The JSON Feed specification mentions that the author (v1.0) and authors (v1.1) fields should contain objects, for instance:

{
    "author": {
        "name": "Brent Simmons",
        "url": "http://example.org/",
        "avatar": "https://example.org/avatar.png"
    }
}

The feed produced by hnrss.org uses author as a string:

{
    "author":  "Brent Simmons"
}

Need RSS feed for https://news.ycombinator.com/best

Exclude results when link matches paywall domains in a text file where the binary runs.

domains.txt

paywall1.com
paywall2.com
paywall3.com

Not just paywalls, there are other reasons to exclude domains.

RFC 4685 provides a way to represent threaded comments in an Atom Feed.

Potential use-cases:

• Republishing comments using Microformat in-reply-to.
• Supporting W3C [WebMention]{https://www.w3.org/TR/webmention/) spec.

For me I'd like to export my own comments and republish them to my Known instance. In the future it would be interesting to see if Bridgy could support the reverse use-case.

Hi,

The docs describe a way to disable the description tag but doesn't seem to work, example:
https://hnrss.org/best?description=0

The returned XML still contains the description tag, is this the expected behaviour?

Thanks

@pageman on Twitter:

HN has http://news.ycombinator.com/classic - does HNRSS support this? Like: https://hnrss.org/classic

Doesn't look like either Algolia or the Firebase API have this, so we're limited. But would be a pretty neat addition.

For example I would like to follow:
https://news.ycombinator.com/from?site=lwn.net
and some other sites.

In <description>, & (and friends) are getting turned into &amp and everything looks all screwy.

If Algolia is escaping stuff properly, should be able to mark it as safe and everything should just work.

Filters would be brilliant - maybe regexs to filter content?

E.g. having 6 out of 7 uber articles for the last few weeks hurts my soul - I would frankly be happy to remove all .uber. articles, and plenty other "hacker" news.

Project looks rad thanks bunches!

Something like http://hnrss.org/newest?count=50.

Maybe cap at 100 or so to prevent overloading the Algolia server.

Hi there - It looks like the frontpage feed does not have any new articles after September 1, 2021? Thanks.

hnrss is using obsolete time zone syntax. GMT is defined obsolete in rfc2822. So this format: Tue, 07 Jul 2015 11:37:52 GMT is obsolete and some parsers (such as strptime(3) in libc) failing to parse this format.

Please use +0000 instead of GMT in rss.py:77

One of the feeds I created is for C++, but this gives me the result for "C " which will include C. "C++" and C++ both gave me "C ".

https://news.ycombinator.com/item?id=23779297

Currently, the channel-wide <link> element always points to "https://news.ycombinator.com/".

This should be changed to better reflect what the RSS feed represents.

For example, RSS feeds for a given thread should point to the thread URL. RSS feeds for a given user should point to the user's page URL, etc.

I wonder whether it's possible to search for foo or bar and not how it's default for foo and bar.

Of course it's possible to use two feeds and let the feed reader filter out, but it would be cooler to have this directly in the feed :)

Suggestion via email:

I'm writing you to suggest the addition of an option to have the
points value show directly in the feed item's title. I already use the
points argument in the feed URL but since I like to have a quick look
even to news whose title doesn't interest me, if the have a very high
score, it would be useful having this feature to avoid having to
always peek at the item body.

I would like to have a feed for comments made in a thread of a specific user.

The replies feed lists comments on comments by a user, but not comments on a thread by a user.

Maybe the replies feed can be improved to include these as well?

