Re: how to satisfy "perl(XXX::YYY)" dependencies?

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

 



Le mercredi 15 dÃcembre 2004 Ã 08:20 -0500, Robert P. J. Day a Ãcrit :
> On Wed, 15 Dec 2004, Nicolas Mailhot wrote:
> 
> > Le mercredi 15 dÃcembre 2004 Ã 07:54 -0500, Robert P. J. Day a Ãcrit :
> >
> > >   note how the list of dependencies doesn't include just other rpms.
> > > it specifically includes *individual* *files*, with no regard to where
> > > they came from (at least, as i read it).  in short, even a stock
> > > standard, unmodified fedora core rpm-based system *already*
> > > incorporates non-rpm information in its own install process.  ergo,
> > > your point is irrelevant.
> >
> > Check again.
> 
> > The files are not taken from the filesystem but from the rpm db ie you
> > can depend on a file, if it has not been installed via rpm your dep
> > won't work.
> 
> hmmm ... ok, i stand corrected.  but then i'm curious as to why the
> list of deps of the "initscripts" package would include both:
> 
>   /bin/sh
>   bash >= 2.0
> 
> it would seem that, once you have bash >= 2.0, that would kind of
> subsume the dependency on a simple bourne lookalike shell, wouldn't
> it?  in fact, "rpm -qf /bin/sh" explicitly shows that it's part of the
> bash package.  so why the apparent redundancy?  (in fact, the list of
> deps for initscripts displays the entry "/bin/sh" five times in
> succession.  not sure what the point of that is.)

The /bin/sh deps were added by the autodeps scripts that read shell
scripts in the package that started with #!/bin/sh

The bash >= 2.0 was added by the human packager that knew he depended on
shell behaviour only bash >= 2.0  could provide.

> 
> i'm going to think on this some more, but i'm still curious as to how
> one would add an extra file like /etc/xinetd.d/cvspserver to the
> system, if you want to maintain strict rpm purity.  are you seriously
> suggesting that one create an rpm whose sole purpose it is to install
> a single config file?  that's kind of the impression i'm getting here.

1. if you don't care about losing your file and can recreate it at will
then sure, add the file manually
2. if you want to archive your work, have interdependent bits and need a
dependency engine, you do an rpm (and save it). Even for a single file.

rpm basics are easy, once you've learnt them creating a package is
scarcely more difficult than creating a tar archive. And with rpm you
can add dependencies on other people's work ie share the load. That's
what big rpm repositories are all about - they're never created by a
single person (even if he does most of the work and has his name on the
main page). They're created by people who find it handy to use packages
created by someone else and focus on their own packages.

-- 
Nicolas Mailhot

Attachment: signature.asc
Description: Ceci est une partie de message =?ISO-8859-1?Q?num=E9riquement?= =?ISO-8859-1?Q?_sign=E9e?=

_______________________________________________
Rpm-list mailing list
Rpm-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/rpm-list

[Index of Archives]     [RPM Ecosystem]     [Linux Kernel]     [Red Hat Install]     [PAM]     [Red Hat Watch]     [Red Hat Development]     [Red Hat]     [Gimp]     [Yosemite News]     [IETF Discussion]

  Powered by Linux