2026-03-16 TSC Meeting notes

Clang format PR

• Remi rebased his PR

• Additional target for the CMake build, follows OIIO process

• Some things not covered, image values and formatting for tests, clang would break all of that

• Also ignores SIMD intrinsics

• OIIO is using version 17, we can start with newer (22) and pin it so it doesn’t auto update and break things

• How to people feel about the current formatting? Are adjustments needed?

• Doug: in theory, the CI will run clang-format right? So if the user doesn’t have clang-formatter installed it will run on the CI and report?

• Remi - would need to take a an action to check that, it wouldn’t auto-fix but it could report it broken

• Doug - as long as the CI fails, that’s fine for now. Doesn’t need to be exact, we can improve as we go on

• Doug - Remi, what’s your preference on how we handle your PR vs Kevin’s?

  • Remi - maybe do Kevin’s first and then see how the automated change affects it?

  • Remi has also split the commits so you can see the different changes easily

  • Will definitely create a lot of code change but that’s to be expected

• Will follow up with Kevin about his PR’s to get those in

Fixing the SonarCloud Action

• https://github.com/AcademySoftwareFoundation/OpenColorIO/issues/2274

• Cuneyt has a PR that works on SonarCloud stuff too - tried to get it working, simple fix didn’t work. Hopefully the info from JF will help, haven’t had a chance to look yet

• Remi - semi related, platform latest is currently broken, with a C++ 23 issue

• Doug also needs to submit a PR for the python bindings in the HueCurveTransform, maybe will fix that at the same time

CI Involvement in General

• In general, would be great if OCIO took a more active approach to being involved in the CI Working group and process. We should try to have an OCIO rep at the meetings, and monitor the ci-wg channel. Open to how the TSC wants to handle this, but having someone other than Carol or Doug take a more active role in this area would be a huge help.

OCIOView & Dev Days

• Carol proposes a focus on OCIOView for dev days

• Michael agrees, and also notes that there is some infrastructure improvements that might benefit from some LLM type analysis

• Lots of opportunities to contribute on the render side of it

• Some parts are OpenGL, some are QT’s 3D, some are more modern frameworks - could definitely use a unification pass for sure

• OCIOView unit test / testing framework - should have that in place before unification / optimization tests

• Carol will get a meeting on the books to figure out next steps

ACES 2 GLES shader compilation issue

• Remi tried to take a look, it might be a false flag

• Carol will ask Kevin to take a look just to be sure

ACES Look transform collaboration

• Doug has been working on it, will have a config for folks to test tomorrow

• There was some discussion about component spaces and whether we could translate this to CLF - Doug’s config will show examples so we can talk concretely at the ACES TSC tomorrow