Re: Debian Weekly News - June 27th, 2000
- To: debian-devel@lists.debian.org
- Subject: Re: Debian Weekly News - June 27th, 2000
- From: Itai Zukerman <zukerman@math-hat.com>
- Date: 01 Jul 2000 11:49:22 -0400
- Message-id: <[🔎] 87wvj57svx.fsf@matt.w80.math-hat.com>
- In-reply-to: Taketoshi Sano's message of "01 Jul 2000 22:28:49 +0900"
- References: <20000628162807.A15031@kitenet.net> <20000629190653.A19984@silly.cloud.net.au> <395B2021.8BC1310F@efore.fi> <20000629224826.B23358@silly.cloud.net.au> <20000629191453.D1607@wyvern> <20000630080910.B31951@silly.cloud.net.au> <20000630001523.A2760@fookidla.PenguinMints.org> <20000630184613.D8818@silly.cloud.net.au> <[🔎] y5ak8f6t1wu.fsf@kgh12351.nifty.ne.jp>
> Suppose a package A in Debian provides some functions, and another package
> B in Debian happenly uses that functions. It is natural for the maintainer
> put the dependency "Depends: A" on his package B.
>
> Also suppose a package A1 out of Debian provides the enough functions
> to use with the package B, and A1 also has some special functions which
> are essential for some limited users, but the package A can not provide
> these functions immediately because of some incompatibility between A
> and A1, or using totally different language for their code, etc.
It looks like your example doesn't preclude having A1 Provides: A. If
that doesn't work (because _A1_ is missing some functionality that
other things depending on A require), why not try to get a virtual
package set up, and ask that B depend on that?
Perhaps if I knew more about the real example behind this abstract
one...
-itai
Reply to: