c2TOI() test case failed

[yuiwong]

try the following case:

c2Poly poly;
poly.verts[0] = c2V(2.0f, 39.0f);
poly.verts[1] = c2V(2.0f, 40.0f);
poly.verts[2] = c2V(1.0f, 40.0f);
poly.verts[3] = c2V(1.0f, 39.0f);
poly.count = 4;
c2MakePoly(&poly);

c2Circle circle;
circle.p = c2V(2.34812546f,40.4136238f);
circle.r = 0.5f;

c2v v0 = c2V(0.0f, -0.106166691f);
c2v v1 = c2V(0.0f, 0.0f);

c2Manifold m0;
c2Collide(&circle, 0, C2_TYPE_CIRCLE, 
          &poly,   0, C2_TYPE_POLY,
          &m0);

i32 iter;
f32 t = c2TOI(&circle, C2_TYPE_CIRCLE, 0, v0,
              &poly,   C2_TYPE_POLY,   0, v1,
              true, &iter);

// advance circle
circle.p = c2V(circle.p.x + v0.x * t, circle.p.y + v0.y * t);

c2Manifold m1;
c2Collide(&circle, 0, C2_TYPE_CIRCLE, 
          &poly,   0, C2_TYPE_POLY,
          &m1);

printf("before %d\n", m0.count);         // output 0
printf("t:%f, iteration:%d\n", t, iter); // output t=1.0, iter=(a value that not make sense).
printf("after %d\n", m1.count);          // output 1

I debug the code and find that c2TOI() return while comparing (velocity_bound), so the 'iteration' is not filled.. Did I do something stupid?

[RandyGaul]

Hi there! I haven't forgotten about this or anything -- Will have some time this week to take a good look.

[RandyGaul]

Took me a while but I put out an updated version of c2TOI. Let me know if you have any problems!