Refine tap disambiguation UMA to track same-node/different-node.

Refine tap disambiguation UMA to track same-node/different-node.

The Android tap disambiguation feature makes users retap on a zoomed screenshot
when the user tapped on tiny close-together links.  This feature is
currently being considered for removal.  If the user's tap would've been
anyway, in the first place, accurately resolved to the node they
ultimately selected after disambiguation, then the popup UI achieved
nothing and we should consider this a failure case.

Notes:
- Because this must be measured in the renderer and the synthetic
taps sent before had nothing to identify them as relating to
disambiguation, I introduced a new ViewMsg_ResolveTapDisambiguation.
I think this is anyway a better architecture -- this is a highly unusual
scenario not much resembling an ordinary touch gesture, and risks
interacting weirdly with ordinary event routing.
- Although handling this in WebViewImpl is broken with OOPIF, this feature
is already broken with OOPIF.  That's one of the reasons it's considered
for removal.
- I'm using hit node top-left offset relative the document as an
identifier the DOM node itself.  This seems less sketchy than persisting
a pointer, and it's good enough for this use case because the page is
generally static during a tap disambiguation (if it self-scrolls, for
instance, the popup will be dismissed with category OTHER).

BUG=705117
Review-Url: https://codereview.chromium.org/2783233004
Cr-Commit-Position: refs/heads/master@{#461290}
Committed: https://chromium.googlesource.com/chromium/src/+/1b55156e5277d39580f4063af924ec7347375ebc

[aelias_OOO_until_Jul13, 2017-03-31 01:40:38]

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

[commit-bot: I haz the power, 2017-03-31 01:41:33]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[aelias_OOO_until_Jul13, 2017-03-31 02:08:16]

aelias@chromium.org changed reviewers:
+ boliu@chromium.org, isherman@chromium.org

[aelias_OOO_until_Jul13, 2017-03-31 02:08:17]

PTAL,

boliu@ for content/browser/android/
isherman@ for tools/metrics/histograms/

[aelias_OOO_until_Jul13, 2017-03-31 02:08:54]

aelias@chromium.org changed reviewers:
+ tsepez@chromium.org

[aelias_OOO_until_Jul13, 2017-03-31 02:09:04]

Also tsepez@ for view_messages.h

[commit-bot: I haz the power, 2017-03-31 03:32:54]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-03-31 03:32:55]

Dry run: This issue passed the CQ dry run.

[Tom Sepez, 2017-03-31 16:04:05]

messages LGTM

[Ilya Sherman, 2017-03-31 17:09:06]

https://codereview.chromium.org/2783233004/diff/1/tools/metrics/histograms/hi...
File tools/metrics/histograms/histograms.xml (right):

https://codereview.chromium.org/2783233004/diff/1/tools/metrics/histograms/hi...
tools/metrics/histograms/histograms.xml:112672: +  <int value="3" label="Tapped
Inside, Same Node as Without Disambiguation"/>
It looks like you've changed the meaning of this bucket -- is that correct?  If
so, please use a different value to back this new semantic meaning, and reserve
the value "3" as a deprecated/no-longer-used value.

[boliu, 2017-03-31 17:36:40]

content/browser lgtm

[aelias_OOO_until_Jul13, 2017-03-31 21:21:46]

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

[aelias_OOO_until_Jul13, 2017-03-31 21:21:57]

https://codereview.chromium.org/2783233004/diff/1/tools/metrics/histograms/hi...
File tools/metrics/histograms/histograms.xml (right):

https://codereview.chromium.org/2783233004/diff/1/tools/metrics/histograms/hi...
tools/metrics/histograms/histograms.xml:112672: +  <int value="3" label="Tapped
Inside, Same Node as Without Disambiguation"/>
On 2017/03/31 at 17:09:05, Ilya Sherman wrote:
> It looks like you've changed the meaning of this bucket -- is that correct? 
If so, please use a different value to back this new semantic meaning, and
reserve the value "3" as a deprecated/no-longer-used value.

Done.

