Loïc Minier wrote: > On Wed, Mar 18, 2009, Raphael Hertzog wrote: >> I also included DEB_VENDOR in the set of variables. This variable is not >> used currently (as I just introduced it with dpkg 1.15.0) but I expect it >> to become more used in the future for things like this: >> - enable additional patch/features depending on the vendor (while still >> sharing a common source package for better cooperation) >> - special purpose derivative could change the behaviour of helpers like >> debhelper/CDBS based on this value (for example to strip doc for >> Emdebian, to extract debug infos in .ddeb, ...) > > While I appreciate the effort of providing DEB_VENDOR, I'd like to > raise a design issue with this approach which we'd best fix now. > > If you implement conditional behavior in your rules, typically based on > lsb_release -is output: > if vendor is Ubuntu: > foo > elif vendor is Debian: > bar > you face a problem when you meet: > else: > > What behavior should one use here? There is a good use case for this that doesn't require a conditional, which is passing --with-package-name=$(DEB_VENDOR) to configure for packages that have this option (e.g. the GStreamer stack, that right now checks lsb_release -si output. It doesn't solve your concerns but perhaps you could do some common action or apply the Debian logic for anything else (which would be the same as not looking at DEB_VENDOR at all). Cheers, Emilio
Attachment:
signature.asc
Description: OpenPGP digital signature