Bug 37692 – [chromium] ignore IME events if the keydown event was prevented

• Status: RESOLVED
• Resolution: FIXED
• Component: New Bugs

[Tony Chang]

[chromium] ignore IME events if the keydown event was prevented

[Tony Chang]

Created attachment 53504 [details]
Patch

[Tony Chang]

Discussion of this bug can be found http://code.google.com/p/chromium/issues/detail?id=9883 and a review by hbono and suzhe can be found http://codereview.chromium.org/1630015/show

[Tony Chang]

Oh, and no layout test because I don't know how to test IME behavior changes.  I guess at some point we should build functionality into eventSender to send IME events.

[Jian Li]

Comment on attachment 53504 [details]
Patch

Generally looks good. Do we have a layout test for this?

[Tony Chang]

(In reply to comment #4)
> (From update of attachment 53504 [details])
> Generally looks good. Do we have a layout test for this?

Since posting comment #3, I learned about textInputController, so I tried to write a layout test.  However, this doesn't work since textInputController call methods directly on WebFrame and this check is in WebView.  I don't think it would make sense to move this check into WebFrame (since WebView knows about the state and does similar checks).

Maybe we could change textInputController.setMarkedText to call WebView::handleCompositionEvent instead, but that seems to deviate from DRT mac.  It's not clear to me that that's what we want either.

[Eric Seidel (no email)]

Comment on attachment 53504 [details]
Patch

How do we test this?

[Tony Chang]

Created attachment 55514 [details]
Patch

[Tony Chang]

(In reply to comment #6)
> (From update of attachment 53504 [details])
> How do we test this?

Since I wasn't able to write a layout test for it, I've added a manual test.

[Tony Chang]

Committed r59735: <http://trac.webkit.org/changeset/59735>