This fixes a case of missing navmesh under a giant mushroom
#rb Mikko.Mononen
#rb Frederic.Doll
#preflight 6294c42712153e8cebf9f12a
#ROBOMERGE-AUTHOR: aris.theophanidis
#ROBOMERGE-SOURCE: CL 20430963 via CL 20430966 via CL 20430972
#ROBOMERGE-BOT: UE5 (Release-Engine-Staging -> Main) (v949-20362246)
[CL 20431400 by aris theophanidis in ue5-main branch]
- add asserts in dtRandomPointInConvexPoly to detect problematic cases
- add special case for point to point links in dtNavMeshQuery::findRandomPointInPoly
#rb guillaume.guay, aris.thoephanidis
#ROBOMERGE-AUTHOR: frederic.doll
#ROBOMERGE-SOURCE: CL 20167439 via CL 20168810 via CL 20169116 via CL 20170041 via CL 20170327
#ROBOMERGE-BOT: UE5 (Release-Engine-Staging -> Main) (v943-19904690)
[CL 20172504 by frederic doll in ue5-main branch]
I've added comments to make it clearer this feature is unsupported and removed the height variable that isn't used to simplify and tidy up the existing code.
#jira UE-135621
#review-19809739
#rb Mieszko.Zielinski
#preflight 625ee03d034d8924cdcc72f1
[CL 19826653 by Stephen Holmes in ue5-main branch]
I need that because I do a floodfill query (with ARecastNavMesh::FindPolysAroundCircle) and then need to test positions to move in the resulting polygons
[REVIEW] aris.theophanidis, mikko.mononen
#ROBOMERGE-AUTHOR: frederic.doll
#ROBOMERGE-SOURCE: CL 19465568 via CL 19465588 via CL 19465610 via CL 19466833 via CL 19475574 via CL 19475839
#ROBOMERGE-BOT: UE5 (Release-Engine-Staging -> Main) (v935-19464907)
[CL 19476450 by frederic doll in ue5-main branch]
(with some extra const)
#rb Maxime.Mercier
#preflight 61fc2b043b22793ca369a98d
#ROBOMERGE-AUTHOR: aris.theophanidis
#ROBOMERGE-SOURCE: CL 18852387 via CL 18852907 via CL 18853018 via CL 18854025 via CL 18854798
#ROBOMERGE-BOT: UE5 (Release-Engine-Test -> Main) (v910-18824042)
[CL 18854824 by aris theophanidis in ue5-main branch]
[FYI] Stephen.Holmes
Original CL Desc
-----------------------------------------------------------------
Memory optimizations for Recast (Detour). I used UE_LOG in recast as it's already there. Also I only added [at]UE BEGIN [at]UE END for the big stuff as a hint that its been changed in case of issues.
#ROBOMERGE-AUTHOR: stephen.holmes
#ROBOMERGE-SOURCE: CL 18417786 in //UE5/Release-5.0/... via CL 18417810
#ROBOMERGE-BOT: STARSHIP (Release-Engine-Staging -> Release-Engine-Test) (v897-18405271)
[CL 18417812 by stephen holmes in ue5-release-engine-test branch]
This represents UE4/Main @18073326, Release-5.0 @18081140 and Dev-PerfTest @18045971
[CL 18081471 by aurel cordonnier in ue5-release-engine-test branch]
This prevents having one cell wide region strips and one cell regions and the 'Region Chunk Splits' setting is now correctly respected.
Before this change, a region split on 2 could result into 3 splits (ex: a 17 cells wide tile would have been divided into 8-8-1 splits, with this change it will be 8-9).
#rb Mikko.Mononen
#jira FORT-420970
#preflight 6179bae1d65ac900014a5a83
#ROBOMERGE-AUTHOR: aris.theophanidis
#ROBOMERGE-COMMAND: FnMain
#ROBOMERGE-SOURCE: CL 17952948 in //UE5/Main/...
#ROBOMERGE-BOT: STARSHIP (Main -> Release-Engine-Test) (v885-17909292)
#ROBOMERGE[STARSHIP]: UE5-Release-Engine-Staging Release-5.0
#ROBOMERGE[bot1]: Main
[CL 17953014 by aris theophanidis in ue5-release-engine-test branch]
This represents UE4/Main @17911760, Release-5.0 @17915875 and Dev-PerfTest @17914035
[CL 17918595 by aurel cordonnier in ue5-release-engine-test branch]
warning V781: The value of the 'XYZ' index is checked after it was used. Perhaps there is a mistake in program logic.
#rb trivial
#jira UE-91644
[CL 15066377 by Tim Smith in ue5-main branch]
