Hello dpkg maintainer people, you might want to wander over to http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=211292 for some context.
On Mon, Sep 22, 2003 at 06:31:27AM +0200, Henning Makholm wrote: > which will unfold to either of > > Depends: libc (>= 2.3.1-1), > Depends: libc (>= 2.3.1-1), debconf(>= 0.5) > > which both work. In the first of these to latter cases, the > remove-empty-subfields feature of dpkg-deb is used. It was the first case I was worried about, if that made it broken too. So you have gone from breaking things if both things defined to breaking things if only one is defined. You also missed a case, how about: Depends: , debconf(>= 0.5) In this case, misc:Depends is defined but shlibs:Depends is not So, to ask the dpkg people, will dpkg be happy with any of the three: Contestant A) Depends: libc (>= 2.3.1-1), Contestant B) Depends: libc (>= 2.3.1-1), debconf(>= 0.5) Contestant C) Depends: , debconf(>= 0.5) The current setup does this, which apparently is now not allowed: Contestant D) Depends: libc (>= 2.3.1-1) debconf(>= 0.5) If not, which ones work? I know contestant B is the right one, but are any of the others permitted or not? - Craig -- Craig Small GnuPG:1C1B D893 1418 2AF4 45EE 95CB C76C E5AC 12CA DFA5 Eye-Net Consulting http://www.enc.com.au/ MIEE Debian developer csmall at : enc.com.au ieee.org debian.org
