The updated text regarding intrinsic sizing of replaced elements might be of interest here: https://www.w3.org/TR/css-sizing-3/#intrinsic https://www.w3.org/TR/css-sizing-3/#min-content-zero
There's also an open issue about sizing iframes to their content: https://github.com/w3c/csswg-drafts/issues/1771 It's tagged against Level 4, but we did just add handling for <textarea> and <input>. ~fantasai -------- Forwarded Message -------- https://lists.w3.org/Archives/Public/www-style/2018Mar/0001.html The CSS WG has published an updated Working Draft of the CSS Intrinsic and Extrinsic Sizing Module Level 3: https://www.w3.org/TR/css-sizing-3/ This module extends the CSS sizing properties with keywords that represent content-based "intrinsic" sizes and context-based "extrinsic" sizes, allowing CSS to more easily describe boxes that fit their content or fit into a particular layout context. Significant changes are listed at: https://www.w3.org/TR/2018/WD-css-sizing-3-20180304/#changes and include * merging in the full definitions of all sizing properties from CSS2.1 (min/max?-width/height) and css-ui (box-sizing) https://www.w3.org/TR/css-sizing-3/#specifying-sizes * more thorough definition of replaced element intrinsic sizes https://www.w3.org/TR/css-sizing-3/#intrinsic and percentage sizing within indefinite containers https://www.w3.org/TR/css-sizing-3/#percentage-sizing * defining min-content and max-content to work on form controls * deferring the 'stretch' and 'fit-content' keywords to Level 4 (whose focus will be defining these keywords and 'contain') Open issues include: * Adding more illustrations (help wanted) https://github.com/w3c/csswg-drafts/issues/1938 * Working out how calc() values including percentages work on margins/padding/width/height/gaps when the container size depends on this child box’s size (input wanted) https://github.com/w3c/csswg-drafts/issues/2297 That is, currently when we are calculating the size of the container we treat a percentage size as zero. Then once the size of the container is established, we resolve the percentage against that size. What should happen if we have a size as calc(20% + 10px)? Do we ignore the 10px or honor it in some way? What about calc(10px - 20%)? See https://github.com/w3c/csswg-drafts/issues?q=is%3Aopen+is%3Aissue+label%3Acss-sizing-3 for all open issues. We expect to transition to CR soon, so this draft effectively marks the beginning of a last call for comments period; we will be accepting comments at least through the end of March, and depending on the state of the draft, aim to transition to CR sometime in April. (We will of course process comments during CR as well, but would prefer to get them sooner rather than later.) (Note that the min-content and max-content keywords have already been officially cleared for shipping prior to CR by the CSSWG https://lists.w3.org/Archives/Public/www-style/2015Aug/0109.html since their syntax was stable and their behavior was tied to behavior exposed in existing CSS2.1 features.) Please review the draft, and send any comments to this mailing list, <www-st...@w3.org>, prefixed with [css-sizing] (as I did on this message) or (preferably) file them in the GitHub repository at https://github.com/w3c/csswg-drafts/issues For the CSS WG, ~fantasai
