16:21 <bozo16> I was using cgal as default backend

16:57 <JordanBrown> bozo16 probably better to use Manifold.

17:01 <gbruno> [github] huskier opened issue #5889 (About counter-intuitive variables and scoping) https://github.com/openscad/openscad/issues/5889

17:06 <gbruno> [github] jordanbrown0 closed issue #5889 (About counter-intuitive variables and scoping) https://github.com/openscad/openscad/issues/5889

19:35 <raydenuni> Hello. I'm trying to use BOSL2 and dotSCAD in the same file, but I'm getting some errors that I'm not having success interpreting.

19:35 <raydenuni> WARNING: undefined operation (vector + number) in file , line 220

19:35 <raydenuni> ERROR: Assertion 'is_int(n)' failed in file , line 137

19:35 <raydenuni> TRACE: called by 'lerpn' in file , line 220

19:35 <raydenuni> TRACE: called by 'bezier_curve' in file Test_Spline.scad, line 156

19:35 <raydenuni> TRACE: called by 'sf_splines' in file Test_Spline.scad, line 158

19:35 <raydenuni> TRACE: called by 'sf_thicken' in file Test_Spline.scad, line 158

19:36 <raydenuni> include <BOSL2/std.scad>

19:36 <raydenuni> include <BOSL2/screws.scad>

19:36 <raydenuni> include <BOSL2/rounding.scad>

19:36 <raydenuni> use <dotSCAD/src/bezier_curve.scad>

19:36 <raydenuni> use <dotSCAD/src/surface/sf_splines.scad>

19:36 <raydenuni> use <dotSCAD/src/surface/sf_thicken.scad>

19:36 <raydenuni> If I don't call `sf_thicken`, it doesn't seem to happen.

19:36 <raydenuni> sf_thicken(sf_splines(ctrl_pts, bezier), thickness);

19:39 <raydenuni> for full source

19:39 <raydenuni> https://pastebin.com/wA1Q93EG

19:41 <raydenuni> oh, is this trace a callstack? good grief

19:42 <raydenuni> there we go, two implementations of bezier_curve()

19:44 <raydenuni> What are my options for namespacing? I didn't find anything in the docs.

19:50 <InPhase> raydenuni: Does it not show the files, and actually shows "file ,"?

19:52 <InPhase> raydenuni: It would sure be easier to understand the issue if we knew exactly which one is tripping up.

19:53 <raydenuni> that is copy/pasted

19:53 <raydenuni> it doesn't say the file

19:53 <InPhase> That's an unfortunate bug in reporting the errors.

19:53 <InPhase> I don't think I've seen that before.

19:54 <raydenuni> but it seems that there's a collision between dotSCAD/src/bezier_curve.scad and BOSL2/beziers.scad.

19:54 <raydenuni> function bezier_curve(t_step, points) = _bezier_curve_impl(t_step, points);

19:54 <raydenuni> vs

19:54 <raydenuni> function bezier_curve(bezier,splinesteps=16,endpoint=true) =

19:54 <raydenuni>     bezier_points(bezier, lerpn(0,1,splinesteps+1,endpoint));

19:54 <raydenuni> let me find that error again

19:54 <InPhase> Yeah. I see the is_int in BOSL2's tests/test_utility.scad line 137

19:55 <raydenuni> although I'm guessing the error handling that prints the file would be somewhere standard

19:56 <InPhase> raydenuni: Is this code just for your use, or do you plan on distributing it?

19:56 <raydenuni> just for me

19:57 <InPhase> raydenuni: What I recommend is opening dotSCAD/src/bezier_curve.scad and editing the module bezier_curve to be ds_bezier_curve. dotSCAD will not be getting any more updates, as unfortunately the maintainer of that impressive work has died. So there will not be anything coming out to clash with your edit.

19:58 <raydenuni> ah, I saw that post on github. :(

19:58 <raydenuni> I believe the assert is coming from BOSL2/math.scad#137

19:58 <raydenuni> function lerpn

19:59 <InPhase> Namespacing is clearly needed on the agenda going forward. The quantity of impressive libraries has been increasing, and the capability to use more of them is increasing with all these performance boosts.

19:59 <raydenuni> I saw there's a post in 2013 discussing namespacing. :P Holding my breath.

20:00 <InPhase> Well there has been discussion of object literals, and progress on this as well. A namespace and an object are not very different in notion.

20:01 <InPhase> If you put everything for a set of includes or uses into an object, then that's a namespace.

20:01 <raydenuni> oh wow, renaming it worked. I expected there to be approximately 800 internal references I would have to clean up

20:01 <InPhase> As long as they can find each other.

20:01 <raydenuni> yep

20:02 <InPhase> Well, there are internal references to clean up, but only if you call those.

20:02 <InPhase> I'll get you the full set.

20:03 <InPhase> raydenuni: There are 5 lines to fix: https://bpa.st/HQNQ

20:05 <raydenuni> thank you

20:07 <raydenuni> now to figure out how I want to define all these splines. wheee

20:08 <InPhase> :)

20:50 <teepee> ok, shall we check if the GSoC selections are publushed?

21:27 <teepee> nice, manifold: 3d-mesh-offset! https://summerofcode.withgoogle.com/programs/2025/projects/LNlkt0XJ

21:28 <teepee> and we have: windows build support: https://summerofcode.withgoogle.com/programs/2025/projects/MPJPiTJb

22:11 <InPhase> Ooo, offset.