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
marcus has quit [Remote host closed the connection]
marcus has joined #openscad
kintel has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
RoyK has quit [Ping timeout: 248 seconds]
RoyK has joined #openscad
<ali1234> kintel's example crashes on the git master build i just did
<ali1234> i mean it segfaults
<ali1234> in Clipper2Lib::GetRealOutRec
<ali1234> however it is not reproducible with OpenSCAD-2025.02.28.ai23859-x86_64.AppImage
<ali1234> this code is just walking a linked list
<ali1234> while (outrec && !outrec->pts) outrec = outrec->owner;
<ali1234> how can that ever segfault??
stealth_ has quit [Quit: Leaving]
stealth_ has joined #openscad
J25k31 has joined #openscad
J25k93 has quit [Ping timeout: 240 seconds]
<InPhase> ali1234: You should compare your libraries between the build that crashes and the build that doesn't, and share this with kintel.
<ali1234> i don't really know how to do that
<InPhase> ali1234: Help, Library Info.
<ali1234> oh, okay
<ali1234> here is the diff between them: https://bpa.st/SYRQ
<ali1234> i'll open a bug for this
<InPhase> My no-segfault lib info: https://bpa.st/PMSQ
<InPhase> Hmm, OpenCSG is different.
<InPhase> But your crash is 1.5 and no crash is 1.7, while my no crash is 1.4.2. So not that, unless it's a bug in just 1.5
<InPhase> Let me try the matched AppImage.
califax has quit [Remote host closed the connection]
snaked has joined #openscad
snaked has quit [Remote host closed the connection]
califax has joined #openscad
snaked has joined #openscad
<gbruno> [github] ali1234 opened issue #5718 (Projection segfault) https://github.com/openscad/openscad/issues/5718
<gbruno> [github] amatulic opened issue #5719 (Allow for range notation in arrays and strings to extract sub-arrays and substrings) https://github.com/openscad/openscad/issues/5719
<gbruno> [github] amatulic edited issue #5719 (Allow for range notation in arrays and strings to extract sub-arrays and substrings) https://github.com/openscad/openscad/issues/5719
<ali1234> InPhase: if it is really memory corruption it could be that merely doing a debug build is enough to make it crash reliably
<gbruno> [github] amatulic edited issue #5719 (Allow for range notation in arrays and strings to extract sub-arrays and substrings) https://github.com/openscad/openscad/issues/5719
<ali1234> and also i have seen a segfault happen with appimage at least once, so maybe it is just harder to trigger
<gbruno> [github] ali1234 edited issue #5718 (Projection segfault) https://github.com/openscad/openscad/issues/5718
<InPhase> ali1234: I forgot the space heater was still running, but the AppImage did not segfault after like 40 minutes.
<InPhase> ali1234: Wait. Your AppImage was the one that did NOT crash. I misread your diff.
<ali1234> yes, correct
<InPhase> ali1234: So you were doing a debug build?
<InPhase> I always do a release build.
<ali1234> yes, only the debug build i did myself crashes
<ali1234> well i wanted to valgrind and gdb it
<ali1234> i will try a release build
<InPhase> That would be a good comparison.
<InPhase> There's a good chance kintel was doing a debug build too. We should find out.
<InPhase> That'd be a handy inverse of the usual Heisenbug. Usually it's release that fails and debug that works. :)
<ali1234> seems like it doesn't crash with a release build
<InPhase> Well then... That's information.
<InPhase> kintel: ^
<gbruno> [github] ali1234 edited issue #5718 (Projection segfault in debug build) https://github.com/openscad/openscad/issues/5718
<InPhase> I will try debug.
