07:24 <phoo> hi. For testing purposes, I wish there was an 'end' command, that stops the parser from contuing

07:29 <gbruno> [github] KungPhoo opened issue #4682 (Column editor) https://github.com/openscad/openscad/issues/4682

07:38 <gbruno> [github] t-paul closed issue #4682 (Column editor) https://github.com/openscad/openscad/issues/4682

07:40 <teepee> phoo: you mean like a cancel button in the GUI? hopefully that comes at some point, needs a little bit of work in the internals so it's not a trivial change

07:43 <gbruno> [github] KungPhoo opened issue #4683 (end() command) https://github.com/openscad/openscad/issues/4683

07:43 <gbruno> [github] KungPhoo edited issue #4683 (end() command) https://github.com/openscad/openscad/issues/4683

07:48 <phoo> Is SCAD internally working with triangles? Or mathemetic shapes, like OpenCascade does?

07:48 <gbruno> [github] t-paul reopened issue #4682 (Column editor) https://github.com/openscad/openscad/issues/4682

07:49 <teepee> at this point everything is mesh based

07:50 <teepee> what do you mean with end() command?

07:50 <phoo> Nice. Is there a way to implement own code that has access to the triangles?

07:50 <phoo> the end() comamnd should stop the parsing. Example: sphere(1); end(); cube(1); // only the sphere is visible

07:51 <teepee> no, and that's unlikely to happen

07:51 <phoo> you might also call it stop or kill

07:51 <teepee> there is the ! modifier for that

07:51 <teepee> !sphere(1); cube(1);

07:51 <teepee> will only show the sphere

07:52 <phoo> I see. But !difference() will still do the difference

07:52 <teepee> currently there's no way to query generated meshes, a feature like that exists as unfinished preview in a Pull-Request

07:52 <teepee> ! means: show only the sub-tree that is marked

07:53 <teepee> an end() command is mostly impossible as we don't process code sequentially

07:53 <phoo> oh! Thank you so much. The ! is exactly the thing I was searching for

07:54 <teepee> parsing generates a tree of operations that will be evaluated in a second step

07:54 <teepee> also check the other modifiers, they are very usefull too :)

07:54 <teepee> modifiers?

07:54 <othx> modifiers are useful helpers for developing/debugging OpenSCAD scripts, for details see https://en.wikibooks.org/wiki/OpenSCAD_User_Manual/Modifier_Characters

07:55 <gbruno> [github] KungPhoo closed issue #4683 (end() command) https://github.com/openscad/openscad/issues/4683

07:55 <gbruno> [github] KungPhoo edited issue #4683 (end() command) https://github.com/openscad/openscad/issues/4683

08:13 <Scopeuk> had a little play with multi line select inline with https://github.com/openscad/openscad/issues/4682 (used alt plus draw a vertical bar as notepad++) oddly it drew the line, then let me type by everything I typed was written backwards on the last row. i.e. I selected three rows and typed hello and ended up with the bottom line saying olleh

08:51 <teepee> ouch, that's not very useful.

08:52 <teepee> I wonder if QScintilla has too old scintilla code or if we somehow suppress the hotkey for the feature

18:04 <joseph_> Here is a file that has strange behavior in F5 mode. https://gist.github.com/jbinvnt/bdb2c63a170c1306e97a5c35373cf2f2 The incline cutout fades away when zoomed in, and appears as expected when zoomed out. So far I tested this on Linux with both stable and nightly, plus on Mac with stable. All have the same behavior

18:05 <joseph_> teepee: Do you know if there is already an open issue for this bug? It might be something relevant for me to know about as I work on rendering improvements. The issue doesn't happen in F6

18:51 <J23k3> joseph_ can't replicate (with manifold and isometric view)  you can try to color the substracted cylinder  sometimes the shading is equal so you can't see the cut  - but if you are close with perspective view some clipping might occure

18:54 <J23k3> if the camera is within the object that is substracted you will see  a wrong preview

19:15 <teepee> joseph_: let me try the example....

19:17 <teepee> yep, that's normal, and not exactly a bug

19:17 <teepee> only happens in perspective mode when you move the camera into the cube doing the cutting

19:18 <teepee> due to the camera being inside, the cut is not drawn

19:19 <teepee> oh, hey, J23k3 already said that :)

19:20 <teepee> this does not happen in F6 as we have the final mesh, no oversized cutting cubes anymore

19:22 <J23k3> that could be added https://en.wikibooks.org/wiki/OpenSCAD_User_Manual/Print_version#Preview_doesn't_appear_to_work_at_all

19:22 <joseph_> Thank you, yes I understand now. It's because of my habit to make a giant cube when I want to make a planar cut without worrying about the necessary size. I guess I just normally wouldn't expect the camera position to factor into what makes the cuts

19:22 <J23k3> didn't find it mentioned

19:23 <teepee> joseph_: I suppose one option would be to explicitely support a cutting plane, but I don't know if that helps with the CSG preview then

19:24 <teepee> it would at least reduce the need for creating a cube

19:28 <joseph_> Perhaps it would be a good exercise for me to try pinpointing in the rendering process what would cause this. Because normally I wouldn't expect that a user would need to worry about clipping of objects that are really just an intermediate step in making their model

19:29 <gbruno> [github] t-paul locked issue #4130 (Dark mode plz.) https://github.com/openscad/openscad/issues/4130

19:31 <teepee> well, moving the camera into the actual object is probably not solvable

19:32 <teepee> the cutting cube is a bit special as it's difficult to produce a general solution, so making it giant is an easy way out for making sure the cut happens but it give the preview issues in perspective mode

19:33 <teepee> there's an old PR that had also user movable X/Y/Z cutting planes for inspecting the inner parts of a model which would also be a very neat feature

19:34 <joseph_> Interesting. Do you mean movable with a UI slider?

19:35 <teepee> yep

19:37 <teepee> right, that new github search is bad

19:37 <teepee> 8 results (in 156 ms)

19:37 <teepee> thanks, very much, information overload

19:37 <teepee> still showing only 5 in the list then

19:40 <teepee> hmm, I thought it had screenshots... https://github.com/openscad/openscad/pull/1300

19:50 <teepee> oh, cool stuff https://www.youtube.com/watch?v=IML9c_MsyQU

19:50 <othx> teepee linked to YouTube video "Making an ESP32-S3 Dev Board for the Framework Laptop" => 1 IRC mentions

20:01 <teepee> lol, #ohl for open hardware license is not ideal. looks like most use is Ontario Hockey League

20:01 <petaflot> whoo clipping planes is going to be nice

20:05 <teepee> indeed, and interactive ones in addition would be awesome. although someone needs to find a working implementation :)

21:00 <kintel> joseph_ The reason for the preview rendering issue is really the CSG rendering algorithm inside OpenCSG. It essentially counts the number of front faces and back faces by shooting a ray from the camera through the object.

21:01 <kintel> ..but since it does this in image space by drawing actual primitives, the primitives drawn needs to fall within the view frustum to render correctly.

21:01 <kintel> Once the camera is inside an object, or even too close to an object, it won't work.

21:02 <kintel> It can always be fixed, but that's a question of inventing as new CSG algorithm :)

21:04 <juri_> CSG systems are hard to write. :)

21:05 <gbruno> [github] kintel closed issue #4673 (Support building on macOS 13 and Xcode >= 14.3) https://github.com/openscad/openscad/issues/4673

21:05 <gbruno> [github] kintel edited issue #4673 (Support building on macOS 13 and Xcode >= 14.3) https://github.com/openscad/openscad/issues/4673

21:05 <gbruno> [github] kintel edited issue #4673 (Support building on macOS 13 and Xcode >= 14.3) https://github.com/openscad/openscad/issues/4673

21:05 <gbruno> [github] kintel edited issue #4673 (Support building on macOS 13 and Xcode >= 14.3) https://github.com/openscad/openscad/issues/4673

21:29 <gbruno> [github] kintel pushed 1 modifications (Bump xcode to 14.3.1) https://github.com/openscad/openscad/commit/abbe51a96f63c60f27b2c224f91fca19941c2da2

21:32 <gbruno> [github] kintel opened pull request #4684 (Support patch version in macosx-build-dependencies + bump xcode) https://github.com/openscad/openscad/pull/4684