Bug 162917 – [Win][Direct2D] Add initial D2D GraphicsContext implementation

• Status: RESOLVED
• Resolution: FIXED
• Component: WebCore Misc.

[Brent Fulgham]

This patch adds a Direct2D GraphicsContext implementation, which will exist in parallel to the existing Cairo and CG implementations.

This patch simply adds the new files. It does not incorporate them into the build.

[Brent Fulgham]

Created attachment 290615 [details]
Patch

[WebKit Commit Bot]

Attachment 290615 [details] did not pass style-queue:


ERROR: Source/WebCore/platform/graphics/Color.h:48:  D2D1_VECTOR_4F is incorrectly named. Don't use underscores in your identifier names.  [readability/naming/underscores] [4]
ERROR: Source/WebCore/platform/graphics/win/ColorDirect2D.cpp:49:  D2D1_VECTOR_4F is incorrectly named. Don't use underscores in your identifier names.  [readability/naming/underscores] [4]
Total errors found: 2 in 11 files


If any of these errors are false positives, please file a bug against check-webkit-style.

[Dean Jackson]

Comment on attachment 290615 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=290615&action=review

rs=me, although I feel like this is so much code it should go through real review with someone who knows D2D :|

> Source/WebCore/ChangeLog:12
> +        features on Windows using Direct2D.
> +
> +
> +        No new tests until complete backend lands.

Blank lines!

> Source/WebCore/platform/graphics/win/ColorDirect2D.cpp:56
> +}
> +
> +

More extra blank lines!

> Source/WebCore/platform/graphics/win/GraphicsContextDirect2D.cpp:177
> +        /*
> +        FIXME: Implement DisplayListRecorder support for drawNativeImage
> +        m_displayListRecorder->drawNativeImage(image, imageSize, destRect, srcRect, op, blendMode, orientation);
> +        */

Why not // comments?

> Source/WebCore/platform/graphics/win/GraphicsContextDirect2D.cpp:214
> +    // ImageOrientation expects the origin to be at (0, 0)

Nit: End with .

> Source/WebCore/platform/graphics/win/GraphicsContextDirect2D.cpp:839
> +        // ShadowBlur contextShadow(m_state);
> +        // contextShadow.drawRectShadow(*this, FloatRoundedRect(rect));

Does this need a FIXME?

> Source/WebCore/platform/graphics/win/GraphicsContextDirect2D.cpp:865
> +        // ShadowBlur contextShadow(m_state);
> +        // contextShadow.drawRectShadow(*this, FloatRoundedRect(rect));
> +        notImplemented();

Same here.

> Source/WebCore/platform/graphics/win/GraphicsContextDirect2D.cpp:890
> +        // ShadowBlur contextShadow(m_state);
> +        // contextShadow.drawRectShadow(*this, rect);
> +        notImplemented();

And here.

> Source/WebCore/platform/graphics/win/GraphicsContextDirect2D.cpp:960
> +        // ShadowBlur contextShadow(m_state);
> +        // contextShadow.drawRectShadow(*this, rect);
> +        notImplemented();

And probably many other places

[Brent Fulgham]

Comment on attachment 290615 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=290615&action=review

>> Source/WebCore/ChangeLog:12
>> +        No new tests until complete backend lands.
> 
> Blank lines!

Whoops!

>> Source/WebCore/platform/graphics/win/ColorDirect2D.cpp:56
>> +
> 
> More extra blank lines!

:-(.  Fixed.

>> Source/WebCore/platform/graphics/win/GraphicsContextDirect2D.cpp:177
>> +        */
> 
> Why not // comments?

Sure! Done.

>> Source/WebCore/platform/graphics/win/GraphicsContextDirect2D.cpp:214
>> +    // ImageOrientation expects the origin to be at (0, 0)
> 
> Nit: End with .

OK!

>> Source/WebCore/platform/graphics/win/GraphicsContextDirect2D.cpp:839
>> +        // contextShadow.drawRectShadow(*this, FloatRoundedRect(rect));
> 
> Does this need a FIXME?

Sure. And all the other places. Done.

>> Source/WebCore/platform/graphics/win/GraphicsContextDirect2D.cpp:865
>> +        notImplemented();
> 
> Same here.

Done.

>> Source/WebCore/platform/graphics/win/GraphicsContextDirect2D.cpp:890
>> +        notImplemented();
> 
> And here.

Done.

>> Source/WebCore/platform/graphics/win/GraphicsContextDirect2D.cpp:960
>> +        notImplemented();
> 
> And probably many other places

Done.

[Brent Fulgham]

(In reply to comment #3)
> Comment on attachment 290615 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=290615&action=review
> 
> rs=me, although I feel like this is so much code it should go through real
> review with someone who knows D2D :|

Sadly, there seems to be no such person. I'm hoping one of the EA or Sony people will jump in at some point with some advice. Until then, I just want to get this prototype in place so others can build it and hack on it.

[Brent Fulgham]

Committed r206773: <http://trac.webkit.org/changeset/206773>

[Csaba Osztrogonác]

(In reply to comment #6)
> Committed r206773: <http://trac.webkit.org/changeset/206773>

It broke the Windows build, see build.webkit.org for details.

[Brent Fulgham]

(In reply to comment #7)
> (In reply to comment #6)
> > Committed r206773: <http://trac.webkit.org/changeset/206773>
> 
> It broke the Windows build, see build.webkit.org for details.

Build correction landed:

Committed r206780: <http://trac.webkit.org/changeset/206780>