<gbruno> [github] ali1234 edited issue #5718 (Projection segfault in debug build) https://github.com/openscad/openscad/issues/5718
<InPhase> ali1234: Hey! Debug crashed. Finally I got a segfault
<InPhase> Switching back to release now. Clearly debug is hazardous. ;)
<gbruno> [github] amatulic edited issue #5719 (Allow for range notation in arrays and strings to extract sub-arrays and substrings) https://github.com/openscad/openscad/issues/5719
<gbruno> [github] amatulic edited issue #5719 (Allow for range notation in arrays and strings to extract sub-arrays and substrings) https://github.com/openscad/openscad/issues/5719
<church_> it's sometimes unpleasant, when rotate_extrude chooses $fs from outer rad. Especially if there is big radius difference between inner. It results in relatively too rough inner. Sometimes i'd wish for there to be param to rotate_extrude, by what to autochoose $f vars
<gbruno> [github] amatulic edited issue #5719 (Allow for range notation in arrays and strings to extract sub-arrays and substrings) https://github.com/openscad/openscad/issues/5719
<gbruno> [github] amatulic edited issue #5719 (Allow for range notation in arrays and strings to extract sub-arrays and substrings) https://github.com/openscad/openscad/issues/5719
<church_> technically i can manipulate with $fn, but hoped to get rid of that & using global $fs/$fa only
<gbruno> [github] amatulic edited issue #5719 (Allow for range notation in arrays and strings to extract sub-arrays and substrings) https://github.com/openscad/openscad/issues/5719
<church_> hmm .. though then again, simplest probably would be something like $fs=$fs/2 within rotate_extrude and forget about it
<gbruno> [github] amatulic edited issue #5719 (Allow for range notation in arrays and strings to extract sub-arrays and substrings) https://github.com/openscad/openscad/issues/5719
joso has joined #openscad
kintel has joined #openscad
<kintel> For me it crashes with debug and release builds, including the last dev snapshot
mmu_man has quit [Ping timeout: 252 seconds]
L29Ah has quit [Ping timeout: 265 seconds]
<gbruno> [github] kintel pushed 2 modifications (Remove PolySet functionality from CGALRenderer) https://github.com/openscad/openscad/commit/4f6878d798079b8181ec6e6d7d5c01a8784d6a96
<gbruno> [github] kintel synchronize pull request #5530 (Introduce PolySetRenderer and add support for wireframe overlay) https://github.com/openscad/openscad/pull/5530
tcurdt has quit [Quit: fBNC - https://bnc4free.com]
tcurdt has joined #openscad
fling has quit [Remote host closed the connection]
fling has joined #openscad
kintel has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
guso78k has joined #openscad
GNUmoon has quit [Remote host closed the connection]
GNUmoon has joined #openscad
califax has quit [Remote host closed the connection]
califax has joined #openscad
califax has quit [Remote host closed the connection]
califax has joined #openscad
kurisu123 has joined #openscad
kurisu123 has quit [Ping timeout: 252 seconds]
stealth_ has quit [Quit: Leaving]
misterfish has joined #openscad
misterfish has quit [Ping timeout: 244 seconds]
misterfish has joined #openscad
mmu_man has joined #openscad
bertronika has joined #openscad
L29Ah has joined #openscad
kurisu123 has joined #openscad
misterfish has quit [Ping timeout: 252 seconds]
ccox_ has quit [Ping timeout: 246 seconds]
ccox has joined #openscad
<gbruno> [github] gsohler synchronize pull request #5717 (Adding Handles(matrices) to objects) https://github.com/openscad/openscad/pull/5717
misterfish has joined #openscad
kurisu123 has quit [Ping timeout: 248 seconds]
kurisu123 has joined #openscad
mmu_man has quit [Ping timeout: 252 seconds]
mmu_man has joined #openscad
kurisu123 has quit [Remote host closed the connection]
snaked has quit [Quit: Leaving]
teepee_ has joined #openscad
teepee has quit [Ping timeout: 264 seconds]
teepee_ is now known as teepee
guso78k has quit [Quit: Client closed]
stealth_ has joined #openscad
guso78k has joined #openscad
mmu_man has quit [Ping timeout: 252 seconds]
mmu_man has joined #openscad
AzAv8r has joined #openscad
bertronika has quit [Quit: Konversation terminated!]
<AzAv8r> question on the color transformation:  I know that it only works on preview, but I understood a color applied to the objects in the scope contained after the color statement.  I have a model with two objects (modules) which nest.  To view them nested, I applied red to object 1, green to object 2.  Only one surface of object 2 is green - those
<AzAv8r> surfaces which are coplanar (but separated by 0.2 mm) are red.  Is this a bug, or is this just a manifestation on limits of using color?  ->  code is:   color("green",1) jig2(); color("red", 1) jig1();   jig1() and jig2() are basically concentric tapered ring shells each about 3mm thick.  jig1 has two ears which fit into corresponding slots
<AzAv8r> on the top of jig2, resulting in the upper (and lower) surfaces being coplanar.
<AzAv8r> Build is the current Ubuntu Nightly 2025.02.25.snap
<teepee> it does work on meshes now with the Manifold backend
<teepee> coplanar surfaces are never good, regeardless of color rendering or mesh calculation
<teepee> well, coplanar surface = random outcome
<AzAv8r> Translating jig1 [0,0,-0.1] was sufficient to create normal behavior.  Thanks!   This was using the manifold backend (which is FABULOUS).  The main model in this project was taking about a second for preview, and about 40 seconds to render.  WIth manifold both operations take under a second.
