00:38 <kintel> bitbasher: What usually happens is that it builds a number of obj files in parallel, and it eventually gets to a point where all compiling files take a lot longer which causes it to halt for a while. If your system is generally low on memory, it can take a bit of time for the build system to make it through this stage. The workaround is typically to reduce parallelization, since the default parallelization of CMake doesn't take memory usage into

00:38 <kintel> account.

00:39 <rapha> InPhase: sorry for taking so long to respond ... took a while for my brain for that to be ready. that's a really nice result, thank you very much! https://imgur.com/a/gzV9iOv

00:40 <gbruno> [github] kintel closed pull request #5967 (Added failing testcase for #5966) https://github.com/openscad/openscad/pull/5967

00:40 <gbruno> [github] kintel pushed 1 additions 1 modifications (Added failing testcase for #5966 (#5967)) https://github.com/openscad/openscad/commit/811ef8cd64ea5ca4e064a7ecb8a65e10f1baf0fd

00:49 <InPhase> rapha: Don't you want something more like h = 30 and removing the -(i*0.05)?

03:06 <church__> when looking on https://blenderartists.org/t/generating-soap-film-or-minimal-surfaces-in-blender-is-it-possible/1398485 i wonder if something like that can be ever implemented in openscad. looks like nice way to add concave fillets at non-simple parts joins. i wonder, what math is behind

03:10 <gbruno> [github] pca006132 closed issue #5956 (Unable to Color Custom Model) https://github.com/openscad/openscad/issues/5956

05:01 <InPhase> https://en.wikibooks.org/w/index.php?title=OpenSCAD_User_Manual/The_OpenSCAD_Language&diff=prev&oldid=4477374 Not sure what was going through the mind of this user from Iran when they attempted this edit which ended up wiping out the manual last month, but I reverted it.

05:02 <InPhase> I guess it's also a little concerning that nobody noticed the entire language reference page was wiped.

05:02 <InPhase> I suppose anyone using from the cheatsheet would have found it still working.

05:03 <InPhase> I was literally in the middle of explaining how handy having a wiki for the manual has been for OpenSCAD when I looked at it and... it was blank. lol

06:31 <sparsh-n> Hi bitbasher, I saw the specific one about OS examples and OS libraries would be useful. Although I'm also not sure how often the wikibooks are used, as I usually look through the readme of any project I clone first.

06:32 <sparsh-n> Was also wondering if you got it built with MSYS2 or with MSVC?

06:32 <sparsh-n> I was able to test some basic functions on the headless version, but the translation library has some issue with working. I'll post the error in a pastebin link and share it here

06:34 <sparsh-n> Also if its an issue with building the obj, my Visual Studio usually has a unreadable/very confusing error, but it usually boils down to a missing include statement or something similar. If you're able to post the error regarding Linking it could be similar to the errors I got while building mine.

06:39 <sparsh-n> On the note of headless builds, when I run my executable it usually just runs a single command at a time, and then closes. I'm under the assumption that a proper headless build shouldn't keep invoking .\openscad.exe for each call?

09:19 <bitbasher> Kintel : i do have a ton of apps running at the same time .. i have closed out a bunch of stuff and will leave the build to chug along as it will

10:19 <rapha> InPhase: no, it needs the height to be able to intersect with the sphere. i tried to use your linear_extrude/offset/sqrt technique to replace the sphere entirely, but wasn't able to find the right math for that.

11:54 <InPhase> rapha: I see that. I'm trying to put it closer to correct. Although I cannot try it as you haven't shared the source. :)

12:06 <rapha> oh! yes, i haven't! let me fix that... o_o

12:13 <rapha> InPhase: https://github.com/sixtyfive/branch-balljoin-adapter

12:54 <InPhase> I totally forgot about helping you with some sort of TKS GTR Top plate almost exactly one year ago. But apparently I already have a "rapha" directory. :)

12:54 <InPhase> I guess you make something every June? ;)

13:15 <InPhase> rapha: https://imgur.com/a/R19QNra and https://bpa.st/RJLA

13:18 <InPhase> rapha: I don't know if you really wanted the base shape protruding through that cylinder. But you had it that way, so I left it. :) Lowering "height" will bring it down. e.g. 19 is sufficient.

