feat: auto update rss feeds

.github/workflows/update-rss.yml

@@ -0,0 +1,28 @@
+name: Update RSS Feed
+
+on:
+  push:
+    branches:
+      - master
+
+jobs:
+  update-rss:
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v2
+
+      - name: Set up Python
+        uses: actions/setup-python@v2
+        with:
+          python-version: 3.11
+
+      - name: Install dependencies
+        run: pip install -r requirements.txt
+
+      - name: Run update 
+        run: python update.py
+        env:
+          MINIFLUX_ENDPOINT: ${{ secrets.MINIFLUX_ENDPOINT }}
+          MINIFLUX_API_KEY: ${{ secrets.MINIFLUX_API_KEY }}

requirements.txt

@@ -0,0 +1,3 @@
+miniflux==1.0.1
+mistune==3.0.2
+Requests==2.31.0

update.py

@@ -0,0 +1,76 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+import miniflux
+import mistune
+import os
+import pprint
+import requests
+
+from mistune.plugins.table import table
+
+
+markdown = mistune.create_markdown(renderer='ast', plugins=[table])
+
+
+def format_rss(blogroll_url):
+    table_line = []
+    rss_urls = []
+    checked_rss_urls = []
+
+    r = requests.get(blogroll_url)
+    ast = markdown(r.text)
+
+    for elem in ast:
+        if isinstance(elem, dict) and elem['type'] == 'table':
+            for tbl_elem in elem['children']:
+                if tbl_elem['type'] == "table_body":
+                    for line_elem in tbl_elem['children']:
+                        table_line.append(line_elem['children'][2])
+    for elem in table_line:
+        raw_data: str = elem['children'][0]['raw']
+        if raw_data.startswith("http"):
+            rss_urls.append(raw_data)
+
+    for url in rss_urls:
+        print("getting {}".format(url))
+        try:
+            req = requests.get(url)
+            if req.status_code != 200:
+                print(f"{url} with status code {req.status_code}")
+                continue
+            if req.text.startswith("<?xml"):
+                checked_rss_urls.append(url)
+        except (requests.exceptions.SSLError, requests.exceptions.ConnectionError) as e:
+            print(f"{url} with exception {e}")
+            continue
+
+    return checked_rss_urls
+
+
+def add_client(valid_rss, client):
+    for url in valid_rss:
+        try:
+            feed_id = client.create_feed(url, category_id=3, crawler=True)
+            print(feed_id)
+        except (miniflux.BadRequest, miniflux.ServerError) as e:
+            print(f"{url} with exception {e}!")
+            continue
+
+
+def get_current_status(client: miniflux.Client):
+    feeds = [(feed['title'], feed['site_url']) for feed in client.get_feeds()]
+    pprint.pp(feeds)
+
+
+if __name__ == '__main__':
+
+    miniflux_endpoint = os.environ["MINIFLUX_ENDPOINT"]
+    miniflux_api_key = os.environ["MINIFLUX_API_KEY"]
+    blogroll_url = "https://raw.githubusercontent.com/NUAA-Open-Source/BlogRoll/master/README.md"
+
+    client = miniflux.Client(miniflux_endpoint, api_key=miniflux_api_key)
+
+    valid_rss = format_rss(blogroll_url)
+    add_client(valid_rss, client)
+    get_current_status(client)