Description of problem: lua defines 'libdir=${exec_prefix}/lib' in its pkgconfig file, lua.pc. This causes problem detecting its library location on say x86_64 (which point to /usr/lib while the libraries are in /usr/lib64). Version-Release number of selected component (if applicable):lua-devel-5.1.2-3.fc9 How reproducible: always Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info:
Thanks for reporting this. lua-5.1.2-4, which fixes this should show up in tomorrows rawhide push. Since this is relatively harmless, I'm only fixing this in rawhide, let me know if you need it fixed for some dependent package in lower Fedora versions.
*** Bug 436760 has been marked as a duplicate of this bug. ***
I need this fix in Fedora 8. It's really a bad thing not to fix such trivial bugs in older *still supported* versions as well.
(In reply to comment #3) > I need this fix in Fedora 8. It's really a bad thing not to fix such trivial > bugs in older *still supported* versions as well. If you can explain the _need_ part... That is explain what goes wrong other then a few warnings, then I'm more then willing to issue an update. People are already complaining about the number of updates, so I'm not going to fix this just to silence a few compile time warnings.
Our test suite will report this as a bug - and in fact warnings are there for a reason. I think it rather annoying that if a bug has been identified it is not fixed just to keep the update volume low. If there are bugs they demand an update, no matter what! Especially for Lua this argument isn't worth much, the package isn't widely used so not many machines will download it, and even if the size is only 224 + 17 kB! Sorry, but not fixing a bug is always a bad thing. If you do not have the time I can do it myself and push the update.
(In reply to comment #5) > Our test suite will report this as a bug - and in fact warnings are there for a > reason. > "Our test suite" still is not a very verbose explanation of why you need this fix, and reporting harmless warnings as a bug could be considered a bug in the test suite :) No please don't respond I agree this is a bug, the only we thing we disagree about is the impact of it. > I think it rather annoying that if a bug has been identified it is not fixed > just to keep the update volume low. There is a balance to keep between impact of the bug, and the additional bandwith / install time consumed for end users. In this case this bug: -only affects developers, not normal end users -even for developers only results in a harmless warning So I really saw, and still see no reason to issue an update, but since this seems important for you an update to the rawhide version (which includes a jump from 5.1.2 to 5.1.3) is on its way to updates-testing, let me know if that works properly for you.
Works just fine, noted in pending update request.