13:18 <InPhase> rapha: You can remove the diagnostic echo statement when you're done examining those numbers to see how it works.

13:20 <InPhase> After flushing the cache, it renders for me in 3 seconds with the nightly, which is quite tolerable.

13:41 <rapha> InPhase: dang! *I* forgot about that ... but oh well, after the GTR crashed into the Mediterranean and got repaired and crashed on the way towards a mountain top and got recovered, it eventually crashed into a brick wall and I decided to switch that quad over to an AlfaRC BabyAPEX frame, which is a bit of a more appropriate match to its power.

13:42 <rapha> Feels good to know there's a folder with my name somewhere in the world, though :)

13:42 <rapha> Oh wow!!!

13:43 <rapha> I looove that for loop that's used for the screw holes. Makes it so much neater.

13:46 <rapha> oh, and the convexity=4 is really good to know about. no more need to hit F6 all the time.

13:47 <rapha> i agree, h=19 is better

13:51 <rapha> i'm tempted to print it right away, but better to save the white PLA+ and wait until the aliexpress package with the new 0.6mm nozzle arrives and straight up do it in wood filament as intended

14:00 <rapha> okay, besides my less elegant programming, it was basically just a matter of adding a fixed (negative) factor to the function. how did you know to do that, though?

14:32 <InPhase> rapha: Well I have some shortcuts. At first I didn't know, and just commented out top() and put a straight call to complex_base() and looked at what I was dealing with. Then I went "wtf?" in my head, and added the echo. Then it was obvious from looking at the numerical values that this was not what I was trying to do in the loop.

14:33 <InPhase> "Just try it, look at results and values, and iterate" is a powerful reasoning shortcut with OpenSCAD. :)

14:34 <InPhase> The values went up when they were supposed to go down. If 0 to 1 values go up when they're supposed to go down, 1- that flips this.

14:34 <InPhase> Which of course one can sit there and reason out. And I've seen enough times that I could have thought it through. Printing out the numbers is faster to catch mistakes though.

15:02 <sculptor> yo

17:41 <teepee> in case anyone is interested in ditching external image upload sites, we have imgs.openscad.org (running slink) to see if that works well

17:42 <teepee> account creation is enabled, but accounts need to be approved

17:44 <teepee> hmm, the public view is a bit strange...

19:09 <InPhase> "Password must contain at least one number." :( CorrectHorse1BatteryStaple

19:10 <InPhase> "Password must contain at least one special character." :((( CorrectHorse!1BatteryStaple

19:26 <teepee> haha, I can disable that :)

19:26 <teepee> give me a minute

19:28 <teepee> I don't mind that as I'm using passbolt to generate and save the password anyway :P

19:29 <teepee> InPhase: ok, if I'm reading the docs correctly, it should now only *require* lowercase

19:30 <teepee> hmm, but where's the button to activate a user?

19:31 <teepee> uff, tiny little thingy totally not where it would be expected :)

19:33 <teepee> can you see the button? https://imgs.openscad.org/image/2e1cc6a2-91da-41ba-bef8-3d80868b3e28.png

19:43 <InPhase> teepee: The dropdown next to teepee to activate rcolyer? ;)

19:44 <InPhase> </logic>

19:44 <teepee> yep, that tiny 2 line angle is the dropdown. obviously.

19:45 <InPhase> Discover Uploads seems to have no way to actually click on the discovered things.

19:46 <InPhase> Well, I guess right-click open image goes right to the full sized image.

19:47 <teepee> that's what I find strange. right click works but it would seem sensible to be able to click on it to get the full picture

19:52 <InPhase> This is some very heavily obfuscated javascript. Was someone selling this?

19:53 <InPhase> I can't see why an open source system would be so obfuscated.

19:56 <teepee> sveltekit generated I think

19:57 <teepee> Built with Symfony and SvelteKit

19:57 <teepee> I don't know Symfony

19:57 <teepee> huh? PHP?

