spx-python 1.0.1a1

A Python library for managing spx server.

spx-python — Quick‑Start Guide

Lightweight Python wrapper for the SPX server — model, simulate and test devices through the SPX REST API.
Designed for effortless use in local development, CI pipelines, and unit‑testing suites.

---

Table of Contents

1. Requirements
2. Installation
3. Running an SPX server
4. Connecting from Python
5. Common operations (step‑by‑step)
6. Full‑system example
7. Using the client in unit tests
8. LLM usage docs
9. CI integration (GitHub Actions)
10. FAQ
11. License

---

Requirements

Requirement	Notes
Python ≥ 3.7	Officially tested on 3.9 – 3.12
Docker & Docker Compose	To launch spx‑server locally
SPX_PRODUCT_KEY	Export in your shell or set as a CI secret

---

Installation

pip install spx-python                         # or add to poetry/requirements.txt

If you develop the package locally:

poetry add --dev spx-python pytest pytest-cov coverage

---

Running an SPX server

1. Create docker‑compose.yml (minimal):

services:
  spx-server:
    image: simplephysx/spx-server:latest
    ports: ["8000:8000"]
    environment:
      SPX_PRODUCT_KEY: ${SPX_PRODUCT_KEY}
    healthcheck:
      test: ["CMD-SHELL", "curl -f http://localhost:8000/ || exit 1"]
      interval: 10s
      timeout: 5s
      retries: 5

2. Start it:

export SPX_PRODUCT_KEY="YOUR_REAL_KEY"
docker compose up -d         # server listens on :8000

---

Connecting from Python

import os, spx_python
from spx_python.client import SpxClient

client: SpxClient = spx_python.init(
    address     = "http://localhost:8000",   # default
    product_key = os.environ["SPX_PRODUCT_KEY"],
)

client now represents the root of the SPX system and behaves like a dictionary plus convenience helpers.

---

Common operations (step‑by‑step)

1 · Inspect an empty system

print(client.keys())      # ['models', 'instances', 'timer', 'polling']

2 · Add a model

client["models"]["TemperatureSensor"] = {
    "attributes": {"temperature": 25.0, "heating_power": 0.0}
}

3 · Create an instance

client["instances"]["sensor"] = "TemperatureSensor"
sensor = client["instances"]["sensor"]

4 · Read / write an attribute

temp_attr = sensor["attributes"]["temperature"]
print(temp_attr.internal_value)      # 25.0
temp_attr.internal_value = 30.0      # update

5 · Delete things

del client["instances"]["sensor"]
del client["models"]["TemperatureSensor"]

---

Full‑system example

The unit‑tests (tests/test_full_spx_python.py) show a PID loop built from three models (TemperatureSensor, PowerSupply, PIDController).
Load the YAML, create instances, wire connections, then run prepare() and run() to watch values propagate.

---

Using the client in unit tests

def test_attribute_roundtrip():
    key = os.environ["SPX_PRODUCT_KEY"]
    w   = spx_python.init(product_key=key)

    w["models"]["Foo"] = {"attributes": {"x": 1}}
    w["instances"]["foo1"] = "Foo"
    inst = w["instances"]["foo1"]

    inst["attributes"]["x"].internal_value = 42
    assert inst["attributes"]["x"].internal_value == 42

---

LLM usage docs

If you are using an LLM to integrate spx-python in another project, start here:

• docs/SPX_PYTHON_LLM.md

---

CI integration (GitHub Actions)

env:
  SPX_PRODUCT_KEY: ${{ secrets.SPX_PRODUCT_KEY }}

steps:
  - uses: actions/checkout@v3
  - uses: actions/setup-python@v4
    with: { python-version: "3.10" }

  - run: |
      python -m pip install poetry
      poetry install

  - run: docker compose up -d        # start server
  - run: |                           # wait until healthy
      for i in {1..10}; do
        curl -fs http://localhost:8000/ && break
        echo "waiting…"; sleep 5
      done
  - run: |                           # run unit tests + coverage
      poetry run python -m unittest discover -s tests
  - if: always()
    run: docker compose down

---

FAQ

Question	Answer
Why not use str|None type‑hints?	Optional[str] keeps compatibility with Python 3.7 – 3.8.

---

License

SPX‑Python is released under the MIT License – see LICENSE.