Re: Validation failing incorrectly on bad value for encoding name

Jaime Iniesta <jaimeiniesta@gmail.com>, 2013-03-10 22:37 +0100:

> 2013/3/10 Michael[tm] Smith <mike@w3.org>
> I've reinstalled it after your changes and I can confirm this has been
> fixed. Thanks a lot!

Super

> On a related note, validating "http://twitter.com" makes the validator
> crash (Java exception) and show a message "Oops. That was not supposed to
> happen. A bug manifested itself in the application internals. Unable to
> continue. Sorry. The admin was notified.".

I can't reproduce that. But if you send me the stack trace it outputs to
the console when that happens, I'll take a look.

By the way, are you running under Java 7? Because there's a known issue
with Java 7 and the validator:

  http://bugzilla.validator.nu/show_bug.cgi?id=949

> It's also failing (with a different message) on validator.w3.org and
> validator.nu

It's not failing, really. That message is expected behavior for the
particular markup case it's reporting.

> but I think that's a harder issue.

It's actually non-issue, because what it's doing is expected behavior.

The thing is, the twitter.com source currently seems to have this:

  </html>
    <input type="hidden" id="init-data" class="json-data"...
    <input type="hidden" class="loadrunner-alias json-data"...
    <input type="hidden" class="swift-boot-module"...
    <input type="hidden" id="swift-module-path" ...
    <script src=...

That is, the source has additional elements after the </html> end tag. The
HTML spec says it is a conformance error to have any element content at all
after the </html> end tag. So what the validator does is, it just stops at
the first such element it finds, then emits that "Any further errors will
be ignored." message -- because anything else that might follow is all
invalid also, so it really doesn't matter what there might be.

-- 
Michael[tm] Smith http://people.w3.org/mike

Received on Monday, 11 March 2013 00:19:01 UTC