W3C home > Mailing lists > Public > www-svg@w3.org > November 2005

RE: [SVGT 1.2] Question on sizing of the outermost <svg>

From: Jon Ferraiolo <jonf@adobe.com>
Date: Tue, 1 Nov 2005 22:04:09 -0800
Message-ID: <6ECA24BE410D994496A2AE995367C5C8301BFE@namail3.corp.adobe.com>
To: "Boris Zbarsky" <bzbarsky@mit.edu>, <www-svg@w3.org>

First off I will point out that I have been arguing that svg:svg should
have separate attributes for intrinsic size (which would not allow
100%/100%) for years and have been rebuffed by the SVG WG.

I agree that the 'width' and 'height' attributes on the svg:svg element
should *not* be considered "presentation attributes" and thus should not
affect the cascade of the 'width' and 'height' properties. In fact, the
SVG WG has been operating on the assumption that 'width' and 'height'
were not part of the CSS cascade since the whole thing stabilized about

Given my failures at persuasion, we have the 'width' and 'height'
attributes on the svg:svg as the only mechanism for defining intrinsic
width and height. With these attribute, a non-percentage defines the
intrinsic value, whereas if you use a percentage value the SVG content
does not have an intrinsic value.

My suggestion is that if there is no intrinsic value, then the intrinsic
value is defined such that it matches the algorithm in CSS 2.1 (i.e.,
300px for width, 150px for height, unless the screen is smaller than
those values). [1]


[1] http://www.w3.org/TR/CSS21/visudet.html#inline-replaced-width

-----Original Message-----
From: www-svg-request@w3.org [mailto:www-svg-request@w3.org] On Behalf
Of Boris Zbarsky
Sent: Tuesday, November 01, 2005 7:05 PM
To: www-svg@w3.org
Subject: Re: [SVGT 1.2] Question on sizing of the outermost <svg>

L. David Baron wrote:
> On Tuesday 2005-11-01 16:57 -0600, Boris Zbarsky wrote:
>>   <svg:svg style="height: 100px" width="50" height="50"/>
> Unless the markup you've written is shorthand for having the width and
> height attributes inside the SVG and the style attribute on the
> that includes it, I don't think the analogy to HTML images holds.

The problem is that HTML images have a way of specifying intrinsic width
of the markup (it's in the image data).  For SVG, it seems that the way
specify intrinsic width is via the width/height attributes on the <svg>

I think we agree that for SVG included via say <xhtml:object> the
attributes give the intrinsic size...  It would make a lot of sense to
me as an 
author if I could get a similar behavior when including SVG inline,
somehow.  At 
the moment, I don't see how -- if we make width/height affect the
computed style 
as presentational attributes in this case, there is no way to set the

Received on Wednesday, 2 November 2005 06:06:52 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 22:54:08 UTC