teepee changed the topic of #openscad to: OpenSCAD - The Programmers Solid 3D CAD Modeller | This channel is logged! | https://openscad.org/advent-calendar-2021/ | 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
<peepsalot> teepee: actually, if you want to just use webbing.scad from that repo, that should be a decent visual
<peepsalot> the default alternative_spheres is too busy
<peepsalot> also, took me a long time to realize those screenshots are actually animated, maybe the delay (looks like 10 or 12s) should be halved?
<gbruno> [github] t-paul pushed 1 additions 1 modifications (Add BOSL2 (beta).). https://github.com/openscad/openscad.github.com/commit/03726c2a0476f0cadff34741ddc3ce5174c4bab4
<teepee> I see only icosahedron_old() but no icosahedron() ?
<teepee> aww, now github is broken too, what's going on today?
ur5us has quit [Ping timeout: 256 seconds]
<gbruno> [github] t-paul pushed 1 modifications (Update NUMCPU.). https://github.com/openscad/openscad/commit/2c2e6dde9ac10062e20deb1ee2ba26db2534de5f
<gbruno> [github] t-paul pushed 1 additions 1 modifications (Add Functional OpenSCAD Library.). https://github.com/openscad/openscad.github.com/commit/e39a547d989854fdfd83bfe536cace8082344ac5
<teepee> aha, github lives! \o/
<peepsalot> the base shapes were all moved into all caps functions, and are passed into generalized sphere_subdiv()
<gbruno> [github] t-paul pushed 1 modifications (Fix build issue with overloaded join().). https://github.com/openscad/openscad/commit/08bf69b4115c989fc5671254e0d05735d01bcca5
J2252938292858 has joined #openscad
<J2252938292858> a shit ..  2 digits more
J22529382928 has quit [Ping timeout: 256 seconds]
J2252938292858 is now known as J22_how_long_can
<InPhase> lol
<teepee> NICKLEN=16
<teepee> I think there's room for 2 more :)
snaked has quit [Remote host closed the connection]
snaked has joined #openscad
snaked has quit [Read error: Connection reset by peer]
snaked has joined #openscad
snaked has quit [Read error: Connection reset by peer]
califax- has joined #openscad
califax has quit [Ping timeout: 276 seconds]
califax- is now known as califax
LordOfBikes has quit [Ping timeout: 240 seconds]
snaked has joined #openscad
LordOfBikes has joined #openscad
<gbruno> [github] Mentors4EDU opened issue #4094 (Feature Request: Elementary OS Support). https://github.com/openscad/openscad/issues/4094
<gbruno> [github] Mentors4EDU closed issue #4094 (Feature Request: Elementary OS Support). https://github.com/openscad/openscad/issues/4094
pie_ has quit [Quit: pie_]
pie_ has joined #openscad
ur5us has joined #openscad
ur5us has quit [Ping timeout: 240 seconds]
ur5us has joined #openscad
ur5us has quit [Ping timeout: 240 seconds]
<J22_how_long_can> added gear examples  ⇒ https://github.com/UBaer21/UB.scad/tree/main/examples/UBexamples     RFC!
<ccox> products link is 404
<ccox> looks like Products.scad hasn't been pushed to the repo.
Alexer has quit [Ping timeout: 256 seconds]
<J22_how_long_can> ccox you are just too fast for me
<J22_how_long_can> just uploaded
<JakeSays> bah. finding usb-c female panel mount sockets is apita
<JakeSays> hmm. maybe i'll go with rj11
Alexer has joined #openscad
lastrodamo has joined #openscad
<ccox> J22_how_long_can: that looks better.
<buZz> 11:56:43 < buZz> wooot, my gf implemented a design i did on paper for her
<buZz> 11:57:16 < buZz> she just finished it :D https://imgur.com/a/HpOANdj
<buZz> 11:57:10 < buZz> so i sketched something for her to openscad and print herself
<buZz> 11:57:00 < buZz> she kept never putting TP on the toilet because her TP mount is so annoying
<buZz> :D
Alexer has quit [Ping timeout: 240 seconds]
luissen has quit [Quit: exited unexpectedly with error code -1]
luissen has joined #openscad
Guest76 has joined #openscad
Guest76 has quit [Client Quit]
Alexer has joined #openscad
<J22_how_long_can> buZz  that is the kind of thinking which 3D-printing  promotes   ( sure you also could have remove the 2 screws and replaced the old one without building a chaotic pendulum)
<JakeSays> buZz: heh. that's cool
<buZz> J22_how_long_can: yeah
<J22_how_long_can> but in the end it is all about problem solving
<buZz> its a quality-of-life hack
<buZz> she hasnt got dexterity to often 'pull that metal apart' , plus the landlord installed em , she would have to safe it forever until she moves to put it back
<buZz> plus she has a 5yo that likes to help but cant do such things
<JakeSays> buZz: i figured it was landlord related
<buZz> she -can- put TP on this print :P
<buZz> JakeSays: its common in dutch houses anyway, to have stuff like this all preinstalled
<buZz> rented houses anyway
<JakeSays> buZz: here too. i have a stack of stuff that i have to re-install when i move
<buZz> hmhm, it takes a ton of organizing imho, i rather just leave stuff like that :P
<J22_how_long_can> isn't there a way to ask the landlord for permission to exchange it for good?
<buZz> but i'm the kinda person that never paints a wall
<buZz> J22_how_long_can: totally , but that takes effort too
<buZz> this print/design just avoids all that effort
<J22_how_long_can> i mean that looks like from the 70s
<buZz> oh yeah likely
<JakeSays> buZz: so your gf did all the openscad?
<J22_how_long_can> and there would be the  https://www.prusaprinters.org/prints/86060-tp-cap
<J22_how_long_can> JakeSays did you notice how sexist that question is?
<JakeSays> J22_how_long_can: it's not at all sexist.
<J22_how_long_can> if she sketched it  and he openscad and printed it -  would you asked " did he did all the openscad?"
<buZz> JakeSays: yeah , she taught herself openscad
<JakeSays> i would not because i know he already knows openscad
<buZz> i sketched it on paper for her, she designed and printed it
<JakeSays> buZz: that's cool
<buZz> yeah she is <3
<J22_how_long_can> i just fear the gender ratio for oSCAD is like 90% male
<JakeSays> while that is unfortunate, don't project your fears on me
<J22_how_long_can> Ü
<buZz> J22_how_long_can: i dont know, never tried to measure
<buZz> at least i know i'll someday try to teach her daughter openscad aswell :P
<J22_how_long_can> buzz it sounds like your gf will already teach her ..  so you can teach her how to cook :-)
<buZz> hehe maybe ;)
<buZz> speaking of which, i need to get milk to try and improve my brioche recipe
<buZz> :D
othx has quit [Ping timeout: 256 seconds]
othx has joined #openscad
<Scopeuk> I have my suspicions that cartoonists could be quite good at openscad, the processes of breaking down an object into constituent primitives to build from is the same. I have no way to test this theory
submariner has quit [Remote host closed the connection]
<JakeSays> bah. the xyz legend thing is hidden when the axis lines are hidden
aiyion has quit [Remote host closed the connection]
<J22_how_long_can> yeah so you can get in image without those ..  but use Pivot(); if you want one visible
aiyion has joined #openscad
<buZz> JakeSays: just design one yourself ? :P
<JakeSays> there should be an option to not hide them
<JakeSays> J22_how_long_can: Pivot()?
<buZz> JakeSays: sure, shouldnt be too hard to implement in the code :) have at it
<JakeSays> buZz: heh. i don't have enough free disk space to build openscad at the moment
<J22_how_long_can> JakeSaxs the Gizmo module from my lib
<buZz> JakeSays: lol wtf
<buZz> i compiled it on a 16GB microSD
<JakeSays> J22_how_long_can: ah
<buZz> incl full OS on that card :P
<buZz> 16GB microSD is basically free nowadays :D
<JakeSays> buZz: i have 2tb total. ~10gb free. lol
* JakeSays needs to do some house cleaning
<buZz> time to dedupe your storage :D
<JakeSays> there's very little duplicating
<buZz> :)
<JakeSays> i'll probably just end up buying another 2tb
<buZz> my last bought 2tb disks were 20 euro i think
<buZz> but 'used' (just ~400 hrs, <10 cycles)
<JakeSays> buZz: i've never seen a m.2 nvme 2tb device anywhere near that cheap
<buZz> oh i thought you said disk :D
<buZz> i wouldnt ever do nvme for bulkstorage tbh
<JakeSays> i like the speed
Guest98 has joined #openscad
Guest98 has quit [Client Quit]
<Scopeuk> bulk + nvme is u2's thing
<Scopeuk> shame the prices are so silly
<Scopeuk> I would love some 15 or so TB slow ish SSD's for my media center
<buZz> i was amazed that microSD dropped so much
<buZz> upgraded my chromebook's rootfs microSD to 128GB for under 20 euro :D
<Scopeuk> yeh once they started using them as generic storage in any old linux + arm box that China is kicking out by the 100k they dropped fast
<buZz> well its mostly just last year? seemingly the chip shortage made ppl produce some chips a lot more/faster :D
<Scopeuk> hmm ok the last year is more recent than that maybe the supply just isn't being eaten as there is nothing to pair them with :P
<buZz> heheh could be :P
<buZz> >128GB doesnt seem to have dropped just as fast
<buZz> but maybe articifially kept higher
<Scopeuk> I guess its also possible something jumped a process and made the small ones cheaper to make
snaked has quit [Ping timeout: 240 seconds]
Alexer has quit [Ping timeout: 256 seconds]
<buZz> perhaps yeah
Alexer has joined #openscad
Virindi has quit [Quit: Leaving.]
<teepee> heh, good start, CGAL agrees with Scopeuk :)
<teepee> now lets cross fingers they have an idea how to fix it :)
<teepee> in totally different news, itch.io is great... https://twitter.com/itchio/status/1490141815294414856
snaked has joined #openscad
<gbruno> [github] t-paul pushed 1 removals (Remove LGTM.). https://github.com/openscad/openscad/commit/118845332568945e2c7175bfd8d9dc7466ff3c50
<Scopeuk> teepee its a start i guess we just hope for a solution or some flags to bully gcc into talking
<teepee> yeah, some extra info would be nice
<teepee> maybe someone could try that with some other compiler
<teepee> seems to be related to some of the windows header files as it does not appear on linux so far
<Scopeuk> My understanding is that windows defines size t differently
<Scopeuk> Linux headers define long long uint as a long uint where as windows has a seperate type
<teepee> right
<teepee> crossing fingers the fast-csg stuff builds everywhere now :)
<gbruno> [github] t-paul closed issue #2360 (Possible Improvement using CGAL "Polygon Mesh Processing" for booleans and more?). https://github.com/openscad/openscad/issues/2360
<gbruno> [github] t-paul pushed 60 additions 16 modifications (Merge pull request #4087 from ochafik/fast_csg_final2
<teepee> aha, I think they found the place in cgal
<gbruno> [github] t-paul pushed 397 modifications (Run uncrustify (via scripts/beautify.sh --all).). https://github.com/openscad/openscad/commit/72eb7d992d6818f653fd729663e90a8c24bedab1
<Scopeuk> "what changed everything and nothing"
<Scopeuk> the uncrustify stuff has been a long time comming
<teepee> well, always lots of discussion
<teepee> I'm thinking we just have to do it eventually
<teepee> ¯\_(ツ)_/¯
<Scopeuk> yeh
<Scopeuk> the only "reason" not to is the impact on working branches, but pulling main to branch and then running over the files you've modified should fix that right up
<teepee> yep, that's what I'm thinking too
<teepee> there's probably always some corner cases but those are hopefully never huge after running the same format on the local stuff
<Scopeuk> and they should only affect current active dev branches
<Scopeuk> its a hit that has to be taken at some point
ur5us has joined #openscad
<teepee> I guess I'll wait for any last words on the formatting, while browsing through the changes for a hopefully last time :)
<peepsalot> indentation changed from 2 to 8 spaces?
<teepee> just tabs, use what you like :)
<peepsalot> so, there's still a mix of tabs and spaces?
<teepee> so far that's the best argument I've seen
<peepsalot> hmm, i don't get how it decides when to use spaces then?
<teepee> tabs up to the brace indent, so it moves blocks
<teepee> alignment inside that block by space, so it moves only relative to the braces
<peepsalot> ok i think i see now
<gbruno> [github] t-paul opened issue #4096 (SVG scaling wrong with width/height given as percentage). https://github.com/openscad/openscad/issues/4096
<J22_how_long_can> the big arrows are still a bit annoying
<teepee> what arrows? in the scad editor?
<teepee> peepsalot: do you think that's not a good idea?
<peepsalot> i'm fine with it. as long as there's some consistent rule. i was just confused at first by the change
<teepee> I tended to say, just use spaces and be done with it
<teepee> but the accessibility argument seems to be a good one, and for that matter there are at least vision impaired people using openscad
<peepsalot> is it going to still be up to user's responsibility to apply formatting, or is there any way we can force it on PRs etc with a github action?
<teepee> for now, I guess user, but we can try to automate the check
<teepee> I've dropped LGTM, so we have plenty of options to make the checks slower again :)
<J22_how_long_can> teepee yes in the scad editor
<teepee> J22_how_long_can: disable them?
