peterhillman
peterhillman
It appears the NaN isn't being detected correctly by `predf()`. Maybe it would be safer to use `isinff(f) || isnanf(f)` [here](https://github.com/AcademySoftwareFoundation/Imath/blob/68cce5db4d8367f5a564845c2b3adefb6659b6b1/src/Imath/ImathFun.cpp#L20) and [here](https://github.com/AcademySoftwareFoundation/Imath/blob/68cce5db4d8367f5a564845c2b3adefb6659b6b1/src/Imath/ImathFun.cpp#L58). I wonder if this is related to...
That sounds reasonable. The branch name becomes less relevant once the changes are merged into the main branch. The commit messages and title of the PR stick around longer, so...
Isn't this just rounding? 5C 28 81 3A is 0.000985394697636 which rounds to 0.000985395 to 6 significant figures 54 C9 as a half is 0.041626 , which is 0.042 to...
I see you updated `exrheader` to dump out the compression type. `exrmaketiled`, `exrenvmap` and `exrmultiview` take a command line string and turn that to a compression type. Should they be...
I got this to build OK, so it looks like the Blosc build is working. Did you research different settings for the number of scanlines in each chunk? 32 scanlines...
I had issues doing `cmake ../openexr -DCMAKE_BUILD_TYPE=Debug` - it seems that causes blosc to generate a `libblosc2_d.a` but OpenEXR still tries to link against `libblosc2.a`. In other news I _think_...
@pleprince I'm a little out of my depth here. I would have thought it was OK to change the fuzz test compile flags, since the purpose of fuzz testing is...
This test is a bit sneaky to simplify the code. It generates a random half, converts it to a float, writes it, reads it back and converts that to a...
@SuperSandro2000 thanks for that report. PR #1490 is an attempt to fix this based on the error message. I wasn't able to reproduce this myself, so it would be helpful...
Yes the reformat caught me out too. The main branch has the update now so you could try building from main to confirm. Or maybe copy the entire source file...