Hi,
>>"Julian" == Julian Gilbey <[EMAIL PROTECTED]> writes:
Julian> We should have a c++-compiler virtual package to match the c-compiler
Julian> package. At present, at least in potato, only g++ Provides this
Julian> virtual package, but there may be others. And policy should encode
Julian> current practice.
Isn't current practice not creating virtual packages until we
actually needs them? I can come up with hordes of potentially useful
virtual packages (actually, several per real package with a little
bit of effort). In the past, we have only created virtual packages
when there really were alternate ``concrete'' packages that could
provide the virtual package.
Indeed, I would lean to actually encode current practice
(which seems like we should first get another C++ compier, thne have
all such C++ compilers and dependents arrange to have a virtual
package name, and _then_ get policy to encode the current practice).
Policy should enerally follow, not lead.
manoj
--
Drilling for oil is boring.
Manoj Srivastava <[EMAIL PROTECTED]> <http://www.debian.org/%7Esrivasta/>
1024R/C7261095 print CB D9 F4 12 68 07 E4 05 CC 2D 27 12 1D F5 E8 6E
1024D/BF24424C print 4966 F272 D093 B493 410B 924B 21BA DABB BF24 424C
