- Basic information about each service is available for consumers
- Information about whether service is enabled or disabled is available
- Interested parties may listen to enabling/disabling of a service
- Information about some core services are made available
James added the Apache Headers module to the makefile, but not the
FreedomBox-Setup configuration script. However, it's ridiculous to
have multiple locations for that same logic, so now the makefile uses
the setup script for the Apache configuration instead.
Also, the modules are now fancy and alphabetized.
There's no point to having two copies of what's essentially the same
file. In the long term, this means the distributions will need to
patch out cfg.py directories instead of plinth.sample.fhs.config. So,
if diff plinth.sample.config plinth.sample.fhs.config reveals
anything relevant, put that in your patch.
Apache configuration was missing a few modules, and the server needs
to be restarted after those modules are enabled. Also, the server
needs to be reloaded after the Plinth site is enabled.
Author: Tzafrir Cohen <tzafrir@debian.org>
Description: These things are easier to install with dh
* Python modules: fighting with dh_python2 is tough
(it changes the /etc/ symlink, for isntance)
* Let's just install man pages ourselves for now.
* symlinks: with dh_link
The makefile has three new targets:
1. hosting: Publish repository and automagically build the archives to
serve to clients.
2. current-checkout.tar.gz: Create an archive of the current project
directory.
3. current-repository.tar.gz: Create an archive of the current project
directory along with the source repository metadata so that the
archive is a full checkout of the project.
* change to u. notation for util, since from foo import * is evil
* make import vendor.foo possible
* vendor dir is part of the repo now, no need to mkdir
* use the vendor.foo notation
* Read python.config from script dir
* Make dirs for pid and run
Maybe making DESTDIR/var/run during make install is not needed because
we can expect the target system to have /var/run, but if it's missing
the CherryPy engine will fail when it tries to write the pid. So I
added it.
* Remove commented out git pull lines. We shouldn't update so automatically.
* Use rsync because it gives us --exclude
* Put docs where the webserver can find them
