pkgin is aimed at being an apt / yum like tool for managing pkgsrc binary packages. It relies on pkg_summary(5) for installation, removal and upgrade of packages and associated dependencies, using a remote repository.
Many so-called GNU/Linux distributions provide a convenient way of searching, installing and upgrading software by using binary archives found on "repositories". NetBSD, and more widely, all operating systems relying on pkgsrc have tools like pkg_add and pkg_delete, but those are unable to correctly handle binary upgrades, and sometimes even installation itself.
This is the purpose of pkgin, to provide the user a convenient way to handle binary packages, using the same working mechanisms than tools like apt-get.
As pkgin CVS code may change quite deeply, please do not forget to look at the CHANGES file if you chose to track development version.
pkgin is developed using the C language. It only depends on databases/sqlite3, a lightweight database used to keep a good speed on older architectures. SQLite has no dependencies, you can simply install it with pkg_add:
# PKG_PATH=ftp://ftp.<mirror>.NetBSD.org/pub/pkgsrc/packages/NetBSD/<arch>/<version>/All pkg_add -v sqlite3
Pkgin is available via CVS, pkgsrc-wip and is available in pkgsrc as pkgtools/pkgin.
While installing via pkgsrc is the preferred method, you may want to track CVS version to help improving pkgin's quality. To check it out, use the following commands:
$ cvs -d :pserver:anoncvs@cvs.gcu.info:/cvs login passwd: anoncvs $ cvs -d :pserver:anoncvs@cvs.gcu.info:/cvs co pkgin
Invoke the configure script, for example:
$ ./configure --prefix=/usr/pkg --with-libraries=/usr/pkg/lib --with-includes=/usr/pkg/include
And finally build the binary:
$ make
pkgin is ready to run !
First thing to do before using pkgin is to setup one or more repositories in the ${PREFIX}/etc/pkgin/repositories.conf file :
$ echo ftp://ftp.fr.netbsd.org/pub/pkgsrc/packages/NetBSD/i386/5.0/All > /usr/pkg/etc/pkgin/repositories.conf
Now you can build the initial database:
# pkgin update
This operation will download the pkg_summary(5) file from the repository and then populate the SQLite database.
Once this is completed, pkgin is fully functionnal.
The pkgin update command must be run regularly in order to keep the database synchronized with the repository.
List available packages:
# pkgin avail
List installed packages:
# pkgin list
Search for a regular expression in the database:
# pkgin search foo.*bar
Install a package and its dependencies:
# pkgin install foo
Install many packages and their dependencies:
# pkgin install foo bar baz
Remove a package and its reverse dependencies:
# pkgin remove foo
Upgrade a package:
# pkgin install foo
Remove orphan dependencies:
# pkgin autoremove
Mark a package as "keepable", i.e. a package that probably has not been installed by pkgin but that is not an orphan dependency:
# pkgin keep foo
Mark a package as "non-keepable":
# pkgin unkeep foo
Upgrade keepable packages to their newer versions present in the repository. If the installed dependencies match the listed needed dependencies, don't upgrade them:
# pkgin upgrade
Upgrade all packages to their newer versions present in the repository:
# pkgin full-upgrade
Delete downloaded packages from the cache directory:
# pkgin clean
Show package direct dependencies:
# pkgin show-deps foo
Show package full dependency tree:
# pkgin show-full-deps foo
Show package reverse dependency tree (packages depending directly or indirectly):
# pkgin show-rev-deps foo
Assume "yes" for all questions:
# pkgin -y install foo
Every command has a shortcut, pkgin -h will show them.
pkgin is known to work and have been tested under the following platforms :
Please do not hesitate to contact me, either by mail at imil@NetBSD.org or iMil on the Freenode IRC Network.