245025 2022-09-10 06:21:48 -0700 [css-grid] span i should not serialize to span 1 i 2022-09-23 23:32:09 -0700 1 1 1 Unclassified WebKit CSS WebKit Nightly Build Unspecified Unspecified RESOLVED FIXED https://github.com/web-platform-tests/wpt/pull/36010 InRadar P2 Normal --- 204611 1 ntim mattwoodrow mattwoodrow webkit-bug-importer oldest_to_newest 1897514 0 ntim 2022-09-10 06:21:48 -0700 See https://wpt.fyi/results/css/css-grid/parsing/grid-area-computed.html?label=master&label=experimental&product=chrome&product=firefox&product=safari&aligned&view=interop&q=label%3Ainterop-2021-grid 1897515 1 webkit-bug-importer 2022-09-10 06:22:11 -0700 <rdar://problem/99778041> 1900275 2 mattwoodrow 2022-09-21 19:06:49 -0700 This test is invalid :( It's checking that `grid-column-start: span 1 i` round-trips exactly, as does `grid-row-start: span i`. The only way to make both of those tests is to preserve whether the '1' was explicitly passed in, and use that to decide how to serialise the computed value, rather than just using the shortest serialisation principle. Gecko is passing this because they parse an implicit integer as 0, and omit serializing the integer if it's 0. They then covert 0 to 1 at used value time. The spec for parsing 'span': If the <integer> is omitted, it defaults to 1. Negative integers or zero are invalid. 1900282 3 mattwoodrow 2022-09-21 19:23:06 -0700 Pull request: https://github.com/WebKit/WebKit/pull/4587 1900829 4 ews-feeder 2022-09-23 23:32:08 -0700 Committed 254821@main (fc4a525dc150): <https://commits.webkit.org/254821@main> Reviewed commits have been landed. Closing PR #4587 and removing active labels.