Also pass unclamped delta and velocity values to GestureEventDetails. (DEPRECATED)

NOTE: this is superseded by https://codereview.chromium.org/258803002. 

Also pass unclamped delta and velocity values to GestureEventDetails.

BUG=332418
TEST=GestureProviderTest.FlingEventSequence,
  GestureProviderTest.FlingAndCancelLongPress
  GestureProviderTest.ScrollUpdateValues

[tdresser, 2014-04-24 19:48:59]

jdduke@, can you take a look at ui/events/gesture_detection/*?

sadrul@, can you take a look at gesture_event_details.cc?

[jdduke (slow), 2014-04-24 19:58:53]

A few nits, and I would love to know how the term "ordinal" became associated
with raw scroll values :)

https://codereview.chromium.org/256593003/diff/1/ui/events/gesture_detection/...
File ui/events/gesture_detection/gesture_provider.cc (right):

https://codereview.chromium.org/256593003/diff/1/ui/events/gesture_detection/...
ui/events/gesture_detection/gesture_provider.cc:322: float ordinal_distance_x =
distance_x;
Nit: Let's make these const above and below suggesting they're unmodified. Also,
do you want the slop removed?

https://codereview.chromium.org/256593003/diff/1/ui/events/gesture_detection/...
File ui/events/gesture_detection/gesture_provider_unittest.cc (right):

https://codereview.chromium.org/256593003/diff/1/ui/events/gesture_detection/...
ui/events/gesture_detection/gesture_provider_unittest.cc:320: EXPECT_EQ(8000,
GetMostRecentGestureEvent().details.velocity_x());
We should probably get 8000 using the |maximum_fling_velocity| constant from
|GetDefaultConfig()| (probably add a method similar to |GetTouchSlop()| above).

https://codereview.chromium.org/256593003/diff/1/ui/events/gesture_detection/...
ui/events/gesture_detection/gesture_provider_unittest.cc:608: EXPECT_EQ(-delta_y
/ 2, gesture.details.scroll_y());
Might as well round this out with 

EXPECT_EQ(-delta_y / 2, gesture.details.scroll_y_ordinal());

https://codereview.chromium.org/256593003/diff/1/ui/events/gesture_event_deta...
File ui/events/gesture_event_details.cc (right):

https://codereview.chromium.org/256593003/diff/1/ui/events/gesture_event_deta...
ui/events/gesture_event_details.cc:79: break;
Should we move the CHECK above to the default case and use a NOTREACHED()?

[sadrul, 2014-04-24 20:00:34]

On 2014/04/24 19:58:53, jdduke wrote:
> A few nits, and I would love to know how the term "ordinal" became associated
> with raw scroll values :)

Some cros-ism.

[sadrul, 2014-04-25 12:13:13]

https://codereview.chromium.org/256593003/diff/1/ui/events/gesture_event_deta...
File ui/events/gesture_event_details.cc (left):

https://codereview.chromium.org/256593003/diff/1/ui/events/gesture_event_deta...
ui/events/gesture_event_details.cc:29: case ui::ET_SCROLL_FLING_START:
Do NOTREACHED() for SCROLL_UPDATE and SCROLL_FLING_START in here instead? (for
readability)