- From: Daniel Holbert <dholbert@mozilla.com>
- Date: Sun, 11 Mar 2012 20:22:15 -0700
- To: "Tab Atkins Jr." <jackalmage@gmail.com>
- CC: www-style list <www-style@w3.org>, Alex Mogilevsky <alexmog@microsoft.com>
On 03/11/2012 07:26 PM, Tab Atkins Jr. wrote: > On Mar 11, 2012 8:49 PM, "Daniel Holbert" <dholbert@mozilla.com > <mailto:dholbert@mozilla.com>> wrote: > > On 03/11/2012 05:29 PM, Alex Mogilevsky wrote: > >> How would this sound as a definition: > >> <ins> > >> Some lengths that are inputs to this algorithm may have values that > are specified exactly or are resolved prior to application of this > algorithm. Such lengths are referred to as <def>Definite</def>. > >> </ins> > > Hmm... So then, what lengths are _not_ definite? Just "auto", or is > there anything else? > > I believe that by "definite" I meant "lengths that don't need layout to > resolve". I think this is probably equivalent to what Alex said. I think if we change Alex's "or are resolved" to "or can be resolved", then yeah, it sounds like those are equivalent. > Counter-examples are 'auto' and percentages. Well -- percentages **with a fixed ("definite"? :)) size to resolve against** would be definite, right? If we've got a width:25% flexbox item inside of a width:400px horizontal flexbox, then I'd say that flexbox item has a "definite size", and we'd want to treat it as such when applying the flexbox layout algorithm.
Received on Monday, 12 March 2012 03:22:44 UTC