I think some char specification may be missing from the RSS feed.

curl https://hnrss.org/best  
<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Hacker News: Best</title><link>https://news.ycombinator.com/best</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v1.1-7-g101d6c0</generator><lastBuildDate>Mon, 21 Dec 2020 11:28:47 +0000</lastBuildDate><atom:link href="https://hnrss.org/best" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[Paper about Herbalife-related patient death removed after legal threats]]></title><description><![CDATA[
<p>Article URL: <a href="https://scienceintegritydigest.com/2020/12/20/paper-about-herbalife-related-patient-death-removed-after-company-threatens-to-sue-the-journal/">https://scienceintegritydigest.com/2020/12/20/paper-about-herbalife-related-patient-death-removed-after-company-threatens-to-sue-the-journal/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=25492038">https://news.ycombinator.com/item?id=25492038</a></p>
<p>Points: 320</p>
<p># Comments: 54</p>
]]></description><pubDate>Mon, 21 Dec 2020 04:06:02 +0000</pubDate><link>https://scienceintegritydigest.com/2020/12/20/paper-about-herbalife-related-patient-death-removed-after-company-threatens-to-sue-the-journal/</link><dc:creator>computer23</dc:creator><comments>https://news.ycombinator.com/item?id=25492038</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=25492038</guid></item><item><title><![CDATA[More challenging projects every programmer should try]]></title><description><![CDATA[
<p>Article URL: <a href="https://web.eecs.utk.edu/~azh/blog/morechallengingprojects.html">https://web.eecs.utk.edu/~azh/blog/morechallengingprojects.html</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=25489879">https://news.ycombinator.com/item?id=25489879</a></p>
<p>Points: 521</p>
<p># Comments: 165</p>
]]></description><pubDate>Sun, 20 Dec 2020 21:53:23 +0000</pubDate><link>https://web.eecs.utk.edu/~azh/blog/morechallengingprojects.html</link><dc:creator>pavehawk2007</dc:creator><comments>https://news.ycombinator.com/item?id=25489879</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=25489879</guid></item><item><title><![CDATA[Journalists Hacked with Suspected NSO Group iMessage ‘Zero-Click’ Exploit]]></title><description><![CDATA[
<p>Article URL: <a href="https://citizenlab.ca/2020/12/the-great-ipwn-journalists-hacked-with-suspected-nso-group-imessage-zero-click-exploit/">https://citizenlab.ca/2020/12/the-great-ipwn-journalists-hacked-with-suspected-nso-group-imessage-zero-click-exploit/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=25489298">https://news.ycombinator.com/item?id=25489298</a></p>
<p>Points: 531</p>
<p># Comments: 212</p>
]]></description><pubDate>Sun, 20 Dec 2020 20:36:22 +0000</pubDate><link>https://citizenlab.ca/2020/12/the-great-ipwn-journalists-hacked-with-suspected-nso-group-imessage-zero-click-exploit/</link><dc:creator>arkadiyt</dc:creator><comments>https://news.ycombinator.com/item?id=25489298</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=25489298</guid></item><item><title><![CDATA[FFmpeg is 20 years old today]]></title><description><![CDATA[
<p>Article URL: <a href="https://twitter.com/ffmpeg/status/1340698413143224320">https://twitter.com/ffmpeg/status/1340698413143224320</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=25487711">https://news.ycombinator.com/item?id=25487711</a></p>
<p>Points: 431</p>
<p># Comments: 90</p>

I've been using your feed via Slack for the last few months and it was working great until recently. Lately, new articles show up only occasionally and there are many articles which show up in the RSS feed but which don't make it into Slack.

I reached out to Slack Support to see what was up. They're not sure as to the exact reason for the issues I'm seeing but they did point out that the feed does not validate.

I'm not sure if making the feed valid will fix my issues but I thought it was worth mentioning nonetheless.

Thanks for providing this service. 😄

Items go back to 2015. Something doesn't seem right.

Hi, I was wondering if there's a undocumented way to get a feed for a given user's list of favorite posts.

The /replies feed only considers replies to my comments, but what about comments on my submissions?

Is it intentional that the entries for "Ask HN" have no stats like upvotes, comments and so on?
I don't know whether other categories have stats or not, just curious whether there's a reason for this :)

I would like to cut down my daily HN reading and instead consume a list of articles once per week. Sort of like a weekly HN magazine.

cperciva has had a weekly top list for some time https://www.daemonology.net/hn-weekly/

It's notable because it uses a fuzzy notion of week, such that a post that is posted right before the deadline that later goes on to receive enough votes will still be featured the next week.

