DataLexicon

DataLexicon is a lightweight, model-driven tool for turning logical data models written in YAML into database-ready SQL, documentation, and ER diagrams.

It is designed for data engineers, analysts, and architects who want to define their data model once and generate consistent outputs for multiple database dialects.

Why DataLexicon?

DataLexicon helps teams:

• keep data models in version control
• generate SQL automatically from logical models
• standardize naming conventions and abbreviations
• generate documentation and ER diagrams
• support multiple dialects such as PostgreSQL, SQL Server, Oracle, and Snowflake

Key features

• YAML-based logical modeling
• Generate SQL DDL for multiple dialects
• Primary key, foreign key, and index generation
• Table and column comments from model descriptions
• Naming configuration via standards/naming.yml
• Docs generation with HTML output and Mermaid ER diagrams
• Works well in Git-based workflows and CI/CD pipelines

How DataLexicon works

1. Create a project folder.
2. Place your logical models in a models directory as YAML files.
3. Add a datalexicon.yml file with your project settings.
4. Optionally configure naming and abbreviations in a standards directory.
5. Run the build command to generate SQL into the target folder.
6. Run the docs command to create an HTML report with the ER diagram.

Suggested project structure

my-project/
  models/
    customer.yaml
    order.yaml
    country.yaml
  standards/
    naming.yml
    abbreviations.yml
  datalexicon.yml
  target/

Example model file

entity: Customer
description: Customer master data
attributes:
  - name: CustomerId
    type: identifier
    required: true
    primary_key: true
    description: Unique customer identifier
  - name: CustomerName
    type: text
    description: Customer display name
relationships:
  - entity: Order
    type: one-to-many
    source_attribute: CustomerId
    target_attribute: CustomerId

Minimal configuration

name: my_project
models_dir: models
standards_dir: standards
dialects:
  - postgres

Naming configuration

You can customize primary key, foreign key, and index naming in standards/naming.yml:

naming:
  primary_key:
    prefix: PK_
    suffix: _PK
  foreign_key:
    prefix: FK_
    suffix: _FK
  index:
    prefix: IX_
    suffix: _IDX

Commands

Initialize your project

Start by creating a new DataLexicon project using the CLI. The initialization command creates a ready-to-use project structure.

datalexicon init .

Creates the recommended folder structure:

my-project/
  models/
  standards/
  datalexicon.yml
  target/

Afterwards, you can create your entities, attributes and relationships using simple YAML files stored in the models directory.

Build SQL

datalexicon build .

This creates SQL files in the target folder per selected dialect.

Validate a model

datalexicon validate models/customer.yaml

Generate documentation

datalexicon docs .

This creates HTML documentation with logical model sections, generated SQL, and an ER diagram.

VS Code extension

The DataLexicon VS Code extension adds commands directly inside VS Code for building models and generating documentation from the open workspace.

License

Copyright © 2026 NvisionData. All rights reserved.

DataLexicon is licensed under the DataLexicon End User License Agreement (EULA).

The EULA defines the permitted use, restrictions, intellectual property rights, and licensing terms for DataLexicon.

For the full license agreement, see:

https://www.nvisiondata.be/products/eula.html