W3C home > Mailing lists > Public > www-svg@w3.org > July 2006

Re: SVG12: conversion to float vs NaN

From: Chris Lilley <chris@w3.org>
Date: Wed, 26 Jul 2006 20:48:53 +0200
Message-ID: <907449323.20060726204853@w3.org>
To: Bjoern Hoehrmann <derhoermi@gmx.net>
Cc: nandini <Nandini.Ramani@Sun.COM>, www-svg@w3.org

On Wednesday, July 26, 2006, 7:53:15 PM, Bjoern wrote:

BH> * nandini wrote:
>>For a property, when you getTrait, it already has the computed value,

BH> That is incorrect.

It would be very helpful if you would say *why* rather than just making
an unsupported statement.

You don't like it when a WG makes a statement with what you consider
insufficient rationale. Similarly, a WG is not really aided by a terse
and unsupported statement.

>>This was discussed within the working group and several implementors 
>>agreed that there are cases where getFloatTrait method will raise an 
>>exception for non-float values and hence we do not see the need to 
>>change the specification. I hope this clarifies your questions.

BH> I, too, agree there are such cases, as the latest draft specifically
BH> mentions two of these cases. It does not, however, provide guidance on
BH> how getFloatTrait and friends behave for other cases, for example:

BH>   * <rect height="100%"
BH>   * <rect height="1.00000"
BH>   * <rect height="1.00001"
BH>   * <rect height="1E999999999999"
BH>   * <rect height="Hello World"
BH>   * attributes not in SVG Tiny 1.2
BH>   * attributes for which trait access
BH>     is not required in SVG Tiny 1.2


I see text such as

> If the attribute is not inherited, or on the root element, the default
> value for the attribute is used, if known. If not known (for example,
> for an unknown attribute, or a known attribute with no specified
> default), 'null' is returned.

Perhaps you could explain why that language does not apply to any of the
cases you just cited.

BH> The draft I originally commented on in fact did
BH> not even mention the <svg width="10%" height="10%" case it covers
BH> now. My concern would easily be addressed by adding something like

BH>   * getFloatTrait and friends must not return values that are not
BH>     equivalent to a real number, specifically NaN and Infinities
BH>     must not be returned.

Thanks for suggesting specific, concrete text that you believe would
address your concerns. We have a problem with that text,though, because
some languages seem to return such values while others never do. Putting
it in a language-independent description rather than a specific language
binding is therefore problematic.

 Chris Lilley                    mailto:chris@w3.org
 Interaction Domain Leader
 Co-Chair, W3C SVG Working Group
 W3C Graphics Activity Lead
 Co-Chair, W3C Hypertext CG
Received on Wednesday, 26 July 2006 18:51:45 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 8 March 2017 09:47:09 UTC