W3C home > Mailing lists > Public > www-validator@w3.org > February 2007

Re: using tidy in Markup Validator

From: olivier Thereaux <ot@w3.org>
Date: Thu, 22 Feb 2007 16:20:59 +0900
Message-Id: <A3B4F073-AD1A-4CA6-BCAD-32DFC021A052@w3.org>
Cc: www-validator Community <www-validator@w3.org>
To: David Dorward <david@dorward.me.uk>

Hi David,

sorry for the late reply on this.

On Feb 8, 2007, at 01:38 , David Dorward wrote:
> < Dorward> Could stick a button at the top of the "Invalid" result
>            page.  "Attempt to fix errors automatically". Run it
>            through tidy, input the result into the validator,
>            revalidate that (we'd need to be able to show source on the
>            Valid page otherwise the exercise would be pointless) and
>            include a notice at the top of both the valid and invalid
>            results pages along the lines of "The code being validated
>            was processed by <a>Tidy</a>, a software project maintained
>            by <a>Foo</a> and not under the control of the W3C"
> This approach probably wouldn't be compatible with file upload or
> direct input validation attempts. Possibly this could be worked around
> using CGI::Session?

As you mention this approach would make it difficult to use with file  
upload and direct input, and would make the option rather hidden: I  
do not have any stats on how many people click on the "validate css"  
and "check links" from the (valid) results page, but I suspect very  
few. Likewise, I suspect that results pages for invalid documents are  
already rather crowded and a link, however useful to tidy, might be a  
little lost there.

I'm currently thinking that the tidy option should be available:
- from the home page
- off by default
- switched off for valid documents
(the latter is arguable)

And when one uses the feature, there should be a strong warning that  
the feature is using third party software and provided without  

The current development version tries to follow this path:

Any thought?

(and what to do about the <meta name="generator" ... ?)

Received on Thursday, 22 February 2007 07:21:27 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 22:59:00 UTC