Gonav
A Source Engine navigation mesh file parser written in Go.
Install / Use
/learn @mrazza/GonavREADME
gonav
A Source Engine bot Nav file parser written in Go. This was written for CS:GO but will likely work with little or no modification for other Source titles. The specifics of the .nav format were reverse engineered using the information on Valve's wiki as a starting point. For more information on Source's Navigation Meshes see Valve's wiki: https://developer.valvesoftware.com/wiki/Navigation_Meshes
Usage
See the _examples folder for examples of how to use this library. The basics, however, are pretty straightforward. Create a parser and pass in any Reader that contains the binary .nav data, then call the Parse() method. This will output a parsed NavMesh object which you can perform operations on. Here's a terse example:
f, _ := os.Open("de_dust2.nav") // Open the file
parser := gonav.Parser{Reader: f}
mesh, _ := parser.Parse() // Parse the file
area := mesh.QuadTreeAreas.FindAreaByPoint(gonav.Vector3{10, 10, 10}, true) // Find the nav area that contains the world point {10, 10} with a Z-value closest to 10
fmt.Println(area)
Path Finding
This library also supports path finding across nav meshes via an A* implementation.
bombsiteA := mesh.GetPlaceByName("BombsiteA")
aCenter, _ := bombsiteA.GetEstimatedCenter()
aArea := mesh.GetNearestArea(aCenter, false)
bombsiteB := mesh.GetPlaceByName("BombsiteB")
bCenter, _ := bombsiteB.GetEstimatedCenter()
bArea := mesh.GetNearestArea(bCenter, false)
path, _ := gonav.SimpleBuildShortestPath(aArea, bArea)
for _, currNode := range path.Nodes {
fmt.Println(currNode.Area)
}
Example output of path finding from the center of A-site on de_nuke to the center of B-site:
Path length: 2381.591
AreaID: 1419 [BombsiteA] @ {{675 -925 -414.96875}, {700 -900 -415.35532}}
AreaID: 7 [BombsiteA] @ {{575 -1325 -415.96875}, {950 -925 -415.96875}}
AreaID: 2285 [BombsiteA] @ {{550 -1300 -415.96875}, {575 -1225 -415.96875}}
AreaID: 3107 [BombsiteA] @ {{500 -1400 -415.96875}, {550 -1225 -415.96875}}
AreaID: 3368 [BombsiteA] @ {{475 -1375 -415.96875}, {500 -1225 -415.96875}}
AreaID: 3076 [BombsiteA] @ {{425 -1375 -415.96875}, {475 -1325 -415.96875}}
AreaID: 3075 [BombsiteA] @ {{425 -1400 -415.96875}, {475 -1375 -415.96875}}
AreaID: 1673 [Vents] @ {{425 -1450 -607.96875}, {625 -1425 -607.96875}}
AreaID: 700 [Vents] @ {{625 -1525 -607.96875}, {650 -1425 -607.96875}}
AreaID: 2478 [Vents] @ {{650 -1450 -607.96875}, {700 -1425 -607.96875}}
AreaID: 2479 [Vents] @ {{700 -1450 -607.96875}, {750 -1425 -607.96875}}
AreaID: 2477 [Vents] @ {{750 -1450 -607.96875}, {825 -1425 -607.96875}}
AreaID: 2300 [Vents] @ {{825 -1450 -607.96875}, {850 -1400 -607.96875}}
AreaID: 2301 [Vents] @ {{825 -1400 -607.96875}, {850 -1350 -607.96875}}
AreaID: 1690 [Vents] @ {{825 -1350 -607.96875}, {850 -1325 -639.96875}}
AreaID: 2265 [BombsiteB] @ {{775 -1325 -639.96875}, {900 -1275 -639.96875}}
AreaID: 3589 [BombsiteB] @ {{900 -1325 -639.96875}, {950 -1275 -639.96875}}
AreaID: 3591 [BombsiteB] @ {{900 -1275 -639.96875}, {950 -1050 -639.96875}}
AreaID: 3593 [BombsiteB] @ {{900 -1050 -639.96875}, {950 -900 -639.96875}}
AreaID: 2490 [BombsiteB] @ {{878 -1050 -597.96875}, {888 -900 -597.96875}}
AreaID: 2690 [BombsiteB] @ {{850 -1050 -767.96875}, {888 -900 -768.9288}}
AreaID: 3500 [BombsiteB] @ {{550 -1075 -770.25446}, {850 -900 -767.96875}}
AreaID: 3502 [BombsiteB] @ {{550 -900 -769.49255}, {850 -725 -767.96875}}
License
This source is licensed under the GNU AFFERO GENERAL PUBLIC LICENSE. If this license is not acceptable for your project, let me know. Additionally, more feature-rich versions of this library exist, written in C# and C++, and can be licensed upon request.
Related Skills
node-connect
349.9kDiagnose OpenClaw node connection and pairing failures for Android, iOS, and macOS companion apps
frontend-design
109.8kCreate distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, or applications. Generates creative, polished code that avoids generic AI aesthetics.
openai-whisper-api
349.9kTranscribe audio via OpenAI Audio Transcriptions API (Whisper).
qqbot-media
349.9kQQBot 富媒体收发能力。使用 <qqmedia> 标签,系统根据文件扩展名自动识别类型(图片/语音/视频/文件)。
