Enable init'ed but unused var warning on windows for closer parity with mac/linux warnings.

Enable init'ed but unused var warning on windows for closer parity with mac/linux warnings.
Committed: https://code.google.com/p/skia/source/detail?r=8307

[bsalomon, 2013-03-20 19:31:15]

This catches warnings that broke the linux/mac builds but went undetected on
windows (r8264-5)

[bsalomon, 2013-03-21 12:57:14]

Adding reed@

[reed1, 2013-03-21 13:14:24]

https://codereview.chromium.org/12536008/diff/1/src/device/xps/SkXPSDevice.cpp
File src/device/xps/SkXPSDevice.cpp (left):

https://codereview.chromium.org/12536008/diff/1/src/device/xps/SkXPSDevice.cp...
src/device/xps/SkXPSDevice.cpp:953: SkShader::GradientType gradientType =
shader->asAGradient(&info);
Why did we eliminate this line, given that we inspect the contents of color 2
lines down?

https://codereview.chromium.org/12536008/diff/1/src/device/xps/SkXPSDevice.cp...
src/device/xps/SkXPSDevice.cpp:969: shader->asAGradient(&info);
Why did we eliminate this call, given that we inspect info in the next line of
code?

https://codereview.chromium.org/12536008/diff/1/src/opts/SkBlitRect_opts_SSE2...
File src/opts/SkBlitRect_opts_SSE2.cpp (right):

https://codereview.chromium.org/12536008/diff/1/src/opts/SkBlitRect_opts_SSE2...
src/opts/SkBlitRect_opts_SSE2.cpp:121: colorA = 255; // skip below if () for
now...
This is backwards from the old code.

The old code *never* took the fast case for opaque (not sure why, perhaps a
bug?). Your new code *always* takes the fast-opaque case.

https://codereview.chromium.org/12536008/diff/1/src/views/win/SkOSWindow_win.cpp
File src/views/win/SkOSWindow_win.cpp (right):

https://codereview.chromium.org/12536008/diff/1/src/views/win/SkOSWindow_win....
src/views/win/SkOSWindow_win.cpp:205: SetDIBitsToDevice(hdc,
can we say

(void)SetDIBitsToDevice

to sorta document that we know this guy returns something?

https://codereview.chromium.org/12536008/diff/1/src/views/win/skia_win.cpp
File src/views/win/skia_win.cpp (right):

https://codereview.chromium.org/12536008/diff/1/src/views/win/skia_win.cpp#ne...
src/views/win/skia_win.cpp:141: GetModuleFileName(NULL, exename, 1024);
same comment about (void)GetModuleFileName

[bsalomon, 2013-03-21 13:45:29]

https://codereview.chromium.org/12536008/diff/1/src/device/xps/SkXPSDevice.cpp
File src/device/xps/SkXPSDevice.cpp (left):

https://codereview.chromium.org/12536008/diff/1/src/device/xps/SkXPSDevice.cp...
src/device/xps/SkXPSDevice.cpp:953: SkShader::GradientType gradientType =
shader->asAGradient(&info);
On 2013/03/21 13:14:24, reed1 wrote:
> Why did we eliminate this line, given that we inspect the contents of color 2
> lines down?

I misunderstood the code and thought these calls were redundant. Updated to just
remove the unused var.

https://codereview.chromium.org/12536008/diff/1/src/opts/SkBlitRect_opts_SSE2...
File src/opts/SkBlitRect_opts_SSE2.cpp (right):

https://codereview.chromium.org/12536008/diff/1/src/opts/SkBlitRect_opts_SSE2...
src/opts/SkBlitRect_opts_SSE2.cpp:121: colorA = 255; // skip below if () for
now...
On 2013/03/21 13:14:24, reed1 wrote:
> This is backwards from the old code.
> 
> The old code *never* took the fast case for opaque (not sure why, perhaps a
> bug?). Your new code *always* takes the fast-opaque case.
> 
oh duh... fixed.

https://codereview.chromium.org/12536008/diff/1/src/views/win/SkOSWindow_win.cpp
File src/views/win/SkOSWindow_win.cpp (right):

https://codereview.chromium.org/12536008/diff/1/src/views/win/SkOSWindow_win....
src/views/win/SkOSWindow_win.cpp:205: SetDIBitsToDevice(hdc,
On 2013/03/21 13:14:24, reed1 wrote:
> can we say
> 
> (void)SetDIBitsToDevice
> 
> to sorta document that we know this guy returns something?

Done.

https://codereview.chromium.org/12536008/diff/1/src/views/win/skia_win.cpp
File src/views/win/skia_win.cpp (right):

https://codereview.chromium.org/12536008/diff/1/src/views/win/skia_win.cpp#ne...
src/views/win/skia_win.cpp:141: GetModuleFileName(NULL, exename, 1024);
On 2013/03/21 13:14:24, reed1 wrote:
> same comment about (void)GetModuleFileName

Done.

[bsalomon, 2013-03-21 19:59:04]

ok to land?

[reed1, 2013-03-21 20:24:20]

lgtm

[bsalomon, 2013-03-21 20:29:12]

Committed patchset #3 manually as r8307 (presubmit successful).