W3C home > Mailing lists > Public > public-svg-ig@w3.org > July to September 2008

Re: [svg-developers] Re: Safari problem with no nought

From: Erik Dahlström <ed@opera.com>
Date: Thu, 11 Sep 2008 10:13:03 +0200
To: svg-developers@yahoogroups.com
Cc: public-svg-ig@w3.org
Message-ID: <op.uha431qbgqiacl@gnorps.linkoping.osa>

On Thu, 11 Sep 2008 02:09:58 +0200, ddailey <ddailey@zoominternet.net> wrote:

> Hi Jeff!
>
> Jeff Schiller wrote:
>
>>On the other hand, I'm not sure about begin/end times in SMIL.  I
>>noticed last night that Opera seems to have problem with missing
>>nought characters in SMIL events (i.e. begin=".25s").  Haven't done
>>any thorough investigation though.
>
> I used to notice (a couple of years ago) that Opera had problems if I used dur=".5s" instead of dur="0.5s", but thought that had all been fixed. Maybe I just got used to using "0." in my time units for SMIL.  So yes, we'll need to make sure to include such cases.

The syntax for the begin attribute is borrowed from SMIL, and the grammar doesn't allow for a "." to start a <Clock-value>. Possibly this is because it can cause confusion with id-values, but I haven't looked at it in detail yet.

Another somewhat related gotcha is dash/minus characters in id:s, that can clash with the <event-value> syntax, e.g. 'someid-0.4s.end' since 'someid-0.4s' is a valid id. What you need to do in such cases is to escape the dash in the id with a backslash.

> There are some boundary problems still existing of that type....
>
> for example in http://srufaculty.sru.edu/david.dailey/svg/SVGOpen2008/filterBlur2.svg , if I put
>
> <filter id="D"  x="0%" y="0%" width="100%" height="100%">
>        <feGaussianBlur stdDeviation="25,0" />
> </filter>
>
> instead of
>
> <filter id="D"  x="0%" y="0%" width="100%" height="100%">
>        <feGaussianBlur stdDeviation="25,0.1" />
> </filter>
>
> (thanks to Erik for helping me figure this out)

It should work, the spec wasn't explicit about this particular case though. If stdDeviation is 0 then nothing should be rendered according to the spec. However when it's > 0 in one direction and 0 in the other then I agree it makes sense to render it. I'll see if I can clarify this in the SVG Filters 1.2 spec, and try to get this fixed in Opera.

Cheers
/Erik

-- 
Erik Dahlstrom, Core Technology Developer, Opera Software
Co-Chair, W3C SVG Working Group
Personal blog: http://my.opera.com/macdev_ed
Received on Thursday, 11 September 2008 08:14:58 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 14 April 2009 16:29:30 GMT