<teepee> or make them thinner?
<J22_how_long_can> thinner as size 1?  ..  and  i still want to see whitespaces
<teepee> well, thinner as 1 seems unlikely, to make it thinner in that case, it would need a higher resolution monitor I guess
<teepee> or a feature request with QScintilla to make it configurable or provide more options
<teepee> all that visual stuff is always difficult as different people have different preferences, like I continue to hate webites switching to all-dark themes
<peepsalot> i never messed with those settings in editor before, but it is maybe slightly strange that spaces display size is configurable, but has not effect on tab arrow display
<J22_how_long_can> hmm size didn't change the arrows .. but they are  font dependent
<J22_how_long_can> and again i am using deja vu mono  which unknowingly has one of  the biggest arrows
<teepee> oh? it's only changing the space dot?
<peepsalot> i think maybe more useful would be an opacity setting, looks 100%, while i feel like other editors use 0.5 or something
<teepee> I have it only enabled for trailing whitespaces so I'm not seeing too much of the markers
<J22_how_long_can> are these configurable via color scheme ?  or tabs and space dots treated same as characters
<teepee> I'm not sure it's possible right now, if possible at all, would need to be checked with the qscintilla code
<teepee> and it ideed only changes the space do but not the arrow
snaked has quit [Ping timeout: 256 seconds]
<teepee> if there's maybe more options available by now, we can add those of course
<gbruno> [github] t-paul pushed 45 modifications (Align catch formatting with else.). https://github.com/openscad/openscad/commit/082a279f3ffbbe853ff1f03c7dc22ea8578f7704
<gbruno> [github] t-paul pushed 1 modifications (Run CodeQL only when changing source code files.). https://github.com/openscad/openscad/commit/3959a14902fc7ad4bc1a6f5d79b69140e77e134c
<teepee> grr
snaked has joined #openscad
<teepee> yaml is annoying :(
<gbruno> [github] t-paul pushed 1 modifications (Run CodeQL only when changing source code files.). https://github.com/openscad/openscad/commit/4c9d3501718200366592171ad7e4db9e7abe9b6a
<teepee> there's QsciScintilla::setWhitespaceForegroundColor
<teepee> so it might be possible to put this into the color scheme
<teepee> maybe even with transparency
<gbruno> [github] t-paul pushed 9 additions 3 modifications 16 removals (Merge pull request #108 from Rotzbua/refactor_cheatsheet
ur5us has quit [Ping timeout: 250 seconds]
<J22_how_long_can> teepee sounds nice ..    really appreciate all your hard work  ..  .oO should say this more often
<teepee> thanks and good point :) thank you (and all the other people sharing code) for the library. that's also an important part of making openscad easily usable
<InPhase> teepee: Ah, good. I had been gradually changing handfuls of files as I worked with them. But it will be nice to have it all standardized. :)
<InPhase> Wait... Did this tab things instead of 4-space them?
<InPhase> I thought the plan was to beautify over to standardized spacing.
<teepee> it's standard with a way to let people configure the tab size
<teepee> because https://alexandersandberg.com/tabs-for-accessibility/ sounds like very reasonable arguments
<teepee> well actually one of the rare cases giving actual *reasons* instead of "I like that better"
<J22_how_long_can> so now we also get a "convert spaces into tabs" option  -  Ü  ( i mean should be possible for the leading spaces)
<teepee> J22_how_long_can: no, this is about the source code of openscad itself, nothing to do with the editor from user side
<J22_how_long_can> sure but "standardizing" means to have this everywhere or?
<teepee> yes, but everywhere in the openscad c++ source code
<teepee> not everywhere as in all scad files or all text files in the world :)
<InPhase> teepee: That same reddit post has a blind person posting and saying they greatly prefer spaces over tabs because it works better with screen readers.
<teepee> so limited everywhere if that exists
<InPhase> teepee: Accessibility seems to be a mixed bag with any approach.
<teepee> yes, it certainly is
<J22_how_long_can> .oO time to let machines write the code without interfering of humans
<teepee> but then, it's not like we can't change it next year slightly :)
<teepee> InPhase: what's your view, go with that for now and review after a while?
<InPhase> Well, I can't edit the code without reconfiguring my editor to special-case this. Which, I can look up how to do. But that also means other people are going to have trouble with all those editors that don't put out tabs for tab presses.
<InPhase> I literally work on zero other codebases that are tab based.
<teepee> why would that need reconfiguring
<InPhase> Well because pressing tab puts spaces.
