n8n-nodes-python-raw

Fork of naskio/n8n-nodes-python - Python Function node for n8n with raw script execution and output.

Execute pure Python scripts in n8n and get raw execution results

Key Differences from Original

This fork provides a fundamentally different approach to Python execution in n8n:

| Original Node | This Fork (Raw) | |--------------|-----------------| | Item-by-item processing | Single script execution | | Returns transformed items | Returns execution metadata | | Uses python-fire wrapper | Direct Python execution | | Hidden stdout/stderr | Full stdout/stderr capture | | Fire-based argument passing | Direct variable injection |

Features

✅ Single Execution: Script runs once regardless of input item count
✅ Raw Output: Access to stdout, stderr, and exit codes
✅ Direct Input: All input items available as input_items variable
✅ Environment Variables: Access to custom environment variables
✅ Configurable Python: Choose Python executable path
✅ Execution Metadata: Timestamps, success status, error details

Output Format

The node returns a single item with execution results:

{
  "exitCode": 0,
  "stdout": "Output from print() statements",
  "stderr": "Error messages and warnings",
  "success": true,
  "error": null,
  "inputItemsCount": 3,
  "executedAt": "2024-01-01T12:00:00.000Z"
}

Usage Example

# Available variables:
# - input_items: list of all input data
# - env_vars: dict of environment variables

import json
import sys

# Process input data
print(f"Processing {len(input_items)} items")

results = []
for item in input_items:
    # Your processing logic here
    processed = {"original": item, "processed": True}
    results.append(processed)

# Output results (will appear in stdout)
print(json.dumps(results, indent=2))

# Exit with success
sys.exit(0)

Installation

Using npm

npm install @zgxsuerwtmrhjzt/n8n-nodes-python-raw

Using Docker

Add to your n8n Dockerfile:

RUN cd /usr/local/lib/node_modules/n8n && npm install @zgxsuerwtmrhjzt/n8n-nodes-python-raw

Local Development

See DEVELOPMENT_SETUP.md for development environment setup.

Requirements

• Python 3.6+ installed and accessible
• n8n running environment

Node Configuration

Python Code

Write your Python script directly. Available variables:

• input_items: List of input data from previous nodes
• env_vars: Dictionary of environment variables

Python Executable

Specify the Python executable:

• python3 (default)
• python
• /usr/bin/python3
• /path/to/conda/envs/myenv/bin/python

Environment Variables (Optional)

Use the PythonEnvVars credential to provide environment variables to your script.

Use Cases

Data Analysis

import json
import statistics

# Analyze input data
numbers = [item.get('value', 0) for item in input_items]
result = {
    "count": len(numbers),
    "mean": statistics.mean(numbers) if numbers else 0,
    "median": statistics.median(numbers) if numbers else 0
}

print(json.dumps(result))

External API Integration

import requests
import json

# Call external API
response = requests.get("https://api.example.com/data")
if response.status_code == 200:
    print(json.dumps(response.json()))
    sys.exit(0)
else:
    print(f"API call failed: {response.status_code}", file=sys.stderr)
    sys.exit(1)

File Processing

import os
import json

# Process files
processed_files = []
for item in input_items:
    if 'filename' in item:
        if os.path.exists(item['filename']):
            processed_files.append(item['filename'])

result = {"processed_files": processed_files}
print(json.dumps(result))

Original Project Credits

This fork is based on n8n-nodes-python by Mehdi Nassim KHODJA.

License

Apache 2.0 with Commons Clause - see LICENSE.md

Contributing

1. Fork this repository
2. Make your changes
3. Update package.json with your npm username
4. Test your changes
5. Submit a pull request

Support

For issues specific to this fork, please create an issue in this repository. For general n8n questions, refer to the n8n documentation.

Note: This is a fork with significant modifications. For the original item-processing behavior, use the original package.