W3C home > Mailing lists > Public > www-style@w3.org > September 2015

Re: [css-align] in which cases does 'left'/'right' compute to 'start' exactly?

From: fantasai <fantasai.lists@inkedblade.net>
Date: Mon, 21 Sep 2015 13:54:15 -0400
To: www-style@w3.org
Message-ID: <56004447.60402@inkedblade.net>
On 09/20/2015 04:48 PM, Mats Palmgren wrote:
> In the description of 'left' (and 'right'), the spec says:
> "If the property's axis is not parallel with the inline axis,
> this value computes to 'start'."
> https://drafts.csswg.org/css-align-3/#positional-values
> Does this include values where those keywords are combined
> with other keywords?  I.e. does the computed value for
> "align-content: left safe" become "start safe" ?
> "align-content: stretch left" become "stretch start" ?
> "align-content: stretch left safe" become "stretch start safe" ?
> etc. or does that sentence just apply when the specified
> value is exactly "left" ?

Yes. Note dbaron filed an issue on computation, so it's probably
going to be the used value that changes over, not the computed

> Also, I want to double-check I understand the first part of
> the sentence correctly: this means that 'left' always becomes
> 'start' for all 'align-*' properties, and always computes to
> itself for all 'justify-*' properties, right?

This is true except in flex layout, because in flex layout
the align/justify mapping depends on the flex direction, not
the writing mode.

(This is why they're named so abstractly--we couldn't map them
to the inline and block axes always.)

> (I would appreciate a quick answer as I am writing code to
> implement this in Gecko right now.)

Thanks, please continue to report anything that's weird or that
you think could be done better. :) The draft has not gotten much
review, and Tab and I would appreciate any feedback you have on
design issues as well as clarifications/fixes.

Received on Monday, 21 September 2015 17:54:43 UTC

This archive was generated by hypermail 2.4.0 : Friday, 25 March 2022 10:08:57 UTC