odd behaviors of BOSL2's offset_stroke() with respect to $fn
// I think this code demonstrates inappropriate handling of $fn in BOSL2's
offset_stroke().
include <BOSL2/std.scad>
include <BOSL2/rounding.scad>
$fn= $preview ? 16 : 128;
function cclip_p(arc, fn, id, od) =
let(
ax2 = (360-arc)0.5,
p0 = difference(
circle(d=id, $fn=fn),
arc(n=2, r=10od, angle=[270-ax2,270+ax2], wedge=true)),
// difference returns a region, strip back to a path without last point
p1 = list_head(p0[0])
) p1;
module cclip(color, arc, fn, id, od, p=0) {
p1 = p ? p : cclip_p(arc, fn, id, od);
w=[0,od-id];
echo(color=color, w=w, p1=p1);
// with $fn=6 end treatments have 2 segments not 3?
// there is weird behavior for which outer path points are rounded
// outer path points are rounded much finer than $fn=6
back(12) color(color) offset_stroke(p1, start="round", end="round",
width=w, $fn=6);
// with $fn=8 end treatments have 3 segments and all outer path points
are rounded
color(color) offset_stroke(p1, start="round", end="round", width=w,
$fn=8);
// with pointed end treatments all outer points are rounded, $fn has no
effect
fwd(12) color(color) offset_stroke(p1, start="pointed", end="pointed",
width=w, $fn=8);
}
d = 6.82;
// only the right most outer point is rounded
right(-12) cclip("yellow", arc=239.0, fn=6, id=d, od=d+1.75);
// with smaller inner diameter all outer points are rounded
right(0) cclip("green", arc=239.0, fn=6, id=d-0.75, od=d+1.75);
// slightly increased arc and only leftmost outer point isn't rounded
right(12) cclip("red", arc=239.001, fn=6, id=d-0.75, od=d+1.75);
// same points as before but not computed and right most outer point is
also not rounded
red_p = [[-2.28771, -1.29435], [-3.035, 0], [-1.5175, 2.62839], [1.5175,
2.62839], [3.035, 0], [2.28771, -1.29435]];
right(24) cclip("blue", arc=0, fn=6, id=d-0.75, od=d+1.75, p=red_p);
[image: offset_stroke.png]
Note that BOSL2 bug reports are best raised as issues on the BOSL2 github.
It appears there is an off-by-one error in the rounding for "round". There
was also a bug in how offset() handled the facet count that I recently
fixed.
On Sun, Sep 7, 2025 at 2:09 PM Todd Allen via Discuss <
discuss@lists.openscad.org> wrote:
// I think this code demonstrates inappropriate handling of $fn in BOSL2's
offset_stroke().
include <BOSL2/std.scad>
include <BOSL2/rounding.scad>
$fn= $preview ? 16 : 128;
function cclip_p(arc, fn, id, od) =
let(
ax2 = (360-arc)0.5,
p0 = difference(
circle(d=id, $fn=fn),
arc(n=2, r=10od, angle=[270-ax2,270+ax2], wedge=true)),
// difference returns a region, strip back to a path without last point
p1 = list_head(p0[0])
) p1;
module cclip(color, arc, fn, id, od, p=0) {
p1 = p ? p : cclip_p(arc, fn, id, od);
w=[0,od-id];
echo(color=color, w=w, p1=p1);
// with $fn=6 end treatments have 2 segments not 3?
