Kai Dietrich

Hi all, I have an internal patch for this. I added a method float Marker::modelScale() const { return glm::max(extent(), 1024.f); } In the model building code I replaced some calls...

So, I have a patch for this issue and made a video. This is the problem:  What you see here is two rectangular extruded polygons. The extrusion property is...

> I can look at the task scheduling to add terrain as a dependent source This would be super nice. I see there is stuff like sub-tasks but I couldn't...

This is a file that crashes with a heap corruption when any of the contained files are decompressed: [id_000003.zip](https://github.com/tfussell/miniz-cpp/files/5420067/id_000003.zip)

Thanks for the correction, I confused the two since the code looks so similar. I changed the issue title accordingly. Is this the authoritative upstream? https://github.com/richgel999/miniz

First potential fix in https://github.com/dtschump/CImg/pull/296

I'm unsure if this is the best fix possible. I can see an alternative fix but am unsure if this is good either: Instead of outright rejecting large dimensions, test...

There also seem to be other recursive operations that will suffer from the same problem. I found: * XMLNode::~XMLNode->XMLNode::DeleteChildren->XMLNode::DeleteChild->XMLNode::DeleteNode->XMLNode::~XMLNode * XMLElement::Accept->XMLElement::Accept I have seen this before in XML handling code....

I wrote a few simple repro test cases to illustrate the problem. These tests currently all crash with a stack overflow. Can I get a comment if this is considered...

Just merged retroarch - works like a charm :) Thanks for your work!