ociodisplay gets an inverse control #3650
Conversation
| @@ -1848,7 +1876,7 @@ bool | |||
| ImageBufAlgo::ociodisplay(ImageBuf& dst, const ImageBuf& src, | |||
| string_view display, string_view view, | |||
| string_view from, string_view looks, bool unpremult, | |||
| string_view key, string_view value, | |||
| bool inverse, string_view key, string_view value, | |||
| ColorConfig* colorconfig, ROI roi, int nthreads) | |||
| { | |||
There was a problem hiding this comment.
Don't know how to select the line 1872, but don't you want to pass inverse there too?
There was a problem hiding this comment.
yes, I think you're right
src/libOpenImageIO/color_ocio.cpp
Outdated
| { | ||
| ImageBuf result; | ||
| bool ok = ociodisplay(result, src, display, view, from, looks, unpremult, | ||
| key, value, colorconfig, roi, nthreads); | ||
| unpremult, key, value, colorconfig, roi, nthreads); |
Fixes 3648
lgritz
force-pushed
the
lg-ociodisplay
branch
from
783b9bf to
1b5b001
Compare
|
Updated the PRs to address @doug-walker's suggestions. |
|
Because this isn't tested, I'm going to hold off on a merge while I do a little tinkering... I'm exploring more direct support for OpenColorIO 2.2 and its built-in configs, which will allow me to test display transforms (merely because it has some, unlike the sample config that's in the testsuite now). If minimal support of this is easy to achieve, I'll add it to this PR (even if the full range of tests only work when we're building against OCIO 2.2). But if that looks like a bigger job that doesn't make sense to append here, I'll merge this PR first (despite not being tested) and the follow up with a separate one with 2.2 support and more testing. |
|
I'm going to merge this now, because the follow-on has grown to be more complex. |
Fixes #3648