KeyboardInputGesture.send should use SendInput rather than Keybd_event

[nvaccessAuto]

Reported by mdcurran on 2014-07-15 07:24
Currently KeyboardInputGesture.send uses multiple Keybd_event calls to send a key press. It should be changed to use one singleSendInput call, which will guarantee that all key downs and key ups complete within the one call disallowing the possibility for the key downs to be sent, and then the user to press a key themselves, and then the key ups to be sent.

[nvaccessAuto]

Comment 1 by mdcurran on 2014-07-16 10:26
Incubating in c6d7b24
Changes:
Added labels: incubating

[nvaccessAuto]

Comment 2 by Michael Curran <mick@... on 2014-07-18 00:33
In [45bb596]:

keyboardHandler.internal_keyDownEvent: when sending the special 0xff vk code when trapping modifiers Re #3472, explicitly provide the modifiers from the blocked gesture to the 0xff gesture as now NVDA releases any unneeded modifiers in KeyboardInputGesture.send, re #4294.

[nvaccessAuto]

Comment 3 by Michael Curran <mick@... on 2014-07-18 00:34
In [fdc8b50]:

Merge branch 't4294' into next. Incubates #4294

[nvaccessAuto]

Comment 4 by jteh on 2014-08-05 01:11
Holding this back because it isn't actually necessary for the purpose for which it was originally intended. It's still probably a good idea, but it needs wider testing on master after the rc.

[nvaccessAuto]

Comment 5 by James Teh <jamie@... on 2014-09-16 03:10
In [c3f28d9]:

Back out t4294.

This is at least a partial cause of #4381. I think this is due to the removal of wx.Yield in KeyboardInputGesture.send. If there is pending stuff in the queue, this would make the first wx.Yield indirectly called by EditableText._hasCaretMoved take longer, so waiting for caret movement would time out sooner than it did before. Technically, this isn't a bug in #4294.
However, we have no current need for #4294 any more and we've had reports that suggest it might be causing modifiers to stick. Therefore, until we have time to debug this or until there is a specific need for it, it's simpler to just back it out.
This reverts commits 6fb9c104..45bb5962.
Re #4294, #4381.

[nvaccessAuto]

Comment 6 by jteh on 2014-09-16 03:11
Changes:
Milestone changed from next to None
Removed labels: incubating

[jcsteh]

Status:

• This was implemented and then backed out due to it causing a regression; see KeyboardInputGesture.send should use SendInput rather than Keybd_event #4294 (comment).
• The underlying cause of the regression might be less relevant after Increase default caret move timeout from 30 to 100 ms and make it configurable #7201.
• While this is probably the "right" thing to do, there's still no specific motivation for this at present, so the effort and risk involved is hard to justify.

Closing as won't fix for now. We can reopen if we find a specific reason to do this.