[metapost] glyph operator and contours order

[Taco Hoekwater]

On 7 feb. 2011, at 21:14, Dan Luecking <luecking at uark.edu> wrote:

> 
> The axis crossing algorithm has no unusual precisionin
> problems. Apart, that is, from such problems inherent in
> metapost. For example, if pnt is close enough to the cycle
> (say within epsilon), then a change in t of epsilon (smallest
> possible) could jump multiple quadrants.
> 
In this case, we are dealing with font glyphs, and these tend to be very
well behaved. Your code should work just fine (but I did not test it).

Best wishes,

Taco