- From: Daniel Holbert <dholbert@mozilla.com>
- Date: Mon, 26 Mar 2012 21:00:47 -0700
- To: "Kang-Hao (Kenny) Lu" <kennyluck@csail.mit.edu>
- CC: WWW Style <www-style@w3.org>
Hi Kenny, Thanks for the feedback! On 03/26/2012 06:23 PM, Kang-Hao (Kenny) Lu wrote: > I'd like to see all these values to be treated the same for flexbox item > generation: a stretch of of whitespace never generates an anonymous > flexbox item. > > This is consistent with how CSS 2.1 handles whitespaces in between > 'table-*' too. My impression (which might be entirely off-base) is that the special rules for whitespace handling in tables are a bit shoehorned for backwards-compatibility with existing HTML <table> content. So, I'm not sure if table quirks are a good thing for us to mimic here, in a brand new spec. Having said that, I share your concern about what behavior authors will expect... > So is your proposal the easiest way to implement this in Gecko? If > that's the case, I think it's acceptable to me too. Collapsing whitespace unconditionally (as you suggest) is marginally easier to implement in Gecko than my proposal, but not by enough to really matter. I'm mostly concerned with what's sane & consistent with the rest of CSS. Moreover -- whichever way this ends up being settled, it will likely end up being a precedent for other CSS specs. In fact, the exact same question applies to CSS3-Grid, which wraps runs of non-replaced inline content in "anonymous grid items". So, we should definitely consider that when deciding. ~Daniel
Received on Tuesday, 27 March 2012 04:01:17 UTC