<AzAv8r> And very nice that it works after rendering now.  That will be useful for placards.
mmu_man has quit [Ping timeout: 268 seconds]
<teepee> yep, it's pretty recent, and so far only 3mf and off support exporting the colors
<teepee> it will take a while to catch up with everything
<teepee> for the overlap I usually have a variable eps that is normally set to 0.01
<teepee> it's annoying but so far there's no common/automatic way of solving the issue. some libraries include dedicated overlap in places where you are supposed to attach things
TheCoffeMaker has joined #openscad
<guso78k> teepee, i think you should not even "union" the two objects, but rather remove the 2 "butting" faces and connect the related vertices instead
Guest20 has joined #openscad
<stealth_> can you test the python version of openscad yet?
mmu_man has joined #openscad
Guest20 has quit [Quit: Client closed]
Guest20 has joined #openscad
<guso78k> or even better: copy the exact vertex coorindates from faceA to faceB before unioning. this will ake sure that there is a ZERO gap in between and even can avoid potentional self-intersection
Guest20 is now known as xaxtric_7
<teepee> stealth_: yes, for the official version you need Linux for now, or you can try the version guso78k provides
<guso78k> teepee, any schedule for next "time slot" and "interest"  to look into, why my  ctests fail
<guso78k> just turned my anycubic printer into klipper resuing the existing display! :b :b
<teepee> the last I saw was just the fn/fs issue that shoulf be fixed now
<teepee> do you have a link to the specific run you mean?
<guso78k> yes, fn works fine now!
<stealth_> teepee, can i just download the latest git repo and it should work?
<stealth_> OpenSCAD version 2025.02.26
<teepee> stealth_: yes, AppImage, Snap, Flatpak and the OBS repos are all updated. Flatpak may lag behind a bit as the builds are not fully automatic
<teepee> or did youu build yourself? then you need to add the flags to enable python
<stealth_> i just built it, OpenSCAD version 2025.03.01
<stealth_> brb
<guso78k> i believe, key to success is beeing able to rerun the test with output enabled.
<teepee> stealth_: you need -DENABLE_PYTHON=ON then with cmake
<xaxtric_7> HI everyone i am new to this community and got to know about this because of gsoc25 so from where should i start i am also new to this gsoc thing
<teepee> guso78k: it has everything enabled, it does output to that download html in the summary section
<guso78k> ok thank you, let me search harder
<teepee> xaxtric_7: did you look through the project proposals for openscad already? anything you are interested in specifically?
<teepee> in general the best 2 thing to start is 1) getting the source code and getting this build and even more important 2) communication! :)
<teepee> -> +AttributeError: module '__main__' has no attribute 'fs'
<teepee> that branch seems to be before the fn/fs fix
<guso78k> yes, i found now myself.
<guso78k> i am able to proceed. thx!
xaxtric_7 has quit [Quit: Client closed]
stealth_ has quit [Remote host closed the connection]
bertronika has joined #openscad
<gbruno> [github] gsohler synchronize pull request #5713 (added missing osimport function) https://github.com/openscad/openscad/pull/5713
<gbruno> [github] scrameta synchronize pull request #5584 (RFC: New general-purpose extrusion mechanism - 2025 refresh) https://github.com/openscad/openscad/pull/5584
<gbruno> [github] gsohler synchronize pull request #5713 (added missing osimport function) https://github.com/openscad/openscad/pull/5713
stealth_ has joined #openscad
teepee_ has joined #openscad
teepee has quit [Ping timeout: 264 seconds]
teepee_ is now known as teepee
mmu_man has quit [Ping timeout: 246 seconds]
<stealth_> teepee, ok built it with -DENABLE_PYTHON=ON
<stealth_> teepee, how to use/test it?
<guso78k> did it actually enable python ? (nettle is requirement)
<teepee> good point
<stealth_> i already have nettle
<teepee> but you can try the mini example shown here: https://fosstodon.org/@OpenSCAD/114067331130565443
<teepee> save as box.py
<teepee> open and see if you get the box
<guso78k> ahmm, enable python in the preferences
<stealth_> "Parser error: syntax error"
<stealth_> o
<stealth_> i don't see any python option in preferences
<stealth_> maybe it didn't compile with python enabled, let me try again
<guso78k> its in Features: python-engine
<stealth_> even though i am adding -DENABLE_PYTHON=ON into cmake the package manager isn't saving it.
<stealth_> shows af it is but its not.
<stealth_> *as
<teepee> did you install nettle-dev?
<teepee> it says in the cmake logs if it's compiling the support
<guso78k> after cmake there should be a message if python is enabled or disabled
<teepee> -- Found Python: /opt/hostedtoolcache/Python/3.13.2/x64/bin/python3.13 (found version "3.13.2") found components: Interpreter Development Development.Module Development.Embed
<teepee> -- Found Nettle: /usr/lib/x86_64-linux-gnu/libnettle.so (found suitable version "3.7", minimum required is "3.4")
<teepee> -- Python enabled, using Nettle 3.7
<teepee> that's from the CI build
<stealth_> its the package manager it wont let me modify cmake, even if i do it ignores the changes and does its own thing.
<stealth_> is there a way to install this using pip or into python?
<guso78k> pythonscad has it, but it  its somewhere between alpha and beta
<stealth_> i see
<guso78k> teepee. the pr with the ctests looks promising
<teepee> stealth_: can you use AppImage/Snap or Flatpak=
<stealth_> teepee, looks like i can use appimage OpenSCAD version 2025.02.28.ai23859 (git ce6161cc0)
<stealth_> at least there is no error using python syntax anymore.
<stealth_> this is totally awesome :D
<gbruno> [github] kintel pushed 5 modifications (Fix a few clang-tidy issues) https://github.com/openscad/openscad/commit/8d9995c76502b5b6a1266c2b4a009abc8ba7d21b
<stealth_> note: python syntax coloring needs to be updated.
<gbruno> [github] kintel opened pull request #5720 (Fix a few clang-tidy issues) https://github.com/openscad/openscad/pull/5720
<guso78k> stelths, its there, but no PR yet
<stealth_> aww... good job guso78k nicely done.
<guso78k> stealth_ , i dont want to decorate myself with others'feathers.  Syntax coloring came from helpers  '=D '=D
mmu_man has joined #openscad
<stealth_> guso78k, no worries, I was talking about adding python to openscad, this really does help me out, I was dreading working with other syntax. I do have a large project I want to focus on so its perfect timing ;) ;)
<stealth_> nice, you can even do `print('hello')` this will help with debugging.
<guso78k> yes of course, redirecting output to openscad log window was quite some challenge ;)
<guso78k> i even plan to add an input line below openscad log to feed in input to python, but not very mature
<stealth_> this is like a game changer, you can import all of pythons stdlib? that is huge!!!
<stealth_> its not a bad start.
<guso78k> stealth, be careful with import of libs. if they have shared objects, they only work, when openscad itself has pip; ed ...
<guso78k> i got some examples of possibilities in pythonscad.org, maybe some examples can be derived at some point in time ...
<stealth_> i wont go too crazy with it, just being able to use itertools and such will be handy https://docs.python.org/3/library/itertools.html
<guso78k> let me know if it works.
<guso78k> initially things like range() did not work until i realized, that i had to import "builtins" to accomplish std python ;)
<teepee> the appimage should work with the default shipped stuff, for more, we need a venv
<teepee> I'm trying that currently
<stealth_> funny you say that, i was just trying to do for loop with range.
<guso78k> range() should work. does it ?
<ali1234> i'll see if i can make a venv
<stealth_> `for i in range(10): print(i)` does give me output
<ali1234> this is python 3.10...
<stealth_> I have to learn a bit more about how pythonscad syntax actually works before playing around more with it.
<stealth_> I did bookmark the websites so will check it out later :)
<stealth_> hmm.. doesn't openscad use the default python installed on your system?
<ali1234> no
<ali1234> and that isn't really viable either
<stealth_> or does it have its own python that comes with it?
<teepee> not possible unless you use the distro install
<ali1234> yes
<guso78k> the AppImage contains openscad, libpython and python's stdlib
<guso78k> if you compile yourself, your actual system's  python is beeing used
<stealth_> it might be worth using the new python since you are shipping your own version of python anyways.
<teepee> not possible right now
<stealth_> o i c, i have to wait for whoever does the compiling to fix that repo to include python version.
<stealth_> python 3.10 reacher end of life next year.
