enforce this mechanism (otherwise ReturnByValue bypasses it).
(use .noalias() to get the old behavior.)
* Remove a hack in Inverse, futile optimization for 2x2 expressions.
Introduction of DenseCoeffBase (revision bfdc1c4973
) meant that non-const
operator() is only defined if DirectAccess is set. This caused the line
"m.reverse()(1,0) = 4;" in MatrixBase_reverse.cpp to fail at compile-time.
Not sure this is correct solution; perhaps we should disallow this? Or make
Reverse DirectAccess with a negative stride - would that break something?
* change test_is_equal so it just checks for equality, doesn't try anymore to allow to ignoring the difference between col and row vectors.
(needed for the upcoming Index types change)
* in ei_assert, don't report on cerr if we're inside of VERIFY_RAISES_ASSERT
* use SelfAdjointView instead of Eigen2's SelfAdjoint flag.
* add tests and documentation.
* allow eigenvalues() for non-selfadjoint matrices.
* they no longer depend only on SelfAdjointEigenSolver, so move them to
a separate file
Rationale: coeffRef() methods should only exist when we have DirectAccess. So a natural thing to do would have been to use enable_if, but since there are many methods it made more sense to do the "enable_if" for the whole group by introducing a new class. And that also that the benefit of not changing method prototypes.
I didn't even put it in Eigen2Support because it requires several other modules. But if you want we can always create a new module, Eigen2Support_LeastSquares...
* Now completely generic so all standard integer types (like char...) are supported.
** add unit test for that (integer_types).
* NumTraits does no longer inherit numeric_limits
* All math functions are now templated
* Better guard (static asserts) against using certain math functions on integer types.
* get rid of BlockReturnType: it was not needed, and code was not always using it consistently anyway
* add topRows(), leftCols(), bottomRows(), rightCols()
* add corners unit-test covering all of that
* adapt docs, expand "porting from eigen 2 to 3"
* adapt Eigen2Support
- Updated unit tests to check above constructor.
- In the compute() method of decompositions: Made temporary matrices/vectors class members to avoid heap allocations during compute() (when dynamic matrices are used, of course).
These changes can speed up decomposition computation time when a solver instance is used to solve multiple same-sized problems. An added benefit is that the compute() method can now be invoked in contexts were heap allocations are forbidden, such as in real-time control loops.
CAVEAT: Not all of the decompositions in the Eigenvalues module have a heap-allocation-free compute() method. A future patch may address this issue, but some required API changes need to be incorporated first.
