00:21 <gbruno> [github] kintel edited issue #4825 (Move manifold out of experimental) https://github.com/openscad/openscad/issues/4825

04:20 <ypy> https://www.reddit.com/r/BambuLab/comments/1c1d1x7/embracing_openscad_community_introducing_the/

04:21 <ypy> a new openscad playground

05:17 <InPhase> Nice. It seems they went public with it.

05:18 <InPhase> "An interactive interface on top of openscad would be epic"

05:19 <InPhase> That reminds me... Granted, reminds me right before bed. But I was planning a make-openscad-pretty PR to change our default color-scheme and default $fa and $fs.

05:20 <InPhase> I do think small things like that would really help give people a better first impression. That's really a large part of what bambu did there to impress people. :)

05:20 <InPhase> Although I have even better colors available than they used.

05:31 <pca006132> curious which version they are using

05:51 <crazy_imp> is there a variable with the openscad version that can be access and used in the model?

05:53 <InPhase> pca006132: Well it runs way too fast to be anything other than manifold.

05:54 <InPhase> pca006132: I tested some classic stuff that takes annoyingly long in cgal, but is instant in manifold, and it generates it instantly.

05:54 <InPhase> pca006132: Therefore, they must have grabbed a nightly.

05:54 <InPhase> Or I mean, up-to-date build, since wasm.

06:25 <J24k87> hm maybe something for the "news" https://openscad.org/news.html

06:26 <J24k87> regarding https://github.com/openscad/openscad/issues/5091  would it make sense to make "language reference" a own sub page (lemma)

06:29 <J24k87> crazy_imp version()?  https://en.wikibooks.org/wiki/OpenSCAD_User_Manual/Other_Language_Features#OpenSCAD_Version

06:39 <crazy_imp> J24k87: nice! pca006132 can you feed it a scad file with the `version()` command inside?

09:45 <pca006132> 2024 3 22

09:46 <pca006132> I hope they will try to keep it up-to-date

13:34 <guso78k> https://github.com/openscad/openscad/blob/master/src/core/parsersettings.cc#L110

13:35 <guso78k> I am getting regular assert crashes here when I pan/move/rotate the model. is it a known issue ? maybe the file is temporarily not available  ?

13:48 <InPhase> That code comment sounds sketchy.

13:48 <InPhase> "We're also not interested in the file's name." So let's crash if it doesn't have one?

13:49 <InPhase> git blames teepee.

13:51 <InPhase> git falsely blames teepee because teepee fixed formatting.

13:52 <guso78k> must be an old relic ...

13:54 <InPhase> https://github.com/openscad/openscad/pull/3276 2020, from here.

13:55 <InPhase> Probably just a comment error that slipped in amidst 1000 lines of changes.

13:56 <InPhase> 2 reviewers, but it was a hefty change. It happens. :)

13:57 <InPhase> So at the very least, that line doesn't do what it says it's doing. The question is, is it doing something wrong?

13:58 <InPhase> That it involves the mouse selector probably explains why something in parser settings is getting called during view manipulation.

13:59 <InPhase> Does it happen when you drag, or when you first click? (If you have noticed the difference.)

13:59 <InPhase> At right click, correct?

14:00 <guso78k> i cant recall but i remember that i was panning rotating

14:00 <guso78k> i can try to prone for the error

14:01 <guso78k> shouldn't be too hard

14:02 <InPhase> I bet it has to do with which part of the model you right clicked on. :)

14:02 <InPhase> It's triggering in MainWindow.cc's line 2396 where get_library_for_path is called, presumably with a location.filePath that is problematic in some manner.

14:04 <InPhase> You must have some portion of your code where the path is set as an empty string?

14:04 <InPhase> Is this merging in python bits?

14:06 <guso78k> yes, i am running my python branch ...  and right now i cannot get the error again i was getting it 3x within 3 minutes

14:06 <InPhase> It seems an appropriate guard would only run that library hunt routine in MainWindow's rightClick if location.filePath is non-empty

14:06 <InPhase> It might have been a proper right click, rather than a right click with drag.

14:07 <InPhase> Perhaps you occasionally and briefly accidentally right click for the right click context menu.

14:07 <guso78k> i have it now

14:08 <guso78k> right click once properly into the model

14:10 <InPhase> Good. Reproducible means fixable.

14:10 <InPhase> Perhaps you can diagnose first why the path is empty.

14:11 <InPhase> Is it perhaps that you python generate something and it has no path set because that makes sense? :)

14:11 <guso78k> yes,  i will offer some "improvement" hints

14:12 <InPhase> If so, you might be the only person triggering this error. But you could still reasonably put a guard in that rightClick function for everyone, since it's invalid to call that function if the path is empty. And we have no benefit to crashing out in such a scenario. It'd be better for the right click menu to quietly not populate.

14:13 <InPhase> Perhaps even something like: fs::path libpath = location.filePath() ? get_library_for_path(location.filePath()) : {};

14:15 <InPhase> I guess you might need location.filePath().empty() ? {} : get_library_for_path(location.filePath())

14:36 <guso78k> yes, your fix works in a derived way . suppose this is my bug only as everybody else has filenames filled in properly :')

14:37 <guso78k> https://imgur.com/a/FSe73c9 and it even knows the toplevel node correctly

14:51 <InPhase> lol. Is the smiley face on purpose?

15:06 <guso78k> nope. suppose before the colon should be the filename which is inadvertently empty. but i take the smiley as a positive indicator

15:07 <guso78k> indicator->sign

16:37 <JordanBrown> WRT #5090 "can't add .scad files to issues"... JamesC1's reluctance to click on a link that starts a download is a good example of why we want to maintain the "no write operations" principle... it would not be surprising if somebody had a configuration where downloading a .SCAD file started OpenSCAD, and where OpenSCAD was set to preview on load.

17:20 <J24k> wait ill open my text files with open scad scintilla

17:53 <InPhase> JordanBrown: Yeah.

21:46 <JordanBrown1> J24k indeed, the particular file in question wouldn't be a problem, because it's not a .scad file.

22:05 <JordanBrown1> The complaint about searching the language reference has made me take a deeper look at the docs, and one low-hanging fruit that seems appropriate to prune is the "Command Glossary", which should be pruned in favor of the cheat sheet.

22:07 <linext_> can opening an openscad .scad file be hazardous?

22:15 <JordanBrown1> Not currently. But people have asked for various features that would allow writing to the local file system, and that *could* be hazardous.

22:17 <JordanBrown1> It's also why the Python-OpenSCAD work, despite being very cool, is terrifying.