03:32 <gbruno> [github] kintel closed pull request #5037 (PolySet: Add setters for isTriangular) https://github.com/openscad/openscad/pull/5037

03:32 <gbruno> [github] kintel pushed 12 modifications (PolySet: Add setters for isTriangular (#5037)) https://github.com/openscad/openscad/commit/db167b1df31fbd8a2101cf3a13dac148b0c2165d

04:00 <pca006132> I think as manifold is updated frequently and openscad wants the latest bug fixes, it is probably a good idea to keep using submodules for a while

04:01 <pca006132> teepee: perhaps thrust is a deprecated package?

04:01 <pca006132> distros probably package cccl instead of thrust now

04:46 <gbruno> [github] kintel pushed 5 modifications (Fix test case and cleanup) https://github.com/openscad/openscad/commit/dfc72b2d1bc494f2423214c2d2835ae669374540

04:46 <gbruno> [github] kintel synchronize pull request #5020 (Maintain manifoldness when converting between PolySet and ManifoldGeometry) https://github.com/openscad/openscad/pull/5020

04:46 <gbruno> [github] kintel edited pull request #5020 (Maintain manifoldness when converting between PolySet and ManifoldGeometry) https://github.com/openscad/openscad/pull/5020

05:49 <slice> Hello! I am new to 3d modelling, and found openscad. Been a learning curve but thankful for those that came before me.

05:49 <slice> I found https://github.com/openscad/openscad/pull/4241/commits/0336a020e4ca20ceb2d26689ea5bd787c8e2e00b and then noticed the "downloadables" are from a 2021 release, so it does not have the feature in the binaries.

05:49 <slice> Are there newer official binaries elsewhere, or I ought to build it from master? Is there a way to snatch a binary from circleci at all?

05:58 <crazy_imp> someone point them to https://openscad.org/downloads.html#snapshots (if slice comes back)

05:59 <crazy_imp> slice: https://openscad.org/downloads.html#snapshots

05:59 <slice> ah wonderful

06:00 <slice> also my bad for being blind as a bat :D

06:03 <crazy_imp> maybe it's also not the best idea to have the 2021 release as first items on the download page :D

06:03 <slice> :)

08:48 <pca006132> teepee what do you think about moving the section about development snapshots to the top of the download page?

08:48 <teepee> pca006132: it's really confusing, ubuntu 22.04 supposedly has thrust 1.15.0, but I see only 1.9.5 when building on OBS, and that simply does not have the cmake files packaged

08:49 <pca006132> ah, weird

08:49 <pca006132> is it some channel thing? e.g. dev vs stable or something?

08:50 <teepee> and cmake is annoying aswell, the automatic find_package in fetch-content is only available starting at 3.24 which is very new

08:51 <teepee> not that I know of, normally debian and ubuntu releases are very much fixed just adding bugfixes for security reasons and "normal" updates only for very very very critical packages like firefox

09:01 <teepee> as for the website, I'm this -><- close to rewriting it in svelte now that leared the basics :) it's very annoying to maintain the current state and changing any layout

10:13 <pca006132> yeah, the automatic find_package in fetch-content is meant for simplifying the workflow, it doesn't make sense to keep it if we need to add fallback for older version...

10:31 <pca006132> is return code -11 indicating segfault?

10:31 <pca006132> https://logs.ofborg.org/?attempt_id=8e7bf9d4-2255-43f9-b9d9-ce1bc590d895&key=nixos%2Fnixpkgs.294911

10:32 <pca006132> csgpngtest_module-recursion failed on aarch64-linux...

11:37 <InPhase> pca006132: Yeah. The last one of those I saw on tests was last year on some tests running manifold.

11:38 <InPhase> Although I think that particular one was all resolved.

11:45 <pca006132> it seems that it is some sporadic test failure, probably due to the stack limit

11:45 <pca006132> using pointer address to check for stack size is not a very safe approach...

11:53 <InPhase> I usually raise the stack for personal use of deep recursion, but I thought all of our tests should be way under stack limits. Could manifold be introducing new deeper recursions? I wouldn't really expect this from a tree structure at the sizes of our tests.

12:05 <pca006132> not sure, cannot reproduce locally

12:05 <pca006132> manifold should not have much recursion

18:38 <peeps[work]> i think StackCheck might need to be modified to do some kind of thread_local check

18:41 <peeps[work]> changing the instance itself from static to thread_local?

18:48 <peeps[work]> then again maybe its fine as long as its not called directly from any multithreaded code, like inside manifold lib

18:49 <peeps[work]> I wonder if a stack limit message is being triggered from StackCheck

20:43 <TylerTork> Is there some way I'm not seeing to enter complex configuration parameters with a JSON-like notation in the source code? Or in an external file? I'd like to be able to refer to, for instance, propellor[i].pitch and propellor[i].width as opposed to having a separate named parameter or using a numbered array index for each value

20:52 <TylerTork> So no, then?

20:57 <Scopeuk> there is no support for structure like constructs at this point

20:57 <Scopeuk> only vectors

20:58 <TylerTork> If the lookup function worked with strings, one could use parameter pairs like [["width",5.5], ["depth",2.03]...]

20:59 <kintel> TylerTork There is some discussion, e.g. in https://github.com/openscad/openscad/pull/4478

20:59 <kintel> ..but it takes time to land such changes

21:01 <TylerTork> thanks for the link, will peruse

21:03 <Smeef> Trying to design a simple band ring (h=7mm, r=16.5mm) with shallow notches along the outer side, similar to a gear or a coin. Is there an easy way to this without creating and placing each notch individually?

21:07 <ali1234> use a for loop

21:15 <Smeef> Thanks, that did the trick!

22:57 <teepee> aww, too late, here *is* JSON import

23:07 <gbruno> [github] t-paul pushed 1 modifications (Remove Ubuntu 22.10, add Ubuntu 23.10.) https://github.com/openscad/openscad.github.com/commit/ad0268ab37161e772eeab44f0c6649beccff9743