993234 - Implement D3E KeyboardEvent.isComposing

Status: RESOLVED FIXED

(Core :: DOM: Events, enhancement)

Comment 1

[Masayuki Nakano [:masayuki] (he/him)(JST, +0900)]

https://dvcs.w3.org/hg/dom3events/raw-file/tip/html/DOM3-Events.html#widl-KeyboardEvent-isComposing

> true if the key event occurs as part of a composition session, i.e., after a compositionstart event and before the corresponding compositionend event.
> 
> The un-initialized value of this attribute MUST be false.

Comment 2

[Masayuki Nakano [:masayuki] (he/him)(JST, +0900)]

Attachment: part.1 Implement KeyboardEvent.isComposing

We can implement this bug very simply.

widget shouldn't set the isComposing value at dispatching key events. Instead of that, EventStateManager::PreHandleEvent() can do it.

Comment 3

[Masayuki Nakano [:masayuki] (he/him)(JST, +0900)]

Attachment: part.2 Add tests for KeyboardEvent.isComposing

Currently, dispatching key events during composition causes some assertions. Therefore, this patch adds |SimpleTest.expectAssertions(0, 16)|. Note that on Linux, the assertion doesn't occur. Therefore, it specifies 0-16.

Comment 4

[Olli Pettay [:smaug][[e-mail]]]

Comment on attachment 8403248 [details] [diff] [review]
part.1 Implement KeyboardEvent.isComposing

I wouldn't add isComposing to the .idl, nor 
KeyboardEvent::GetIsComposing.

New stuff can go to .webidl only, unless there is some good reason
to add to .idl too.

So, remove the .idl stuff, r=me

Comment 5

[Olli Pettay [:smaug][[e-mail]]]

Comment on attachment 8403251 [details] [diff] [review]
part.2 Add tests for KeyboardEvent.isComposing

I'd like to see at least some test for keypress too, at least to check that
it returns sane value, and not some random value for isComposing.

Comment 6

[Masayuki Nakano [:masayuki] (he/him)(JST, +0900)]

Attachment: part.1 Implement KeyboardEvent.isComposing (r=smaug)

Thanks, then, this doesn't need sr.

Comment 7

[Masayuki Nakano [:masayuki] (he/him)(JST, +0900)]

Attachment: part.2 Add tests for KeyboardEvent.isComposing

Although, keypress event should never be fired during composition even if after bug 354358 is fixed. But it's okay to include the test for something wired case.

Comment 8

[Masayuki Nakano [:masayuki] (he/him)(JST, +0900)]

And I hope VK_INSERT will never be handled by our editor...

Comment 9

[Masayuki Nakano [:masayuki] (he/him)(JST, +0900)]

https://hg.mozilla.org/integration/mozilla-inbound/rev/839bf025c534
https://hg.mozilla.org/integration/mozilla-inbound/rev/d40a8916b7e7

Comment 10

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/839bf025c534
https://hg.mozilla.org/mozilla-central/rev/d40a8916b7e7

Comment 11

[Jean-Yves Perrier [:teoli]]

Documented:
https://developer.mozilla.org/en-US/docs/Web/API/KeyboardEvent.isComposing
https://developer.mozilla.org/en-US/docs/Web/API/KeyboardEvent
and
https://developer.mozilla.org/en-US/Firefox/Releases/31#Interfaces.2FAPIs.2FDOM