W3C home > Mailing lists > Public > public-script-coord@w3.org > July to September 2011

Re: Removing 'caller' from WebIDL

From: Sam Weinig <weinig@apple.com>
Date: Thu, 15 Sep 2011 11:52:08 -0700
Cc: Cameron McCormack <cam@mcc.id.au>, public-script-coord@w3.org
Message-id: <C6761129-9826-4207-9674-68C5A0E269E1@apple.com>
To: Jonas Sicking <jonas@sicking.cc>

On Sep 9, 2011, at 10:11 AM, Jonas Sicking wrote:

> On Fri, Sep 9, 2011 at 1:50 AM, Cameron McCormack <cam@mcc.id.au> wrote:
>> On 26/08/11 6:21 AM, Jonas Sicking wrote:
>>> 
>>> The 'caller' annotation in WebIDL makes an object callable. This is
>>> needed for legacy support at least for the document.all (non-firefox
>>> browsers also support it on other objects, but it's not been
>>> established that this is actually needed for webcompat).
>>> 
>>> My understanding is that this convention is not implementable in
>>> Javascript.
>>> 
>>> I think we should remove this from WebIDL as to prevent the pattern
>>> from spreading any further. Apparently the DAP contacts API has
>>> started using it, as has the microdata DOM API. So it appears that
>>> putting it in WebIDL has made it easier for the pattern to spread.
>>> 
>>> Prose can always be added to the document.all object (and other
>>> objects too if needed) as to make them callable.
>> 
>> I'd like to wait to see what the results of the WebKit experiments are, and
>> whether that indicates whether the non-HTMLAllCollection callers will be
>> removed from the HTML spec.
>> 
>> For the moment, I will give Anne's suggestion a try -- I'll leave it in the
>> spec, and rename it to legacycaller.
>> 
>> http://dev.w3.org/cvsweb/2006/webapi/WebIDL/Overview.html.diff?r1=1.394;r2=1.395;f=h
>> 
>> Jonas, can you indicate whether this resolution is satisfactory?
> 
> I'm ok with this for now, but I don't want to leave things in this
> state forever without getting feedback from webkit that legacycaller
> is really needed in as many interfaces as it's currently used.
> 
> / Jonas
> 

I have landed the experiment of removing caller from collections (except document.all) in WebKit (http://trac.webkit.org/changeset/95203 for the curious) and it should be available in a nightly sometime later today.  It will take a little while to know what the compatibility story is, but I will try to keep you all informed if data comes in.

-Sam
Received on Thursday, 15 September 2011 18:52:38 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 8 May 2013 19:30:04 UTC