Hacker News Official API

[goranmoomin]

Only using the official API is quite cumbersome, as fetching a story with comments requires one network request per each comment.

Instead, try considering the (more unknown) Algolia API[0]. It supports fetching the entire tree in one request. Unfortunately the comments aren’t sorted, so you’ll still have to use the official one (or parse the HN HTML) if you need that.

For my HN client[1], I parse the HN HTML (as it’s required for voting), and sort the comments from the Algolia API. It’s much faster than recursively requesting the official API.

[0]: https://hn.algolia.com/api

[1]: https://github.com/goranmoomin/HackerNews

[catnip09]

Yeah, I built a graphql API for HN for this reason. It will let you fetch all comments, users, and posts in a single request.

https://hngraphql.fly.dev/graphql

[CodesInChaos]

If you already parse the html, what's the point in calling the Algolia API at all?

[adamredwoods]

I found this interesting, looking at the Firebase API. I guess one could get the min/max item ids for the child posts, and query the range of ids (startAt(), endAt()).

Van Puffelen answers some of this here, but he doesn't get into specific queries: https://www.youtube.com/watch?v=66lDSYtyils

He also mentions here that Firestore offers a bit more for WHERE type clauses over Firebase:

https://stackoverflow.com/questions/26700924/query-based-on-...

[dang]

In response to people's complaints about the usability of the HN Firebase API: yes. We're going to eventually have a new API that returns a simple JSON version of any HN URL. At that point we'll phase out the Firebase API, with a generous deprecation period. I'd be curious to hear people's thoughts about what a generous deprecation period might be.

[somishere]

The great thing about the Firebase API is that it's live, so you can subscribe to updates to a post/comment/profile and see them as they're made (no polling, just websockets). I'd hope any future version would retain this feature.

I'm reluctant to lose it as I'm about to release a browser extension that relies on it pretty heavily :)

[Edit]: this feature is undocumented in the link above, but it's a standard feature of FB and works as expected in all client libraries https://firebase.google.com/docs/database

[MarcellusDrum]

I'd say it depends on how straightforward the new API will be. The app I use (Materialistic) is both free and ads-free. So I don't expect the developer to drop everything they are doing to update it if will require a lot of work. So 6 months minimum if everything is well documented, a year if it will be a hassle.

[arthurcolle]

I would say a full calendar year is "generous" from the announcement date

[catnip09]

Can you allow login and write operation via the API?

I want to build a HN web client but it's not possible without scrapping the site which won't scale.

[ksec]

6 to 12 months AFTER we have dark mode.

[yamrzou]

I'd say about 6 months.

[krapp]

Honestly, it would be nice if the Firebase API were updated. You could even build the former on the latter, so JSON urls pull from the API.

Just being able to get an entire thread in a single request alone, and filter queries would make things much easier.

[saagarjha]

How difficult would it be to host the same Firebase API on top of whatever will be driving the new JSON stuff?

[livinglist]

As a dev who has built a Hacker News client app, I would say at least 12 months.

[O__________O]

Wish API would show comment votes instead of only showing them up until 8 years ago when displaying them was turned off on the main site. Aside from allowing users to enable them via their on interface, do analysis, etc — it broke HN’s comment search based on comment popularity.

Related:

https://news.ycombinator.com/item?id=32535637

[chrismorgan]

I’d like to be able to conveniently access my most popular comments. There are quite a few things in there that I might like to write longer-form articles about.

(https://james.darpinian.com/blog/scraping-my-own-hacker-news... gives one way of doing it, by crawling your /threads?id=‹username› page, which does contain the point counts for each comment by you.)

[1123581321]

I don’t want vote information to be publicly available. Hiding those numbers was good for the site. Users have ways to get their own data. That’s enough.

[metadat]

What difference would exposing counts make? Why should this data be mineable?

We're already helping to create a training set of slightly above average English speaker discourse for free.

[superb-owl]

I wrote a script that will allow you to use cookies to download your comment scores [1]. I also did some analysis of my own scores [2]

[1] https://github.com/superb-owl/hacker-news-comments

[2] https://superbowl.substack.com/p/commenting-on-hacker-news

[AtroxDev]

I found the official API quite a pain to work with while building myself an alternative "hn ui"[0]. I mostly switched over to using the Algolia API[1], which is a lot more "fun" to work with. I can only recommend checking that one out as well.

[0]: https://hnhub.dev/

[1]: https://hn.algolia.com/api/

[avg_dev]

I wrote a fairly simple "Who is hiring?" browser (using this API) that fetched the post listings and stored them locally in an HTML file accessible by pointing your browser at file:///... . It supported filtering by "Remote", "Interns", and "Visa" with buttons, had an ad hoc Regex filter, and the ability to remove posts using a delete button and not see them again (or to restore all deleted posts) using the localStorage api.

