Daniel @DPA@awoo.space
@kasiandra Ich wünschte meine Bank würde solche Standards nutzen, statt proprietäre Lösungen. Statt QR Codes haben sie diese "color codes", eine matrix aus farbigen punkten, und ein dazugehöriges Gerät, um die einzuscannen, welches dann Nummern anzeigt...
@mntmn This reminds me, I should probably resend my kernel patch, I think it was something for allowing the CardLongName to be set based on the device tree, so it wouldn't matter what bootloader/firmware is used and how the kernel is booted, because that affects which UCM2 gets loaded.
(This is why sound doesn't work if mainline uboot is used with uefi to boot the kernel on Pinephone Pro. (Aside from no UCM being upstreamed yet))
@sir_pepe The width depends on the font and its glyphs, but wcswidth only knows about the codepoints, but nothing about the font, so it can only guess.
@foone Just write a css rule for blink containing a css keyframe animation changing the opacity property to make it work again. I think it's pretty much the only html4 tag that no longer works. I mean, there was that tag for augmenting context menus, but I think that was already html5.
@whitequark The true name of a system is it's root password. It gives those who know it complete control over the system, and as such must be protected by any means necessary.
@foone Easy. Just fill every second line with blanks.
@thephd This is nice, but my bigger problem with _Generic is that it checks the semantics of non-matched expressions too.
ˋˋˋ
_Generic(x, struct a: x.x, int: x)
ˋˋˋ
This currently fails if x is an int. There are workarounds, but they are really annoying, hard to read and error prone.
ˋˋˋ
_Generic(x, struct a: _Generic(x, struct a: x, default: (struct a){0}).x, int: x)
ˋˋˋ
