能否通过Python脚本执行Allrun脚本？是否可在Python中模拟终端环境运行命令？

Absolutely! You can definitely run your OpenFOAM Allrun script from a Python script—Python has great tools for spawning terminal commands and interacting with system processes. Here are the most common and reliable methods to do this on Ubuntu:

Using the subprocess Module (Python Standard Library) #

The subprocess module is Python's built-in tool for running external commands, and it's perfect for this use case. It lets you mimic terminal behavior, including setting working directories, capturing output, and handling environment variables (critical for OpenFOAM).

1. Basic Run (with Environment Setup) #

OpenFOAM relies on specific environment variables to work correctly, so you'll need to load its environment before running the Allrun script. Here's how to do it in one command:

import subprocess

# Replace with your actual OpenFOAM case directory
case_directory = "/path/to/your/openfoam/case"

# Command to load OpenFOAM env and run Allrun (adjust the OpenFOAM path to match your installation)
run_command = "source /opt/openfoam10/etc/bashrc && ./Allrun"

# Execute the command
result = subprocess.run(
    run_command,
    cwd=case_directory,  # Set working directory to the case folder
    shell=True,          # Required to run multiple commands with &&
    stdout=subprocess.PIPE,  # Capture output
    stderr=subprocess.PIPE,  # Capture errors
    text=True            # Return output as string instead of bytes
)

# Print results for debugging/verification
print("=== Allrun Output ===")
print(result.stdout)
if result.stderr:
    print("\n=== Errors/Warnings ===")
    print(result.stderr)
print("\nReturn Code:", result.returncode)  # 0 = success, non-zero = error

2. Run with Isolated Environment Variables #

If you want to avoid using shell=True (for security reasons, if your script handles untrusted input), you can first capture OpenFOAM's environment variables and pass them to the subprocess:

import subprocess
import os

# Capture OpenFOAM environment variables
env_capture_cmd = "source /opt/openfoam10/etc/bashrc && env"
env_result = subprocess.run(
    env_capture_cmd,
    shell=True,
    stdout=subprocess.PIPE,
    text=True
)

# Parse environment variables into a dictionary
openfoam_env = {}
for line in env_result.stdout.splitlines():
    if "=" in line:
        key, value = line.split("=", 1)
        openfoam_env[key] = value

# Merge with system environment variables (optional but recommended)
openfoam_env.update(os.environ)

# Run Allrun with the prepared environment
case_directory = "/path/to/your/openfoam/case"
result = subprocess.run(
    ["./Allrun"],  # No shell=True needed here
    cwd=case_directory,
    env=openfoam_env,
    stdout=subprocess.PIPE,
    stderr=subprocess.PIPE,
    text=True
)

# Process output the same way as before
print("=== Allrun Output ===")
print(result.stdout)

3. Real-Time Output (For Long-Running Cases) #

If your Allrun script takes a long time to execute, you might want to see the output in real-time instead of waiting for it to finish:

import subprocess

case_directory = "/path/to/your/openfoam/case"
run_command = "source /opt/openfoam10/etc/bashrc && ./Allrun"

# Spawn the process
process = subprocess.Popen(
    run_command,
    cwd=case_directory,
    shell=True,
    stdout=subprocess.PIPE,
    stderr=subprocess.STDOUT,  # Merge errors with output
    text=True,
    bufsize=1,  # Line-buffered output
    universal_newlines=True
)

# Print output line by line as it's generated
for line in process.stdout:
    print(line, end='')  # end='' avoids duplicate newlines

# Wait for the process to finish and get return code
process.wait()
print("\nAllrun completed with return code:", process.returncode)

Key Notes #

• OpenFOAM Path: Adjust the path to bashrc (e.g., /opt/openfoam9/etc/bashrc) to match your installed OpenFOAM version.
• Working Directory: Always set cwd to your case folder—OpenFOAM scripts rely on being in the correct directory to find case files.
• Security: Avoid shell=True if your script uses user-provided input (it can lead to command injection). Use the environment variable method instead in that case.

内容的提问来源于stack exchange，提问作者user842424