Skip to content

python.md

Latest commit

 

History

History
127 lines (81 loc) · 4.95 KB

python.md

File metadata and controls

127 lines (81 loc) · 4.95 KB
Raw
title Python
description Use a module from any supported technology in your Python service with Graftcode - no REST wrapper, no rewrite, no custom interop. Install it as a typed Graft and call it like any other dependency.

Goal

Use a module from any supported technology directly in a Python service with Graftcode - no REST wrapper, no rewrite, no custom interop. In this tutorial we use a .NET NuGet module as the example.

What You'll See

  • Install a module from another technology as a strongly-typed Graft using pip - just like any other package.
  • Configure the generated client to point at the in-memory host.
  • Import and call it from your Python code as if it were a native pip package.

Prerequisites

  • Python installed locally
  • .NET SDK (or a Microsoft.NETCore.App runtime) installed locally - inMemory loads this Graft on Netcore, not Node

Step 1. Create a project folder

Create a new folder for your project:

mkdir py-js-module-demo
cd py-js-module-demo

Step 2. Install a module from another technology

For this example we'll use a .NET sample library from NuGet (sdnTestSimpleCar / SimpleCar), but the same approach works with any module from a supported repository - npm, PyPI, Maven or NuGet.

python -m pip install --extra-index-url https://grft.dev/simple graft-nuget-sdnTestSimpleCar==0.1.0

This installs a Graft - a strongly-typed Python client generated from the module. You import and call it like any other pip package, regardless of which technology the module was originally written in.

Step 3. Set the SDK key

Set the HYPERTUBE_KEY environment variable in your terminal before running the project:

PowerShell:

$env:HYPERTUBE_KEY="f7GZ-De6k-Mx4p-t7FN-q5DC"

Bash (macOS / Linux):

export HYPERTUBE_KEY="f7GZ-De6k-Mx4p-t7FN-q5DC"

Windows CMD:

set HYPERTUBE_KEY=f7GZ-De6k-Mx4p-t7FN-q5DC

The first successful activation creates a hypertube.lic file. Later runs (and other local projects) can reuse that license. Activating the same key again in a clean environment may return ERROR:Key already used.

Step 4. Call the cross-language module and run it

Create main.py:

import os
from graft_nuget_sdntestsimplecar.simple_car import SimpleCar, GraftConfig

GraftConfig.host = "inMemory"

car = SimpleCar("Toyota", "Corolla", 2022, "car-001")

print(car.to_string(), flush=True)
print("Running:", car.is_running, flush=True)

# Now let's call some method!
car.start()
print("Running:", car.is_running, flush=True)
os._exit(0)

Run it:

python main.py

The Python Graft exposes snake_case members (to_string, is_running). Use flush=True before os._exit(0) so printed output is not lost when the process exits.

You should see the car description and running state printed in your terminal. SimpleCar comes from a .NET NuGet package, but your code reads like a regular method call - no HTTP request, no response parsing, no serialization. GraftConfig.host = "inMemory" tells Graftcode to load and execute the .NET assembly inside the same process.

Step 5. Run with a Project Key (recommended for real-world usage)

Everything above works without any account - perfect for learning and local development. When you're ready for real-world usage, create a free account at portal.graftcode.com, set up a project, and copy its Project Key.

With a Project Key, point GraftConfig.host at your project's stable registry URL instead of a raw WebSocket address. A Project Key gives you:

  • Stable registry URL - the address for your Grafts stays permanent, so install commands don't change when you redeploy.
  • Portal visibility - see all your gateways and services in one place at gateways.graftcode.com.
  • Access control - decide who can download your Grafts using package manager authentication and permissions.

Without Graftcode

Using a module from another language in Python typically requires:

  • Building a separate service around the module with REST or gRPC endpoints
  • Defining request and response DTOs for every operation
  • Hosting and deploying that service independently
  • Writing HTTP or gRPC client code in your Python project
  • Maintaining the integration layer as the module evolves
  • Or rewriting the module entirely in Python

With Graftcode

  • Install any public module from npm, NuGet, or PyPI as a strongly-typed Graft with one command
  • Import and call it like a regular Python dependency - no wrappers, no clients
  • The module runs in-process or on a remote host, controlled by a single configuration line

Technology choice stops being an integration constraint. You can keep writing Python and use the best libraries from any ecosystem - JavaScript, .NET, Java - as if they were native pip packages.