README.md

feedcreator

• This project aims to turn any website to RSS Feed which we can then subscribe to using RSS readers.
• This link explains what are RSS feeds pretty well.

A little Theory

• The following is structure of a simple RSS feed.

<rss version="2.0">
    <channel>
        <title>CareerVault</title>
        <link>https://www.careervault.io/remote?category=software</link>
        <description>Careervault Jobs Feed</description>
        <item>
            <title>Senior Software Engineer</title>
            <link>
                https://careervault.io/remote/beamery-10955/software/senior-software-engineer-remote-london-uk-5157167
            </link>
            <description></description>
            <pubDate>Thu, 17 Oct 2024 10:27:41 GMT</pubDate>
            <guid>
                https://careervault.io/remote/beamery-10955/software/senior-software-engineer-remote-london-uk-5157167
            </guid>
        </item>
    </channel>
</rss>

• Going through above we will know that to create RSS Feed we mainly need two elements: title and link. description is optional
• Now, a webpage consists of elements like <li>, <a>, <article>, <div class="panel">. class attributes are used to apply css to a bunch of elements together. They also helps in uniquely identifying elements on the webpage along with ids
• So, for creating RSS feed of any website we need to identify items with their css selector i.e class attribute on a webpage which can give us the title and link.
• On the form you will find following 4 fields :
  • itemSelector : for selecting the item from which we can get title and link.
  • titleSelector: for selecting the title from the item
  • linkSelector : for selecting the link
  • descSelector : for selecting the description element

Example of selecting items for creating RSS feed

• This has 4 items in list format. In a day these items keep getting updated and we can use RSS feeds to track them.
• A single item is something like :

• To identify the tag and class associated with the list item we can right-click on the list title and select inspect
• The result is shown below:

• itemSelector would be span with class attribute titleline which will be written as span.titleline
• titleSelector will be a
• linkSelector will be also a
• Now we can fill the below form

• Title is the title of the feed with which you would like to track the feed. Here it could be like - HackerNews Feed
• URL is the link of the feed. Here https://news.ycombinator.com/newest
• Description is optional but could be something to describe the feed.
• Name is something to uniquely identify the feed one is tracking. It could be a short name like hn
• Now we need to fill the extractor parameters we identified earlier. From the above example:
  • For item selector we have to use span.titleline
  • For title selector we have to use just a.
  • For link selector we have to use just a.
• After filling the form, it should look like this:

• Once we submit it we will get a page which will have the feed url with other details and will look like this:

• Copy the hn.xml link - http://localhost:8080/static/rss/hn.xml and add it to any of the RSS Feed readers for tracking

QuickStart

• make init to setup the database and download the dependencies.
• make run to start the application locally.
• make build to build the application locally.
• make clean to purge the contents of the database
• Don't forget to add the path to your browser executable for path flag. The default now is Brave Browser in Mac

Design & Implementation

• This project is using colly library for scraping the webpage.
• The feeds are saved under ui/static/rss directory and there is a feeds.db table where we save feed metadata.
• Two functions are scheduled at a configurable interval in main.go. Using the metadata it rescans all urls to update or clean the feed items.
• For cleaning, the scheduler runs every 90 minutes and it deletes feeds which are more than 3 days old which can be changed.
• For updating, the scheduler runs every 60 minutes.
• For loading dynamic websites it uses chromedp

Hosting

• There is a github workflow which creates a docker image for self-hosting.

Remarks

• Few feeds along with their metadata is included.
• Currently, we rescan all the feeds at a fixed interval. We can optimize it to scan each site at a fixed interval.
• We are also not passing request headers like If-Modified-Since or checking response headers like Last-Modified.