W3C home > Mailing lists > Public > www-style@w3.org > July 2010

Re: [css3-values] proposal for handling division and units in calc() [ACTION-205]

From: L. David Baron <dbaron@dbaron.org>
Date: Sun, 25 Jul 2010 21:00:47 -0700
To: Brad Kemper <brad.kemper@gmail.com>
Cc: "www-style@w3.org" <www-style@w3.org>
Message-ID: <20100726040047.GA8796@pickering.dbaron.org>
On Sunday 2010-07-25 17:27 -0700, Brad Kemper wrote:
> On Jul 25, 2010, at 11:48 AM, "L. David Baron" <dbaron@dbaron.org> wrote:
> 
> > My tentative conclusion is that supporting division by values is not
> > among the top use cases for calc(), and since it poses considerable
> > difficulty for both specification and implementation,
> 
> Can you elaborate on why this is difficult? It seems to me that if
> any value that is in units allowed by the property could be
> coerced into a percentage, and from there it is a short hop to
> being a unitless number. I assume that dividing by percentages is
> still OK anyway, isn't it, since a percentage is just another way
> to represent a digital number? 

Percentages aren't just like numbers; they mean whatever percentages
mean for the property, which is something with units.

It's difficult from a specification and implementation perspective
because it requires figuring out what to do with division-by-zero
after the cascade has already happened (so there's no fallback).

It's difficult (in the sense of being a good bit of work) from an
implementation perspective because one has to write code to track
the different dimensions represented by units so that the
implementation can determine that
  calc((500px / 10Hz + 50px * 3s) / 1s)
computes to
  200px
and is therefore a valid value for 'width'.  But I'd feel like I was
wasting my time writing code to do that, because I don't see why
anyone needs it.

-David  

-- 
L. David Baron                                 http://dbaron.org/
Mozilla Corporation                       http://www.mozilla.com/
Received on Monday, 26 July 2010 04:01:19 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:29 GMT