<teepee> no, it need a very different build setup
<guso78k> if you compile from source, you can choose *any* python version
<teepee> choose within reason depending on available libraries and compilers
<guso78k> not python2.7 , however
<stealth_> I am just saying, i am not really worried that much about python version. Just have python stdlib is a huge thing for me.
<stealth_> glad its not python 2 :D
<teepee> yeah, going back would not make much sense
<teepee> AppImage is built against Ubuntu 22.04 now
<teepee> so at some point we can use any later python that will build on that
<guso78k> but what decided that python3.10 was chosen ?
<teepee> right now it's fixed on what is installed by that Ubuntu
mmu_man has quit [Ping timeout: 246 seconds]
<stealth_> whats the point of adding venv ?
<teepee> looks like we may need to explicitely add the security repo
<teepee> with venv you can then install *any* package from pip that is available for that python version
<ali1234> the point of adding a venv is to be able to use libraries like numpy
<stealth_> if you are shipping your own version of python venv isn't needed but if you are using systems python version sure, if best to keep things separated.
<ali1234> venv is always needed
<teepee> it's impossible to ship the system version
<ali1234> you can't install things inside an appimage
<ali1234> except when building the appimage
<stealth_> Didn't you say if you compile openscad will use the systems python?
<guso78k> yes, i said so
<teepee> stealth_: * except the one case when using distro packages
<stealth_> you can just include Python.h and compile python with OpenScad.
<guso78k> but those openscad is not very portable.
<stealth_> If it was me, I would just include python with OpenScad and do not bother with the OS python and wont use venv either.
<guso78k> you can just use it yourself
<teepee> sorry, but that makes no sense
<teepee> if you build yourself ignoring the system python, you have exaclty 2 options
<teepee> 1) don't use anything other than the builtin modules from python (stdlib or whatever that is called)
<teepee> 2) use a venv to place the packages, e.g. installed via pip
<ali1234> 3) install any extra libraries you want at build-time
<stealth_> I normally just compile my own version of python into /opt and use that, I don't use Python that comes with OS, since its really for OS and other software to use.
<ali1234> 4) don't build an immutable style package and install whatever you want in the package's system env
<stealth_> this way I don't have to mess with venv as well, since those have cause many bugs and just wastes time.
<teepee> 3 is no option for the official builds
<ali1234> neither is 4
<teepee> there is always hacks that may or may not work
<teepee> I'm not counting those as options
<ali1234> python is nothing but hacks... sorry :(
<ali1234> it's hacks all the way down
<teepee> once we get python nicely integrated, adding other languages might be simple
<guso78k> Elixir :') :') :') :')
<teepee> I don't see that happening
<guso78k> just kidding
<ali1234> rust, for the masochists
<teepee> that repo was 6 years abandoned
<stealth_> guso78k, any way to get openscad preview to work within sublimetext?
<stealth_> I want to use sublime for coding since the editor is way better.
<guso78k> i think  you just need to use the external editor feature
<stealth_> I suppose, its gets weird when you are working with multiple files and you need to render, it will close the editor and open a new one.
<guso78k> nope, you  can hide openscad's editor and just edit the source file with your sublime editore
<stealth_> I know, I didn't mean that, but its no big deal... I am just happy I can code 3d in python.
<ali1234> okay so running venv module makes a venv, but it is the system python
mmu_man has joined #openscad
<stealth_> in that cause its good to use venv if you are using system python, you don't want to install stuff from pip into your system python but venv is ok .
<ali1234> you still have that bug when if the script exits, the whole program exits
<ali1234> and then you can never open that script again, lol
<stealth_> ?
<stealth_> with venv?
<guso78k> yeah, that is still an issue. :')
<ali1234> i think the only sensible way to make this work is by adding "venv" and "pip" commands to openscad's cli somehow
<ali1234> "--python -m venv" or something
<ali1234> and make "--python" be default if executable name starts with "python"
<teepee> just trying that, the python init has a dedicated field for running a module
<stealth_> when you say use python, it should create default venv to your python code. Also pip that only works inside that venv
<ali1234> the first step is getting venv to work *at all*
<ali1234> UI is a totally different problem
<ali1234> apparently it is possible to ship two binaries in an appimage
<ali1234> oh, but the main binary has to run the secondary one
<stealth_> you just have to include venv & pip files inside site-packages
<stealth_> or any folder and make sure that folder is included using sys.path
<ali1234> and supply a way of running them, which we currently don't
<stealth_> just like you run python now like `/usr/bin/python` you run it `/path/venv/python` maybe there is an init process (not sure i don't use venv)
<ali1234> how you going to run python when it isn't shipped?
