[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: strange preinst tests about --assert-working-epoch



On Fri, Feb 15, 2002 at 01:01:47AM +0100, Wichert Akkerman wrote:
> Previously Martin Quinson wrote:
> > It seems bad to me: It's a kind of dependency there, no ? But on the other
> > hand, they just depend a recent dpkg to be installed, not to be used...
> 
> It is a dependency that might not be possible to be resolved by an old
> dpkg: if it is a versioned dependency using epochs and you are using an
> dpkg which does not support epoch you need an assertion like that.
> 
> At this point in time epochs have been in use for so lang that doing
> that assertion check doesn't make any sense anymore though.

It didn't make much sense in maintainer scripts in the first place,
unless you implement --assert-running-dpkg-and-installed-dpkg-are-the-same
(or a saner implementation of assertions)

-- 
  .''`.  ** Debian GNU/Linux ** | Andrew Suffield
 : :' :  http://www.debian.org/ | Dept. of Computing,
 `. `'                          | Imperial College,
   `-             -><-          | London, UK



Reply to: