added static method based on 'geometryId' to the (Singleton)TopologyFactory (taken from version in dune-localfunctions)

moved the deprecated topology classes to a separate header. The header is included in types.hh for the current release without deprecation warnings but after the 2.8 release the header 'deprecated_topology.hh' will need to be included which contains deprecation warnings.

Add Impl::toGeometryTypeConstant helper

See merge request !164

This helper maps a dynamic `GeometryType` to a compile time
`std::integral_constant<GeometryType::Id,...>`. Thus is functionally
replaces the `IfGeometryType` helper. In contrast to the latter it
does not rely on the undocumented implementation of `GeometryType::Id`
but on `LocalGeometryTypeIndex` and the generic `Hybrid::switchCase`
helper. Furthermore the new helper uses a lambda-friendly callback
interface.

Cleanup Impl namespace

See merge request !159

Replace one left-over DUNE_UNUSED_PARAMETER by maybe_unused

See merge request !165

Use [[maybe_unused] instead of DUNE_UNUSED / DUNE_UNUSED_PARAMETER

See merge request !158

Except for one occurence of DUNE_UNUSED_PARAMETER in type.hh due
to GCC bug 81429 which got fixed in GCC 9.3

construction in "dune/geometry/topologyfactory.hh" and "dune/geometry/type.hh".

Feature/topology to geometry type

See merge request !161

from `Topology` to `GeometryType`.

Andreas Dedner authored 4 years ago and Henrik Stolzmann committed 4 years ago

(10.2 is fine with the test)

Prepared transition from `Topology` template parameter to
`GeometryType`.

The old Topology construction based on template recursion via the
structs `Point`, `Pyramid` and `Prism` is dated.

use the dune docker registry for the python ci test

See merge request !162

Use std::void_t instead of Dune::void_t

See merge request !160

fix typo in header name in CMakeLists.txt

See merge request !157

Feature: Gauss-Radau quadratures

See merge request !154

use SCRIPT instead of COMMAND in dune_python_add_test

See merge request !156

matched the quad order in the maxima scripts to the definitions in quadraturerules.hh, final cleanup, tested everything

refactored the writequad maxima script, it produces (apart from comments) the same files as before for jacobi rules and gauss lobatto rules