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

Re: New packaging manual draft



Hi,

	Yet another version is up at 
 http://master.debian.org/%7Esrivasta/new-packaging.txt


>>"Jason" == Jason Gunthorpe <jgg@ualberta.ca> writes:

 Jason> AFIAK this is an error:

 Jason>      All but `Pre-Depends' (discussed below) take effect
 Jason>      _only_ when a package is to be configured.  They do not
 Jason>      prevent a package being on the system in an unconfigured
 Jason>      state while its dependencies are

 Jason> It should be All but 'Pre-Depends' and 'Conflicts'.

	Done.

 Jason> The rest of the paragraph is also misleading. At any arbitary
 Jason> time it is possible for any Depends relations to become un
 Jason> satisfied either due to an invalid version of a newly
 Jason> installed package or due to the target package being unpacked.

	Ummm, could you propose corrected wording, then? Are you
 saying that pre depends and conflicts can now prevent a package being
 on the system in an unconfigured state? Or are all bets off when a
 new package is unpacked? 

 Jason>           `dpkg' will not configure packages whose
 Jason>           dependencies aren't satisfied.  If it is asked to
 Jason>           make an installation which would cause an installed
 Jason>           package's dependencies to become unsatisfied it will
 Jason>           complain [1], unless `--auto-deconfigure' is
 Jason>           specified, in which case those packages will be
 Jason>           deconfigured before the installation proceeds.

 Jason> IIRC this is just completely false. Dpkg breaks reverse dependencies
 Jason> without any error or warning - that is how it must operate.

	Hmm. Reverse dependencies are never mentioned here, so I think
 completely false is a trifle strong.


 Jason>           However, when a package declaring a predependency is
 Jason>           being unpacked the predependency can be satisfied
 Jason>           even if the depended-on package(s) are only unpacked
 Jason>           or half-configured,

 Jason> I don't think this is true..

 Jason> There seems to be missing a discussion on how dependencies and
 Jason> essential packages relate to maintainer scripts. There are
 Jason> some unobvious side effects here that should be noted
 Jason> explicitly - like removal scripts can't rely on
 Jason> pre-depends/depends and pre-inst can not rely on depends, etc.

	Are you sure that is policy? Seems to me that belongs in a
 devel guide, since it is the only correct way way to create the
 scripts. 

 Jason> The list of fields section is massively incomplete.. I have
 Jason> the full list someplace if someone would like to document them
 Jason> all.

	It is meant to be inconplete. I have now mentioned the
 fact. This is only a list of fields for which we have a policy
 dictum, and is nowhere near exhaustive.

	manoj
-- 
 Hurewitz's Memory Principle: The chance of forgetting something is
 directly proportional to... to... uh.....
Manoj Srivastava   <srivasta@debian.org>  <http://www.debian.org/%7Esrivasta/>
1024R/C7261095 print CB D9 F4 12 68 07 E4 05  CC 2D 27 12 1D F5 E8 6E
1024D/BF24424C print 4966 F272 D093 B493 410B  924B 21BA DABB BF24 424C



Reply to: