Reprise is an easy to use Debian repository and package management tool - a simple, easy to setup, and daemon-less software with an intuitive interface.
Reprise is meant to be run on the actual package server, not the client machine uploading and managing packages. This way an unnecessarily complicated client-server model is avoided. No daemon is running and Reprise itself knows nothing about network communications.
With Reprise you can easily manage as many Debian repositories as you like.
All you need to do is to install Reprise itself and provide SSH access to the
machine it runs on. For simplicity reasons the classical Debian stages like
stable, testing, unstable etc. are left out. Instead it is recommended
to create repositories with different names.
Repositories can be updated from one another by simply overwriting. So if you are about to decide that the packages in your repository work properly together use the update command to overwrite your old releases and put your new development packages into a new repository.
Furthermore repositories can be pulled from other machines which might be useful to deliver packages to Reprise instances which might run, for example, on an EC2 instance.
Because Reprise will write files to /srv, you should run all of its
commands as root.
First, create a default gpg key if you don't already have one:
gpg --gen-key
This will take a long time, especially on virtual machines, which typically don't have much entropy available.
Note: Reprise depends on python3-docopt, which you can find in this PPA.
To bootstrap Reprise you need to upload an initial package to a server and
install it using gdebi (install gdebi-core, NOT the gdebi package). From
then on you can use Reprise to host its own packages and this way update
itself. Once Reprise is installed, you can initialize the server:
reprise init
See: reprise --help or man reprise
It's easy to use Reprise from the local command line with a simple shell alias. Just define:
alias rp='ssh <user>@<host> reprise'
Type rp and the Reprise help will be displayed. Use the rp command like
a local program and all commands will be sent to the server via SSH.
Uploading packages is simply done via scp. Copy the Debian package into
/srv/Reprise/incoming like:
scp <package> <user>@<host>:/srv/reprise/incoming/
Afterwards run reprise include <repository> <package> to make the package
available in your repository.
To benefit from your new repository as usual an apt source needs to be added.
The corresponding line can be generated with reprise source <repository>.
Since it is impossible to auto detect the FQDN of the server running Reprise
you still need to set the proper host name in the source line.
You might have also realized that SSH is used in the source line. This means that every computer that wants to access packages needs SSH access to the machine which runs Reprise. This is a simple way to provide security against unwanted access to the hosted packages. If public access via HTTP is wanted a regular web server needs to be set up.
The test suite is located in reprise/tests/. You can run the tests by
typing make test (or make coverage to also ensure full code coverage).
Reprise also comes with a Vagrantfile and Puppet manifest, letting you set up a testing environment using Vagrant:
vagrant up
vagrant ssh
make test
make deb