Intent to Implement: Replace DOMString with DOMSettableTokenList for {HTMLAnchorElement, HTMLAreaElement}.ping.

hyunjune

unread,

Jun 5, 2015, 8:23:10 AM6/5/15

to blin...@chromium.org

Contact emails

hyunju...@samsung.com, hyki...@gmail.com

Spec

https://html.spec.whatwg.org/#the-area-element

https://html.spec.whatwg.org/#the-a-element

Motivation

If we change to DOMSettableTokenList, can make the data structure reasonable. Because currently
every time the blink send pings, it should work to split string for URLs. It is stupid.
Like this.
void HTMLAnchorElement::sendPings(const KURL& destinationURL) const
{
(... omit ...)
      SpaceSplitString pingURLs(pingValue, SpaceSplitString::ShouldNotFoldCase);
      for (unsigned i = 0; i < pingURLs.size(); i++)
      PingLoader::sendLinkAuditPing(document().frame(), document().completeURL(pingURLs[i]), destinationURL);
}
So I suggest that replace DOMString with DOMSettableTokenList.

Compatibility Risk

This feature is discussing on the WHATWG. and doesn't support on gecko(firefox).

Philip Jägenstedt

unread,

Jun 5, 2015, 1:02:38 PM6/5/15

to hyunjune, blink-dev

LGTM, but please create the OWP launch tracking bug and chromestatus.com entry.

Because DOMSettableTokenList has a stringifier and these attributes have [PutForwards=value], typical uses that assume it is a string should continue to work as before.

To unsubscribe from this group and stop receiving emails from it, send an email to blink-dev+...@chromium.org.

Boris Zbarsky

unread,

Jun 6, 2015, 8:11:42 AM6/6/15

to hyunjune, blin...@chromium.org

On 6/5/15 8:23 AM, hyunjune wrote:
> This feature is discussing on the WHATWG.

Might I ask where? I see nothing on the WHATWG mailing list.

-Boris

hyunjune

unread,

Jun 6, 2015, 9:36:59 AM6/6/15

to blin...@chromium.org, bzba...@mit.edu

2015년 6월 6일 토요일 오후 9시 11분 42초 UTC+9, Boris Zbarsky 님의 말:

Sorry, my say seems wrong. Just this mean should be that are not yes reflected in the w3c about {HTMLAnchorElement, HTMLAreaElement} interface.
But on WHATWG this feature was reflected on {HTMLAnchorElement, HTMLAreaElement} interface.

hyunjune

unread,

Jun 6, 2015, 9:41:59 AM6/6/15

to blin...@chromium.org, hyunju...@samsung.com

but please create the OWP launch tracking bug and chromestatus.com entry.

Ok, I am going to create them.

PhistucK

unread,

Jun 6, 2015, 9:44:32 AM6/6/15

to hyunjune, blink-dev, Boris Zbarsky

Here -

https://html.spec.whatwg.org/multipage/semantics.html#htmlanchorelement

https://html.spec.whatwg.org/multipage/embedded-content.html#htmlareaelement

☆PhistucK

To unsubscribe from this group and stop receiving emails from it, send an email to blink-dev+...@chromium.org.

Philip Jägenstedt

unread,

Jun 9, 2015, 4:56:28 AM6/9/15

to PhistucK, hyunjune, blink-dev, Boris Zbarsky

Let's treat this as an Intent to Implement and Ship, as there isn't much sense in having a flag for this, we should just change the behavior and ship it.

hyunjune

unread,

Jun 9, 2015, 8:54:56 AM6/9/15

to blin...@chromium.org, phis...@gmail.com, bzba...@mit.edu, hyunju...@samsung.com

I made the issue.
https://code.google.com/p/chromium/issues/detail?id=498219

2015년 6월 9일 화요일 오후 5시 56분 28초 UTC+9, Philip Jägenstedt 님의 말:

TAMURA, Kent

unread,

Jun 16, 2015, 5:56:41 PM6/16/15

to Philip Jägenstedt, hyunjune, blink-dev

LGTM2

--

TAMURA Kent
Software Engineer, Google

Philip Rogers

unread,

Jun 16, 2015, 6:10:11 PM6/16/15

to TAMURA, Kent, Philip Jägenstedt, hyunjune, blink-dev

Boris, did you have any remaining concerns here? If not, I think think this seems reasonable, LGTM.

Reply all

Reply to author

Forward