- From: olivier Thereaux <ot@w3.org>
- Date: Thu, 15 Feb 2007 09:05:59 +0900
- To: Karl Dubost <karl@w3.org>
- Cc: QA Dev <public-qa-dev@w3.org>
Hi Karl, On Feb 14, 2007, at 13:17 , Karl Dubost wrote: > I think it would be very useful to have a uniform output format for > validators. Indeed, that's something I have been discussing quite a lot recently, within the team (notes from work with Yves coming up soon) and on www- validator [1]. > I'm more interested for now of the abstract representation, then we > can choose a format. Right. Worth doing in parallel, I believe, is keep an eye on EARL (any other equivalent we should be looking at) as it has a interesting modeling for reporting, where we could find ideas to add to the format. > CSS validator: > > - physical context: Line number > - logical context: class names or ids > - Message: descriptive with variations > - type of result: warning, error, level OK, this is the modeling of a single message. Something Yves and I have been working on is the modeling of the observation itself, which encompasses a group of messages. I guess an alternative would be to have each message be a single observation, but that would perhaps be a problem when several messages relate to each others, e.g: Error: element foo not closed Info: start tag was here > Maybe we should ask Henri Sivonen, Sam Ruby as well. Yes, I think we should have this discussion in www-validator. I'll be sending something with my notes taken while working with Yves in the days to come. Maybe we can take it from there. -- olivier
Received on Thursday, 15 February 2007 00:06:04 UTC