So does the website itself, though, right?
this post was submitted on 20 Jul 2025
-23 points (27.5% liked)
Programming
22147 readers
9 users here now
Welcome to the main community in programming.dev! Feel free to post anything relating to programming here!
Cross posting is strongly encouraged in the instance. If you feel your post or another person's post makes sense in another community cross post into it.
Hope you enjoy the instance!
Rules
Rules
- Follow the programming.dev instance rules
- Keep content related to programming in some way
- If you're posting long videos try to add in some form of tldr for those who don't want to watch videos
Wormhole
Follow the wormhole through a path of communities !webdev@programming.dev
founded 2 years ago
MODERATORS
No, favicons are usually encrypted and not publically visible. You have to connect your website to the google favicon service for it to be shown to users. /s
Last night, somewhere in Mountain View, Sundar Pichai sat bolt upright in bed, realizing that there was a possible profit center yet untapped...
I know that KeepassXC has an automatic favicon downloader built-in and it's open source. Maybe check out it's code.
Why would you ever want to route your request like this through a third party especially an ad company? To get the favicon of a site you just request www.example.com/favicon.ico.
Not necessarily. You need to fetch the HTML of the web page you want the icon for and see if there's a <link rel=icon> or equivalent HTTP header.
And yes, this means different pages on the same site can have different icons.
@argv_minus_one@mastodon.sdf.org @FrostyPolicy@suppo.fi @Pro@programming.dev
AFAIK there's also a special file name Safari uses for bigger PNG icons. That Google thing probably just facilitates all this machinery, but I agree — I wouldn't rely on Google for such things myself.
It doesn’t have to be an *.ico file. So you might need to try different file extensions.
True. I've just seen requests to that file in my demo env for web apps I have and I've seen my browser request that file by itself when running those locally.
So you might need to try different file extensions.
No, you don't. The HTTP response header will tell you what type it is. Anything using file name suffixes to determine content type on the web (unless it's just a fallback guess) is broken.
So you might need to try different file extensions.
No, you don’t. The HTTP response header will tell you what type it is.
I’m not sure what you’re talking about. You need to know the name of the file before you can download it.
For instance, my lemmy client, says in its HTML that its favicon is <link id="favicon" rel="shortcut icon" type="image/x-icon" href="https://sopuli.xyz/pictrs/image/9c6eeb58-bf66-4a15-9537-0a822f3c4feb.png">. If I were to blindly download /favicon.ico, I’d naturally get a 404 page:
$ curl https://sopuli.xyz/favicon.ico -i
HTTP/2 404
# more data
You need to know the name of the file before you can download it.
Indeed. And /favicon.ico might not be a Windows ICO file, but instead be a PNG, GIF, or some other image format. I thought that's what you were (correctly) pointing out when you wrote "it doesn’t have to be an *.ico file". In such cases, the HTTP Content-Type field tells what image format it is.
If I were to blindly download /favicon.ico, I’d naturally get a 404 page:
Ah, so it turns out you were thinking of cases where /favicon.ico doesn't exist at all. That can also happen, but your suggestion to "try different file extensions" is not the answer, as you can see if you try to curl /favicon.png, /favicon.gif, etc. The correct approach is to parse a web page's HTML in search of a favicon URL, which you did in your above reply, but that's not the same thing as what you originally suggested.
duckduckgo, icon.horse and favicongrabber.com also have their own APIs for this
I bet there is a way this could be abused in order to proxy/tunnel arbitrary requests via google IPs. Even if it's slow and must be done through e.g. PNG image data and dynamic subdomains or something, it would still be possible.
What makes this URL an API?
Do they disclose it as free use anywhere?
I would be more interested in themed favicon sets like on Android or Linux DE's, and any way that would avoid Google or lock-in/forced updates to such as with the Play store or Windows I would avoid like the plague.
Gimme back .dll's full of icons I can create and edit to my liking. For modernization, meta-text(url or application name) for each icon instead of referencing "icon #x in this .dll", and a format more usefull than .dll's full of icons that all have to be the same size(and .tiff, .giff, or bitmaps at that, when jpeg and vector formats exist), but that's about it.
All this re-inventing-the-wheel/web2.0/3.0 shit is geared towards creating dependency and lock-in in the stupidest ways possible.
I mean, on F-Droid and Linux are plenty of icon sets. SVG too.
Indeed.