<kintel>
I don't have the credentials any longer, but you can just reset the passwd :)
<teepee>
probably from travis times or something
<kintel>
There may be better mechanisms to push these days, but haven't looked into it
<teepee>
I have not found much, at least not for docker hub
<kintel>
right, creating an access token for the robot user is probably the easiest
<teepee>
in the end it's pretty much the same if it's a password or an access token
<teepee>
they not seem to allow binding specific actions to tokens or so
<teepee>
so far I'm playing with concourse as CI which seems pretty neat, I could already build the ARM64 appimage base which was not possible on dockerhub
kintel has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
teepee has quit [Remote host closed the connection]
teepee has joined #openscad
teepee_ has joined #openscad
teepee has quit [Ping timeout: 246 seconds]
teepee_ is now known as teepee
kintel has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
othx has quit [Read error: Connection reset by peer]
othx has joined #openscad
little_blossom has quit [Quit: little_blossom]
little_blossom has joined #openscad
mmu_man has joined #openscad
ubitux has quit [Ping timeout: 245 seconds]
ubitux has joined #openscad
guso78k has joined #openscad
<guso78k>
teepee, the windows build errors on pyton-pr3-win(where i actually forgot to enable python for windows) some time again are exactly the same ones as the windows build errors with my python-pr3 yesterday. suppose this is an unrelated issue.
<_SKiTZO>
I am trying to use openscad from commandline. I am lloking at the documentation for the -p and -P arguments and scratching my head. Does the json file need to exist or will openscad generate it?
<_SKiTZO>
I am writing a service that uses customizer and that makes it needed to parse opencad files to look for parameters. I would rather not have to do this manually and so was hopeful that -p /-P parameters could take as input an openscad file and spit out a json outlining the parameters as openscad parsed them
omegatron has quit [Remote host closed the connection]
<joseph_>
kintel: Thanks for the reminder about OpenCSG. I think my bouncer got kicked from Libera so for some reason I didn't get a notification. But I saw the message in logs
<teepee>
joseph_: what's the state of that PR, last commit sounds quite promising
teepee_ has joined #openscad
teepee has quit [Ping timeout: 246 seconds]
teepee_ is now known as teepee
<joseph_>
teepee: It did to me too, until I double checked and remembered that the legacy functions are actually still allowed in OpenGL 3.0. It would have made more sense to remove legacy support on a major version change but that's not how they did it. The distinction of core vs compatability shows up in 3.2
<teepee>
ah, right, strict core only started in 3.2
<teepee>
so it's still using some legacy calls in the PR? specifically the opencsg stuff mentioned earlier?
<joseph_>
Yes it turns out I still missed some legacy functions. For example, it appears that there's currently no modern code written for drawing the axes (GLView.cc). So what I'm currently testing is whether I can just have my macro disable the axes entirely but still render the model. Obviously someone will eventually need to rewrite it
<teepee>
yep, I think it's ok to disable axes and indicator, I assume those would later go into a dedicated shader
<joseph_>
teepee: Now I have added the capability to truly disable ALL the legacy calls with a macro. I verified this with a successfull build under 3.2 Core profile. Perhaps unsurprisingly, actually trying to preview/render anything (even with experimental VBO features enabled) will cause a segfault. This means at least one critical bit of functionality doesn't yet have a modern OpenGL alternative implemented