19:58 <teepee> I thought the *Kit* part of SvelteKit means JavaScript/TypeScript on the backend side, but who knows

19:59 <teepee> hmm, that is definitely sveltekit https://github.com/andrii-kryvoviaz/slink/blob/main/services/client/src/routes/info/%5Bid%5D/%2Bpage.server.ts

20:00 <teepee> oh, https://github.com/andrii-kryvoviaz/slink/blob/main/screens/d3905318-f8ad-48d5-bb9a-09e1e45dbb8e.png has some link icons top right of the image

20:00 <teepee> haha, totally invisible with my 2 white background images :)

20:05 <InPhase> "User Approval [enabled] - Toggle whether users need to be approved before they can access the application" "Unauthenticated Access [enabled] - Toggle whether users can access the application without being authenticated"

20:05 <InPhase> I guess the first probably meant to say for uploading, and the second meant to say for viewing?

20:06 <teepee> yep, unauthenticated is the /explore page

20:06 <teepee> that's empty if disabled

20:06 <teepee> so you need to login to explore too

20:08 <InPhase> Uhm... Let's say some user named teepee went off the rails and started posting porn or something.

20:08 <InPhase> I see I have admin access, but I don't see a way to actually delete problematic content. Wouldn't that be important? :)

20:08 <teepee> hence the approval

20:09 <InPhase> Can you even delete your own posts?

20:09 <teepee> hmm, good question

20:09 <teepee> I guess "remove user" might also work

20:10 <InPhase> How much you want to bet that leaves the files still on the server.

20:11 <teepee> we can try

20:14 <teepee> I see it in upload-history

20:15 <teepee> with trash button

20:15 <teepee> not very scalable but I guess we won't have 1000s of users anytime soon

20:15 <teepee> oh, it's gone

20:16 <InPhase> I still see it.

20:16 <teepee> lol, cache bug?

20:16 <teepee> it was in the history list, but gone after refresh

20:17 <InPhase> I still see it in a new browser...

20:17 <InPhase> We're talking about aaa's logo image, right?

20:17 <teepee> yeah, but only as "normal" image not in the history where it has the delete button

20:17 <InPhase> lol. So it's gone from the history but still displayed as an image?

20:18 <teepee> I think that was a local browser cache thing as I uploaded from the same browser for both users

20:19 <teepee> so lets delete the user and see what happens with the image

20:20 <InPhase> That does not look meritorious as a feature set...

20:20 <teepee> account gone, but marked as deleted and the image remains, so there seems to be no way of doing moderation

20:21 <teepee> we can keep it for a bit for a known round of people but I'll continue searching for something with a better matching feature set

20:23 <InPhase> Yeah.

20:23 <InPhase> Were you able to delete one of your own uploads even?

20:25 <teepee> yep, that works and offers physical deletion as option

20:25 <teepee> confirmed, it's also removed from the image folder

20:26 <InPhase> Okay. So it works okay in a trust environment.

21:46 <teepee> oh, indeed, humble bundle: https://www.humblebundle.com/books/coding-for-curious-no-starch-books

21:46 <teepee> with OpenSCAD book from kintel too

21:48 <teepee> I got it a while ago in the "Linux" bundle :)

22:25 <InPhase> That's an eclectic collection.

22:53 <sparsh-n> https://bpa.st/PTMA

22:53 <sparsh-n> I was testing a simple logo.scad using the headless version and was wondering if anyone else has experienced errors similar to the ones above?

23:02 <teepee> sparsh-n: that means it's not finding the path to the resource, like the shaders it complains about

23:03 <teepee> I guess for this directory structure we may want another case to find those files

23:04 <teepee> right now for windows we have: https://github.com/openscad/openscad/blob/master/src/platform/PlatformUtils.cc#L46-L51

23:05 <teepee> looking at the path shown, it probably needs and additional "../.." in the windows section, like that: https://github.com/openscad/openscad/blob/master/src/platform/PlatformUtils.cc#L58