Re: Re: Keyboard events for accessible RIAs and Games

This looks fine to me, I have question though. Are we sure that the locale
dependent printable keys are uniquely identified by code alone (and not
also location on the keyboard)?


On Sun, Feb 24, 2013 at 5:36 PM, Gary Kacmarcik (Кошмарчик) <
garykac@chromium.org> wrote:

> I've updated the UIEvents document with an initial draft for
> queryKeyCap/queryLocale
>
> https://dvcs.w3.org/hg/d4e/raw-file/tip/source_respec.htm (Section 4.1)
>
>
> On Mon, Feb 18, 2013 at 8:09 AM, Gary Kacmarcik (Кошмарчик) <
> garykac@chromium.org> wrote:
>
>> I'll be updating the document this week. I'll send an update to the list
>> after that happens.
>>
>>
>> On Sat, Feb 16, 2013 at 7:40 AM, Florian Bösch <pyalot@gmail.com> wrote:
>>
>>> Any progress on the speccing of queryKeyCap?
>>>
>>>
>>> On Fri, Feb 1, 2013 at 9:14 PM, Gary Kacmarcik (Кошмарчик) <
>>> garykac@chromium.org> wrote:
>>>
>>>> On Fri, Feb 1, 2013 at 11:42 AM, Travis Leithead <
>>>> travis.leithead@microsoft.com> wrote:
>>>>
>>>>>  I think we should give it another try by including it in our UI
>>>>> Events spec. I like the idea of adding the static queryKeyCap(code) API to
>>>>> Keyboard events. I wonder about the name though. A "key capability" doesn't
>>>>> sound right. Are we querying for a key's locale name? e.g.,
>>>>> queryKeyLocaleName(code)?
>>>>>
>>>>
>>>>  SGTM. I'll add a section to the spec for this.
>>>>
>>>> The KeyCap name refers to the "cap" placed over the keyswitch of the
>>>> physical keyboard.  It's not a great name since there's no guarantee that
>>>> the physical keyboard matches the current locale (although they usually
>>>> do). However, the other (more accurate) names that I was able to come up
>>>> with at the time were all rather unwieldy.
>>>>
>>>> Taking your name as a base, I think we'd need something like
>>>> queryLocaleChar(locale, code) or queryLocaleKey since we'd be returning the
>>>> equivalent of the 'char' (or 'key') attribute.
>>>>
>>>> Thinking briefly about this now:
>>>> * If we return 'char' equivalents, we won't return dead keys or other
>>>> virtual keys.
>>>> * If we return 'key' values, then we need to address how to handle
>>>> non-printable keys like "Shift" and "Home" that people might expect to be
>>>> translated.
>>>> * I think we'll also want the ability to specify modifier keys to apply
>>>> to the 'code' (to generate shifted or AltGr'ed versions).
>>>>
>>>> I'll formalize this a bit more and send something out for comments.
>>>>
>>>> -Gary
>>>>
>>>>
>>>
>>
>