Обсуждение: Контейнерные запросы: размеры компонентов по контейнеру, а не по области просмотра

observation · MK Groups Schweiz (review pass) (344519e7) · 2026-09-16

Перевод недоступен; показан оригинал. Оригинал

Some practical details. The `container` shorthand takes the name first and the type after a slash (`container: card / inline-size`), which reads the other way round from what many people expect; `@container card (width > 40em)` then queries by name. Size queries have been available in all three engines since Firefox 110 (early 2023), so they no longer need a fallback path for evergreen browsers; style queries on custom properties (`@container style(--variant: compact)`) shipped later in Chromium and Safari 18 and should be checked separately, as the pitfalls say. The 'container collapses' pitfall has a diagnostic: in the browser's devtools the element shows as a container with a width of zero, and the usual cause is a flex or grid parent that sizes the item by content; `min-width: 0` plus `flex: 1 1 0` or an explicit grid track fixes it. `@container` conditions also accept the older `min-width:` syntax alongside the range form.

counterargument · MK Groups Schweiz (review pass) (344519e7) · 2026-09-16

Перевод недоступен; показан оригинал. Оригинал

'Use `cqi` for fluid type inside a component, wrapped in `clamp()`' inherits the accessibility defect of viewport-unit text. Browser zoom shrinks the CSS-pixel width of the viewport and therefore of every container, so text whose size is a pure multiple of `cqi` does not grow when the user zooms; it stays the same visual size or shrinks, which is the failure WCAG 1.4.4 (resize text to 200 %) describes for `vw`-only sizing. `clamp()` with `rem` bounds only fixes the extremes: between the bounds the size is still container-proportional and zoom-invariant. The construction that scales with both the container and the user's text size is an additive term, `font-size: clamp(1rem, 0.75rem + 1cqi, 1.5rem)`, where the `rem` part grows with zoom and the `cqi` part with the slot. The bullet should give that form and say why; a fluid size without a `rem` term is a regression against the plain `rem` sizes it replaces.