Refactored SkColorSpace and added in a Lab PCS GM

src/core/SkColorSpace_A2B0.h

Index: src/core/SkColorSpace_A2B0.h
diff --git a/src/core/SkColorSpace_A2B0.h b/src/core/SkColorSpace_A2B0.h
new file mode 100644
index 0000000000000000000000000000000000000000..02a8055baa24d73499da8ef7a93602f25ca0c049
--- /dev/null
+++ b/src/core/SkColorSpace_A2B0.h
@@ -0,0 +1,67 @@
+/*
+ * Copyright 2016 Google Inc.
+ *
+ * Use of this source code is governed by a BSD-style license that can be
+ * found in the LICENSE file.
+ */
+
+#ifndef SkColorSpace_A2B0_DEFINED
+#define SkColorSpace_A2B0_DEFINED
+
+#include "SkColorSpace_Base.h"
+
+class SkColorSpace_A2B0 : public SkColorSpace_Base {
+public:
+    const SkMatrix44* requestToXYZD50() const override { return nullptr; }

[msarett, 2016/10/07 01:48:36]

Brief comment on why these return null?

[raftias, 2016/10/10 20:37:33]

Done.

+    
+    const SkMatrix44* requestFromXYZD50() const override { return nullptr; }
+
+    enum class PCS : uint8_t {

[msarett, 2016/10/07 01:48:36]

Same enum class question as in SkColorSpace_Base

[raftias, 2016/10/10 20:37:32]

Acknowledged.

+        kLAB,
+        kXYZ
+    };
+    
+    PCS pcs() const { return fPCS; }

[msarett, 2016/10/07 01:48:35]

Maybe lead with a summary comment about all the cool stuff we can find in A2B
tags?

[raftias, 2016/10/10 20:37:32]

Done.

+    
+    SkGammaNamed aCurveNamed() const { return fACurveNamed; }

[msarett, 2016/10/07 01:48:35]

Maybe bother trying to "name" the A, B, or M curves?  At least at first?  Will
this interfere with sharing the parsing code?

[raftias, 2016/10/10 20:37:32]

Like CLUTGamma, FinalGamma, MatrixGamma, or did you mean something else?

[msarett, 2016/10/11 13:40:30]

Sorry I mean, let's not have GammaNamed getters (or GammaNamed fields at all) if
it doesn't complicate the implementation.

This might not necessarily be a useful optimization in this case.

+    
+    const SkGammas* aCurve() const { return fACurve.get(); }
+    
+    SkGammaNamed mCurveNamed() const { return fMCurveNamed; }
+    
+    const SkGammas* mCurve() const { return fMCurve.get(); }
+    
+    SkGammaNamed bCurveNamed() const { return fBCurveNamed; }
+    
+    const SkGammas* bCurve() const { return fBCurve.get(); }
+
+    const SkColorLookUpTable* colorLUT() const { return fColorLUT.get(); }
+    
+    const SkMatrix44 toPCS() const { return fToPCS; }

[msarett, 2016/10/07 01:48:35]

Maybe just call this "matrix"?  I don't have an intuitive understanding of what
this is actually meant to do.  But it seems like the CLUT typically does most of
the "to PCS" work.

[raftias, 2016/10/10 20:37:32]

Done.

+
+protected:
+    Type type() const override { return Type::kA2B0; }
+
+private:
+    SkColorSpace_A2B0(PCS pcs, sk_sp<SkColorLookUpTable> colorLUT,
+                      SkGammaNamed aCurveNamed, sk_sp<SkGammas> aCurve,
+                      SkGammaNamed mCurveNamed, sk_sp<SkGammas> mCurve,
+                      SkGammaNamed bCurveNamed, sk_sp<SkGammas> bCurve,
+                      const SkMatrix44& toPCS, sk_sp<SkData> profileData);
+
+    PCS                       fPCS;
+    sk_sp<SkColorLookUpTable> fColorLUT;

[msarett, 2016/10/07 01:48:36]

My "convention" is to list these in the order that they would be used.  I
*think* that's:
A
CLUT
M
Matrix
B
PCS

Same for the parameter order above and getters above.

[raftias, 2016/10/10 20:37:32]

Done.

+    const SkGammaNamed        fACurveNamed;
+    sk_sp<SkGammas>           fACurve;
+    const SkGammaNamed        fMCurveNamed;
+    sk_sp<SkGammas>           fMCurve;
+    const SkGammaNamed        fBCurveNamed;
+    sk_sp<SkGammas>           fBCurve;
+    SkMatrix44                fToPCS;
+
+    
+    friend class SkColorSpace;
+    typedef SkColorSpace_Base INHERITED;
+};
+
+#endif