Provide a choice of predicate kernel. Make AdaptivePredicates.jl the new default (#165)

Author: DanielVandH

.github/workflows/CI.yml

@@ -12,7 +12,7 @@ concurrency:
   cancel-in-progress: ${{ startsWith(github.ref, 'refs/pull/') }}
 jobs:
   test:
-    name: Julia ${{ matrix.version }} - ${{ matrix.arch }} - ${{ github.event_name }} - ExactPredicates ${{ matrix.preferences }} - ${{ matrix.os }}
+    name: Julia ${{ matrix.version }} - ${{ matrix.arch }} - ${{ github.event_name }} - ${{ matrix.os }}
     runs-on: ${{ matrix.os }}
     env:
       JULIA_NUM_THREADS: ${{ matrix.julia-threads }}
@@ -27,10 +27,6 @@ jobs:
           - ubuntu-latest
         arch:
           - x64
-        preferences:
-          - 'DEFAULT'
-          - 'EXACT'
-          - 'INEXACT'
     steps:
       - name: Checkout repository for access
         uses: actions/checkout@v4
@@ -50,8 +46,6 @@ jobs:
 
       - name: Run the tests
         uses: julia-actions/julia-runtest@v1
-        env:
-          PREDICATES: ${{ matrix.preferences }}
 
       - name: Process the coverage
         if: always()

NEWS.md

@@ -2,7 +2,18 @@
 
 ## Main (upcoming v.1.1.0)
 
-- Added the option to disable ExactPredicates.jl using Preferences.jl. See [#131](https://github.com/JuliaGeometry/DelaunayTriangulation.jl/pull/131) and [#137](https://github.com/JuliaGeometry/DelaunayTriangulation.jl/pull/137).
+There are a lot of changes in this release, most of them irrelevant for the user. The most important change is the following:
+
+- We now support a choice between fast, exact, and adaptive predicates via `FastKernel()`, `ExactKernel()`, and `AdaptiveKernel()`, respectively. The default is now `AdaptiveKernel()`. Moreover, triangle areas are now computed using the adaptive `orient` predicate to be more robust. See [#165](https://github.com/JuliaGeometry/DelaunayTriangulation.jl/pull/165).
+
+Previously, ExactPredicates.jl was used everywhere, which can be slow and not necessary for certain point sets. The `FastKernel()` option 
+has no exact or adaptive arithmetic and so should be used with caution. The documentation discusses these choices in more detail. 
+
+To actually configure the choice of predicate, you can e.g. in `triangulate` use the `predicates` keyword argument and pass one of 
+`DelaunayTriangulation.FastKernel()`, `DelaunayTriangulation.ExactKernel()`, or `DelaunayTriangulation.AdaptiveKernel()`. If you are computing a predicate manually, then the predicate is instead passed as the first argument.
+
+Some other changes: 
+
 - Added `DelauanyTriangulation.validate_triangulation` for validating triangulations. See [#131](https://github.com/JuliaGeometry/DelaunayTriangulation.jl/pull/131).
 - Fixed a bug with the currently unused `orient(p, q, r, s)` predicate. See [#131](https://github.com/JuliaGeometry/DelaunayTriangulation.jl/pull/131).
 - Added private functions `getz`, `_getz`, `getxyz`, and `_getxyz`. See [#131](https://github.com/JuliaGeometry/DelaunayTriangulation.jl/pull/131).
@@ -17,6 +28,7 @@
 - Added a CONTRIBUTING.md file and issue templates. See [#160](https://github.com/JuliaGeometry/DelaunayTriangulation.jl/pull/160).
 - Added `is_point2` and `is_point3` to detect if a given input is a point. This allows vector coordinates to be passed to `convert_boundary_points_to_indices`. See [#161](https://github.com/JuliaGeometry/DelaunayTriangulation.jl/pull/161).
 - Removed an allocation from `add_vertex!`. See [#163](https://github.com/JuliaGeometry/DelaunayTriangulation.jl/pull/163).
+- Fixed an issue with the user-supplied `rng` not being passed to `lock_convex_hull!`.
 
 ## v1.0.5