A mouse event created from the same native event will be marked as repeated.

A mouse event created from the same native event will be marked as repeated.

If a mouse event has been created from the same native event with the
same time stamp, it shouldn't be marked as repeated.

R=sadrul@chromium.org
BUG=474282
TEST=events_unittests --gtest_filter=EventTest.*

Committed: https://crrev.com/e904708872a797ce2ecbca3d33427d388a89f3fc
Cr-Commit-Position: refs/heads/master@{#324271}

[afakhry, 2015-04-06 19:15:22]

Hi Sadrul,
Please review. Thanks!

[sadrul, 2015-04-07 17:16:22]

sadrul@chromium.org changed reviewers:
+ oshima@chromium.org

[sadrul, 2015-04-07 17:16:22]

+oshima@

We compare timestamps to check for key-repeats, so I suppose this is fine. lgtm

But I don't think we actually have any guarantee from the underlying platform
that two events cannot have the same timestamp. Are there other fixes/heuristics
we could use for this?

[sadrul, 2015-04-07 17:16:35]

The CQ bit was checked by sadrul@chromium.org to run a CQ dry run

[sadrul, 2015-04-07 17:16:41]

The CQ bit was unchecked by sadrul@chromium.org

[oshima, 2015-04-07 17:51:44]

On 2015/04/07 17:16:22, sadrul wrote:
> +oshima@
> 
> We compare timestamps to check for key-repeats, so I suppose this is fine.
lgtm
> 
> But I don't think we actually have any guarantee from the underlying platform
> that two events cannot have the same timestamp. Are there other
fixes/heuristics
> we could use for this?

I first tried to compare native event pointers, but this didn't work on Windows
because
NativeEvent is MSG struct.

Another approach is to inject a flag to native event in dedicated factory
method, and use
that when checking if it's repeated. Downside is that it'll be platform
dependent, and I don't
know the safe way to do that on Win/Mac. (On X, we probably can use send_event)

On the other hand, we're already using timestamp to detect repeat with
assumption that
timestamp is reasonably accurate, so I'm fine with this too.

[afakhry, 2015-04-08 00:01:02]

Modified EventsXTest.ClickCount to account for the change.

[oshima, 2015-04-08 00:12:54]

lgtm

[afakhry, 2015-04-08 17:46:36]

The CQ bit was checked by afakhry@chromium.org

[afakhry, 2015-04-08 17:46:36]

The patchset sent to the CQ was uploaded after l-g-t-m from sadrul@chromium.org
Link to the patchset: https://codereview.chromium.org/1068463003/#ps20001
(title: "Modified EventsXTest.ClickCount to account for the change.")

[commit-bot: I haz the power, 2015-04-08 17:47:50]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1068463003/20001

[commit-bot: I haz the power, 2015-04-08 19:32:23]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2015-04-08 19:33:08]

Patchset 2 (id:??) landed as
https://crrev.com/e904708872a797ce2ecbca3d33427d388a89f3fc
Cr-Commit-Position: refs/heads/master@{#324271}