snapserve 1.3.0

Turn your Python functions, classes, objects, and variables into remotely accessible artifacts with zero boilerplate.

SnapServe

Turn your Python functions, classes, objects, and variables into remotely accessible artifacts with zero boilerplate.

SnapServe is a lightweight library for exposing Python functions, classes, objects, and variables as remotely accessible artifacts in a single command line of code without writing API boilerplate. Whether you want to build a simple microservice, share a machine learning model, or provide a remote interface to your Python code, SnapServe makes it easy to expose and use Python functionality over the network.

📦 Install

pip install snapserve

🚀 Quick Start

Define your Python artifacts

Create a file with anything you want to expose: functions, classes, objects, or variables:

# calculator.py

def add(a: float, b: float) -> float:
    return a + b

class Calculator:
    def __init__(self, bias: float = 0.0):
        self.bias = bias

    def multiply(self, a: float, b: float) -> float:
        return a * b + self.bias

calc = Calculator(bias=2.0)

z = 10

Serve them without messing with API boilerplate

Expose anything with a single command:

snapserve serve calculator \
--expose add,Calculator,calc,z \  # List of artifacts to expose (comma-separated)
--host localhost \
--port 8000

Use them remotely

Connect from any Python script:

from snapserve import Remote

with Remote("http://localhost:8000") as remote:
  # 🔹 Functions: Call remote functions like local ones
  result = remote.add(5, 3)
  print(result)   # → 8

  # 🔹 Classes: Instantiate and use remote classes
  calc = remote.Calculator(bias=1.0) # Instantiate with custom arguments
  result = calc.multiply(5, 3)
  print(result)   # → 16 (5 * 3 + 1.0)

  # 🔹 Objects: Use pre-initialized remote objects
  result = remote.calc.multiply(5, 3)
  print(result)   # → 17 (5 * 3 + 2.0)

  # 🔹 Variables: Access shared variables directly
  print(remote.z) # → 10
  remote.z = 20   # Remote variable can be updated and will reflect across all clients
  remote.z += 10  # Remote variable can be updated with operations
  print(remote.z) # → 30
# After the with block, the instantiated objects created on the server will be automatically cleaned up.

🛠️ CLI Commands

Serve Python functions, classes, objects, and variables as an API.

Usage: snapserve serve [OPTIONS] MODULE_PATH
Arguments:
  MODULE_PATH  The module path to serve.
Options:
  --expose STRING         Comma-separated list of artifacts to expose (functions, classes, objects, variables).
  --host STRING           The host to bind the server to. [default: localhost]
  --port INTEGER          The port to bind the server to. [default: 8000]
  --workers INTEGER       The number of worker threads to handle requests. [default: 2 × CPU cores]
  --max-concurrency INTEGER  The maximum number of concurrent requests the server can handle. [default: 100]
  --timeout INTEGER       The request timeout in seconds. [default: 30]
  --allow-cache           Enable caching of function results to improve performance for repeated calls with the same arguments.
  --cache-size INTEGER    The maximum number of entries to store in the cache. [default: 10000]
  --daemon                Run the server as a background daemon process.

List all running snapserve servers.

Usage: snapserve ps

Stop a running snapserve server.

Usage: snapserve stop [OPTIONS] SERVER_ID
Arguments:
  SERVER_ID  The ID of the server to stop.
Options:
  --all                   Stop all running servers.
  --delete                Delete the server after stopping it.

Start a stopped snapserve server.

Usage: snapserve start [OPTIONS] SERVER_ID
Arguments:
  SERVER_ID  The ID of the server to start.
Options:
  --host STRING           The host to bind the server to.
  --port INTEGER          The port to bind the server to.