- From: Jon Ferraiolo <jferraio@Adobe.COM>
- Date: Sun, 12 Nov 2000 13:12:20 -0800
- To: "Kirk Jensen" <kjensen@palestone.com>
- Cc: <www-svg@w3.org>
- Message-Id: <4.2.2.20001106094945.00d7f300@mailsj.corp.adobe.com>
Kirk, Thanks for pointing this out. I agree with your analysis and recommendation. I will recommend to the SVG working group that percentages be disallowed on feColorMatrix as you suggest. Jon Ferraiolo SVG Editor Adobe Systems Incorporated At 11:07 AM 11/6/00 -0600, Kirk Jensen wrote: ><http://www.w3.org/TR/2000/CR-SVG-20001102/filters.html#feColorMatrix>http: >//www.w3.org/TR/2000/CR-SVG-20001102/filters.html#feColorMatrix > >I believe that the definition of values in feColorMatrix has a >contradictary definition. The spec states that the type of values is >"list of Numbers", yet for the case of type=saturate, the spec states the >values can be a number from 0-1 or an optional percentage value (e.g. 50%). > >In chapter 4 (Basic Data Types) list of xxx is defined as a list of type >xxx only, not a variation of types. Therefore, it should be safe to >assume that if an attribute is defined as a list of xxx, only xxx types >will appear. The inconsistency with feColorMatrix values causes some >difficulty in parsers set up to parse values bases on a single basic data >type. In this case, the parser has to be designed to allow a change in >the definition of the attribute depending on the value of another attribute. > >I know that in the past opacity attributes followed the same >inconsistency, but I believe in updated specs this has been changed to >only allow numbers in the range 0-1. (Correct?) > >Although I cannot recall off the top of my head, I believe there are >other cases where the same is true. Eliminating this special case would be >a good start. (I am not including animate from,to,by,values in this >statement; they are completely flexible types by design.) > >Sincerely, >Kirk Jensen
Received on Sunday, 12 November 2000 16:13:01 UTC