Wrong face measured by AreaMeasurement

[CabeDigicorp]

Describe the bug 📝

I am measuring faces on my usual sample model.
When I measure a window pane in a two pane window, whether I click on the left or right one, only the right one is highlighted and measured.
See these videos:
https://github.com/user-attachments/assets/10368186-6301-4243-b197-6179ac08c860
https://github.com/user-attachments/assets/6babc3ab-b91f-4aa6-b4b1-8d108669f79c

Reproduction ▶️

No response

Steps to reproduce 🔢

No response

System Info 💻

System:
    OS: Windows 11 10.0.26200
    CPU: (16) x64 AMD Ryzen 9 6900HX with Radeon Graphics
    Memory: 11.84 GB / 31.24 GB
  Binaries:
    Node: 22.12.0 - C:\Program Files\nodejs\node.EXE
    Yarn: 1.22.21 - C:\Program Files\nodejs\yarn.CMD
    npm: 10.9.0 - C:\Program Files\nodejs\npm.CMD
    pnpm: 9.12.1 - C:\Program Files\nodejs\pnpm.CMD
  Browsers:
    Chrome: 148.0.7778.179
    Edge: Chromium (148.0.3967.70)
    Firefox: 151.0 - C:\Program Files\Mozilla Firefox\firefox.exe
    Internet Explorer: 11.0.26100.8115

Used Package Manager 📦

npm

Error Trace/Logs 📃

No response

Validations ✅

• Read the docs.
• Check that there isn't already an issue that reports the same bug to avoid creating a duplicate.
• Make sure this is a repository issue and not a framework-specific issue. For example, if it's a THREE.js related bug, it should likely be reported to mrdoob/threejs instead.
• Check that this is a concrete bug. For Q&A join our Community.
• The provided reproduction is a minimal reproducible example of the bug.

[agviegas]

Fixed in 49793d8c, will go out in the next patch. Reproduced it on your Condominio model, thanks for the videos, the two-pane window made the cause obvious once I looked at the right place.

Face snapping was choosing a face purely by how close the pick point was to each candidate face's plane, and never checking whether the point actually landed inside that face's polygon. Two panes of the same window are coplanar, so both candidates sit at a distance of roughly zero from the pick point and nothing distinguishes them. The winner was then decided by nothing more than iteration order, which is why clicking either pane always measured the same one.

Now, among the candidate faces, the tool prefers one whose polygon actually contains the pick point, and only falls back to the nearest plane when none does. So picks that land slightly off a polygon still resolve the way they used to.

Worth noting the same code path backs face snapping generally, not just area measurement, so any tool snapping to a face on coplanar geometry was affected by this.