Merge branch 'beanstalk'

README.md

@@ -8,6 +8,8 @@ This is based on my [Express ActivityPub Server](https://github.com/dariusk/expr
 
 This requires Node.js v10.10.0 or above.
 
+You also need `beanstalkd` running. This is a simple and fast queueing system we use to manage polling RSS feeds. [Here are installation instructions](https://beanstalkd.github.io/download.html). On a production server you'll want to [install it as a background process](https://github.com/beanstalkd/beanstalkd/tree/master/adm).
+
 ## Installation
 
 Clone the repository, then `cd` into its root directory. Install dependencies:
@@ -44,7 +46,7 @@ Go to `https://whateveryourdomainis.com:3000/convert` or whatever port you selec
 
 ## Sending out updates to followers
 
-There is also a file called `updateFeeds.js` that needs to be run on a cron job or similar scheduler. I like to run mine once a minute. It queries every RSS feed in the database to see if there has been a change to the feed. If there is a new post, it sends out the new post to everyone subscribed to its corresponding ActivityPub Actor.
+There is also a file called `queueFeeds.js` that needs to be run on a cron job or similar scheduler. I like to run mine once a minute. It queries every RSS feed in the database to see if there has been a change to the feed. If there is a new post, it sends out the new post to everyone subscribed to its corresponding ActivityPub Actor.
 
 ## Local testing
 

package.json

@@ -10,6 +10,7 @@
     "cors": "^2.8.4",
     "express": "^4.16.3",
     "generate-rsa-keypair": "^0.1.2",
+    "jackd": "^1.2.4",
     "parse-favicon": "^2.0.0",
     "pug": "^2.0.3",
     "request": "^2.87.0",

queueFeeds.js

@@ -0,0 +1,26 @@
+const Database = require('better-sqlite3');
+const db = new Database('bot-node.db');
+const Jackd = require('jackd');
+const beanstalkd = new Jackd();
+
+
+async function foo() {
+
+  // get all feeds from DB
+  let feeds = db.prepare('select feed from feeds').all();
+
+  console.log('!!!',feeds.length);
+
+  let count = 0;
+
+  await beanstalkd.connect()
+
+  for (feed of feeds) {
+    await beanstalkd.put(feed.feed)
+  }
+
+  await beanstalkd.disconnect()
+
+}
+
+foo()

updateFeeds.js

@@ -7,29 +7,37 @@ const db = new Database('bot-node.db'),
       crypto = require('crypto'),
       parser = new Parser({timeout: 2000});
 
-// get all feeds from DB
-let feeds = db.prepare('select * from feeds').all();
+const Jackd = require('jackd');
+const beanstalkd = new Jackd();
 
-console.log('!!!',feeds.length);
+beanstalkd.connect()
 
-let count = 0;
-
-doFeed();
-
-function doFeed() {
-  let feed = feeds[count];
-  console.log(count, feed.feed);
-  if (feed === undefined) {
-    return;
+async function foo() {
+  while (true) {
+    try {
+      const { id, payload } = await beanstalkd.reserve()
+      console.log(payload)
+      /* ... process job here ... */
+      await beanstalkd.delete(id)
+      await doFeed(payload)
+    } catch (err) {
+      // Log error somehow
+      console.error(err)
+    }
   }
+}
+
+foo()
+
+function doFeed(feedUrl) {
+return new Promise((resolve, reject) => {
   // fetch new RSS for each feed
-  parser.parseURL(feed.feed, function(err, feedData) {
+  parser.parseURL(feedUrl, function(err, feedData) {
     if (err) {
-      console.log('error fetching', feed.feed, err);
-      doFeed(++count);
+      reject('error fetching ' + feedUrl + '; ' + err);
     }
     else {
-      //console.log(feedData);
+      let feed = db.prepare('select * from feeds where feed = ?').get(feedUrl);
       // get the old feed data from the database
       let oldFeed = JSON.parse(feed.content);
 
@@ -72,15 +80,14 @@ function doFeed() {
         // update the DB with new contents
         let content = JSON.stringify(feedData);
         db.prepare('insert or replace into feeds(feed, username, content) values(?, ?, ?)').run(feed.feed, acct, content);
-        count = count + 1;
-        setTimeout(doFeed, 100);
+        return resolve('done with ' + feedUrl)
       }
       else {
-        count = count + 1;
-        setTimeout(doFeed, 100);
+        return resolve('done with ' + feedUrl + ', no change')
       }
     }
   });
+}).catch((e) => console.log(e));
 }
 
 // TODO: update the display name of a feed if the feed title has changed
@@ -139,10 +146,8 @@ function transformContent(item) {
   });
 
   // couple of hacky regexes to make sure we clean up everything
-  item.content = $('body').html().replace(/^(\n|\r)/,'').replace(/>\r+</,' ').replace(/  +/g, '');
-  item.content = item.content.replace(/^(\n|\r)/,'').replace(/>\r+</,' ').replace(/  +/g, '');
-  // remove whitespace nodes from in between paragraphs
-  item.content = item.content.replace(/p>\s<p/g,'p><p');
+  item.content = $('body').html().replace(/^(\n|\r)/,'').replace(/>\r+</,'><').replace(/  +/g, '');
+  item.content = item.content.replace(/^(\n|\r)/,'').replace(/>\r+</,'><').replace(/>\s*</g,'><').replace(/>\u200B+</g,'><').replace(/  +/g, '').replace(/<p><\/p>/g,'');
   return item;
 }