This bug report was migrated from our old Bugzilla tracker.

These attachments are available in the static archive:

• 0001-X11-Add-support-for-fd.o-Idle-Inhibition.patch (0001-X11-Add-support-for-fd.o-Idle-Inhibition.patch, text/plain, 2013-10-21 15:03:19 +0000, 6166 bytes)
• 0001-X11-Add-support-for-fd.o-Idle-Inhibition.patch (0001-X11-Add-support-for-fd.o-Idle-Inhibition.patch, text/plain, 2014-01-10 11:05:57 +0000, 7081 bytes)

Reported in version: HG 2.1
Reported for operating system, platform: Linux, x86_64

Comments on the original bug report:

On 2013-10-21 15:00:05 +0000, Bastien Nocera wrote:

The attached patch will fix the problem for GNOME 3, and any desktop environments that implement the specification.

On 2013-10-21 15:03:19 +0000, Bastien Nocera wrote:

Created attachment 1389
0001-X11-Add-support-for-fd.o-Idle-Inhibition.patch

Add support for the freedesktop.org Idle Inhibition specification,
as per:
http://people.freedesktop.org/~hadess/idle-inhibition-spec/

This makes screensaver inhibition work with GNOME 3 and other
desktop environments that implement the specification.

On 2014-01-07 20:15:19 +0000, Ryan C. Gordon wrote:

Do you know if there's a way to decide if the system has org.freedesktop.ScreenSaver support (without having to parse a bunch of XML through the D-Bus introspection API)?

Ideally, I'd like to know at runtime if we should use the old way or the new way, instead of blasting out both and hoping for the best. Ideally, we don't decide what we should do based on whether session_cookie is zero.

I don't know if that's actually solvable. Otherwise, this patch looks pretty good.

--ryan.

On 2014-01-08 12:19:07 +0000, Bastien Nocera wrote:

(In reply to Ryan C. Gordon from comment # 2)

Do you know if there's a way to decide if the system has
org.freedesktop.ScreenSaver support (without having to parse a bunch of XML
through the D-Bus introspection API)?

You don't need to parse XML. The D-Bus call will simply fail if there's no such interface implemented. I can make SDL_dbus_screensaver_inhibit() return a boolean and only "tickle" if that failed.

Ideally, I'd like to know at runtime if we should use the old way or the new
way, instead of blasting out both and hoping for the best. Ideally, we don't
decide what we should do based on whether session_cookie is zero.

I don't know if that's actually solvable. Otherwise, this patch looks pretty
good.

If the above is OK with you, I can make the changes.

On 2014-01-10 11:05:57 +0000, Bastien Nocera wrote:

Created attachment 1518
0001-X11-Add-support-for-fd.o-Idle-Inhibition.patch

The only thing I'm missing is a good application description. We currently have "My SDL application" which I guess is better than nothing, but another identifier of any sort would be useful.

On 2014-03-02 20:01:28 +0000, Ryan C. Gordon wrote:

This patch is now https://hg.libsdl.org/SDL/rev/fee216d949e3, thanks!

I'm planning to add an application metadata API at some point, which will help with the "My SDL application" string. For now, it's good enough.

--ryan.