MRCPP-v1.5.0: Minor release

Added

• Imported basic MPI functionality from MRChem
• Haar basis MW filters (only functions, no operators)
• Possibility for anisotropic MWOperator
• CartesianConvolution operator (e.g. xy/r^3)

Fixed

• Reintroduces operator build_prec needed by VAMPyR
• Bug in getQuadraturePoints (row/col mixup)

MRCPP-v1.4.2: Bugfix release

Fixed

• Invalid read in coef block for nodes without coefs
• Node count error for negative scales with PBC
• Missing includes resulting in compile errors for recent gcc

MRCPP-v1.4.1: Bugfix release

Changed

• Catch version 2.13.5

Fixed

• Buffer overflow in NodeAllocator
• Polynomial integration bound error message

MRCPP-v1.4.0: Minor release

Added

• New constructors for BoundingBox and MRA, accepting box=[-L,L] argument
• Add FunctionTree::evalf_precise which evaluates both scaling and wavelet parts
• Possibility for empty tree skeletons without allocated coefficients
• Possibility to manually set location of MW filters at configure time

Changed

• Eigen version 3.4.0
• Improvements under the hood of the Tree classes (non-API):
  • remove GenNode and ProjectedNode
  • simplified NodeAllocator class
  • new improved OMP locking

Fixed

• Get node center and (lower/upper) bounds

MRCPP-v1.3.6: Bugfix release

Fixed

• Put OpenMP and MPI in CMake INTERFACE
• Fix --arch-flags CMake option
• Fix kramdown security issue
• Document variuos packaging options

MRCPP-v1.3.5: Bugfix release

Fixed

• Remove final qualifier GaussFunc/GaussPoly (fixing pybind11 bindings)

MRCPP-v1.3.4: Bugfix release

Fixed

• Allow to remove arch-specific optimization flags
• Undo export of compiler flags with MRCPP CMake target
• Encapsulate OpenMP within MRCPP and export MRCPP_HAS_OMP flag
• Encapsulate MPI within MRCPP and export MRCPP_HAS_MPI flag
• Fix faulty OpenMP locks
• Minor inlining optimizations

MRCPP-v1.3.3: Bugfix release

Fixed

• Export compiler flags with MRCPP CMake target

MRCPP-v1.3.2: Bugfix release

Fixed

• Miscellaneous fixes for building on conda-forge

MRCPP-v1.3.1: Bugfix release

Fixed

• Uninitialized (max)norms used in space-varying precision algorithms