W3C home > Mailing lists > Public > whatwg@whatwg.org > January 2012

[whatwg] Feedback on Meta referrer

From: Adam Barth <w3c@adambarth.com>
Date: Thu, 26 Jan 2012 01:45:24 -0800
Message-ID: <CAJE5ia8Acc6RX-m9YHm+one+SaKg8D24SrWk=LxQO5AdkhK7=w@mail.gmail.com>
On Thu, Jan 26, 2012 at 1:35 AM, Boris Zbarsky <bzbarsky at mit.edu> wrote:
> On 1/26/12 9:12 AM, Adam Barth wrote:
>>> Should the speculative parser have knowledge of<meta name=referrer>?
>>
>> That's not what's currently specified. ?Like many other browser
>> features, this feature lets web sites detect that the browser is
>> speculatively prefetching resources. ?If that's a big issue, it's
>> something we can try to address.
>
> It seems like a bigger problem is that if speculative prefetches don't know
> about this <meta> then they will leak the referrer, which is something the
> site did NOT want to happen. ?So it seems like either this <meta> needs to
> disable prefetch altogether or be taken into account when prefetching.
> ?Either way, the prefetch code needs to know about it.

I've added a TODO to http://wiki.whatwg.org/wiki/Meta_referrer to add
this requirement.  I'm slightly unsure how to specify it because the
preload scanner isn't part of the spec's machinery, but we'll figure
out a way.

>> I'm not sure all implementations have the speculative parser
>> understand<base>. ?For example, WebKit's preload scanner does not
>> appear to understand the<base> ?element:
>>
>> http://trac.webkit.org/browser/trunk/Source/WebCore/html/parser/HTMLPreloadScanner.cpp
>
> That's not as big a deal, because it will just mean you prefetch the wrong
> thing and have to do a second fetch. ?(That said, I think we may have had
> bug reports about the prefetch not understanding <base>; Henri would know
> for sure).

Makes sense.

Thanks,
Adam
Received on Thursday, 26 January 2012 01:45:24 UTC

This archive was generated by hypermail 2.3.1 : Monday, 13 April 2015 23:09:10 UTC