"error in function split_at_vertex" what gives?

Trying to export the Gcode of this file on 1.1.7 and get that error in function error. Gcode is only approx 2 mb.

Tried updating to experimental 1.2.6, and the process finishes without error, about 6.66 mb, but failed to complete last night.
Not sure what happened, but when I woke up, only 1/3 of the part was printed, and the end gcode ran like it was done. Nozzle was back to home, and fan and heat were all off.

Anyone know why, or could look at the file for me?

It's the last piece I have to print on an ENORMOUS kit I've been printing for a week straight.

Anyone?

So the version I've been using all this time is 32 bit Slic3r 1.1.7 on a Windows 7 64 bit machine.
I've since tried downloading 64 bit 1.1.7, and 64 bit 1.2.6.

All yield a gcode file of different completion. Not good, this software should be 100% reliable, and has been to this point.

So I installed 1.1.7 on a windows xp VM at work today, and ran this .STL file through it, and got NO errors, a gcode file DOUBLE the size from any version I got at home, including the version I've been using all this time with absolutely NO issue up until this ONE file.

I'm running this file now, to see if it prints properly or not. Even if it does print correctly, how do I trust this application anymore?

The fact that an application is in "stable" version, does not mean that it has no issues at all.
On the other hand, the fact that you get an error for 1 out of 100 files, does not mean your model file is free from errors.
Some errors are corrected automatically, some generates erroneous code and some may crash the application.
Though apparently is, trust is not an absolute thing