FlareSolverr XML Bridge - Transparently proxy RSS request through FlareSolverr

app.py

import os
from flask import Flask, request, Response
import requests
import logging

# Force all requests to be IPv4
requests.packages.urllib3.util.connection.HAS_IPV6 = False

app = Flask(__name__)
FS_URL = os.getenv("FS_URL", "http://foobar.local:8191/v1")

logging.basicConfig(level=logging.INFO)
logger = logging.getLogger('waitress')
logger.setLevel(logging.INFO)


@app.route('/get')
def proxy():
    # Use: http://bridge-domain:5191/get?mime=text/html&url=https://site.com
    # target URL better to be url-encoded to ensure querystring are properly parsed

    target_url = request.args.get('url')
    if not target_url:
        return "Missing url parameter", 400

    expected_mime = request.args.get('mime')
    if not expected_mime:
        expected_mime = 'text/html'

    payload = {
        "cmd": "request.get",
        "url": target_url,
        "session": "bridge-session",
        "maxTimeout": 60000
    }

    try:
        fs_handle = requests.post(FS_URL, json=payload)
        fs_response = fs_handle.json()
        logging.info("FS request.get: " + fs_handle.text[:350])
    except Exception as e:
        return str(e), 500

    if expected_mime != "application/xml":
        return Response(
            fs_response['solution']['response'],
            status=fs_response['solution']['status'],
            mimetype=expected_mime
        )
    else:
        # re-request XML directly from original website as FS returns false XML output
        xml_response = request_true_xml(fs_response)
        return Response(
            xml_response.text,
            status=xml_response.status_code,
            mimetype=expected_mime
        )


def create_fs_session():
    # Create a session for future use
    payload = {
        "cmd": "sessions.create",
        "session": "bridge-session"
    }

    try:
        request = requests.post(FS_URL, json=payload)
        logging.info("FS sessions.create: " + request.text)
    except Exception as e:
        logging.error(e)


def request_true_xml(fs_response):
    logging.info("Request is XML, re-retrieve raw data")
    url = fs_response['solution']['url']
    headers = {
        "User-Agent": fs_response['solution']['userAgent'],
        "Accept": "application/xml"
    }
    cookies = {}
    for cookie in fs_response['solution']['cookies']:
        cookies[cookie["name"]] = cookie['value']

    response = requests.get(url, headers=headers, cookies=cookies)
    return response


if __name__ == '__main__':
    # create or refresh FS session to ensure it's usable later
    create_fs_session()

    # and start server
    from waitress import serve

    serve(app, host='0.0.0.0', port=5191)

docker-run.sh

#!/bin/sh

docker run -d \
  --name flaresolverr-xml-bridge \
  -p 5191:5191 \
  -e FS_URL="http://<your-fs-address:port>/v1" \
  --restart unless-stopped \
  python:3-slim \
  sh -c "apt-get update && apt-get install -y curl && \
         mkdir /app && cd /app && \
         curl -sSL https://gist.github.com/x-magic/990568de4cd72890e663aeb687c4dfe9/raw/requirements.txt -o requirements.txt && \
         curl -sSL https://gist.github.com/x-magic/990568de4cd72890e663aeb687c4dfe9/raw/app.py -o app.py && \
         pip install --no-cache-dir -r requirements.txt && \
         python app.py"

requirements.txt

flask
requests
waitress