What I want is the ability to configure the number of posts in this list. cperciva's is fixed to 10.

Is this possible with hnrss'?

To the extent possible, every endpoint here (https://github.com/minimaxir/hacker-news-undocumented#all-public-urls-with-user-generated-content) should have a matching endpoint on hnrss.

Refs: https://news.ycombinator.com/item?id=23779892

Algolia allows to exclude keywords from search with the "-" sign ("firefox -mozilla").

HNRSS being a wrapper for Algolia, it should allow the same, but https://hnrss.org/newest?q=firefox-mozilla returns results for "firefox AND mozilla" while https://hnrss.org/newest?q=firefox+-mozilla returns no results.

Can this be implemented? I can try if I'm pointed in the right direction.

https://news.ycombinator.com/item?id=15293578

The first code block is:

  $ dig crates.io @ns-a1.io +norecurse

  ; <<>> DiG 9.11.2 <<>> crates.io @ns-a1.io +norecurse
  ;; global options: +cmd
  ;; Got answer:
  ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 18874
  ;; flags: qr; QUERY: 1, ANSWER: 0, AUTHORITY: 3, 
  ADDITIONAL: 1

  ;; OPT PSEUDOSECTION:
  ; EDNS: version: 0, flags:; udp: 4096
  ;; QUESTION SECTION:
  ;crates.io.			IN	A

  ;; AUTHORITY SECTION:
  crates.io.		86400	IN	NS	dns3.easydns.ca.
  crates.io.		86400	IN	NS	dns1.easydns.com.
  crates.io.		86400	IN	NS	dns2.easydns.net.

  ;; Query time: 83 msec
  ;; SERVER: 194.0.1.1#53(194.0.1.1)
  ;; WHEN: Wed Sep 20 15:33:13 CEST 2017
  ;; MSG SIZE  rcvd: 127

But renders as:

I'm not sure why it's happening, but whenever I try and perform a search, an error saying "Invalid JSON received from Algolia" happens.
Here's a for instance:
https://hnrss.org/newest?q=windows
PS, I hope I'm filing this in the right place.

If you subscribe to replies to your own posts, the replies are devoid of context, so it's usually several clicks to get to the post referenced by your reply.

This can also be solved with a direct link to the parent post in the content, just what is done now for the frontpage, for example.

this is a python app, how do I get Hacker news feed url?

This is request for rss feed for https://news.ycombinator.com/best

It's the same request as https://github.com/edavis/hnrss/issues/25 which was closed by its author for an unknown reason.

Love this service! Until recently, I was using this for HN RSS. One thing I liked about those feeds was that the description just contained a link to the comment feed. This worked well with my RSS reader: I could read the article from the link, view the comments from the description.

Would it be possible to add a flag to HNRSS allowing the description to instead link to the comments?

Hi!
First of all, thanks for hnrss, it is awesome! :)
I just discovered that there is the count parameter, but it seems like it doesn't affect the alternative feeds at all.
Here is an example:

https://hnrss.org/best?count=5

It still gives back 30 feeds (the default value)

Sorry if the title doesn't make much sense. I have built a pipeline for automatically scraping new hackernews comments and I'm wondering how often I can call the server to check for new comments. I'm currently checking for new comments every 10 seconds, in which I make a new call using python's feedparser to check for any new comments not included in the last call. It seems like the parser only returns a maximum of 20 new comments every ~5/6 minutes. I'd like to get comments more frequently, and I'm wondering if I am reaching the limit on the number of calls to the server.

Otherwise, this is a great project and it has been very useful for my work. Thank you for all you've been doing!

What do you think about the idea of adding a context link to comment items? Or possibly this would be a query parameter?

https://github.com/edavis/hnrss/blob/849d3f5f7323e0b9f390921e6b1c8c2790cd9341/rss.py#L67-L71

The idea is that when viewing someone's comment, you often need to go up two or three levels of parent before you can actually see what is going on.

Concretely, this changes suggests adding a link (or maybe the default?) which would take you to https://news.ycombinator.com/item?id=17031306#17034806 instead of https://news.ycombinator.com/item?id=17034806

After reading through the Octotree hubbub, I should probably figure out the licensing situation for hnrss...

