Show HN: Extract an RSS feed from almost anything | Hacker News Reader

Show HN: Extract an RSS feed from almost anything

rssfeedasap.com

151 pointsroks0n54 comments

Howdy! RSSfeedASAP scratches my own itch. I run a regional podcasting directory which gets dozens of messy submission for podcasts. Often they don't even include an xml file and me being a good samaritan I sometimes do the manual work and find it myself. I got tired of that manual work and decided to build a microapp.

RSSfeedASAP is this app and I decided to release it in case someone else finds any use in it.

Open articleView on HN

Maybe I misunderstood, this tries to find RSS feeds from sites, not generate RSS feeds for sites that don't have them, right?

I've been using RSSPreview addon on Firefox for a couple years to perform this, but I really see the value of a Web service to do this anywhere. Kudos!

The primary objective was to access RSS feeds from various platforms or to create them with known structures to facilitate access to feeds, such as iTunes, SoundCloud, YouTube, and Google Podcasts. Recently, I have incorporated the ability to generate RSS feeds for Spotify shows, as they currently do not provide access to an XML file. However, this feature is limited to shows only and does not include playlists or artists.

The main focus was on accessing podcast RSS feeds, which is why the initial efforts were concentrated on these platforms. However, plans are in place to expand support to other platforms and different types of content based on user demand.

May I ask what purpose you are using the RSSPreview addon for?

> May I ask what purpose you are using the RSSPreview addon for?

Mostly, to have an RSS button in the navigation bar, but I found it useful when checking different feeds for the same page. For instance, I've recently started following some software projects on GitHub, some of which I like to follow by release, while some I like to follow by commit, for example. The preview let me see what each feed shows, and get a feel on the frequency of "news" for each feed.

Hope this can help you, I'm no podcast listener so can't really give you a hint on which podcast-related features I would look for in a news feed.

Feed43 used to be that.[0]

[0]: https://web.archive.org/web/20230326224112/https://feed43.co...

Nice, but this seems to only detect some existing RSS feeds on a given website.

* Shameless plug * Our feed generator can detect existing RSS feeds, and combines a scraper and data transformer to help create custom RSS feeds for any public webpage.

https://newsloth.com/product/feed-builder

41$ or 59$ per month to get content in RSS form rather than visiting a couple websites and that's just the basic, not the professional version? SOLD!

I appreciate the visits and people testing things out, I'm observing the logs and I've already fixed a few problems and added a few things to my todo list. \o/

There's also RSS-Bridge: https://github.com/RSS-Bridge/rss-bridge

"The RSS feed for websites missing it"

There are also a bunch of similar projects in the comments of my Show HN [1] where I shared a similar project.

[1] Show HN: RSS feeds for arbitrary websites using CSS selectors - https://news.ycombinator.com/item?id=27739568

I tried https://news.ycombinator.com and it's saying an error occurred.

Tried on https://openai.com/blog and got an error `Houston, we have a problem!.` Looks like this is just a search for existing rss feeds

Can I suggest removing recently extracted or having a way to hide whatever you extract from the home page?

Good suggestion. I see people already started getting naught ideas. Change is being deployed now.

Maybe a hand curated list of feeds that could be extracted would be better? :)

Nice work, I was able to extract out a feed URL from a Listnr exclusive podcast and can now add it to my normal podcast app without needing to use their stupid propriatary one. Win for openness!

I think your tagline is misleading.

Maybe a bit tangential, but are there any “RSS to email” tools? Meaning — whenever I update my blog, the rss tool detects it and auto-sends an email version of it (maybe a teaser snippet) to a certain mailing list (e.g from Substack)

https://github.com/rss2email/rss2email

Do you use alternate meta tag to detect the rss feed?

It uses several techniques one of which is also meta tags.

Strange. I tried it with my blog and it said there was no feed https://kevincox.ca. I definitely have a alternate link tag set up.

What it's worth my own RSS reader also fails to find any feeds from your site because my parser expects the link type to be quoted with either single or double quotes. Not sure what the offical specs say though.

  type="application/atom+xml"

This is a valid unquoted attribute. So it looks like a bug in your reader (and quite possibly the linked service as well)

https://www.w3.org/TR/2012/WD-html-markup-20120329/syntax.ht...

Ha! I've added few more mime types and now it's able to detect it. Thanks for letting me know! :)

Is there a way to use it with Google News (with all advanced search options)? Thanks

Cool project! Dovetails nicely with a recent project of mine.

Thank you! I am intrigued, what project are you currently working on? I would be interested to know if you come across any limitations or see any areas for improvement in my app. Feel free to reach out on X @asapship

https://news.ycombinator.com/item?id=37362151

This is great! I've had htmlx on my "things to look into" list for a few weeks now and funny enough I use Go on the backend of rssfeedsasap as well. :)

Thanks! I'm trying to make xtemplate into a 'Rapid Application Development' RAD tool, as one of the commenters pointed out. lmk if you feel like trying it out; @infogulch everywhere

i play a lot with broken feeds. for your tool i imagine one could examine previous or following editions of a page or the wayback machine.

