W3C home > Mailing lists > Public > www-archive@w3.org > April 2015

Re: Need secure Web Socket support for w3c-test.org

From: Arthur Barstow <art.barstow@gmail.com>
Date: Mon, 27 Apr 2015 10:27:55 -0400
Message-ID: <553E476B.2020105@gmail.com>
To: "Michael[tm] Smith" <mike@w3.org>, Philippe Le Hegaret <plh@w3.org>
CC: James Graham <james@hoppipolla.co.uk>, "team-webapps@w3.org" <team-webapps@w3.org>, "www-archive@w3.org" <www-archive@w3.org>
Hi Mike, Philippe,

What is the status of this request (aka 
<https://github.com/w3c/web-platform-tests/issues/1690>)?

-Thanks, Art

On 4/1/15 7:09 AM, James Graham wrote:
> On 01/04/15 02:18, Michael[tm] Smith wrote:
>> +jgraham
>>
>> Hi Art, Hi Philippe,
>>
>> I’ve talked to James about the wss: support and based on that discussion I
>> think the ETA is that we could get the limitation removed within one month.
>>
>> But maybe James can chime in to let me know if that’s too optimistic.
> So the deal is that the way that the tests run in wptrunner in gecko
> uses autoproxy which is kind of a pain with TLS (it first sends a
> CONNECT message unencrypted and then starts the encrypted session). This
> is needed because I can't modify the hosts file on our test machines.
> It's a pain to hack pywebsocket to support that flow, which is why we
> don't have wss already.
>
> However I recently discovered that we have a gecko pref to set up a
> profile-specific equivalent of /etc/hosts that I can use instead of all
> this autoproxy stuff. If I change the gecko runner to use that enabling
> wss should be trivial (I hope!).
>
> How soon this happens basically depends on how soon I get to work on
> this, which I'm not sure about. I expect it will either be "almost right
> away" or "not for a while", depending on whether I get some large
> high-priority item I have to do first.
Received on Monday, 27 April 2015 14:28:28 UTC

This archive was generated by hypermail 2.3.1 : Monday, 27 April 2015 14:28:28 UTC