noaa/app/app.py

import logging
import os
import re
import time
from datetime import datetime

import requests
from flask import Flask, Response, render_template, request, send_from_directory
from prefect.deployments import run_deployment

PORT = 9021
PROTO = "https"
BARE = True

app = Flask(__name__)

logging.basicConfig(level=logging.DEBUG)


def deploy_name():
    return datetime.utcnow().strftime("%Y-%m-%dT%H:%M:%S") + "Z"


def get_host():
    host = os.environ.get("LIGHTNING_CLOUDSPACE_HOST")
    if host is None:
        default_host = os.environ.get("HOST_NAME", "0.0.0.0")
        return f"{default_host}:{PORT}" if not BARE else f"{default_host}"
    else:
        return f"{PORT}-{host}" if not BARE else f"{host}"


@app.route("/iframe")
@app.route("/iframe/")
@app.route("/iframe/<path:subpath>")
def home(subpath="images/animations/", proto=PROTO):
    host = get_host()
    initial_url = f"{proto}://{host}/{subpath}"
    api_url = f"{proto}://{host}/api"
    return render_template(
        "index.html", initial_url=initial_url, host=f"{proto}://{host}", api_url=api_url
    )


@app.route("/api", methods=["POST"])
def handle_api():
    data = request.json  # This assumes you're sending JSON data.
    url = data.get("url")
    if not url.endswith("/"):
        url += "/"

    logging.debug(f"Received URL: {url}")
    params = {"url": url, "limit": 24 * 60, "ext": None}
    response = run_deployment(
        name="create-animations/noaa-animate",
        parameters=params,
        flow_run_name=f"{deploy_name()}.webapp.{url}",
    )
    # response is a FlowRun - need to get what we want from it.

    # Process the data as needed.
    return {
        "status": "success",
        "message": f"{url} processed successfully",
        # "response": response,
    }, 200


@app.route("/videos/<path:filename>")
def custom_static(filename):
    return send_from_directory("../out", filename)


@app.route("/", methods=["GET"])
@app.route("/<path:url>", methods=["GET"])
def proxy(url="", proto=PROTO):
    original_base_url = "https://services.swpc.noaa.gov"
    host = get_host()
    proxy_base_url = f"{proto}://{host}/"

    target_url = f"{original_base_url}/{url}"
    logging.debug(f"Fetching URL: {target_url}")

    try:
        headers = {
            "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/58.0.3029.110 Safari/537.36"
        }
        response = requests.get(target_url, headers=headers, stream=True)
        excluded_headers = [
            "content-encoding",
            "content-length",
            "transfer-encoding",
            "connection",
        ]
        headers = [
            (name, value)
            for (name, value) in response.raw.headers.items()
            if name.lower() not in excluded_headers
        ]

        if "text/html" in response.headers.get("Content-Type", ""):
            content = response.content.decode("utf-8")
            content = re.sub(r"'http://", "'https://", content)
            content = re.sub(
                r"https?://services.swpc.noaa.gov", proxy_base_url, content
            )

            content = content.replace(
                "</body>",
                f"""
                <script>
                    window.addEventListener('load', function() {{
                        var hasSufficientImages = false;
                        var observer = new MutationObserver(function(mutations) {{
                            mutations.forEach(function(mutation) {{
                                if (mutation.type === 'childList') {{
    								console.log('observer');
                                    checkImages();
                                }}
                            }});
                        }});
                        observer.observe(document.body, {{ childList: true, subtree: true }});

                        function checkImages() {{
                            var links = document.querySelectorAll('a');
                            var imageLinkRegex = /\.(jpg|jpeg|png)(\\?.*)?(#.*)?$/i;
                            var numImages = Array.from(links).filter(link => imageLinkRegex.test(link.href)).length;
                            console.log('Number of eligible links:', numImages);
                            hasSufficientImages = numImages >= 60;
                            window.parent.postMessage({{ type: 'urlUpdate', url: '{original_base_url}/{url}', eligible: hasSufficientImages }}, '*');
                        }}

                        // Run initial check in case content is already loaded
   						console.log('initial');
                        checkImages();
                    }});
                </script>
                </body>""",
            )
            content = content.encode("utf-8")
            return Response(content, status=response.status_code, headers=headers)
        else:
            return Response(
                response.content, status=response.status_code, headers=headers
            )

    except Exception as e:
        logging.error(f"Error fetching URL: {e}")
        return Response(f"Error fetching URL: {e}", status=500)


if __name__ == "__main__":
    app.run(host="0.0.0.0", port=9021, debug=True)