- From: olivier Thereaux <ot@w3.org>
- Date: Tue, 24 Apr 2007 09:58:33 +0900
- To: Brett Bieber <brett.bieber@gmail.com>
- Cc: www-validator@w3.org
Hi Brett, On Apr 24, 2007, at 00:17 , Brett Bieber wrote: > At first I considered some post-processing within my api interface > library ( http://pear.php.net/package/Services_W3C_HTMLValidator/ ) > when the result is an invalid page - but in my mind it makes some > sense that the validator returns the error context in the report if > possible. I think this is a good idea. Adding this to the existing SOAP format output would not be a problem, as it would not break implementations of older versions of the format. > The major problem I could see is providing valid XML output while > preserving the invalid code --- this might be somewhat tricky --- and > if the marked error is returned, there would need to be documented > decode procedures so the code with errors could be recovered. providing valid XML output should be ok if the source snippet is properly escaped. Marking of the error location would indeed need to be properly documented. > Any thoughts on this? Is this feasible, and worth a feature request > in bugzilla? Certainly. Thanks a lot for your suggestions. -- olivier
Received on Tuesday, 24 April 2007 00:58:38 UTC