Skip to main content

Generate type-safe communication protocols for embedded systems from a single YAML specification

Project description

main lint and test docs build Discord Code style: black

Stop writing serialization code twice. Stop debugging protocol mismatches.

Avlos generates type-safe communication protocols for embedded systems from a single YAML specification.

  • Write once → Generate C firmware code, Python clients, docs, and CAN databases
  • Optimized for constrained systems → Minimal overhead, no dynamic allocation
  • 🔒 Type-safe across the boundary → Catch errors at build time, not runtime
  • 🎯 Battle-tested → Production-proven in Tinymovr motor controllers

Αυλός (Avlόs) - flute, also channel.

Documentation

🌟 Why Avlos?

For Embedded Systems

  • Minimal footprint for resource-constrained devices (microcontrollers, motor controllers, sensors)
  • No dynamic memory allocation in generated C code
  • Efficient binary serialization optimized for bandwidth-limited channels
  • Built-in validation with checksum for spec matching

For Development Teams

  • Single source of truth prevents client/firmware drift
  • Type-safe communication catches errors at compile time, not runtime
  • Automatic documentation generation for all endpoints
  • Physical unit handling built-in (volts, amps, celsius, radians, etc.) via Pint
  • Multiple output formats (C, C++, Python, RST docs, CAN DBC) from one spec

Production Ready

Originally developed as the communication layer for Tinymovr, a high-performance motor controller running on ARM Cortex-M microcontrollers. Now used in production embedded systems where reliability and efficiency are critical.

💡 Illustrative Use Case

Let's make a protocol to control a toaster. First we generate a spec file containing the structure we want the toaster to expose:

name: toaster
remote_attributes:
- name: sn
    dtype: uint32
    getter_name: toaster_get_sn
    summary: The unique device serial number.
- name: heater
  remote_attributes:
  - name: temperature
      dtype: float
      unit: celsius
      getter_name: toaster_get_heater_temp
      summary: The toaster heater temperature.
- name: relay
  remote_attributes:
  - name: relay_state
      dtype: bool
      getter_name: toaster_get_relay_state
      setter_name: toaster_set_relay_state
      summary: The toaster heating relay element state.

Given the above, Avlos can generate the following:

  • C implementation of the spec, to be included in the device firmware. The implementation offers data validation, (de-)serialization, getter/setter function calls, and a entry function to call with channel data. The actual comms channel implementation is left to the user.

  • A Python object reflecting the spec, to be used in the client. The object includes data validation, (de-)serialization, units integration and pretty presentation. The root node needs a comms channel to realize communication with a remote device.

  • RestructuredText-based documentation for each endpoint.

  • CAN DBC file (CAN database), for every endpoint, for use with CAN-based comm channels.

In addition, Avlos will compute a checksum for the spec and add it as a variable to the implementation so that it can be retrieved by the client for comparing client and device specs.

The output location, as well as many other attributes of the files are flexible and easily configurable.

🎁 Installation

pip install avlos

⚙️ Project Configuration

Device Spec

The Device Spec is a YAML file that defines how the device is structured. It consists of a tree-like structure. For an example of Spec file see the tests/definition/good_device.yaml file.

Output Config

The output config defines the output modules that will be used and their options. Example, showing C code generation for embedded devices:

generators:
    generator_c:
        enabled: true
        paths:
            output_header: outputs/header.h
            output_impl: outputs/header.c
        header_includes:
        - src/header.h
        impl_includes:
        - src/test.h

⚡ Usage

Ensure a device spec and an output config exist in the current folder.

avlos from file device.yaml

This will generate the outputs according to the configuration in the output config file.

Example Project

A complete project example using Avlos is available at example/. Note that all the output paths defined in the output config are relative to that file. In contrast, includes are parsed as is.

💎 Available Generators

  • generator_c: C Embedded Code
  • generator_cpp: C++ Client Code
  • generator_rst: RST-based Docs
  • generator_dbc: CAN Bus Database Format

In addition, the object resulting from the deserialization of the spec can be used as a Python object for RPC, by supplying a channel as follows:

import yaml
from avlos import deserialize
from myProject import myChannel # update this

device_description = ...
obj = deserialize(yaml.safe_load(device_description))
obj.set_channel(myChannel())

The myChannel class is a subclass of BaseChannel that needs to implements recv, send, max_ep_id, max_packet_size and serializer. The abstract BaseChannel class is defined in channel.py.

📝 Various Notes

Avlos offers:

  • A simple straightforward tree structure description, sufficient for most device types out there
  • A flexible templating system with several built-in generators, and a simple unassuming system to extend
  • Tight integration with physical units through the Pint module.

Avlos does not offer:

  • An implementation of the comms channel, this is left to the user.

  • Segmentation of data into packets (this is planned)

  • The Avlos_Command enum is structured so as to be compatible with CAN bus RTR field (i.e. 0 -> write, 1 -> read)

  • Even though Avlos generators generate a protocol hash for both device-side (as a variable) and client-side implementations (as an object attribute), the way the hash is retrieved/checked/enforced is not included. This is due to the fact that each comms channel may implement different means of performing the above.

📦 Versioning

Avlos uses git tags for version management via setuptools-scm. Version numbers are automatically derived from git tags.

To release a new version:

  1. Commit your changes
  2. Tag the release: git tag v0.X.Y
  3. Push the tag: git push origin v0.X.Y

Between releases, development versions are automatically generated (e.g., 0.8.7.dev3+g1234abc).

🔑 License

MIT

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

avlos-0.8.9.tar.gz (138.6 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

avlos-0.8.9-py3-none-any.whl (34.2 kB view details)

Uploaded Python 3

File details

Details for the file avlos-0.8.9.tar.gz.

File metadata

  • Download URL: avlos-0.8.9.tar.gz
  • Upload date:
  • Size: 138.6 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.12.12

File hashes

Hashes for avlos-0.8.9.tar.gz
Algorithm Hash digest
SHA256 6fdccba1ec70c2b0720d77b65f8d27123fe0275453b60864d1de1f6c48acbb3f
MD5 4828a18fad38d27a76b060f515f46c1c
BLAKE2b-256 8afe4fff901e314261079aa468ec9bdebf3f9ae261c1725620fa8bed0ff266f4

See more details on using hashes here.

File details

Details for the file avlos-0.8.9-py3-none-any.whl.

File metadata

  • Download URL: avlos-0.8.9-py3-none-any.whl
  • Upload date:
  • Size: 34.2 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.12.12

File hashes

Hashes for avlos-0.8.9-py3-none-any.whl
Algorithm Hash digest
SHA256 c4c84f936ff318437b5fa2ddbd7335329a66697205457d5d4f1b06f607941bb4
MD5 6b71eaadfe6b8f39cb44bbbf38d1c25c
BLAKE2b-256 8d9fef1c9b0f998cac014975fa56c9468e656c0ebb267633c179271a06bc0c37

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page