If you want to create new RSS feeds from almost anything, my rssparser.lisp is still there: https://code.rosaelefanten.org/rssparser.lisp

Thanks

This is very cool! I will check it out in more detail. I appreciate your recommendation!

Looking forward to feedback! I posted it on HN a while ago, but it can't hurt to have new input.

Thanks!

Is there a docker image for this?

What exactly would be the point of shoving a script that is nothing more than a simple cronjob and that could theoretically just be compiled as a static binary into a Docker container?

Had you replied in a respectful manner, you would have gotten a reply from me with what I think are legitimate reasons for having this script in a docker image.

The required degree of respect was exhausted by the question of what the advantages were. This rampant epidemic of wanting to dump everything into some kind of "container" and "image" was quite funny for a while, but is now just annoying.

It took more energy to respond in such an abrasive way, than it would've to just ignore the comment if you found it so silly. Go take your social ineptitude elsewhere

> Go take your social ineptitude elsewhere

I hereby humbly apologise for daring to participate in the comments section of Hacker News despite my autism. Will not happen again.

(But then where are "people like me" supposed to go?).

Dude, it was only a question. You could've just said no instead of going on some weird rant. There are plenty of reasons for containerization, but it seems like engaging in any sort of conversation with you is pointless.

> You could've just said no instead of going on some weird rant.

"Dude", I could, but then a valid follow-up would have been something like "why not?", and I think that I have sufficiently covered this.

> There are plenty of reasons for containerization

"Bro", I understand that containerization would make a small amount of sense under the following circumstances:

- The software in question was a giant package with a quadrillion dependencies which all interfere with each other. rssparser.lisp is not.

- The software in question will need to be deployed on a metric ton of very different machines. rssparser.lisp can easily be deployed without this additional layer.

> it seems like engaging in any sort of conversation with you is pointless.

You haven't even tried yet. I'm open for any kind of discussion, but it would probably help both sides if the subject was "how to improve the software", don't you think?

> You haven't even tried yet.

Someone asked a simple question.

You replied with a comment that ended like so "into a fucking Docker container?"

As a third party here (and not the person you asked), it's fairly easy to assume that someone who pretends to be "Looking forward to feedback" and then responds with "fucking" and being rude is really just trolling here and not interested in reasonable and adult conversations.

At the very least, you know you were wrong, because you've been editing your comments to make it seem less aggressive.

My honest suggestion: apologize, and get offline. There is no reason to be acting like this.

I thought that being rude towards Docker containers - because I /really/ don't like the containerization of every single shell script that is happening these days - will hurt nobody's feelings. It seems that I was wrong. But now that I rephrased the comment, could we please talk about the subject now?

Actually, since I am unfamiliar with the requirements to run a lisp script I thought a docker image would provide all necessary dependencies (SBCL? Quicklisp?) without the necessity to investigating myself and installing them directly on my machine.

Secondly, after this would be obviously installed for checking and testing functionality for meeting my personal the requirements, it is much more easy to uninstall all dependencies by stopping and removing a container then manually removing all packages in case of not needing it.

Thirdly, a service like that would be installed on my home server, which separates the OS installation and other extra services by having all extra services inside docker containers keeping the OS "clean".

Makes sense. Are you familiar with how to create one? Because I’m not, but I’d volunteer to link to one.

SBCL is a package on most Linux distributions, Quicklisp is a package manager for SBCL: https://www.quicklisp.org/beta/#installation

I'm old enough to remember when browsers would highlight in the url bar that a site had an RSS feed. Well, I'm old enough to remember upgrading from Mosaic to Netscape, so I'm v old.

Cool little project.

Ironically, chrome(ihm) is starting to show that again. An RSS icon next to the URL if an RSS feed is provided.

This is a good application. However, I think that among similar products, https://github.com/DIYgod/RSSHub is a more usable choice(able to generate RSS). In addition, using rsshub with the https://github.com/DIYgod/RSSHub-Radar browser extension would be more convenient.

I recently installed self-hosted FreshRSS, an rss reader which can _create_ rss feeds from sites that don't provide them using techniques like xpath selectors.

https://www.freshrss.org

Shameless plug on something similar: https://feedreader.xyz

Comes as a bookmarklet and it will not only find a page's feed, it will also display it nicely directly.

Demo on Verge's feed: https://feedreader.xyz/?url=https%3A%2F%2Fwww.theverge.com%2... (you just visit the site, hit the bookmarklet & there you go)

This was primarily built after realizing that the official Google RSS plugin for Chrome was never gonna be updated and was "stuck" to that annoying layout while displaying a page's feed.

By examining the submissions he gets through this web form, maybe the author will discover some new podcasts.

Reply on news.ycombinator.com