- From: Dr. Olaf Hoffmann <Dr.O.Hoffmann@gmx.de>
- Date: Wed, 10 Jan 2007 14:43:06 +0100
- To: www-svg@w3.org
Hello, If I understand SMIL correct, description and (at present) PNG of the subtest for wallclock in animate-elem-62-t seems to be wrong. For this: " <rect x="-6" y="-6" width="12" height="12" fill="#F33"> <set attributeName="fill" to="#3F3" end="wallclock(2200-06-10T12:34:56Z)" dur="indefinite"/> </rect> " it is explained: " The eight's <set> target has its end attribute set to 'wallclock()'. Therefore, the target should never turn green because the target wallclock time is in the very distant future, so the animation should not end (unless this test is used for close to two hundred years, in which case the end date should be updated for an additional two or three hundred years). " I think, correct would be something like (and in the PNG the rectangle has to be green for the next years): " The eight's <set> target has its end attribute set to 'wallclock()'. The result depends on the presentation time. If the document is viewed completely before 2200-06-10T12:34:56Z, the rectangle has to be always green. begin is not explicitely set, therefore it is zero, dur is indefinite and end is in the future. If the document is viewed completely after 2200-06-10T12:34:56Z, the only end value is before the implicitely given only begin value and therefore the set does not start, the rectangle remains red. If the document is viewed in a time interval started before 2200-06-10T12:34:56Z and ended after this date, the rectangle will start green at the beginning, change to red at 2200-06-10T12:34:56Z and will remain red until the end of presentation. " See (3.2.1) http://www.w3.org/TR/2001/REC-smil-animation-20010904/#AnimFuncTiming Because wallclock values are converted into document time, a wallclock value in the future is the same as a positive offset for end. The viewer has just to determine the offset at the begin of the presentation, and can replace the wallclock value for end with this offset. The description of the first subtest has a minor problem too: " The first <set> has an unspecified end attribute. That value defaults to an offset of 0s so the animation should apply as soon as the document is loaded. " If end is not explicitely given, the default is not 0s or something else. A default value for unspecified end may cause problems or another behaviour as specified in SMIL in more complicated situations. Happy testing...
Received on Wednesday, 10 January 2007 15:09:15 UTC