Uncaught TypeError: t.split is not a function about bezierjs HOT 4 CLOSED

that is very curious - thanks for reporting it, I'll have a look at why this happens.

from bezierjs.

it looks like the inflections function got overwritten at some point to only test cubic inflections... no idea why, but restoring that to cover the other cases again now.

from bezierjs.

Interestingly that's only part of the problem. Solving for inflections, there is a super fascination split result:

{ left:
   { [String: '[330/592, 330/592, 330/592, 330/592]']
     _3d: false,
     order: 3,
     points: [ [Object], [Object], [Object], [Object] ],
     dims: [ 'x', 'y' ],
     dimlen: 2,
     _linear: true,
     _t1: 0,
     _t2: 0,
     dpoints: [ [Object], [Object], [Object] ],
     clockwise: false },
  right:
   { [String: '[330/592, 330/557, 315/522, 315/485]']
     _3d: false,
     order: 3,
     points: [ [Object], [Object], [Object], [Object] ],
     dims: [ 'x', 'y' ],
     dimlen: 2,
     _linear: false,
     _t1: 0,
     _t2: 1,
     dpoints: [ [Object], [Object], [Object] ],
     clockwise: true },
  span:
   [ { x: 330, y: 592 },
     { x: 330, y: 557 },
     { x: 315, y: 522 },
     { x: 315, y: 485 },
     { x: 330, y: 592 },
     { x: 330, y: 557 },
     { x: 315, y: 522 },
     { x: 330, y: 592 },
     { x: 330, y: 557 },
     { x: 330, y: 592 } ],
  _t1: -0,
  _t2: 0 }

I can't remember the last time I saw -0 in JavaScript...

from bezierjs.

Turns out the error came from extrema(), which would happily return a list like [-0, 0, 0.5, 1], which would then lead to the code forming segments between each of those t values, which would lead to it trying to form a segment by calling thing.split(-0,0) which broke all the things.

I've extended the tests to explicitly check the result of extrema() using your curve parameters to prevent any code changes in the future screwing this back up, and published v2.0.2 to npm =)

from bezierjs.