TSC Meeting Notes 2020-11-05

Owned by Cary Phillips
Nov 04, 2022
2 min read

Attending:

  • Cary Phillips

  • Christina Tempelaar-Lietz

  • Eskil Steenberg

  • Joseph Goldstone

  • Kimball Thurston

  • Larry Gritz

  • Nick Porcino

  • Owen Thompson

  • Peter Hillman

Discussion:

  • State of the RC-3 branch:

    • Kimball: Has anyone looked at the branch? Was going to just merge, couldn’t figure out ow to do a pull request.

    • MacOS and Windows now work in the CI, for the RC-3 branch.

    • Python is off by default, so default OpenEXR build avoids all reference to boost.

    • Kimball: Some more cleanup that I’d like to do.

    • Majority of the changes are eorganization, redoing directory structure.

    • Cleaned out the IlmBase, removed config dependencies, simpler config path.

    • Code should still work the same way.

    • To get Windows build to work, I had to fix some of the DLL exports tht weren’t being defined correctly.

    • Imath is also now building on Windows.

    • Removed non-c++11 stuff, may cause some people some grief. Removes so much code. We're willing to accept the consequences.

  • The Imath python bindings still need some cleanup

    • Still references some IlmBase stuff.

    • Most of the PyImathConfig.h and PyImathConfigInternal.h symbols are unnecessary and can be removed.

    • No need to duplicate HAVE_LARGE_STACK with ImathConfig

    • Need to clean out PyIex stuff. If PYIMATH_ENABLE_EXCEPTIONS is on, the code won't compile becuase it references Iex. We need to clearly state that we're entirely removing PyIex from distribution.

  • Joseph: No further progress with community on Standard Optional Attributes. They need a comment that says what they are.

    • Have not done: the alignment of camera viewing direction, camera up vector direction, should be linked. Have not had time to think about it.

  • A bunch of fuzz reports have come in.

    • Peter: It's current running my newly configured test, plus the original tests. But my new test isn’t finding any. It handles memory allocation better, so that may be why.

    • Some of the out of memory things are worth investigating.

    • The most recent repott was an overflow where you can have more than INT_MAX pixels in a file.

  • CLA:

    • Cary: Lucasfilm has finally agreed to the new CLA, with Disney's approval.

    • Kimball: Weta is still confued about why they need to sign something new, but it should be quick.

    • Nick: Pixar is close to approval on OpenTimelineIO.

    • Larry: Sony still working on OSL.

    • Christina: Has Epic signed? Not sure, but check with Michael Dolan, he'll know. And beware that your individual name also needs to be added to the approved list by your CLA manager.