You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
CSG sometimes seemingly gets stuck in an infinite loop. I suspect this is caused by unioning shapes with barely touching polygons (happens when working with spheres but not cubes). There may be some relation to floating point error that remains consistent between runs (changing the structure of the program while theoretically keeping the intended render the same, such as by abstracting a function or creating another shape that never gets rendered, changes whether it gets stuck).
Here is an example snippet. It's modified from #292 (no longer does what was originally intended):
The snippet as it is will be able to render, but if I uncomment the outer_layer shape (that never gets used for rendering), then it gets stuck.
However, if I then keep outer_layer but move the contents of the function out, the program no longer gets stuck (so it doesn't seem creating outer_layer is causing it):
When the program gets stuck, there is this error in the console (may just be generic when any program does so):
main.cffef817.js:2 TypeError: Cannot set properties of undefined (setting 'state')
at eval (eval at <anonymous> (eval at <anonymous> (main.cffef817.js:65:41813)), <anonymous>:13944:51)
at eval (eval at <anonymous> (eval at <anonymous> (main.cffef817.js:65:41813)), <anonymous>:13947:3)
at eval (eval at <anonymous> (main.cffef817.js:65:41813), <anonymous>:3:26)
at main.cffef817.js:65:41911
at Generator.next (<anonymous>)
at o (main.cffef817.js:65:37563)
CSG sometimes seemingly gets stuck in an infinite loop. I suspect this is caused by unioning shapes with barely touching polygons (happens when working with spheres but not cubes). There may be some relation to floating point error that remains consistent between runs (changing the structure of the program while theoretically keeping the intended render the same, such as by abstracting a function or creating another shape that never gets rendered, changes whether it gets stuck).
Here is an example snippet. It's modified from #292 (no longer does what was originally intended):
The snippet as it is will be able to render, but if I uncomment the
outer_layer
shape (that never gets used for rendering), then it gets stuck.However, if I then keep
outer_layer
but move the contents of the function out, the program no longer gets stuck (so it doesn't seem creatingouter_layer
is causing it):When the program gets stuck, there is this error in the console (may just be generic when any program does so):
See also:
The text was updated successfully, but these errors were encountered: