teepee changed the topic of #openscad to: OpenSCAD - The Programmers Solid 3D CAD Modeller | This channel is logged! | Website: http://www.openscad.org/ | FAQ: https://goo.gl/pcT7y3 | Request features / report bugs: https://goo.gl/lj0JRI | Tutorial: https://bit.ly/37P6z0B | Books: https://bit.ly/3xlLcQq | FOSDEM 2020: https://bit.ly/35xZGy6 | Logs: https://bit.ly/32MfbH5
mtm has quit [Ping timeout: 260 seconds]
mtm has joined #openscad
dalias has quit [Ping timeout: 252 seconds]
dalias has joined #openscad
<JordanBrown> Guest69: No. Difference or intersect with a big cube.
mtm has quit [Ping timeout: 248 seconds]
mtm has joined #openscad
kintel has joined #openscad
kintel has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
othx has quit [Remote host closed the connection]
othx has joined #openscad
foul_owl has quit [Ping timeout: 255 seconds]
foul_owl has joined #openscad
<Guest69> Thanks Jordan, yes that's what I did
kintel has joined #openscad
<kintel> f0lkert Fail in what way?
<kintel> Note that quads in floating point space are often technically non-planar. If some downstream code has a too strict check, things may go south
kintel has quit [Client Quit]
<meshugga> is there a way to "pointify" a projection() so i can use the resulting 2d as a path for path_extrude from dotSCAD?
<InPhase> meshugga: No, not without a somewhat outdated PR that hasn't been integrated.
<meshugga> nooooo i was counting on being able to do that
<meshugga> well, crap
<InPhase> There is probably a way to rethink the problem.
<InPhase> There is one way to sort of do it. You can export as svg, then grab the points from the svg, and put them back in as input data.
<InPhase> But this is not automated unless you write a little wrapper to help you out.
<InPhase> The points are human-readable though in an svg.
<InPhase> But generally the OpenSCAD style approach to such problems is to figure out how to generate your points functionally.
<meshugga> yeah, the trig is to complicated and it's actually already done for me
<meshugga> it's right there :)
<meshugga> there is no way to inspect the result of a projection runtime?
<meshugga> shit, that really limits design options drastically
<meshugga> InPhase: what patch are we talking about? gh issue#?
<InPhase> meshugga: The data = render() one from some time back.
<InPhase> meshugga: I forget the number, but it'd probably be searchable.
J24k26 has joined #openscad
peepsalot has quit [Remote host closed the connection]
peepsalot has joined #openscad
J24k has quit [Ping timeout: 256 seconds]
Guest49 has joined #openscad
Guest49 has quit [Client Quit]
<gbruno> [github] IamLRBA opened pull request #5462 (Added case-sensitive toggle to Find and Replace panel) https://github.com/openscad/openscad/pull/5462
snaked has quit [Quit: Leaving]
snaked has joined #openscad
mmu_man has quit [Ping timeout: 252 seconds]
Reisga2449 has quit [Ping timeout: 244 seconds]
Reisga2449 has joined #openscad
guso78k has joined #openscad
<guso78k> @kintel, i have this issue:
<guso78k>        const QAction *action ;
<guso78k>  const auto form = info.format;
<guso78k>  const auto map = mainWindow->export_map;
<guso78k>  action = map[form];
<guso78k> then:
<guso78k>   267 | action = map[form];
<guso78k> link is better
<f0lkert> kintel: found that there's kind of a major problem: polygons in povray can only be 2D. That's why sometimes it renders black spots. So for now we need the triangulate stuff while I'll think of a solution to this.
guso78k has quit [Quit: Client closed]
GNUmoon has joined #openscad
<gbruno> [github] folkertvanheusden synchronize pull request #5460 (Fix for #5459.) https://github.com/openscad/openscad/pull/5460
<gbruno> [github] folkertvanheusden synchronize pull request #5460 (Fix for #5459.) https://github.com/openscad/openscad/pull/5460
<f0lkert> kintel: ok what I now do is for each polygon, convert it into a triangulated mesh2-object. now I only need to see how I can determine the 'normal' for each triangle so to make it into a smooth object.
guerd871 has joined #openscad
guerd87 has quit [Ping timeout: 245 seconds]
mtm has quit [Ping timeout: 245 seconds]
mtm has joined #openscad
Guest28 has joined #openscad
Guest28 has quit [Quit: Client closed]
Guest28 has joined #openscad
mmu_man has joined #openscad
<Guest28> Hi, someone above helped me create this openscad model https://pasteboard.co/RNkD22UFP7c8.png
<Guest28> I can't see the link anymore because my libera cha reset can anyone see the link in the chat history
<Guest28> or is there a way I can view the chat history here ?
<Guest28> ah just seen the link to the logs
Guest4 has joined #openscad
guso78k has joined #openscad
Guest4 has quit [Quit: Client closed]
dicot has joined #openscad
RoyK has quit [Ping timeout: 252 seconds]
<gbruno> [github] dhyananbho opened issue #5463 (Copy/paste does not work in file dialogue windows under macOS 15.1.1) https://github.com/openscad/openscad/issues/5463
RoyK has joined #openscad
<Guest28> InPhase your created the openscad code for this object https://pasteboard.co/RNkD22UFP7c8.png
<Guest28> the link has expired any chance you can send it to me again
<InPhase> Guest28: Yeah, I still have that one in my temp files: https://bpa.st/V3EQ
cart_man has joined #openscad
<gbruno> [github] kintel edited issue #5444 (macOS 15: Quitting the app does not warn about unsaved files under macOS 15.1) https://github.com/openscad/openscad/issues/5444
<gbruno> [github] kintel edited issue #5445 (macOS 15: The size and position of the window on the screen is not saved for the next time the app opens under macOS 15.1) https://github.com/openscad/openscad/issues/5445
<cart_man> Anybody maybe have a library for DIN Rails in OpenSCAD?
<J24k26> https://en.wikipedia.org/wiki/DIN_rail doesn't look very complex
<cart_man> J24k26: No it is not but ther are so many different kinds. Thought someone might have a library of sorts for it
<J24k26> cart_man https://bpa.st/ZZTA
<J24k26> cart_man here with round edges https://bpa.st/XHCQ
TheAssassin has quit [Remote host closed the connection]
TheAssassin has joined #openscad
<cart_man> J24k26: Gheezzz dude. I literally just finished my primitive polygon trace and quite chuffed with it and you slam out this absolute banger! I do not even know where to begin to be thisn good at OpenSCad : /
<teepee> just do it 10 years daily like J24k26 ;-)
<teepee> the best approach is to use things and learn along, if it's useful and/or fun enough, you will learn. nobody starts with trying to compete with world best sports people or musitions either
<J24k26> you also learn some tricks and "best practice" over time ( "i stand on shoulders of giants")
<teepee> hmm, still unsure what to do about scadvent :-/
<J24k26> and i used some lazy shortcut for the rounded version - as i am not using my library
<J24k26> cart_man here a different approach using turtle grafic https://bpa.st/26CA
<J24k26> teepee maybe this could be scadvent  4× a task and  showing 7 approaches how to do it
<J24k26> 6
<teepee> I like the idea, but difficult to get those so late
<cart_man> What does the $fs=.5;$fa=1; mean? I always just $fa=200 and get done with it
<J24k26> InPhase loves code golf
<teepee> for the late start my idea was weekdays = tutorial, weekend some designs like last years
<teepee> I know :-)
<J24k26> cart_man i assume you mean $fn=200..  $fa is the max angle to limit $fs (fragment size)
<teepee> but there's not even a november 31th, so december probably started somewhere already
<J24k26> didn't you had something from jordan
<teepee> yes, there's one from last year on github
<teepee> well, for first day, maybe a simple tree design is fine and doable on short notice
<teepee> also I guess if start a bit late, that's also not a huge deal
<teepee> I'm not so much concerned about first day, but the 23 after that :)
<guso78k> code golf is shortes program to goal ?
<teepee> yep
Guest28 has quit [Ping timeout: 256 seconds]
<guso78k> so basically obfuscated openSCAD ;)
<teepee> that is usually a side effect
<teepee> in pretty much all languages
<teepee> except maybe APL and Perl
<teepee> ({⍵,+/¯2↑⍵} ⍣ 20) (1 1)
<teepee> guess what :-)
<teepee> so for a tutorial edition of scadvent, we'd need 15 lessons, let me see if I can come up with 15 topics
<teepee> it would have been great to integrate the web stuff, but it's way too late for that this year
<J24k26> so when do we get apl extension for scad - Ü
<teepee> heh, guso78k probably just started with that
<J24k26> so i guess we just need to solve the timemachine problem to make this working
<teepee> that would work, yes
<J24k26> we could use the wiki tips n tricks section for each day
<teepee> first week would be easy
<J24k26> using the cheatsheet as template
<teepee> that's what I have open in parallel
<teepee> was there any other design in that reddit thread? with the one from jordan we would have 2 more complex ones so far
<J24k26> day 3  circle(10,$fn3);
<J24k26> day 4 circle(10,$fn4);
<J24k26> can't remember seeing any on reddit
<teepee> hm? any? what's with that fuzzy font thingy, I do remember trying that one. don't tell me that was last year ;-)
<teepee> yep, that last one I meant
<teepee> if it's not too much to get the needed stuff from the library
<teepee> I still would like to try getting my "schwibbogen" idea to come to life
<teepee> that could take care of the 3 saturday slots
<J24k26> probably just 2 modules - but maybe those are not suited for "tutorial" to learn how to make it
<J24k26> the "airvent" is without lib
val has joined #openscad
<teepee> I'd say not a problem if the weekday tutorial part and the other "nice things" have no relation
<teepee> not directly anyway, they could still serve as interesting challenge
<gbruno> [github] kintel ready_for_review pull request #5455 (Use Qt6 (6.5.3) for macOS) https://github.com/openscad/openscad/pull/5455
<gbruno> [github] kintel edited pull request #5455 (Use Qt6 (6.5.3) for macOS) https://github.com/openscad/openscad/pull/5455
<gbruno> [github] kintel edited pull request #5455 (Use Qt6 (6.5.3) for macOS) https://github.com/openscad/openscad/pull/5455
<gbruno> [github] kintel pushed 4 modifications (Merge branch 'master' into kintel-qt6) https://github.com/openscad/openscad/commit/f1e9a9ff8c995b1ef14465f07292b9598ed28bf7
<gbruno> [github] kintel synchronize pull request #5455 (Use Qt6 (6.5.3) for macOS) https://github.com/openscad/openscad/pull/5455
teepee_ has joined #openscad
<gbruno> [github] kintel pushed 1 modifications (Build for macOS 11) https://github.com/openscad/openscad/commit/5036bfaaac26d9f4f352868b3d66881e61745bae
<gbruno> [github] kintel synchronize pull request #5455 (Use Qt6 (6.5.3) for macOS) https://github.com/openscad/openscad/pull/5455
<gbruno> [github] kintel pushed 1 modifications (Build for macOS 11) https://github.com/openscad/openscad/commit/0d5fabfb4d4f0d51105af86cc6adc49c694bc91a
<gbruno> [github] kintel synchronize pull request #5455 (Use Qt6 (6.5.3) for macOS) https://github.com/openscad/openscad/pull/5455
teepee has quit [Ping timeout: 260 seconds]
teepee_ is now known as teepee
<gbruno> [github] kintel pushed 1 modifications (Qt5 comment) https://github.com/openscad/openscad/commit/494072e3939ec4dd14ad8c36b12ac2c73273f3d7
<gbruno> [github] kintel synchronize pull request #5455 (Use Qt6 (6.5.3) for macOS) https://github.com/openscad/openscad/pull/5455
kintel has joined #openscad
<gbruno> [github] kintel closed pull request #5455 (Use Qt6 (6.5.3) for macOS) https://github.com/openscad/openscad/pull/5455
<gbruno> [github] kintel pushed 2 modifications (Use Qt6 (6.5.3) for macOS (#5455)) https://github.com/openscad/openscad/commit/b72b049dcae204af93b895c21d20269de20eac10
<kintel> guso78 You cannot use the [] operator on a const map, as it will insert an object if it doesn't exist
<kintel> If you're 100% certain that the object exists, use map.at(), otherwise, map.find() and check the iterator
snaked has quit [Remote host closed the connection]
Drizzt321 has quit [Quit: Leaving]
<kintel> FYI: Making macOS snapshots Qt6-only going forward. Too many Qt5 issues with recent OSes
germ_ has quit [Read error: Connection reset by peer]
germ_ has joined #openscad
guso78k has quit [Quit: Client closed]
guso78k has joined #openscad
<gbruno> [github] gsohler synchronize pull request #5461 (Sorry, reopening Fileformat mapper) https://github.com/openscad/openscad/pull/5461
<gbruno> [github] folkertvanheusden synchronize pull request #5460 (Fix for #5459.) https://github.com/openscad/openscad/pull/5460
guso78k has quit [Quit: Client closed]
kintel has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
mmu_man has quit [Read error: Connection reset by peer]
ali12341 has joined #openscad
<gbruno> [github] kintel pushed 1 modifications (Build OpenSCAD with Qt6 for macOS) https://github.com/openscad/openscad/commit/7f3c8f394ca19416b536fc491b59149cb9eefb1c
<gbruno> [github] kintel opened pull request #5464 (Build OpenSCAD with Qt6 for macOS) https://github.com/openscad/openscad/pull/5464
ali1234 has quit [Remote host closed the connection]
<gbruno> [github] kintel closed pull request #5464 (Build OpenSCAD with Qt6 for macOS) https://github.com/openscad/openscad/pull/5464
<gbruno> [github] kintel pushed 1 modifications (Build OpenSCAD with Qt6 for macOS (#5464)) https://github.com/openscad/openscad/commit/fc15a971d863b47ceaa8ab81288ba8891e29b62a
<gbruno> [github] gsohler synchronize pull request #5461 (Sorry, reopening Fileformat mapper) https://github.com/openscad/openscad/pull/5461
TheCoffeMaker has quit [Remote host closed the connection]
TheCoffeMaker has joined #openscad
TheAssassin has quit [Remote host closed the connection]
TheAssassin has joined #openscad
Guest69 has quit [Ping timeout: 256 seconds]
<dicot> so uh, I was messing with the recursion example... how do I stop it trying to render?
<InPhase> dicot: What do you mean by that?
<InPhase> teepee: So are we doing scadvent? I was sick for much of the last week, so didn't really get started on thinking about it. But I'm recovered.
pie_ has quit []
pie_ has joined #openscad
<InPhase> Looks like we have 1 design in the repo... So I guess no one else added any yet. :)
<dicot> I tried to change the stop condition and then pressed the preview button, and now I it won't stop, it's been going for two minutes.
<InPhase> dicot: Oh. Rendering doesn't really stop in the middle through the gui.
<InPhase> dicot: A problem yet to be addressed. Probably we could look into how to do this since the upgrade to Manifold. It was a non-starter with CGAL for a long time.
<InPhase> For Preview, I'm not sure. But I'm confident Preview is going to go away at some point (probably after the upcoming release) and be phased out in favor of Manifold-all-the-things.
<dicot> :) I'll be careful with recursion then, or upgrade to a CLI, if I read between your lines correctly?
<InPhase> dicot: I can add that the failure to stop is kind of irrelevant if you use an external editor. :)
<InPhase> dicot: I usually just terminate OpenSCAD and reopen it, and my file I'm editing stays open.
<InPhase> And with external editor usage, preview is auto-triggered at save anyway, so you'll never lose progress.
<InPhase> Although OpenSCAD saves a backup when preview is hit so it's always recoverable.
<teepee> InPhase: no final decision yet, I'm toying with that plan: https://pads.domainepublic.net/p/scadavent-2024_c5283e0a-99cc
<InPhase> Dec. 1 is tomorrow, so we should get ourselves going soon. :) Let's see that link.
<dicot> Well, I'm certainly learning a lot from the examples
<InPhase> Never heard of this Schwiboggen thing.
<teepee> I wanted to do that last year already and failed :)
<InPhase> Flammable handle holder. How has this not caught on?
<InPhase> s/handle/candle/
<teepee> well, people put candles into christmas trees too, which is probably even more dangerous once the trees (and needles!) are dried up
<teepee> but then todays pieces are led obviously
<InPhase> Yeah. :) I remember hearing about that going wrong a lot. And remember the era of hot incandescents being a risk. :)
<InPhase> In my childhood we used to time cutting down a tree so it wasn't too tried out by Christmas so that the big incandescent bulbs wouldn't light it on fire. And we'd never leave the lights on unattended.
<InPhase> s/tried/dried/
<teepee> we never had real candles even in my childhood, I do know some people who did that though
<InPhase> Yeah, that phased out where I was before I was born, as it seemed to go wrong a lot.
<teepee> real candles were in the pyramids, mostly safe if not left unattended
<teepee> tree I'm not happy with yet https://pastebin.com/UhuUwBZb
<InPhase> We really did no candles at all in the house other than for birthday cakes. The kerosene lamp only came out for power outages. I think I bought my first candles after graduating from college.
TheCoffeMaker has quit [Remote host closed the connection]
<InPhase> That's a nice tree for the code size.
<teepee> we mostly used tea lights which are pretty safe when used sensibly
<teepee> I was looking for a printable one, maybe 3 different sized parts stackable, but that's not really working out yet
<teepee> maybe that's just not important
<f0lkert> trying to figure out how to calculate normals. the latest patch uses mesh2 objects as polygons in povray are 2D only I found out. https://www.povray.org/documentation/view/3.70/68/ promises that triangles become smooth when appropriate normals are defined.
TheCoffeMaker has joined #openscad
<teepee> face or vertex normals?
<teepee> face normals are just 2 vectors of 3 points multiplied
<teepee> the STL code might have the code for that
<InPhase> teepee: for (i = [2:13]) { d = 4e-2*(21-i)^2-1e-7*(21-i)^6;
<InPhase> teepee: Suggested math tweak to your design.
<f0lkert> it says "For each vertex there is one normal vector listed in normal_vectors{}" so I believe that's vertex normals.
<f0lkert> teepee: ^
<InPhase> teepee: The green spire goes a bit too high there yet, but yeah, that's definitely tree-like. :)
<teepee> true, cutting back the loop to 14
<teepee> or 15 :)
<InPhase> Good old principle-driven math-fudging.
<f0lkert> tree ^ rendered with povray
<InPhase> f0lkert: Looks like it could be in a 1997 era game. :)
<f0lkert> :)
<teepee> yep, math teacher called that "PI mal Fensterkreuz", which may translate to something like "PI times fenster cross" and probably does not make too much sense ;-)
<InPhase> 14 probably works a bit better than 15. Room for a star yet. :)
<teepee> cool, I'm tempted to declare that good enough for day one, so if we think we can get the tutorial sorted on the way, we still may have a scadvent 2024
<teepee> and J24k26 has a neat idea for next year, but that probably requires a bit of headstart that time
<InPhase> How would we structure tutorials. Final-result image and guiding comments?
<InPhase> Animated with steps maybe?
<InPhase> Maybe code written like 10 steps as: if ($t < 0.05) { .... } else if ($t < 0.15) { ... }
<teepee> maybe just a normal variable and the image is generated externally as animation
<teepee> animation directly in openscad is probably not too useful
<InPhase> So one file, but a step selector variable controlled externally?
<teepee> unfortunatly too late for the web stuff
<teepee> yeah, I think one file per day is best as we don't have any new infrastructure, only the web stuff from previous years
<InPhase> Yeah. I'm trying to think about how to hammer it into the same format. :)
<InPhase> Also then we can mix and match. It doesn't need to be all tutorial stuff if we don't come up with enough of that.
<teepee> see that link ^ - 15 tutorials for the week days, filling the other days with other designs
<teepee> at least right now it sounds doable :-)
<teepee> famous last words of every time estimation of IT projects
<InPhase> Yeah. We just have to see what it comes to as divisions once we have to think of real stuff to demonstrate. :)
<InPhase> I could for example probably think of multiple for math/functions.
<InPhase> It probably deserves a separate demo for basic math and functions for polygons, for example, versus things like recursive functions which are more involved.
<teepee> well, week 3 still has 2 open slots in my current list
<InPhase> Ideas not fully formed in my head yet, but just thinking about how I might break things down.
<teepee> feel free to mess with that etherpad link, it should be editable
* InPhase nods.
<teepee> so maybe I should get the infrastructure up before going to bed
<InPhase> Primitives and translate/rotate would be hard to do on separate days, as you can't do much with primitives without adding translation and rotation.
<teepee> but we have the step=1, step=2 so the primitives are always in default place
<InPhase> But I might leave deciding on those to someone else. My head is drawn to some week 2 stuff.
<teepee> perfect :-)
<InPhase> Maybe I could spell out function-design thinking in a stepwise manner. Usually we just plunk the final answer, and people are like "How did you get to this magic??"
<InPhase> 4e-2*(21-i)^2-1e-7*(21-i)^6 for example was like 20 steps in 2 minutes. :)
<InPhase> Hiding the mental sauce makes it hard to grok.