// there is weird behavior for which outer path points are rounded
// outer path points are rounded much finer than $fn=6
back(12) color(color) offset_stroke(p1, start="round", end="round",
width=w, $fn=6);
// with $fn=8 end treatments have 3 segments and all outer path points
are rounded
color(color) offset_stroke(p1, start="round", end="round", width=w,
$fn=8);
// with pointed end treatments all outer points are rounded, $fn has
no effect
fwd(12) color(color) offset_stroke(p1, start="pointed", end="pointed",
width=w, $fn=8);
}
d = 6.82;
// only the right most outer point is rounded
right(-12) cclip("yellow", arc=239.0, fn=6, id=d, od=d+1.75);
// with smaller inner diameter all outer points are rounded
right(0) cclip("green", arc=239.0, fn=6, id=d-0.75, od=d+1.75);
// slightly increased arc and only leftmost outer point isn't rounded
right(12) cclip("red", arc=239.001, fn=6, id=d-0.75, od=d+1.75);
// same points as before but not computed and right most outer point is
also not rounded
red_p = [[-2.28771, -1.29435], [-3.035, 0], [-1.5175, 2.62839], [1.5175,
2.62839], [3.035, 0], [2.28771, -1.29435]];
right(24) cclip("blue", arc=0, fn=6, id=d-0.75, od=d+1.75, p=red_p);
[image: offset_stroke.png]
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
Thanks Adrian!
Note that BOSL2 bug reports are best raised as issues on the BOSL2
github.
My apologies. My frame of mind was still somewhat confused rather than
this is definitely a bug meriting a bug report. The documentation said
roughly "the number of segments for end rounding and offset rounding are
computed in the usual way with respect to special variables." I did not
have certainty as to what that meant. I wondered, is each rounding formed
from equal length segments with the number of segments set by a floor or
ceiling? I tried to gain clarity with experimentation and the results
made no sense to me but I still wasn't certain what the results should be
or that I wasn't making a mistake in the invocation. I turned to this
email list as a place welcoming of confused user thoughts and I was
hesitant to pollute the expert developer level discourse on github.
On Sun, Sep 7, 2025 at 10:18 PM Adrian Mariano via Discuss <
discuss@lists.openscad.org> wrote:
Note that BOSL2 bug reports are best raised as issues on the BOSL2
github.
It appears there is an off-by-one error in the rounding for "round".
There was also a bug in how offset() handled the facet count that I
recently fixed.
On Sun, Sep 7, 2025 at 2:09 PM Todd Allen via Discuss <
discuss@lists.openscad.org> wrote:
// I think this code demonstrates inappropriate handling of $fn in
BOSL2's offset_stroke().
include <BOSL2/std.scad>
include <BOSL2/rounding.scad>
$fn= $preview ? 16 : 128;
function cclip_p(arc, fn, id, od) =
let(
ax2 = (360-arc)0.5,
p0 = difference(
circle(d=id, $fn=fn),
arc(n=2, r=10od, angle=[270-ax2,270+ax2], wedge=true)),
// difference returns a region, strip back to a path without last
point
p1 = list_head(p0[0])
) p1;
module cclip(color, arc, fn, id, od, p=0) {
p1 = p ? p : cclip_p(arc, fn, id, od);
w=[0,od-id];
echo(color=color, w=w, p1=p1);
// with $fn=6 end treatments have 2 segments not 3?
// there is weird behavior for which outer path points are rounded
// outer path points are rounded much finer than $fn=6
back(12) color(color) offset_stroke(p1, start="round", end="round",
width=w, $fn=6);
// with $fn=8 end treatments have 3 segments and all outer path
points are rounded
color(color) offset_stroke(p1, start="round", end="round", width=w,
$fn=8);
// with pointed end treatments all outer points are rounded, $fn has
no effect
fwd(12) color(color) offset_stroke(p1, start="pointed",
end="pointed", width=w, $fn=8);
}
d = 6.82;
// only the right most outer point is rounded
right(-12) cclip("yellow", arc=239.0, fn=6, id=d, od=d+1.75);
// with smaller inner diameter all outer points are rounded
right(0) cclip("green", arc=239.0, fn=6, id=d-0.75, od=d+1.75);
// slightly increased arc and only leftmost outer point isn't rounded
right(12) cclip("red", arc=239.001, fn=6, id=d-0.75, od=d+1.75);
// same points as before but not computed and right most outer point is
also not rounded
red_p = [[-2.28771, -1.29435], [-3.035, 0], [-1.5175, 2.62839], [1.5175,
2.62839], [3.035, 0], [2.28771, -1.29435]];
right(24) cclip("blue", arc=0, fn=6, id=d-0.75, od=d+1.75, p=red_p);
[image: offset_stroke.png]
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
