Exécuter des commandes dans un environnement sandbox

In this part of the tutorial, you execute commands inside a sandbox and capture their output. The following examples show common patterns, including error checking, timeouts, and working directories.

# Basic command
sandbox.exec(["echo", "Hello"]).result()

# Raise SandboxExecutionError if command returns non-zero exit code
sandbox.exec(["ls", "/nonexistent"], check=True).result()

# Raise SandboxTimeoutError if command exceeds timeout
sandbox.exec(["sleep", "60"], timeout_seconds=5).result()

# Run in a specific directory
sandbox.exec(["ls"], cwd="/app").result()

Parallel execution

Run multiple commands in parallel when you don’t need to wait for each one to finish before starting the next. This reduces total runtime when commands are independent.

import cwsandbox

# exec() returns a Process immediately without waiting for the command to finish
p1 = sandbox.exec(["sleep", "1"])
p2 = sandbox.exec(["sleep", "1"])
p3 = sandbox.exec(["sleep", "1"])

# Call .result() when you need the output - ~1 second total, not 3
cwsandbox.result([p1, p2, p3])

For streaming, error handling, and process control, see the Execution guide.

​Parallel execution

Parallel execution