To anybody reading this: Any easy to follow guides out there on how to license the code behind something like hnrss.org?

Hello,

Thanks for your implementation, it really looks great!
I know this is more a general Linux/Apache question, but since I'm not that deep into Linux, I dare to ask....

I have setup hnrss on my Linux web server, it's running but only with internal access via 127.0.0.1.
I'd like to make the service available for external access, to integrate it into 3rd party news aggregators like Flipboard.

Do you have any instructions at hand, how to do this?
(I'd like to do this with my own installation of hnrss, in order to be able to increase the hard coded maximum of items to retrieve from the default 100 to a greater number)

I tried setting a IPTable rule to map external port 5000 to internal port 5000, but that did not work.

Any help is welcome!
Thanks in advance.

Sascha.

For the past few days I'm seeing many entries in the users feed show the title of the Hacker News story in the item title as just "". For example, in the feed https://hnrss.org/user?id=kibwen, the feed now contains an entry like this:

<item>
<title>
<![CDATA[ New comment by kibwen in "" ]]>
</title>
<description>
<![CDATA[ <p>Are we leaping to conclusions by assuming this is referring to an impossible burger/beyond burger sort of product? A "plant-based" burger could also just be the sort of plain old veggie burger that e.g. Burger King has been offering for decades. Is there more detailed information somewhere?</p> ]]>
</description>
<pubDate>Mon, 09 Nov 2020 22:52:28 +0000</pubDate>
<link>https://news.ycombinator.com/item?id=25040947</link>
<dc:creator>kibwen</dc:creator>
<comments>https://news.ycombinator.com/item?id=25040947</comments>
<guid isPermaLink="false">https://news.ycombinator.com/item?id=25040947</guid>
</item>

Please consider adding a feed preview.

When you open an RSS feed directly in a browser, it will show the raw XML by default [...]. That is not a pleasant way to preview a feed, even if you know XML. And to the rest of us, feeds are not directly accessible at all. Raw XML looks like gibberish to almost everybody.

You could display a simple preview by adding the one line below or, of course, you could create a custom preview. But please consider adding one. You can find more context at https://github.com/frontaid/feed-ui

+ <?xml-stylesheet type="text/xsl" href="https://raw.githubusercontent.com/frontaid/feed-ui/main/preview.xsl"?>
<rss version="2.0">
  <!-- your feed content -->
</rss>

Is it possible to add link preview like below

Hi,

Great work on this project. I am curious if you have any plans to add a feed for these threads: who is hiring and seeking freelancer? I would personally find this very useful.

I was just looking over the code for this app and I'm pretty sure I could implement it myself. I just wanted to make sure I wouldn't be stepping on your toes if I did. Do you accept PRs?

Thanks!

Let's Encrypt should work just fine for this

Hey, thanks for this service!
Some feed readers which will opt to use the GUID of an item as the item's link (even if there is a link specified). I believe this is because the RSS 2.0 spec says that the isPermalink attribute of GUID is true by default (and overrides link even if it's specified - though that's a debated area from the few searches I've done).

The feed reader in question is Thunderbird's built-in RSS reader-
Original bug which had them change this behavior: https://bugzilla.mozilla.org/show_bug.cgi?id=327378

Feedly also seems to use the GUID instead of the link and I believe Google Reader used to use it as well before it closed down (thought they may have changed the behavior just before they closed).

Would it be possible to either change the element to or to add an option to append something to the URL so it changes it?

Not at all a critical problem -- using ?link=comments gives you something not too far off but just thought I'd throw it out there -- thanks!

Right now hnrss.org/ redirects to the old docs site at https://edavis.github.io/hnrss/.

Update it so hnrss.org/ redirects to the new docs site at https://hnrss.github.io/.

(right now edavis.github.io/hnrss/ does a <meta> redirect to hnrss.github.io so it "works" but is clunky.)

https://hnrss.org/frontpage

the last story u see is of September 19, 2020, 1:57 PM