Re: binNMU request for geany
On Thu, Jun 07, 2007 at 10:10:03PM -0300, Damián Viano wrote:
> On Thu, Jun 07, 2007 at 02:01:14AM -0700, Steve Langasek wrote:
> > Hi Damián,
> > On Sun, Jun 03, 2007 at 04:57:29PM -0300, Damián Viano wrote:
> > > Could someone schedule this binNMU please?
> > > geany_0.11-1, Rebuild against gtk 2.10 needed for dnd support, 1, i386
> > If geany is supposed to be built with dnd support and this requires gtk
> > 2.10, why does geany not have a versioned build-dependency on gtk which
> > enforces this?
> Because actually the dependency is just to use the same gtk runtime
> library that geany was built against (i.e building against 2.8, and
> running against 2.10 wont work), unfortunately I've built the i386
> binary on a outdated pbuilder with gtk 2.8 and gtk 2.10 is now on sid
> which makes geany dnd code unhappy.
If your package has to run with the same version of gtk that it's built
with, then I think you have a sourceful bug here because your package's
dependencies don't reflect this.
> > How should we recognize if the binNMU is successful? The only possibly
> > relevant symbol that's referenced by the alpha binary and not by the i386
> > binary, for example, is 'gtk_notebook_set_tab_reorderable'. Is this how one
> > should know that the package is built correctly?
> As stated above, if geany is built against gtk2.10 everything would be
> fine, at least for now. I found about this problem recently, because of
> this upload, and will evaluate better way of handling this in future
> versions of the package (i.e. Depends: libgtk >= (2.10), libgtk < (2.11)
Yes, that sounds like the correct fix to me; please make this change in the
package, then no binNMU is needed here.
Steve Langasek Give me a lever long enough and a Free OS
Debian Developer to set it on, and I can move the world.