On Mon, Feb 18, 2013 at 06:40:44PM +0200, Zeeshan Ali (Khattak) wrote: > On Mon, Feb 18, 2013 at 6:31 PM, Christophe Fergeau <cfergeau@xxxxxxxxxx> wrote: > > On Mon, Feb 18, 2013 at 05:22:23PM +0200, Zeeshan Ali (Khattak) wrote: > >> From: "Zeeshan Ali (Khattak)" <zeeshanak@xxxxxxxxx> > >> > >> New udev/blkid expose more ISO9660 properties > > > > I think by 'new' here, you mean unreleased versions of these. Given that > > libosinfo releases end up as stable updates in released distros (this is > > the case in fedora at least), I think it's premature to remove the rule > > altogether. It seems to already be disabled be default unless --with-udev > > is used. Maybe we can add a warning that this is deprecated when using > > util-linux vXX and systemd vYY and keep the rule for now? We could also > > potentially add another switch --with-deprecated-udev rule switch to make > > sure distros notice this is going away. Another way would be a udev version > > check, and error out with new udev and --with-udev. > > > > All in all, I think just removing it is very premature, especially as there > > are users of this rule (boxes). I'm all for documented it's deprecated one > > way or another. > > As I said in the other mail, I share your concern but not as deeply. > Once systemd/util-linux are released, we can merge this. And break Boxes through stable updates in released distros that choose to update lo latest libosinfo? > There will be > a week or two in between systemd release and libosinfo release. > Knowing that some distros missed the 'enable-udev' option (at least in > the beginning), I don't think renaming the option would grab attention > from disto packagers. It really depends, not installing a file that is listed in a .spec %files section will cause a package build failure on distros using RPM, which is a sure way of grabbing the packager attention. Anyway, I've made other suggestions in my email, Christophe
Attachment:
pgpNEVu5CNgea.pgp
Description: PGP signature
_______________________________________________ Libosinfo mailing list Libosinfo@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/libosinfo
