Hi Konstantin,

Thanks for your prompt response :-)

I didn't notice this bug and the respective patches.

I read the report and still find it hard to understand, if "User Agent sniffing 
is a terrible way to determine whether a browser supports certain features", 
what is the correct way of doing it?

Sent: Thursday, April 26, 2018 at 1:16 PM
From: "Konstantin Tokarev" <annu...@yandex.ru>
To: "Ricky Young" <ri...@gmx.com>, "webkit-dev@lists.webkit.org" 
Subject: Re: [webkit-dev] New iOS versions sending bogus User-Agent build data

26.04.2018, 13:08, "Ricky Young" <ri...@gmx.com>:
> Until iOS 11.3, User-Agent build data contained the build that corresponds to 
> the device (e.g. Mobile/15D100 if the requesting device was running iOS 
> 11.2.6).
> Since iOS 11.3 was released I noticed that the User-Agent that is reported to 
> the brower is reporting a bogus build, i.e Mobile/15E148, regardless to the 
> iOS version that the requesting device is running. I would expect that a 
> device that runs 11.3 will reprot a different User-Agent than a device that 
> runs 11.3.1 and certianly than a device that runs 11.4 (Beta 1 & 2).
> I just wonder if my observation is an expected behavior, and if so, what is 
> the intention behind it. If this behavior is not intended and expected, I 
> would love to research it and propose a patch (would be my first Open Source 
> contribution).


User-Agent string is frozen since https://bugs.webkit.org/show_bug.cgi?id=180365

webkit-dev mailing list

Reply via email to