omxplayer should respect already existing dbus environment variables

[fredmorcos]

Currently, from what I understand from the omxplayer executable script, it only checks whether the OMXPLAYERDBUS* is set before executing a new dbus instance. I think DBUS_SESSIONBUS* environment variables should also be checked first, what is the reason to start an omxplayer-only dbus instance?

This is making it hard for me to run omxplayer and control it over dbus from my application. If omxplayer checks the DBUS_SESSIONBUS* variables as well, I can launch my application with dbus-launch and that bus will be used by my application and omxplayer as well.

Please let me know if more clarification is required.

[ingwinlu]

I would like to pile on this issue. Is there a reason why omxplayer has to have it's own dbus session instead of using normal user sessions?

EDIT: Just going to call omxplayer.bin directly, nvm

[matthijskooijman]

I was already running into this issue while working on the python-omxplayer-wrapper dbus control, see https://github.com/willprice/python-omxplayer-wrapper/issues/107

I would also think it would be useful if omxplayer would only start its own dbus if there is no session dbus present already, that would make it easier to talk to it without having to rely on the somewhat fragile reading of the PID/bus address files in /tmp.

[SolidTux]

I also ran into this issue and wondered why I couldn't connect to omxplayer. It would be at least good to document this (in the man page for example).