add initial scraper for dashing

add initial scraper for dashing

BUG=skia:

Committed: http://code.google.com/p/skia/source/detail?r=13843

[reed1, 2014-03-17 21:02:35]

sample output

drawPoints = 3060/3060
drawPath = 3/10250
histogram of point-counts for all drawPoints calls
[2] = 3060
vert = 684, hori = 2376

[bsalomon, 2014-03-17 21:10:52]

lgtm (w/ minor suggestion)

https://codereview.chromium.org/196603027/diff/1/include/utils/SkLua.h
File include/utils/SkLua.h (right):

https://codereview.chromium.org/196603027/diff/1/include/utils/SkLua.h#newcode48
include/utils/SkLua.h:48: void pushArray(const SkPoint[], int count, const char
key[] = NULL);
pushArrayPoint?

[reed1, 2014-03-17 21:12:23]

On 2014/03/17 21:10:52, bsalomon wrote:
> lgtm (w/ minor suggestion)
> 
> https://codereview.chromium.org/196603027/diff/1/include/utils/SkLua.h
> File include/utils/SkLua.h (right):
> 
>
https://codereview.chromium.org/196603027/diff/1/include/utils/SkLua.h#newcode48
> include/utils/SkLua.h:48: void pushArray(const SkPoint[], int count, const
char
> key[] = NULL);
> pushArrayPoint?

Yea, thought about that. Do you think we need that given C++ type checking? I
can add it, or perhaps remove the suffix from pushArrayU16

[bsalomon, 2014-03-17 21:17:17]

On 2014/03/17 21:12:23, reed1 wrote:
> On 2014/03/17 21:10:52, bsalomon wrote:
> > lgtm (w/ minor suggestion)
> > 
> > https://codereview.chromium.org/196603027/diff/1/include/utils/SkLua.h
> > File include/utils/SkLua.h (right):
> > 
> >
>
https://codereview.chromium.org/196603027/diff/1/include/utils/SkLua.h#newcode48
> > include/utils/SkLua.h:48: void pushArray(const SkPoint[], int count, const
> char
> > key[] = NULL);
> > pushArrayPoint?
> 
> Yea, thought about that. Do you think we need that given C++ type checking? I
> can add it, or perhaps remove the suffix from pushArrayU16

Nope, definitely not necessary. We could consider some template mojo in the
future so that you write a type-specific func to push an individual element and
then get the array version for free. I'm indifferent about whether the current
code should have "Point" in the name as long as it's consistent with the other
array pushers.

[reed1, 2014-03-17 21:18:31]

On 2014/03/17 21:17:17, bsalomon wrote:
> On 2014/03/17 21:12:23, reed1 wrote:
> > On 2014/03/17 21:10:52, bsalomon wrote:
> > > lgtm (w/ minor suggestion)
> > > 
> > > https://codereview.chromium.org/196603027/diff/1/include/utils/SkLua.h
> > > File include/utils/SkLua.h (right):
> > > 
> > >
> >
>
https://codereview.chromium.org/196603027/diff/1/include/utils/SkLua.h#newcode48
> > > include/utils/SkLua.h:48: void pushArray(const SkPoint[], int count, const
> > char
> > > key[] = NULL);
> > > pushArrayPoint?
> > 
> > Yea, thought about that. Do you think we need that given C++ type checking?
I
> > can add it, or perhaps remove the suffix from pushArrayU16
> 
> Nope, definitely not necessary. We could consider some template mojo in the
> future so that you write a type-specific func to push an individual element
and
> then get the array version for free. I'm indifferent about whether the current
> code should have "Point" in the name as long as it's consistent with the other
> array pushers.

Done

[reed1, 2014-03-17 21:19:45]

The CQ bit was checked by reed@google.com

[commit-bot: I haz the power, 2014-03-17 21:19:48]

CQ is trying da patch. Follow status at
https://skia-tree-status.appspot.com/cq/reed@google.com/196603027/20001

[commit-bot: I haz the power, 2014-03-17 23:09:54]

Change committed as 13843