<guso78k>
kintel sorry, this was not my intent to push something. can you help out ?
<teepee>
guso78k: I think you just recreated the branch, that should not hurt anything, we can just click the delete-branch again
mmu_man has joined #openscad
<teepee>
did you actually "git push" anything, or just click the "restore" button in the web gui?
teepee_ has joined #openscad
<guso78k>
teepee, yes i just clicked on the restore button. i mean to reopen the PR because i was not sure if my post would be visible on closed PR's
teepee has quit [Ping timeout: 252 seconds]
teepee_ is now known as teepee
<guso78k>
kintel, sorry, i am late here is my paste with the VERSION_LESS thing https://bpa.st/SY5Q
<teepee>
it's sending out notification mails anyway, if they are enabled. I got one :)
<teepee>
that's unrelated to the branch itself, after merging, that part is done
<teepee>
I just deleted the branch again
<guso78k>
ok, thank you
adigitoleo has quit [Read error: Connection reset by peer]
adigitoleo has joined #openscad
kintel has joined #openscad
<kintel>
guso78 The CMake error is because it cannot identify the OpenCSG version. See the previous line: "OpenCSG:" - that should display the version number of OpenCSG
<kintel>
Do you know which version you've got installed?
<kintel>
Alternatively, one workaround is "cmake -DUSE_BUILTIN_OPENCSG=1"
<kintel>
I can try setting up a fedora docker image and test, but it will probably take me until the week-end as I'm out traveling. If you want to try debugging that would be very helpful though!
<guso78k>
thank you, investigating and collecting answers :)
<guso78k>
no docker image :)
<kintel>
some message statements scattered around should tell a good story
<kintel>
We should set up some CI runners against various distros to avoid such breakage in the future
kintel has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
kintel has joined #openscad
lastrodamo has joined #openscad
kintel has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<teepee>
well, OBS seems to be all red too, but we can extend our ci as much as we want
misterfish has quit [Ping timeout: 244 seconds]
J23k98 has quit [Quit: Client closed]
J23k98 has joined #openscad
J23k98 has quit [Quit: Client closed]
J23k98 has joined #openscad
J23k98 has quit [Client Quit]
J23k98 has joined #openscad
erectus has quit [Remote host closed the connection]
erectus has joined #openscad
mmu_man has quit [Ping timeout: 240 seconds]
<guso78k>
when a function returns std::shared_ptr<AbstractNode> and it actually returns NULL, can i expect to read NULL , if it printf it with "%p" ? or get true if i compare it with "result == NULL" ?
<guso78k>
or is it an indication of memory corruption ?
JakeSays_ has quit [Ping timeout: 240 seconds]
JakeSays has joined #openscad
guso78k has quit [Quit: Client closed]
guso78k has joined #openscad
guso78k has quit [Client Quit]
Guest33 has joined #openscad
Guest33 has quit [Client Quit]
<peepsalot>
guso78, its nullptr rather than NULL in idiomatic C++ (technically std::nullptr, but people tend to not bother with the prefix)
<peepsalot>
comparing with NULL still works but is not as type-safe as using nullptr. compiler might print a warning about it, depending on how strict the flags are
fancsali has quit [Quit: Client closed]
snaked has quit [Remote host closed the connection]
<Guest70>
I'm trying to write a function that would accept a 3d or 2d shape, I would like the function to poke holes in that shape (think like a mesh, hexagons, circles, etc). Is that possible?
<teepee>
with some limitations yes
<Guest70>
what type of limitations?
<teepee>
you can not inspect the shape for example
<teepee>
doing this to the final 3d form is probably not going to work
<teepee>
appying some function to the sides separately and assembling the object later might be ok
<Guest70>
yeah, that is what I was thinking
<Guest70>
other way i was thinking was passing in some coordinates and generating the shapes to punch out. Its easy enough to do this one off (which i have), trying to make this a reusable module is proving difficult
<InPhase>
Guest70: Fully general solutions would never work for such a thing anyway, as someone would always be able to come up with a shape to thwart your smart algorithm. As the maxim goes, "If you make something idiot-proof, someone will just make a better idiot." Instead, maybe think about solving it with nice parameterization for a class of shapes that are interesting to you. Like maybe, radially arranged
<InPhase>
holes with a parameterized diameter, angular count, and vertical count, or something like that.
<InPhase>
Once you accept perfect is not possible, usefully acceptable has a much wider space of opportunities available. :)