rss-directory

Rss from directory listing

Run as microservice in container

docker run --rm -d -p5000:5000 -v ~/YourPodcast:/pub/ netpascal0123/rss-directory

By default, this application will list all podcast file in the /pub/ directory and publish a RSS report on port 5000

Run as wsgi service

docker run --rm -d -p3031:3031 -v ~/YourPodcast:/pub/ netpascal0123/rss-directory uwsgi --ini uwsgi.ini

Sample configuration to run with a reverse proxy

You must forward host and original protocol to backend

   location /podcast/ {

       proxy_pass         http://127.0.0.1:5000/;
       proxy_set_header X-Forwarded-Host $host/podcast;
       proxy_set_header X-Forwarded-Proto $scheme;
   }

This is similar if you prefer the wsgi backend

    location /podcast {
        # solve lot of problem caused by
        # - uwsgi's manage-script-name
        # - nginx uwsgi_modifier1 30;
        rewrite /podcast(.*) $1 break;

        include uwsgi_params;
        uwsgi_pass backend:3031;
        uwsgi_param X-Forwarded-Host $host/podcast;
        uwsgi_param X-Forwarded-Proto $scheme;
    }
    

You must have a clean document root for the backend

        uwsgi_param DOCUMENT_ROOT /pub/;

Or build

sudo apt-get install libssl-dev
CFLAGS="-I/usr/local/opt/openssl/include" LDFLAGS="-L/usr/local/opt/openssl/lib" UWSGI_PROFILE_OVERRIDE=ssl=true venv/bin/pip3 install uwsgi -I --no-cache-dir

git clone https://github.com/PascalNoisette/rss-directory.git
pip install -r requirements.txt
python ./directoryrss.py 

Batch

You can also generate the rss file in command line, outside a server context

python ./background.py 

Use broker instead of thread

If the broker url is supplied in the environment variable BROKER celery will be used instead of a thread. In that case workers must also be started

version: '3'

services:
  backend:
    build: .
    volumes:
      - "~/YourPodcast:/pub/"
    depends_on:
      - rabbitmq
    user: uwsgi
    command: sh -c "python -u ./directoryrss.py & python ./venv/bin/celery -A background worker --loglevel=debug"
    environment:
      - BROKER=pyamqp://guest@rabbitmq//
    ports:
      - 5000:5000

  rabbitmq:
      image: rabbitmq