Skip to content

Latest commit

 

History

History
64 lines (38 loc) · 2.73 KB

hop.md

File metadata and controls

64 lines (38 loc) · 2.73 KB
Raw

The hop command - The GitOps package manager for half_orm

[WIP][alpha] This work is in progress and subject to major changes.

hop is a command-line tool provided with the half_orm library. It allows you to manage the full lifecycle of your PostgreSQL/Python projects, following a GitOps approach.

Overview

The hop tool distinguishes between two environment modes: development and production. It offers a Git-based workflow for the development and deployment of your half_orm projects.

hop workflow

Initializing a hop project

To create a new hop project from an existing PostgreSQL database, use the new command:

hop new <database_name> [--devel]

The --devel option is important. It allows adding specific tables and views to the database model, which are necessary for version tracking with hop.

Development

Once the project is initialized, you can start working on your data model and Python code in a hop directory.

Version management

Version management with hop is done incrementally by preparing successive "releases":

  • hop prepare [-l <level>] [-m <message>]: Prepares a new release (level: patch, minor, or major)
  • hop apply: Applies the changes for the release in preparation
  • hop release [--push]: Validates and pushes the release in preparation

You can also:

  • hop undo: Cancel the changes for the release being prepared

Workflow example

  1. hop prepare -l patch -m "Bug fixes"
  2. Modify the SQL files in Patches/<version>/...
  3. Modify your Python code
  4. hop apply to apply the changes
  5. git add/commit ...
  6. hop release --push to validate and push the release

Production

In a production environment, hop allows you to deploy and manage different versions of your project.

  • hop upgrade: Updates the database to the latest published release
  • hop restore <version>: Restores the database to a specific release version

Automated testing

Before validating a new release with hop release, the tool automatically runs the tests defined in your project (using pytest by default). This ensures the integrity of your code and data model at each step.

Additional resources

With hop, you benefit from a powerful tool to manage the full lifecycle of your half_orm projects, following modern development best practices such as continuous integration, automated testing, and continuous deployment.