TamedPy

Tamed Python - Sandboxed execution of arbitrary Python code

Usage

from tamedpy.sbc import SandboxedContext
with SandboxedContext() as sbc:
  code = 'print(2**4)'
  result = sbc.execute(code)
  print(result)

Placing input files into Sandbox

sbc.execute(code, ['path/to/inputfile1', 'path/to/inputfile2',])

Reading files back from Sandboxed execution

result = sbc.execute(code)
op = result.readFile('output.txt')

Reading stderr from Sandboxed execution

result = sbc.execute(code)
print(result.stderr())

Specify number of workers

with SandboxedContext(num_workers=6) as sbc:
  # do stuff

Setup

System Dependencies: python2.7, virtualenv, pip, Docker

Steps

Clone and cd TamedPy

In a python 2.7 env / virtualenv, install requirements.
pip install -r requirements

Build docker image
docker build -t tamedpy tamedpy/container/

Start using from within a python script / shell by importing tamedpy.driver.Driver or tamedpy.sbc.SandboxedContext

Schematic Diagram