[commit-bot: I haz the power, 2017-03-31 21:22:53]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[Ilya Sherman, 2017-03-31 21:37:13]

Metrics LGTM, thanks.

[aelias_OOO_until_Jul13, 2017-03-31 21:43:44]

The CQ bit was unchecked by aelias@chromium.org

[aelias_OOO_until_Jul13, 2017-03-31 21:43:47]

The CQ bit was checked by aelias@chromium.org

[aelias_OOO_until_Jul13, 2017-03-31 21:43:47]

The patchset sent to the CQ was uploaded after l-g-t-m from tsepez@chromium.org,
boliu@chromium.org
Link to the patchset: https://codereview.chromium.org/2783233004/#ps20001
(title: "Deprecate existing field")

[commit-bot: I haz the power, 2017-03-31 21:44:17]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-04-01 00:46:09]

CQ is committing da patch.

Bot data: {"patchset_id": 20001, "attempt_start_ts": 1490996627036640,
"parent_rev": "b332a9d54d5e06c6013c293e293f625f840d711d", "commit_rev":
"1b55156e5277d39580f4063af924ec7347375ebc"}

[commit-bot: I haz the power, 2017-04-01 00:47:24]

Description was changed from

==========
Refine tap disambiguation UMA to track same-node/different-node.

The Android tap disambiguation feature makes users retap on a zoomed screenshot
when the user tapped on tiny close-together links.  This feature is
currently being considered for removal.  If the user's tap would've been
anyway, in the first place, accurately resolved to the node they
ultimately selected after disambiguation, then the popup UI achieved
nothing and we should consider this a failure case.

Notes:
- Because this must be measured in the renderer and the synthetic
taps sent before had nothing to identify them as relating to
disambiguation, I introduced a new ViewMsg_ResolveTapDisambiguation.
I think this is anyway a better architecture -- this is a highly unusual
scenario not much resembling an ordinary touch gesture, and risks
interacting weirdly with ordinary event routing.
- Although handling this in WebViewImpl is broken with OOPIF, this feature
is already broken with OOPIF.  That's one of the reasons it's considered
for removal.
- I'm using hit node top-left offset relative the document as an
identifier the DOM node itself.  This seems less sketchy than persisting
a pointer, and it's good enough for this use case because the page is
generally static during a tap disambiguation (if it self-scrolls, for
instance, the popup will be dismissed with category OTHER).

BUG=705117
==========

to

==========
Refine tap disambiguation UMA to track same-node/different-node.

The Android tap disambiguation feature makes users retap on a zoomed screenshot
when the user tapped on tiny close-together links.  This feature is
currently being considered for removal.  If the user's tap would've been
anyway, in the first place, accurately resolved to the node they
ultimately selected after disambiguation, then the popup UI achieved
nothing and we should consider this a failure case.

Notes:
- Because this must be measured in the renderer and the synthetic
taps sent before had nothing to identify them as relating to
disambiguation, I introduced a new ViewMsg_ResolveTapDisambiguation.
I think this is anyway a better architecture -- this is a highly unusual
scenario not much resembling an ordinary touch gesture, and risks
interacting weirdly with ordinary event routing.
- Although handling this in WebViewImpl is broken with OOPIF, this feature
is already broken with OOPIF.  That's one of the reasons it's considered
for removal.
- I'm using hit node top-left offset relative the document as an
identifier the DOM node itself.  This seems less sketchy than persisting
a pointer, and it's good enough for this use case because the page is
generally static during a tap disambiguation (if it self-scrolls, for
instance, the popup will be dismissed with category OTHER).

BUG=705117

Review-Url: https://codereview.chromium.org/2783233004
Cr-Commit-Position: refs/heads/master@{#461290}
Committed:
https://chromium.googlesource.com/chromium/src/+/1b55156e5277d39580f4063af924...
==========

[commit-bot: I haz the power, 2017-04-01 00:47:26]

Committed patchset #2 (id:20001) as
https://chromium.googlesource.com/chromium/src/+/1b55156e5277d39580f4063af924...