What's the rationale behind picking 1001 as the status code? I suppose you wanted something outside the usual range so it wouldn't overlap with any other potentially meaningful status code

#TIL aussi ! Une idée d'use case ?

Six on Linux*, but should be about the same on other systems

Kind of. In x86-64, the first six arguments to a function call are handled using registers. Any extra arguments are pushed onto the stack, which can lead to a stack overflow if there are too many.

Observed on Firefox, Chrome, and Safari. The consistent behavior makes me think there might be an explanation in the standard, though I couldn't find anything

inb4, yes, this is invalid HTML as you're not supposed to have an end tag for a void element. What has been puzzling me is that the wrong end tag is usually just ignored, except for the <br> element, which ends up repeated

We had a use case for this at work today. Hardcoding the list of void elements is not ideal since it may change over time. The above is the first solution I came up with. If anyone has a better idea, I'm all ears!

La table ASCII est une merveille de design que C permet d'exploiter pleinement

What do you think is better about XPath?

Malheureusement, j'ai testé et elles ne sont clairement pas à la hauteur de leur nom 😞

tweet-qui-a-mal-vieilli% en seulement 3 h, c'est un nouveau record !

Contexte : www.vatican.va/content/leo-...

HTML imports était dépendant du JavaScript pour injecter le contenu au bon endroit. En fin de compte des APIs comme fetch permettent déjà d'en faire autant. C'est apparemment l'une des raisons pour lesquelles ç'a été abandonné, si on en croit ce billet de Mozilla : hacks.mozilla.org/2015/06/the-...

Quand j'essaye d'imaginer l'implémentation, je pense que je comprends pourquoi ça n'existe toujours pas. Comment fetcher les includes ? De façon synchrone ? Ça impliquerait de bloquer le parsing. Asynchrone ? Layout shift.

C'est probablement plus compliqué qu'il n'y paraît 😕

That would be really neat indeed. I guess one of the hurdles to its implementation might be how you load the included parts. Is it synchronous? Then it'll block the parser until the subpart is fetched. Asynchronous? Then what about layout shift? Odds are it'll also undo some parsing optimizations

You might find this page interesting 🙂
unicode.org/emoji/charts...