Anyone know what cause this crash : FbxGeometryConverter::Triangulate() ?

[kuzume7WAC6]

Hi there.

We are using FbxGeometryConverter::Triangulate() and it's causing a crash in

some kind of FBX files.

I couldn't attach examples for some reasons but it's sure that it doesn't include any NURBs.

Version:

Autodesk FBX SDK 2016.1.2 Release, September 10th 2015

Does anyone know of a way to solve a problem ?

Thanks,

[viviane.rochon]

Hi kuzume7WAC6,

To attach files in this forum, you must zip them beforehand.

Please attach the file - it will help us investigate the issue.

Thanks!

Viviane

[kuzume7WAC6]

Hi Viviane

Here is triangulate.zip which contains a triangle.sln that always reproduces the problem.

> triangulate.exe!main(int argc, char * * argv) Line 49 C++
triangulate.exe!__tmainCRTStartup() Line 536 C
triangulate.exe!mainCRTStartup() Line 377 C
kernel32.dll!75c3338a() Unknown
[Frames below may be incorrect and/or missing, no symbols loaded for kernel32.dll]
ntdll.dll!77e29a02() Unknown
ntdll.dll!77e299d5() Unknown

Required tools

• VisualStudio 2012
• Autodesk FBX SDK 2016.1.2 Release, September 10th 2015

Thanks

[kuzume7WAC6]

Hi Viviane.

Could you reproduce the problem with our attached file ?

Please let me know if there's anything we can do.

Thanks

kuzume

[viviane.rochon]

Hi Kuzume,

Yes, we could reproduce your issue using your file.

We logged it internally as FBXX-1145.

Thank you!

Viviane

[kuzume7WAC6]

Hi viviane,

Thanks a lot for your investigating and we're looking forward to fixing the issue FBXX-1145.

kuzume

[kuzume7WAC6]

Hi Viviane.

Please let me know when does it fix if you can.

Thanks!

kuzume

[viviane.rochon]

Hi Kuzume,

The crash is actually the consequence of an assert in debug mode. The file has bad data. An assert is raised because the line's start and end points are the same, and the FBX SDK cannot compute the direction of the line. The same program built in Release will complete the triangulation calls, but the result is going to be a mesh with some holes caused by the degenerated data that causes the assert to raise in the first place.

I hope this clears things up!

[kuzume7WAC6]

It was nice of you to take the time to solve this problem.

Thnaks a lot.

kuzume