<joseph_>
teepee: The midterm evaluation forms will open tomorrow with a deadline of July 14th. I'm just getting a few more minor things together before my PR #4675 is ready for additional feedback. Its scope is still limited to non-functional refactoring, which should simplify review.
<joseph_>
After the second phase starts, I think it will be valuable to discuss how I should proceed with marking legacy vs. modern rendering calls at compile and runtime. I expect that once I get momentum on that, the volume of my code changes will increase. This first half has been more deliberative/exploratory. Eventually, I want to polish up my notes on the codebase and perhaps publish them as documentation, but this is GSoC not GSoD :)
<kintel>
joseph_ some fly-by comments; I'm not online very much these days: I think it would be beneficial for review to let us know when your PR is feature complete, or document the expected scope of the PR, so we don't continuously review a changing PR.
<kintel>
In terms of legacy vs. modern calls; I think you're unlikely to be able to move everything to modern calls in this project, but one good step forward could be to identify and document where legacy calls exist, and choose your battles from there. In most places you might be able to simply port to modern, while in other places it might be trickier and we may have to live with a compile-time flag.
<kintel>
I was hoping to land my https://github.com/openscad/openscad/pull/4582 before the summer, but it still needs some GUI testing across platforms. That PR should make it easier to request GL contexts without legacy support
<teepee>
yep, declaring a state for the current PR would be great
<teepee>
would be a nice sync point with the mid-term being this week
<joseph_>
I updated the top comment/summary of my PR to provide an overview of what it changes. Is that how you'd generally recommend I define the scope and add necessary explanations? Also, when a PR is ready for "final review" preceding a merge, I'll unmark it as draft. Before then, if I have specific questions about one part then I can send a message here
<joseph_>
Strangely, one test case has failed. After resolving that, I will unmark it as a draft
<teepee>
which one is it? there's still an unstable one, IIRC example024
<joseph_>
Yes it's that one
mmu_man has quit [Ping timeout: 245 seconds]
<teepee>
ok, let me check, I think you can ignore that for now. I hoped the Manifold upgrade can finally fix it, but it seems it did not :(
<joseph_>
teepee: I didn't realize you pushed an upgrade one hour ago. That one isn't yet on my branch, so maybe the upgrade did actually fix the problem
<teepee>
no, it failed in that PR to
mmu_man has joined #openscad
<teepee>
I had to re-run the action in that PR to get it to green