It was indeed barebones but I'd like to think it helped me find a job. I would link the code (it's on GitHub) but today the "Who is hiring?" posts include a few considerably more advanced and capable "searchers" right in the top of the post:

> Searchers: try https://kennytilton.github.io/whoishiring/, https://hnjobs.emilburzo.com, https://news.ycombinator.com/item?id=10313519.

It was fun; I enjoyed working on it.

[maxique]

I'd still like to see yours if you're up to sharing! :)

[ashish01]

I wrote a simple hn data downloaded based on this api. Last I remember I was able to download the entire corpus by letting it run overnight.

https://github.com/ashish01/hn-data-dumps

[breck]

Oh this might be exactly what I need. Giving it a try now. Thanks!

[reneberlin]

Firebase is owned by google. There are plenty of OSS solutions to create this exact realtime thing without getting too close to the elephant in the room.

[easrng]

I don't think it was owned by google when hn started using it for the API.

[somishere]

Switch to Supabase?

[greatgib]

This is so awesome. I like very much the spirit of giving access to the api as-is. With no fuss.

[Asdrubalini]

I'm currently writing an HN posts score tracker (basically it plots score variations over time since they were in new for each popular post) and it's really handy that you can just subscribe with eventsource [0] and get realtime update when something changes without having to manually poll each post wasting resources. Not yet sure about the latency though.

[0] https://github.com/jpopesculian/reqwest-eventsource/

[password4321]

https://hnrankings.info/

[imnitishng]

Tiny project I built using the same API https://github.com/imnitishng/hkrnws

[banana_giraffe]

If you want to see a small example of this API being used, here's a utility to generate an RSS feed of replies to comments you post that you can self-host:

https://gist.github.com/Q726kbXuN/15e61acc003bb6d46a458001fe...

[BeyondSchwifty]

Anyone know of a good tool to get something like push notifications when specific topics come up on HN? I've been thinking maybe I should build one, but I also feel like someone must have already done it by now.

[mavci]

You can check this [1] service that I prepared for myself then made it publicly available. For now it’s only sending popular stories as notification. Planning to add keyword feature.

[1] https://hnn.avci.me/

[tailspin2019]

Don’t have any suggestions but as an alternative to notifications, I wrote a script to check the HN API for keywords such as “outage” “downtime” “breached” and show the results on a TV screen in my office.

Has been quite useful for the ops side of what I work on!

[djbusby]

Cronjob to the Algolia Search API, send to your email.

[dinkledunk]

There's been several, but they always get abandoned after a year or two.

[plibither8]

Similar to this, I created "HN Faves API", which is an API to fetch a user's favorited posts and comments on HN. Interestingly, HN doesn't have an official API for this.

On GitHub: https://github.com/plibither8/hn-faves-api

You can try it out here: https://hn-faves.mihir.ch

[adamredwoods]

I've never used Firebase before. Took a little while to dig through the docs, but finally would the web/js version for observing and get:

https://firebase.google.com/docs/database/web/read-and-write

[overshard]

I use this API in a few places such as a custom New Tab extension I made for myself on Chromium based browsers:

https://github.com/overshard/newtab/blob/master/newtab.js#L1...

Haven't had any issues with it!

[superb-owl]

If you're looking to get your comment scores, the API sadly will not oblige.

I wrote a script that uses your browser cookies to scrape them from the website here: https://github.com/superb-owl/hacker-news-comments

[shkaj]

Annoying how it doesn't show the text or anything for dead items.

Is there any way to query this API as if showdead is switched on?

[avg_dev]

I used to query flagged comments from time to time using the API, just out of some morbid curiosity. It definitely used to work for that; I am not sure if flagged items are in the same class as dead ones, though, or if they have since deliberately eliminated this capability. I eventually found more interesting things to do :)

[fileeditview]

For a few years I am working on the side on a HN client. One thing that is really sad is that there is now way to handle authentication properly as a 3rd party. The user basically has to trust you with her/his password or you have to "steal" the auth token from a login.

[wslh]

Again, it would be also great if we have a user's personal API to get our private info either.

[teleclimber]

Agreed. An read only API for my upvoted stories and comments would be great so I can preserve this data and use it outside of HN.

[O__________O]

Also be nice if per private user behavior, that a user was able to authorize API access to it without risking account lose.

[kogir]

It always amuses me when the API link in the footer floats up as a submission every few years.

[krapp]

It's almost as if a color scheme of grey on grey makes features hard to notice.