} as {{ rename_columns[col] }}
    {%- else %}
    {{ col }}
    {%- endif %}
{%- endfor %}

Full Example

A complete custom staging template with audit columns and renaming:

staging-with-audit.json

{
  "name": "staging-with-audit",
  "version": "1.0.0",
  "description": "Staging model with audit columns and column renaming",
  "category": "staging",
  "variables": [
    { "name": "source_name", "type": "string", "required": true },
    { "name": "source_table", "type": "string", "required": true },
    { "name": "columns", "type": "array", "required": false },
    { "name": "rename_map", "type": "object", "required": false },
    { "name": "add_audit", "type": "boolean", "default": true }
  ],
  "content": "{{/*\n  Staging: stg_{{ source_name }}__{{ source_table }}\n  Generated by DataChonk\n*/}}\n\nwith source as (\n    select * from {{ source('{{ source_name }}', '{{ source_table }}') }}\n),\n\nrenamed as (\n    select\n        {%- if columns %}\n        {%- for col in columns %}\n        {%- if rename_map and col in rename_map %}\n        {{ col }} as {{ rename_map[col] }}{% if not loop.last %},{% endif %}\n        {%- else %}\n        {{ col }}{% if not loop.last %},{% endif %}\n        {%- endif %}\n        {%- endfor %}\n        {%- else %}\n        *\n        {%- endif %}\n        \n        {%- if add_audit %}\n        -- Audit columns\n        , current_timestamp as _dbt_loaded_at\n        , '{{ source_name }}.{{ source_table }}' as _dbt_source\n        {%- endif %}\n    from source\n)\n\nselect * from renamed"
}

Using Custom Templates

Generate code with your custom template:

bash

# Use a custom template by name
datachonk generate staging \
  --template staging-with-audit \
  --source raw.stripe.payments \
  --name payments

# Or in the web app, select template from dropdown
# when creating a new Chonk

Best Practices

Version Your Templates

Store templates in your dbt project repo. Changes are tracked and reviewed like code.

Document Variables

Add clear descriptions to all variables. Your team will thank you later.

Start from Built-ins

Customize the built-in templates rather than starting from scratch. They encode dbt best practices.

Pro Tips

Template Registry

Enterprise users can create a shared template registry. Teams can publish and consume templates organization-wide, ensuring consistency across all dbt projects.

Self-Hosting

Plugins

Configuration

Code Templates

Customize how DataChonk generates dbt code to match your team's standards and conventions.

Built-in

Default templates bundled with DataChonk. Good starting point for customization.

Project

.datachonk/templates/
Project-specific templates in your repo.

User

~/.datachonk/templates/
Personal templates across all projects.

Viewing Templates

List all available templates with the CLI:

bash

# List all templates
datachonk templates

# Filter by category
datachonk templates --category staging

# Output:
# Staging Templates:
#   staging-model - Standard staging model that cleans and renames columns
#   staging-snapshot - Staging model with SCD Type 2 tracking
#
# Mart Templates:
#   dimension-model - Dimension table following Kimball methodology
#   fact-model - Fact table with foreign key joins
#   ...

Template Structure

Templates are JSON files with metadata and Jinja-like content:

my-staging-template.json

{
  "name": "my-staging-template",
  "version": "1.0.0",
  "description": "Custom staging model with audit columns",
  "category": "staging",
  "author": "Data Team",
  "tags": ["staging", "audit", "custom"],
  
  "variables": [
    {
      "name": "source_name",
      "type": "string",
      "required": true,
      "description": "Name of the source"
    },
    {
      "name": "source_table",
      "type": "string",
      "required": true,
      "description": "Source table name"
    },
    {
      "name": "columns",
      "type": "array",
      "required": false,
      "description": "Columns to select"
    },
    {
      "name": "add_audit",
      "type": "boolean",
      "required": false,
      "default": true,
      "description": "Add audit columns"
    }
  ],
  
  "content": "... (template content)"
}

Template Syntax

Templates use Jinja-like syntax for variable substitution and control flow:

Variable Substitution

-- Basic variable
select * from {{ source('{{ source_name }}', '{{ source_table }}') }}

-- With default value
{{ model_name | default: 'stg_default' }}

-- From config
{{ dbt_utils.generate_surrogate_key(['{{ primary_key }}']) }}

Conditionals

{%- if add_audit %}
    -- Audit columns
    , current_timestamp as _loaded_at
    , '{{ source_name }}' as _source
    , md5(row_to_json(source.*)::text) as _row_hash
{%- endif %}

{%- if scd_type == '2' %}
    -- SCD Type 2 columns
    , valid_from
    , valid_to
    , is_current
{%- endif %}

Loops

select
{%- for col in columns %}
    {{ col }}{% if not loop.last %},{% endif %}
{%- endfor %}
from source

-- With rename mapping
{%- for col in columns %}
    {%- if rename_columns and col in rename_columns %}
    {{ col }} as {{ rename_columns[col] }}
    {%- else %}
    {{ col }}
    {%- endif %}
{%- endfor %}

Full Example

A complete custom staging template with audit columns and renaming:

staging-with-audit.json

{
  "name": "staging-with-audit",
  "version": "1.0.0",
  "description": "Staging model with audit columns and column renaming",
  "category": "staging",
  "variables": [
    { "name": "source_name", "type": "string", "required": true },
    { "name": "source_table", "type": "string", "required": true },
    { "name": "columns", "type": "array", "required": false },
    { "name": "rename_map", "type": "object", "required": false },
    { "name": "add_audit", "type": "boolean", "default": true }
  ],
  "content": "{{/*\n  Staging: stg_{{ source_name }}__{{ source_table }}\n  Generated by DataChonk\n*/}}\n\nwith source as (\n    select * from {{ source('{{ source_name }}', '{{ source_table }}') }}\n),\n\nrenamed as (\n    select\n        {%- if columns %}\n        {%- for col in columns %}\n        {%- if rename_map and col in rename_map %}\n        {{ col }} as {{ rename_map[col] }}{% if not loop.last %},{% endif %}\n        {%- else %}\n        {{ col }}{% if not loop.last %},{% endif %}\n        {%- endif %}\n        {%- endfor %}\n        {%- else %}\n        *\n        {%- endif %}\n        \n        {%- if add_audit %}\n        -- Audit columns\n        , current_timestamp as _dbt_loaded_at\n        , '{{ source_name }}.{{ source_table }}' as _dbt_source\n        {%- endif %}\n    from source\n)\n\nselect * from renamed"
}

Using Custom Templates

Generate code with your custom template:

bash

# Use a custom template by name
datachonk generate staging \
  --template staging-with-audit \
  --source raw.stripe.payments \
  --name payments

# Or in the web app, select template from dropdown
# when creating a new Chonk

Best Practices

Version Your Templates

Store templates in your dbt project repo. Changes are tracked and reviewed like code.

Document Variables

Add clear descriptions to all variables. Your team will thank you later.

Start from Built-ins

Customize the built-in templates rather than starting from scratch. They encode dbt best practices.

Pro Tips

Template Registry

Enterprise users can create a shared template registry. Teams can publish and consume templates organization-wide, ensuring consistency across all dbt projects.