<ali1234> how you going to run venv/bin/python when you haven't made a venv yet?
<stealth_> you are using systems python right? just add a check to see where its located, also you can just include venv files and point it to where python is.
<ali1234> no
<stealth_> you don't have to "install" venv, you need the .py files.
<ali1234> as previously mentioned, we cannot use system python
<stealth_> you can use the same idea with both, one you include and system python.
<ali1234> python executable is not included
<ali1234> it is libpython
<ali1234> it's a library
<ali1234> you can't run it
<stealth_> well for openscad to run python code, it needs to use python.exe from somewhere! either you compiled or system compiled.
<ali1234> no, it doesn't
<stealth_> ya, you include python.h but when you create openscad.exe python is already included in openscad.exe also
<stealth_> how is it that when I type `print('hello')` it outputs `hello` without python's runtime?
<teepee> guso78k: any idea why PyUnicode_FromString() crashes? what's the way of creating the config lists, the doku is a bit confusing
<teepee> I can do: openscad --python-pip
<guso78k> there are some samples in pyopenscad.cc.  you need to supply a char* string. did you do that ?
<guso78k> wow - you started coding embedded python :)
bertronika has quit [Quit: Konversation terminated!]
<teepee> yes
<teepee> (and yes)
<teepee> I did the pre-config thing
<guso78k> hmm, not much exeriecne with preconfig
<guso78k> not sure if you can use PyUnicodeFromString before python is fully intialized .
<guso78k> maybe preconfig just requires char * strings ?
<teepee> I need to PyWideStringList_Append() for adding the argv values
<teepee> so I need some way of converting char* to wchar_t*
<guso78k> it might be  a good exercise to git clone cpyrthon. watch Modules/main.c o see how python interpreter starts up!
<guso78k> cpython
<guso78k> github.com/python/cpython.git
<gbruno> [github] kintel synchronize pull request #5720 (Fix a few clang-tidy issues) https://github.com/openscad/openscad/pull/5720
<gbruno> [github] kintel pushed 1 modifications (Fix for Python build) https://github.com/openscad/openscad/commit/1440f73321ff2c2e9742c0fae99311e511a5fe68
<guso78k> teepee:  it reads: status = PyConfig_SetArgv(&config, args->argc, args->wchar_argv);
<teepee> just using std::mbstowcs() now
<guso78k> ok i think you plugin this function
<teepee> calling pip works https://imgur.com/9S3PLn6
<guso78k> great. do you receive any strange warnings during installation of packages ?
<teepee> it does not respect venv yet, so no installation so far
<guso78k> got some warnings, because pip did feel that it got "misplaced" but in total it worked
<guso78k> it appears that python devs for openscad grows a big community :')
<teepee> ./build/openscad-nightly --python-module venv /tmp/venv-1
<teepee> Error: Command '['/tmp/venv-1/bin/openscad-nightly', '-m', 'ensurepip', '--upgrade', '--default-pip']' returned non-zero exit status 1.
<teepee> hmm, I think we use -m already
<guso78k> https://github.com/forabi/quick-installer/blob/master/install-pip.sh works even better , but not sure if its is applicable here
<teepee> so maybe we indeed need to switch command line parsing when called as "python"
<guso78k> better than ensurepip
<guso78k> sorry, i got very little experience with venvs .
<teepee> it fails as -m is something else in openscad
<teepee> lets see if it works stealing the flag
<guso78k> does venv create the -m command line argument ?
<ali1234> i think normally it just makes symlinks
<ali1234> and pip executable is this python script: https://bpa.st/SI6Q
<ali1234> although there is more than one way to make a venv
<ali1234> virtualenv, pyenv etc may do it differently
<guso78k> yes,  native pip functionality is in a module
<gbruno> [github] kintel pushed 1 modifications (Fix for Python build) https://github.com/openscad/openscad/commit/d62425e63c998f9d5f38b8cf46bcf7d7f809ffff
<gbruno> [github] kintel synchronize pull request #5720 (Fix a few clang-tidy issues) https://github.com/openscad/openscad/pull/5720
<guso78k> BTW: it feels great, that  manpower for python in openscad has more than doubled =*
Guest23 has joined #openscad
Guest23 has quit [Client Quit]
<teepee> I would not mind working on some of the topics on my own list ;-)
mmu_man has quit [Ping timeout: 252 seconds]
mmu_man has joined #openscad
misterfish has quit [Ping timeout: 268 seconds]