Hi everyone, below you will find the updates from the Desktop team from the last week.
If you’re interested in discussing a topic please start a thread in the Desktop area of the Community Hub (this site).
We also have our weekly meeting on IRC. We meet on Tuesday at 13:30 UTC in #ubuntu-desktop on Freenode. There will be an “Any Other Business” section at the end where you are welcome to raise topics. These topics might be discussed during the meeting, or afterwards depending on the time, depth of conversation, topic and so on.
Last week’s notes are here: Monday 17th June 2019
Next week’s notes are here: Monday 1st July 2019
snapd user session agent (PR #6954):
- Received in-principle approval to have the user session agent run as an independent process to
snap userd: trying to merge the two basically broke the D-Bus services provided by userd on systems where the session bus wasn’t launched by a user instance of systemd (i.e. Ubuntu 14.04, Ubuntu 16.04, CentOS 7).
- Asked for and received a second round of review feedback that I am working my way through. This included requests to split some changes out into their own PRs, leading to the creation of #7027 and #7028 (both since merged).
- I hope to get the main PR merged this week.
Desktop notifications for the session agent:
- One of the features the session agent will need is the ability to post desktop notifications. There are a number of Go libraries to handle this, but none felt appropriate. Most were simple wrappers for
notify-send, with no support for actions, and one was part of a larger glib/gobject cgo binding.
- So I put together a rudimentary native implementation using the D-Bus binding already in use by snapd. At 350 lines, I’ve got something that approximates the
GNotification API and supports both the GTK and freedesktop.org notification APIs. This covers all desktop environments we care about, and allows for persistent notifications on GNOME 3 based desktops.
- While working on that, I discovered a possible deadlock in the godbus code. I submitted PR #161 to fix that, which has been accepted upstream.