* Build with cmake in appveyor
* Install content files and tests binary to bin
* Install SDL2.dll on windows
* Build RecastDemo as GUI application on Windows and Mac OS
* Copy SDL2.dll to current binary dir
The GCC version used by Travis is old and appears not to support all the
warnings we would like to get, so reenable Clang. Additionally make lots
of cleanups to Travis config and make the build a lot faster too.
Also change both Travis and AppVeyor builds to treat warnings as errors,
and include VS2017 in the matrix for AppVeyor. I have also removed
VS2010 because it is ancient at this point.
Updated Premake to alpha12 and fixed the premake build file
appropriately.
Add rcNew/rcDelete, template functions that allocate/deallocate objects using rcAlloc/rcNew.
The intention is to use these to move construction logic out of rcAllocateFooObject/rcFreeFooObject and into constructors/destructors, giving users flexbility in how objects are allocated. In particular, this allows stack allocation.
This implements most of the std::vector interface, with some minor elisions of
infrequently used methods such as "insert", and some STL bookkeeping.
Performance is favorable compared to std::vector, and, in some cases, somewhat
better than naive code to manually manage a malloc'd dynamic array.
BM_FlatArray_Push: 100 iterations in 72889721 nanos: 728897.21 nanos/it
BM_FlatArray_Fill: 100 iterations in 2700746 nanos: 27007.46 nanos/it
BM_FlatArray_Memset: 100 iterations in 2687810 nanos: 26878.10 nanos/it
BM_rcVector_Push: 100 iterations in 51114316 nanos: 511143.16 nanos/it
BM_rcVector_PushPreallocated: 100 iterations in 9532106 nanos: 95321.06 nanos/it
BM_rcVector_Assign: 100 iterations in 2371534 nanos: 23715.34 nanos/it
BM_rcVector_AssignIndices: 100 iterations in 2573052 nanos: 25730.52 nanos/it
BM_rcVector_Resize: 100 iterations in 2455761 nanos: 24557.61 nanos/it
BM_stdvector_Push: 100 iterations in 71739167 nanos: 717391.67 nanos/it
BM_stdvector_PushPreallocated: 100 iterations in 15723737 nanos: 157237.37 nanos/it
BM_stdvector_Assign: 100 iterations in 2564985 nanos: 25649.85 nanos/it
BM_stdvector_AssignIndices: 100 iterations in 5174923 nanos: 51749.23 nanos/it
BM_stdvector_Resize: 100 iterations in 2520054 nanos: 25200.54 nanos/it
Remove dstReg and dstDist from rcBuildRegions entirely -- just use temporary scratch space in expandRegions.
This saves a considerable amount of time when the number of spans is large. A sample map tried has 16M spans, and only writes at most 4000 of them.
* Implement saveAll loadAll for Sample_SoloMesh
- Solo mesh sample is loaded and saved as a tiled mesh containing
an unique tile, so as to remain compatible with tiled mesh
- navmesh is saved to/loaded from `solo_navmesh.bin` binary file
- add and connect the corresponding 'Save/Load' GUI elements
* Fix `implit-fallthrough` gcc warnings
On linux with gcc 7.2.0+ and all warnings treated as errors,
implicit fallthroughs in case statements are considered as errors.
If the comment just below the next `case` or `default` matches
various strings, one of which is `falls through` the fallthrough
is then considered explicit.
Ground color is changed back to the previous default color, and the road
color is changed to a dark shade of brown that is a bit easier to
differentiate with normal input geometry.
In U and A shaped terrain it was possible that polygons extruded their
BV-tree nodes significantly. This happened because all points for the
polygon (when excluding the detail mesh) were either on the high-level
terrain, or on the low-level terrain. This fixes the long-standing todo
about this.
As a result of this change, findNearestPoly and queryPolygons should
return more accurate results.
Fix#230
When a point is outside the polygon being queried in
dtNavMesh::closestPointOnPoly and dtNavMeshQuery::closestPoinOnPoly,
Detour will interpolate along one of the edges to find the closest point
on the poly. However, Detour did not properly use the interpolated point
to find the correct height according to the detail mesh.
Fix#227
findPath was unconditionally overriding the status it used to indicate
DT_OUT_OF_NODES for later. Restore this behavior so it properly can
return this failure flag.
Fix#221