<teepee> other than "allow tabs"
<teepee> well, honestly I don't mind too much, and it seems peepsalot is on a similar page
<teepee> so if you give me a number for the space I can just revert back to that and we start with that setting :)
<InPhase> Also I have to figure out if I want to keep these tab markers enabled that I use just to warn me when someone leaked tabs into a file and it's messing up alignment: https://i.imgur.com/osjOuuK.png
<teepee> so what would be the magic number of spaces?
<InPhase> I usually use 2, but I had 4 configured for OpenSCAD because we had standardized on that.
<teepee> 2 was actually the value in the older uncrustify config :)
<teepee> I don't mind putting that back and mention that accessibilty article as "considered, when suggesting to go with that please file an issue"
<InPhase> Like such: https://bpa.st/UN4Q
<InPhase> So, proof that I can adapt. :) But it might cause some issues out there with contributers.
<InPhase> I think this plot shows the proportion of people that are setup for each. https://www.7pace.com/blog/programming-spaces-vs-tabs
<InPhase> In C++ github code, tab usage is almost non-existent.
<teepee> which does not say much in regard if it's a good idea for accessibility
<teepee> for gov stuff I think germany has a rule to be accessible, solution is that all files just add a note "file not compatible with accessibility"
<InPhase> Correct, and a stipulated point. Although it would be worth having some survey numbers on how that actually impacts accessibility on average for people who need it.
<teepee> true, which is why I'd be fine to put that comment for speaking up into the config file and use spaces as originally configured
<teepee> so what's the vote? peepsalot InPhase ?
<teepee> I'm ok with both :)
<teepee> but I do want to merge *something*
<teepee> there's no point in talking another 5 years ;-)
<ccox> you could merge more bug fix pull requests ;-)
<teepee> ccox: there's a chance for that too, I assume you mean the context fix?
<InPhase> I've googled for tabs, spaces, and accessibility, and really all I'm finding about this is that one article, which consists of one person sharing anecdotes about two people with vision impairment preferring tabs configured in two ways, and one reddit poster preferring spaces for vision impairment.
<InPhase> Mostly I'm seeing almost nobody talking about this as an accessibility issue before that post, and everyone talking about it shortly after seems to have referred to that post.
<ccox> teepee: yep, that's the last one I have open.
<teepee> right, still the *only* one giving at least some reason
<teepee> all the other talk about that is basically vi vs. emacs
<teepee> InPhase: so is your vote for 2 or for 4 spaces?
lastrodamo has quit [Quit: Leaving]
<peepsalot> I guess I generally prefer 2 spaces, but I'd rather just move forward with whatever is least contentious, than argue over it.
<InPhase> teepee: I like 2 better, but I acknowledge 4 seems more popular on average, so I can work with 4 without complaining. :)
<teepee> looks like it's 2 for todays PR :D
<InPhase> As long as we all agree 3 is wrong. It's a compromise, but it's just odd.
snaked has quit [Ping timeout: 240 seconds]
<ccox> <-- finds the least friction across teams and systems with 4 spaces, most friction with 8 spaces, and catches bloody heck from blind researchers when using tabs.
<InPhase> ccox: So you've also experienced accessibility complaints about tabs? Is it that mentioned screenreader issue?
<ccox> InPhase: yes. I did eventually help one Adobe/IBM researcher fix his preprocessing scripts to remove tabs so the screenreader went faster.
<InPhase> Good to know.
<ccox> (though that meant that he and his guide dog visited my office less often -- loved his sense of humor, and the dog)
<ccox> (And apparently he's left IBM and gone to Google now)
<teepee> yes, I might have vetoed 3 :)
<gbruno> [github] t-paul pushed 1 modifications (Run CodeQL only when changing source code files.). https://github.com/openscad/openscad/commit/4fbccae6228bc58f3455a47880f6e26920607f59
<teepee> I know some people use it and it's horrible
<InPhase> I wish we could all agree on 3, but even I can't agree on 3, because it's just not in me as a programmer. So 2 and 4 is what we're left with. :)
<teepee> I always liked that java came with an official formatting guide
<teepee> there's enough stuff to fight about still, like people still insisting on 80 char limits and putting empty // comments at the end of wrapped lines because the eclipse autoformatter is just plain stupid
snaked has joined #openscad
snaked has quit [Ping timeout: 256 seconds]
<teepee> heh, that quite some more files changed with the 2 spaces, but lets just give git something to do :)
zauberfisch has quit []
zauberfisch has joined #openscad