Version 1.2.1 is three years old, and incompatible with SOFA files generated by Mesh2HRTF. This prevents them from working in Pipewire. Please update the library to v1.3.2. Reproducible: Always Steps to Reproduce: 1. Attempt to load a SOFA file generated by Mesh2HRTF Actual Results: It fails Expected Results: It works
Thanks for the report. The problem with 1.3.2 is that unit tests are massively failing, so I was relunctant to update. I've just reported the issue at https://github.com/hoene/libmysofa/issues/224 A new release is been prepared, (although the tag is wrong ie latest), so I hope to have a formal release soon: https://github.com/hoene/libmysofa/issues/223
Ah, I think I see the problem. The test code generates tmp1.json for each SOFA file, and compares them to the .json.bz2 files. Since ctest is being run with -j6, multiple threads are overwriting the temp file at the same time. Only a single test fails when I build 1.3.2 on my machine (MIT_KEMAR_normal_pinna), but that same test fails when I build 1.2.1, so it shouldn't cause any regressions.
FEDORA-2024-9d4b36e46c (libmysofa-1.3.2-3.20240917git2297dd8.fc42) has been submitted as an update to Fedora 42. https://bodhi.fedoraproject.org/updates/FEDORA-2024-9d4b36e46c
FEDORA-2024-9d4b36e46c (libmysofa-1.3.2-3.20240917git2297dd8.fc42) has been pushed to the Fedora 42 stable repository. If problem still persists, please make note of it in this bug report.
OKay, I've updated a snapshot of libmysofa for f40+ and epel10, I hope it would be good enough
Installing that build on Fedora 40 solved my problem. Thank you! :)