Pingsville Ping

A tiny Deno server that:

1. Accepts XML-RPC weblogUpdates.extendedPings,
2. Parses the RSS/JSON feed, then
3. Saves blogs, posts, and tags to a database

Set up

MongoDB

1. Install MongoDB:

   brew install mongodb-community@7.0
   

2. Run as a macOS service:

   brew services start mongodb-community@7.0
   

Deno

1. Install Deno:

   brew install deno
   

Pingsville Ping

1. Run server:

   deno task dev
   

Send test ping

Postman

Using Postman, you can send a POST request to localhost:1417 with the following raw XML in the body:

<?xml version="1.0"?>
<methodCall>
  <methodName>weblogUpdates.extendedPing</methodName>
  <params>
    <param>
      <value>
        <string>Example Blog</string>
      </value>
    </param>
    <param>
      <value>
        <string>https://blog.example.com/</string>
      </value>
    </param>
    <param>
      <value>
        <string>https://blog.example.com/feed/</string>
      </value>
    </param>
  </params>
</methodCall>

The Deno task should log the values to the console, and add an entry to the blogs database collection, if the URL is not a duplicate.

If a feed URL is provided, it will grab the feed XML/JSON and add all the entries to a database collection called blogPosts.

If these posts have categories or tags, those will be saved to a tags collection as well.

WordPress

You can also try this with real-world data from a local WordPress install (I recommend wp-now) by adding localhost:1417 to the list of Update Services in wp-admin/options-writing.php.

Viewing records

1. Open mongo Shell (mongosh)
2. List databases: show databases
3. Switch to the Pingsville database: use pingsville
4. Show collections: show collections

Blogs

1. Show documents in blogs collection: db.blogs.find()
2. Clear all blogs from the collection: db.blogs.deleteMany({})
3. Count the number of blogs: db.blogs.countDocuments({})

Blog Posts

1. Show documents in blogPosts collection: db.blogPosts.find()
2. Clear all blogs from the collection: db.blogPosts.deleteMany({})
3. Count the number of blogs: db.blogPosts.countDocuments({})
4. Find all blog posts in reverse chronological order: db.blogPosts.find().sort({ pubDate: -1 })
5. Find all blog posts with a particular tag: db.blogPosts.find({tags: ObjectId("XXXXXXXXXXXXXXXXXXXXXXXX")})
6. Find all blog posts from a particular blog: db.blogPosts.find({"blog.id": ObjectId("XXXXXXXXXXXXXXXXXXXXXXXX")})

Tags

1. Show documents in tags collection: db.tags.find()
2. Clear all tags from the collection: db.tags.deleteMany({})
3. Count the number of tags: db.tags.countDocuments({})

Cleanup

1. Remove all instances of tag from all blog posts: db.blogPosts.updateMany({}, { $pull: { "tags": { id: ObjectId("XXXXXXXXXXXXXXXXXXXXXXXX") }}})

To Do

• Normalize URLs (remove trailing slash, expand short URLs, etc.)
• Look into using @foxglove/xmlrpc
• Better error handling
• Use GraphQL, to be database-agnostic
• Prevent spamming?
• A job queue?
• Create Pingsville API, for querying data
• Create Pingsville Web, for browsing content
• Scrape feed data and store blog posts
• Store tags
• Normalize tags
• Return something meaningful to the pinger (spec)

Resources

• XML-RPC Spec
• Rebooting XML-RPC
• XML-RPC Debugger
• davexmlrpc
• WordPress Update Services
• What happened to Technorati?

How you can help

Please help me! I don't know what I'm doing!!

You can reach me on Mastodon at @jsit@social.coop

License

Boy, I haven't really thought about this.