I don't know why, but marking the output of copy constructor of any
subclass or output of conversion operator of any class as constexpr
causes MSVC to complain about non-constant expression.
Probably just another bug.
Useful for squeezing out last bits of performance, e.g. in this case:
Vector3 a;
a[0] = something++;
a[1] = something++;
a[2] = something++;
In the code all elements are first zeroed out and then overwritten
later, thus it might be good to avoid the zero-initialization:
Vector3 a{Math::NoInit};
a[0] = something++;
a[1] = something++;
a[2] = something++;
This will of course be more useful in far larger data types and arrays
of these.
Some classes are by default constructed zero-filled while other are set
to identity and the only way to to check this is to look into the
documentation. This changes the default constructor of all classes to
take an optional "tag" which acts as documentation about how the type is
constructed. Note that this result in no behavioral changes, just
ability to be more explicit when writing the code. Example:
// These two are equivalent
Quaternion q1;
Quaternion q2{Math::IdentityInit};
// These two are equivalent
Vector4 vec1;
Vector4 vec2{Math::ZeroInit};
Matrix4 a{Math::IdentityInit, 2}; // 2 on diagonal
Matrix4 b{Math::ZeroInit}; // all zero
This functionality was already present in some ugly form in Matrix,
Matrix3 and Matrix4 classes. It was long and ugly to write, so it is
now generalized into the new Math::IdentityInit and Math::ZeroInit tags,
the original Matrix::IdentityType, Matrix::Identity, Matrix::ZeroType
and Matrix::Zero are deprecated and will be removed in the future
release.
Math::Matrix<7, Int> m{Math::Matrix<7, Int>::Identity}; // before
Math::Matrix<7, Int> m{Math::IdentityInit}; // now
The only places where they aren't absolute are:
- when header is included from corresponding source file
- when including headers which are not part of final installation (e.g.
test-specific configuration, headers from Implementation/)
Everything what was in src/ is now in src/Corrade, everything from
src/Plugins is now in src/MagnumPlugins, everything from external/ is in
src/MagnumExternal. Added new CMakeLists.txt file and updated the other
ones for the moves, no other change was made. If MAGNUM_BUILD_DEPRECATED
is set, everything compiles and installs like previously except for the
plugins, which are now in MagnumPlugins and not in Magnum/Plugins.
Unlike Matrix3/Matrix4 these don't have any transformation-related
functions. Deprecated the old Matrix2 typedef, which is now replaced
with Matrix2x2 and will be removed in future release.
Needed to adjust the test cases slightly, because they were firing the
assert even if they shouldn't and the "expect fail" case wasn't working
at all. I now badly need proper floating-point equality comparison.
Also fixed unary RectangularMatrix::operator-() and Vector::operator-()
documentation (was stating that the operation is done in-place, which is
impossible.
* Merged constExpressions() into other test cases, reducing duplicates
and simplifying the checks.
* Fixed old-and-forgotten operator[] overload in Matrix subclasses, it
was reinterpret_cast on T* array, it is now sufficient to do only
static_cast. Constexpr operator[] overload returns const copy to make
constexpr operations working even on returned value, e.g.:
constexpr Matrix4 a;
constexpr Vector3 b = a[2].xyz();
It's now possible to conveniently transform 2D vectors and points with
3x3 matrices and 3D vectors/points with 4x4 matrices. Previous most
low-level solution:
Matrix4 m;
Vector3 v;
Vector3 a = (m*Vector4(v, 1.0f)).xyz();
Vector4 b = (m*Vector4(v, 0.0f)).xyz();
Another, more generalized solution for points was with Point2D/Point3D,
adding a lot of confusion (what is that class and what does .vector()?):
Vector3 a = (m*Point3D(v)).vector();
And the worst solution was with generic 2D/3D code (WTF!):
auto a = (m*typename DimensionTraits::PointType(v)).vector();
Now it is just this, similar for both dimensions:
Vector3 a = m.transformPoint(v);
Vector3 b = m.transformVector(v);
Note that transformation three-component vectors with 3x3 matrices or
four-component vectors with 4x4 matrices is easy enough so it doesn't
need any special convenience functions whatsoever:
Vector3 c = m.rotation()*v;
Was causing improper implicit conversions, such as here (example
directly from unit tests, where it was unintentionally used):
Vector3 normal;
Matrix4 transformation;
auto transformedNormal = transformation*normal;
Not only that it was possible to multiply 4x4 matrix with 3-component
vector, but the resulting type was Point3D which was absolutely
confusing. Currently it must be explicitly converted:
transformedNormal = transformation*Point3D(normal);
Overall architecture is simplififed with this change and also it's not
needed to use reinterpret_cast in matrix internals anymore, thus there
is no need for operator() and [][] works now always as expected without
any risk of GCC misoptimizations.
On the other side, constructing matrix from list of elements is not
possible anymore. You have to specify the elements as list of
column vectors, which might be less convenient to write, but it helps to
distinguish what is column and what is row:
Matrix<2, int> a(1, 2, // before
3, 4);
Matrix<2, int> a(Vector<2, int>(1, 2), // now
Vector<2, int>(3, 4));
For some matrix specializations (i.e. Matrix3 and Matrix4) it is
possible to use list-initialization instead of explicit type
specification:
Matrix<3, int>({1, 2, 3},
{4, 5, 6},
{7, 8, 9});
I didn't yet figure out how to properly implement the general
(constexpr) constructor to also take lists, so it's a bit ugly for now.
Matrix operations are now done column-wise, which should help with
future SIMD implementations, documentation is also updated accordingly.
I also removed forgotten remains of matrix/matrix operator*=(), which
can be confusing, as the multiplication is not commutative. Why it is
not present is explained in d9c900f076.
Long-standing TODO, can be used for in-game mirrors etc. I give up with
shearing, as I think that it makes sense only in 2D and I can't find any
reasonable use case for that yet.
Removing of another <*stream> #include leads to more compilation time
saving, now from ~5:12 to ~4:55. Another compilation time improvements
will now be possible only by using Clang's modules, I don't know where
to optimize further (except for getting rid of <sstream> in tests).
Vladimír Vondruš