Allan McRae wrote:
Since when it is a good packaging quality to upload packages which can't be installed?
They can be installed. Or do you mean can not be installed by an AUR helper? In which case you are still wrong due to the extra dependency line.
I think this point should be stressed. Yaourt and other AUR helpers do not determine the validity of a PKGBUILD. If you can download the tarball from the AUR and build the package with makepkg, then the PKGBUILD is *valid* as far as the AUR is concerned. Having said that, I disagree that the criteria for a *good* PKGBUILD is only that it build. PKGBUILDs should try to conform to certain patterns and not exploit the fact that they are written in Bash (unless absolutely necessary, and even then only reluctantly). All metapackaging tools would have been much easier to write if PKGBUILDs were perfectly parsable without arbitrary code execution. The choice of Bash was myopic and lazy in my opinion, and something that should be reversed as far as possible, not glorified.