Re: [epg@progeny.com: Bug#154142: dhcp-client conflicts]
Previously Moshe Zadka wrote:
> Ummm.....if a C compiler doesn't support a /usr/bin/cc which supports -o
> and -c, it shouldn't "Provide: c-compiler"
A virtual package is a means to indicate a package provides a certain
interface, not some functionality. Functionality is useless if you
can't use it in a standard way. If a random package X can not rely
on and use expected behaviour by random package Y that provides a
virtual package Z without hardcoding specific on Z we might as well
ditch the concept of virtual packages.
For MTAs the standard interface is /usr/sbin/sendmail with a couple
of standard commandline options (LSB has a nice list). For a C compiler
the interface is /usr/bin/cc with a few common options (which definitely
include -c and -o).
If policy is currently unclear on this we should improve the policy
text. It definitely makes sense for each virtual package to specify
the exact interface it represents.
Wichert.
--
_________________________________________________________________
/wichert@wiggy.net This space intentionally left occupied \
| wichert@deephackmode.org http://www.wiggy.net/ |
| 1024D/2FA3BC2D 576E 100B 518D 2F16 36B0 2805 3CB8 9250 2FA3 BC2D |
--
To UNSUBSCRIBE, email to debian-policy-request@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Reply to: