ash-jc-allen / favicon-fetcher

A Laravel package for fetching favicons from websites.

Hey, I'm Ash Allen! 👋

I'm a freelance Laravel web developer from Preston, UK

• 💥 I build systems and websites for small businesses using Laravel.
• 📚 I have a blog that focuses on helping other developers learn more about Laravel, PHP and web development.
• 🖥 I have contributed to Laravel and other open-source projects.
• 🎉 I currently have 5 Laravel packages that are being used in commercial systems.
• ✍️ I write about Laravel, PHP and other web developments topics on the Ash Allen Design blog.
• ✨ I wrote the FREE 70-page ebook called The Clean Coder's Guide to Laravel.
• ⚡ I'm open to collaborating on other Laravel projects.
• 🥅 2022 Goal: Contribute to more open-source projects and write a new e-book.

I Write Guest Blog Posts

• Laravel News - How to get a Websites Favicons in Laravel
• Honeybadger - An Introduction to Caching in Laravel
• Honeybadger - A Guide To Error & Uptime Monitoring In Laravel Using Honeybadger
• Honeybadger - A Complete Guide To Managing User Permissions In Laravel Apps

I Was Featured on Laravel News!

• Laravel Exchange Rates API Package
• Simplify Installing and Updating your App With Laravel Executor
• Short URL Package for Laravel Applications

Languages and Tools

---

⚡ My Open Source Projects

• Laravel Exchange Rates
• Laravel Executor
• Short URL
• Laravel Config Validator
• Laravel Mailbox Layer
• Favicon Fetcher

---

📘 Some of My Blog Posts

• Cleaning Up Laravel Controllers
• Interfaces vs Abstract Classes in PHP
• Using Interfaces to Write Better PHP Code
• Using the Strategy Pattern in Laravel
• How to Make Your Laravel App More Testable

---

favicon-fetcher's Issues

How do we handle HTTP timeouts?

Hi, I'm using the http driver here for fetching favicon. When a favicon can't be retrieved, the HTTP driver seems to wait for a long time which is slowing down my front-end, any chance we can customise the timeout?

HttpDriver findLinkElement method

Hello,

currently findLinkElement() function pattern is
$pattern = '/<link.*rel="(icon|shortcut icon)"[^>]*>/i';

however this does not work in all cases, because it might assume that this is favicon link:

when we remove dot and asterisk (.*), then it works much better. However, I can not test all possible cases.
$pattern = '/<link rel="(icon|shortcut icon)"[^>]*>/i';

Any ideas why this pattern is how it is?

Call to a member function cache() on null

Hello,

If I run the following code which looks like it should work:

use AshAllenDesign\FaviconFetcher\Facades\Favicon;
Favicon::fetch("https://example.com/")->cache(now()->addWeek())->getFaviconUrl();

It will crash with:

[Error] Call to a member function cache() on null

How to handle cURL error 6: Could not resolve host

Hi! First of all, thanks for this great package.

I'm building service which handles 500k+ domains and there are many domains that are not available. When this happens, we will get cURL error for "Could not resolve host".

Any suggestions how to handle this error so that we will get favicon as null?

Thanks!

Wrong file extension when storing the favicon.

Hi thanks for this nice package, but I think there is a problem.
When storing the favicon in the storage I got this result :

$favicon = Favicon::fetch('https://headlessui.com')->store('favicons', 'public');
// 6f886546-edc2-4041-9a5c-8803f236ac25.com

$favicon = Favicon::fetch('https://vuejs.org')->store('favicons', 'public');
// favicons/bf4d32f3-fd86-4696-9a01-482d810da7f9.org

$favicon = Favicon::fetch('https://remote.io')->store('favicons', 'public');
// favicons/d0e9c6d4-2cd5-4abd-ac53-2fcad9ff817a.io

The file extension follows the URL domain. that's should not be the case, it should be .png or other image formats.

Doesn't handle urls that redirects to path

When the Uri provided to the Scan function responds with a 301 or 302 redirect, It's not working when the redirect location is a path.

Specifically, I'm having an issue where my local Jellyfin instance is at 192.168.x.x:x, but that address returns a 302 with a location of /web/index.html. The favicons are then publicly available at 192.168.x.x:x/web/asset.ico. However, the ScanResults will have a location of 192.168.x.x:x/asset.ico (missing the web path). The fetch is also returning null, which I believe has the same root issue.

Redirects seem to be handled correctly when it's a different domain (for example, https://youtu.be has some scan results for youtube.com).

Issue on parsing from element

Hi, I have tried this package for favicons but on some website have got error, i.e if I try to get icon for that site https://www.smartsupp.com/

Possibility to convert ico to png image

Hello,

just idea that would be nice to have image file to display on html page. Or is it possible already somehow?