<html> <head> <meta content="text/html; charset=UTF-8" http-equiv="Content-Type"> </head> <body bgcolor="#FFFFFF" text="#000000"> <div class="moz-cite-prefix">On 09/04/2012 05:47 PM, Caolán McNamara wrote: </div> <blockquote cite="mid:1346773650.2694.573.camel@Brinn" type="cite"> <blockquote type="cite"> <pre wrap="">1. Port my code to dbus-glib and always build dbus-glib? Or port my code to dbus-glib and leave everything else as is, meaning bluetooth on Linux needs the user to use --enable-dbus? </pre> </blockquote> <pre wrap=""> I feel this is the best approach. Except that I'll toggle enable-dbus on by default rather than off when I get the chance. Which would mean that the rhel-4 based universal-build-generating box would either have to start using --disable-dbus or get upgraded to something less archaic. </pre> </blockquote> I've ported my code to dbus-glib now (wrapped in ENABLE_DBUS) -- for the moment I don't think it's too critical to have dbus enabled, but would be nice to have for 3.7. I've found that the other code using dbus ( vcl/unx/gtk/window/gtkframe.cxx ) spits out: ** (soffice:28499): WARNING **: Inhibit method failed ** (soffice:28499): WARNING **: Inhibit problem : The name org.gnome.SessionManager was not provided by any .service files ** (soffice:28499): WARNING **: Invalid cookie (Presumably since I'm on kde.) I'm not building with any of the debug options (I only have debug=t for sd). Would it not be more appropriate to use g_debug rather than g_warning for these messages, as currently they'll appear on any system that doesn't have gnome? Cheers, Andrzej </body> </html>