Avatar
Eric A. Meyer
6a32e47d258bb692f15a085dbdcb60ed381bb2921af6b4db1107a7ea545109cc
Armchair rocket scientist graffiti existentialist. Old enough for it to hurt when I stand up, not old enough to have stopped noticing. Agnostic in principle, atheist in practice, #CSS guy, mzungu. I’m only here for the food.

Yeah, I think I’ll stick with this videochat background for a while.

nostr:npub1pepws2dru2yyhe55e0v9d0va4hsdrmf708jfwf4vjxwxmmpq02hsqjnts5 I’d like to know where Google got that information, given Polaris is not nearly that far from us (~450 light years by the latest estimates).

nostr:npub1hvdfn5ne7zvks3ztfewr00vdjv3kve4nddht43p8fjepc8rn6yasz69sq9 I just read the article and nothing like that quote appears in it.

Just came across another instance where an algorithm was inadvertently cruel — incredibly so, this time.

And so it goes.

This was a fun and fascinating conversation with nostr:npub1a48z5ak88sr9e39qppahrs7ysaw7y3408wnqt5dwnwwnrwf90exqwhwfsc about undersea internet cables, and I think it’s well worth an hour of your time this weekend as you do chores or run errands or what have you. https://floss.social/@igalia/110872055046442348

Replying to Avatar Dave Rupert

nostr:npub1c353w465eg94sjctq7kh03f9zajjqqsg9wzmg20gc8tvqwm6u54sm89pmh I will send thoughts and prayers instead and tell you "you can't put the genie back in the bottle" and then recommend 72 different build pipelines and editor plugins for you to install

Replying to b7c7240f...

nostr:npub1c353w465eg94sjctq7kh03f9zajjqqsg9wzmg20gc8tvqwm6u54sm89pmh nostr:npub16mupqs08pvwk8crlddfd8wuag79ldtqss09gslalvwksne2u6r0qy2qy4m I am looking to see if it's in a spec, but I'd imagine "canvas" might behave like "img" with respect to how width and height work, e.g. relative to content. Also, "canvas" has squirrely default dimensions that might need to be overridden.

nostr:npub145cfddznynfez4da8l0lcp6gk25fax97lesn33hvl78plf6xqjgsys9l48 nostr:npub16mupqs08pvwk8crlddfd8wuag79ldtqss09gslalvwksne2u6r0qy2qy4m Would that mean the same problem if the `canvas` is wrapped in a `div` and 100% heights are enforced? I kept hitting problems trying to do that as well.

Replying to Avatar Eric A. Meyer

nostr:npub16mupqs08pvwk8crlddfd8wuag79ldtqss09gslalvwksne2u6r0qy2qy4m Probably not? The problem is it’s too tall, not too short. And in my experiments, `max-height: 100%` didn’t help either. I just added it to be sure, and nope, still overflowing in WebKit and Chromium.

Replying to 714c2ce1...

nostr:npub1c353w465eg94sjctq7kh03f9zajjqqsg9wzmg20gc8tvqwm6u54sm89pmh you are probably looking for min-height:0 on the canvas element?

nostr:npub16mupqs08pvwk8crlddfd8wuag79ldtqss09gslalvwksne2u6r0qy2qy4m Probably not? The problem is it’s too tall, not too short. And in my experiments, `max-height: 100%` didn’t help either. I just added it to be sure, and nope, still overflowing in WebKit and Chromium.

So I boiled my problem down a simplified test case and it’s still driving me bonkers — the more so because Gecko does what I expect, but Chromium and WebKit do not. In those two, when the rendering window gets short (that is, not much height), the `canvas` element overflows its grid row, despite being set to `height: 100%`. In Gecko, it always stays within its grid row, as I want it to do. Nor can I find a way to get the Gecko behavior across the board. Argh!

https://codepen.io/meyerweb/pen/eYQXERZ?editors=1100

Replying to Avatar mattg

nostr:npub1c353w465eg94sjctq7kh03f9zajjqqsg9wzmg20gc8tvqwm6u54sm89pmh That is rad, but that gross selector hack would make me revert to a js solution

nostr:npub1r9jx94fxwtqjm9rm7hdclq5tju3yhqvlukdpfezvd3qvsu6d0yysf20dh4 Use JS to write out the gross selector, problem solved.

Replying to Avatar Eric A. Meyer

If you have a browser using an engine with a full `:has()` implementation (WebKit, Chromium) then this is kind of nifty, despite the gross selector hack, if I do say so myself: https://codepen.io/meyerweb/pen/jOQZoaw

(In Gecko, there will be some hover shadows as you mouse around the table, but not the down-and-across effect, due to their partial implementation of `:has()`.)

Of note: I *think* the gross selector hack would be obviated by https://github.com/w3c/csswg-drafts/issues/4559, if I’ve understood those proposals correctly. If not, then… not, I guess.

If you have a browser using an engine with a full `:has()` implementation (WebKit, Chromium) then this is kind of nifty, despite the gross selector hack, if I do say so myself: https://codepen.io/meyerweb/pen/jOQZoaw

(In Gecko, there will be some hover shadows as you mouse around the table, but not the down-and-across effect, due to their partial implementation of `:has()`.)

I wish I could think of a way to monetize the incidence of crypto rugpulls and related scams, because I would be *filthy* rich.