W3C home > Mailing lists > Public > www-style@w3.org > January 2012

RE: [css3-flexbox] Computed value and flex-align/flex-item-align.

From: Alex Mogilevsky <alexmog@microsoft.com>
Date: Fri, 27 Jan 2012 19:54:21 +0000
To: Tab Atkins Jr. <jackalmage@gmail.com>, Boris Zbarsky <bzbarsky@mit.edu>
CC: "www-style@w3.org" <www-style@w3.org>
Message-ID: <D51C9E849DDD0D4EA38C2E539856928412E20457@TK5EX14MBXC218.redmond.corp.microsoft.com>
>From: Tab Atkins Jr. [mailto:jackalmage@gmail.com]
>Sent: Friday, January 27, 2012 8:48 AM
exbox] Computed value and flex-align/flex-item-align.
>
>I've thought about it a bit, and I don't think there's any good reason for
>'inherit' to see ''auto''.
>
>If your parent's flex-item-align is set to a non-auto value, using
>'inherit' ensures that you're aligned the same way as they are: if they're
>start-aligned in their flexbox, you'll be start-aligned; if
>they're end-aligned, you'll be end-aligned.    This seems potentially
>useful.  However, if 'auto' inherits as 'auto', this relationship breaks -
>you'll now align according to your parent's 'flex-align'
>property, which can be different than their actual alignment.  So, 'auto'
>should inherit as the alignment it corresponds to.

Now, this sounds like a sensible reason to do the work. There is actually a place where it makes a difference.

Note btw that you will get exactly same result by setting "flex-item-align:auto; flex-align:inherit" on the middle flexbox. But inheriting actual alignment makes sense here. I agree to the text you already have in the spec.
Received on Friday, 27 January 2012 19:55:10 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:49 GMT