Description of problem: With the upgrade to pandoc-2.19.2, pypandoc seems to be unable to properly detect pandoc on ppc64le. This has caused notcurses to FTBFS. Version-Release number of selected component (if applicable): 2.19.2-19.fc38 How reproducible: Always Steps to Reproduce: 1. Run a scratch build of notcurses for ppc64le Actual results: It fails to detect pandoc during the cffi part of the build, and crashes. Expected results: It succeeds at detecting pandoc and the build completes successfully. Additional info: This was originally detected by Koschei: https://koschei.fedoraproject.org/build/14992201 However, I encountered this while trying to rebuild notcurses for the ffmpeg 6.0 upgrade: https://koji.fedoraproject.org/koji/taskinfo?taskID=98612357
See bug 2172771 for more details. (Hopefully you don't need the LUA support in Pandoc) I will try disabling the hslua version check on ppc64le.
Changing to Post for the workaround (I will keep the ghc bug open anyway)
FEDORA-2023-ea05c2acee has been submitted as an update to Fedora 38. https://bodhi.fedoraproject.org/updates/FEDORA-2023-ea05c2acee
FEDORA-2023-ea05c2acee has been pushed to the Fedora 38 testing repository. You can provide feedback for this update here: https://bodhi.fedoraproject.org/updates/FEDORA-2023-ea05c2acee See also https://fedoraproject.org/wiki/QA:Updates_Testing for more information on how to test updates.
FEDORA-2023-ea05c2acee has been pushed to the Fedora 38 stable repository. If problem still persists, please make note of it in this bug report.
Just noting that it seems necessary to disable the lua version output for ppc64le also for F39. See eg https://koji.fedoraproject.org/koji/taskinfo?taskID=106825976 (I thought earlier it had gone away, so it is possible the problem is intermittent or rather some race condition, so that it occasionally doesn't happen perhaps? I believe the previous F39/Rawhide builds had it enabled though.)