[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Conflicting packages not of extra priority.



Somebody said that following the current policy would make "too many" 
packages to become extra. Others said that applying this policy "does not
scale", given the high number of packages in the distribution.

Well, as an exercise, this weekend I have tried to modify the override
file for hamm so that it conforms to policy (or what I and a lot of other
people think it is what policy says).

This has been the result:


# override file for the override file for hamm/main (Debian2.0r2)
# to make  required+important+standard+optional  self-consistent
#
libpthread0      extra    conflicts with libc6 (this package should be removed)
libpthread0-dev  extra    conflicts with libc6 (this package should be removed)
libg++272-dev    extra    conflicts with libstdc++2.8-dev
gobjc            extra    depends on egcc which is extra
kterm            extra    depends on locale-ja which is extra and "incorrect"
xcdroast         extra    depends on cdrecord which is extra
egcs-docs        extra    conflicts with gcc-docs which is optional
dhcp-beta        extra    conflicts with dhcp-relay-beta which is the client
gnats            extra    conflicts with gnats-user which is the client
journal-dev      extra    used only to build packages containing online magazine
lprng            extra    conflicts with lpr which is standard
nmh              extra    conflicts with mh which is standard
squid-novm       extra    conflicts with squid and needs a patched kernel
hyperref         extra    tetex-extra conflicts with it
xlockmore-gl     extra    conflicts with xlockmore and has security holes
cti-ifhp         extra    depends on lprng which is extra
yorick-gist      extra    special for people who want gist but not yorick
exim             extra    conflicts with mail-transport-agent provided by smail
eximon           extra    depends on exim which is extra
libpng0g-dev     extra    conflicts with libpng-dev provided by libpng2-dev
tcl7.6-dev       extra    conflicts with tcl-dev provided by tcl8.0-dev
tk4.2-dev        extra    depends on tcl7.6-dev which is extra
tkstep4.2-dev    extra    provides tk-dev with which tk8.0-dev conflicts
tkstep8.0-dev    extra    provides tk-dev with which tk8.0-dev conflicts
boa              extra    conflicts with httpd which apache should provide
cern-httpd       extra    conflicts with httpd which apache should provide
dhttpd           extra    conflicts with httpd which apache should provide
ncsa             extra    conflicts with httpd which apache should provide
wn               extra    conflicts with httpd which apache should provide
slang0.99.38-pic optional boot-floppies depends on it and it's optional
diald            optional dialdcost depends on it and it's optional
gmp2             optional python-mpz depends on it and it's optional
rpm		 optional rpm2html depends on it and it's optional
alien            optional rpm recommends alien
#
# Follows entries to make required+important+standard self-consistent
#
data-dumper      standard libnet-perl depends on it and it's standard
zlib1g           standard lynx depends on it and it's standard
libpng0g         standard tetex-bin depends on it and it's standard
libdb2           standard man-db depends on it and it's standard
ppp              standard ppp-pam depends on it and it's standard
libpam0g         standard ppp-pam depends on it and it's standard
libpam0g-util    standard libpam0g depends on it and it's standard
libpwdb0g        standard libpam0g-util depends on it and it's standard


In this example, I have modified only 43 package priorities.

Please note that this is just an example, but I think most (if
not all) of the changes are reasonable and well-justified.

Since hamm has 1525 packages, this would be only 2.8% of them, and this
is a *tiny* fraction.


I believe, therefore, that this policy does not require to change so many
package priorities.

Thanks.

-- 
 "914ebd589780e9e1d4d9067588887993" (a truly random sig)


Reply to: