- From: olivier Thereaux <ot@w3.org>
- Date: Wed, 18 Apr 2007 10:55:05 +0900
- To: Jacques Distler <distler@golem.ph.utexas.edu>
- Cc: www-validator Community <www-validator@w3.org>, www-html-editor@w3.org, Masayasu Ishikawa <mimasa@w3.mag.keio.ac.jp>
Hi Jacques, I am copying the www-validator mailing-list, which is the proper channel for feedback and discussion on the validator. Please send your feedback through it in the future: not only does it scale better because there are a lot of knowledgeable people on the list, but it's also important that it is *archived* and others facing the issue can find answers without asking the same questions over and over again. On Apr 18, 2007, at 00:37 , Jacques Distler wrote: > > Can you please explain why the W3C Validator flags > http://golem.ph.utexas.edu/~distler/blog/files/validator-svg.xhtml > as invalid? Short answer: because it is not valid per the DTD for XHTML+MathML+SVG. Longer answer: because 1) you're using an invalid construct per this profile, and because 2) there is a bug in the published (flattened) DTD for the XHTML+MathML+SVG Working Draft. 1) your xlink xmlns is declared on the svg:defs element. The profile allows the xlink namespace to be declared only on the svg, html or mathml root element, AFAICT. Moving your xmlns:xlink="http:// www.w3.org/1999/xlink" to <svg:svg will fix this problem. 2) There is a problem with the declaration of the svg:stop element in the flat DTD for XHTML+MathML+SVG. The DTD declares the element "stop" as not being prefixed. I think this is a bug, which is not shocking - remember, XHTML+MathML+SVG is a *Working Draft*. Below is a patch for the flat DTD, courtesy of Karl Dubost who helped me debug the issue: --- xhtml-math-svg-flat.dtd 2002-08-08 21:41:02.000000000 +0900 +++ xhtml-math-svg-flat2.dtd 2007-04-18 10:25:04.000000000 +0900 @@ -754,7 +754,8 @@ <!ENTITY % SVG.linearGradient.qname "%SVG.pfx;linearGradient" > <!ENTITY % SVG.radialGradient.qname "%SVG.pfx;radialGradient" > -<!ENTITY % SVG.stop.qname "stop" > +<!ENTITY % SVG.stop.qname "%SVG.pfx;stop" > + <!-- module: svg-pattern.mod ........................... --> > I would not care, but for the fact that I actually USE a local copy > of the W3C Validator in my blog software. So, the fact that it > flags valid documents as invalid is a PROBLEM for me. Let me just remind you that the markup validator is open source software, free/libre and open to contributions and patches, and its distribution terms include: THIS SOFTWARE AND DOCUMENTATION IS PROVIDED "AS IS," AND COPYRIGHT HOLDERS MAKE NO REPRESENTATIONS OR WARRANTIES, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO, WARRANTIES OF MERCHANTABILITY OR FITNESS FOR ANY PARTICULAR PURPOSE [...] -- olivier Thereaux - W3C - http://www.w3.org/People/olivier/ W3C Open Source Software: http://www.w3.org/Status
Received on Wednesday, 18 April 2007 01:55:24 UTC