I just cvs removed it from the F-9 and devel branches. It needs to blocked from F-8, F-9, and devel (I think the latter two are already done but it still exists in F-8 where nspluginwrapper exists and solves the problem).
On Mon, 2008-10-27 at 10:36 -0700, Christopher Aillon wrote:
I just cvs removed it from the F-9 and devel branches. It needs to blocked from F-8, F-9, and devel (I think the latter two are already done but it still exists in F-8 where nspluginwrapper exists and solves the problem).
It's been blocked. I assume nspluginwrapper has the appropriate Provides/Obsoletes in it?
Jesse Keating wrote:
On Mon, 2008-10-27 at 10:36 -0700, Christopher Aillon wrote:
I just cvs removed it from the F-9 and devel branches. It needs to blocked from F-8, F-9, and devel (I think the latter two are already done but it still exists in F-8 where nspluginwrapper exists and solves the problem).
It's been blocked. I assume nspluginwrapper has the appropriate Provides/Obsoletes in it?
No. To be more precise, firefox itself does the specific thing that firefox-32 intended. The person who created the package did not think to check 'setarch i386 firefox' to see if that would work. It does.
nspluginwrapper solves a different problem really, but has a by-product that less people will want to run 32 bit browsers.
I see no reason to allow stupid packages to clutter specfiles with Obsolete/Provides, especially with the history of this package (me having asked the packager to not go through with the package in the first place, etc).
On Mon, 2008-10-27 at 11:04 -0700, Christopher Aillon wrote:
No. To be more precise, firefox itself does the specific thing that firefox-32 intended. The person who created the package did not think to check 'setarch i386 firefox' to see if that would work. It does.
nspluginwrapper solves a different problem really, but has a by-product that less people will want to run 32 bit browsers.
I see no reason to allow stupid packages to clutter specfiles with Obsolete/Provides, especially with the history of this package (me having asked the packager to not go through with the package in the first place, etc).
Well I'm mostly concerned about the user experience for people that already have firefox-32 installed. What does it depend on, what will show up as broken deps if firefox itself gets updated, etc... Will this package just sit quietly and never harm the user by being installed?
Jesse Keating wrote:
On Mon, 2008-10-27 at 11:04 -0700, Christopher Aillon wrote:
No. To be more precise, firefox itself does the specific thing that firefox-32 intended. The person who created the package did not think to check 'setarch i386 firefox' to see if that would work. It does.
nspluginwrapper solves a different problem really, but has a by-product that less people will want to run 32 bit browsers.
I see no reason to allow stupid packages to clutter specfiles with Obsolete/Provides, especially with the history of this package (me having asked the packager to not go through with the package in the first place, etc).
Well I'm mostly concerned about the user experience for people that already have firefox-32 installed. What does it depend on, what will show up as broken deps if firefox itself gets updated, etc... Will this package just sit quietly and never harm the user by being installed?
There is no harm to the user having it installed. It essentially is a packaged up shell script that runs setarch i386 before invoking firefox (though in a backward way). nspluginwrapper will still work with this package installed. So will manually invoking setarch. There is no reason anyone needs it installed, but it will not harm anything to have it installed.
On Mon, 2008-10-27 at 14:49 -0700, Christopher Aillon wrote:
There is no harm to the user having it installed. It essentially is a packaged up shell script that runs setarch i386 before invoking firefox (though in a backward way). nspluginwrapper will still work with this package installed. So will manually invoking setarch. There is no reason anyone needs it installed, but it will not harm anything to have it installed.
Actually I'm seeing only one build done of this, and not shipped in F8, or any updates, so I don't think this went anywhere except for maybe a few rawhide composes way back when.
Where are you seeing this package come up?
Jesse Keating wrote:
On Mon, 2008-10-27 at 14:49 -0700, Christopher Aillon wrote:
There is no harm to the user having it installed. It essentially is a packaged up shell script that runs setarch i386 before invoking firefox (though in a backward way). nspluginwrapper will still work with this package installed. So will manually invoking setarch. There is no reason anyone needs it installed, but it will not harm anything to have it installed.
Actually I'm seeing only one build done of this, and not shipped in F8, or any updates, so I don't think this went anywhere except for maybe a few rawhide composes way back when.
Where are you seeing this package come up?
mcepl said he heard from someone who had the package, and didn't know that he didn't need it. If it's not shipped, he must have grabbed it from koji or an old rawhide. Either way, I'm not too worried about it, just going by the rulebook for the EOL process.
rel-eng@lists.fedoraproject.org
-
Christopher Aillon -
Jesse Keating