<html> <head> <base href="https://bugzilla.gnome.org/" /> </head> <body><a href="page.cgi?id=describeuser.html&login=mclasen%40redhat.com" title="Matthias Clasen <mclasen@redhat.com>"> Matthias Clasen</a> changed <a class="bz_bug_link bz_status_NEW " title="NEW - wayland: Use g_get_prgname as the xdg surface application id" href="https://bugzilla.gnome.org/show_bug.cgi?id=746435">bug 746435</a> <table border="1" cellspacing="0" cellpadding="8"> <tr> <th>What</th> <th>Removed</th> <th>Added</th> </tr> <tr> <td style="text-align:right;">CC</td> <td> </td> <td>mclasen@redhat.com </td> </tr></table> <div> <a class="bz_bug_link bz_status_NEW " title="NEW - wayland: Use g_get_prgname as the xdg surface application id" href="https://bugzilla.gnome.org/show_bug.cgi?id=746435#c5">Comment # 5</a> on <a class="bz_bug_link bz_status_NEW " title="NEW - wayland: Use g_get_prgname as the xdg surface application id" href="https://bugzilla.gnome.org/show_bug.cgi?id=746435">bug 746435</a> from <a href="page.cgi?id=describeuser.html&login=mclasen%40redhat.com" title="Matthias Clasen <mclasen@redhat.com>"> Matthias Clasen</a> <pre>(In reply to Jonas Ådahl from <a href="show_bug.cgi?id=746435#c3">comment #3</a>) > (In reply to Matthias Clasen from <a href="show_bug.cgi?id=746435#c2">comment #2</a>) > > Review of <a href="attachment.cgi?id=299790&action=diff" name="attach_299790" title="wayland: Use g_get_prgname as the xdg surface application id">attachment 299790</a> <a href="attachment.cgi?id=299790&action=edit" title="wayland: Use g_get_prgname as the xdg surface application id">[details]</a> <a href='review?bug=746435&attachment=299790'>[review]</a> [review] [review]: > > > > If you look at gdk_pre_parse, the two are more or less the same. I'm not > > sure this is sufficiently better. > > Right, but GApplication sometimes override prgname which made it > occasionally work. > > > > > What we really need is the application id that gets passed as the second > > argument to gdk_wayland_window_set_dbus_properties_libgtk_only - but ware > > already using that, right ? At least if the application in question is using > > GtkApplication. > > True. We call gdk_wayland_window_set_dbus_properties_libgtk_only too late > though. We can make it call earlier (handle_window_realize instead of > handle_window_map) but then we should maybe duplicate the data in > gdkwindow-wayland.c and set it after mapping (as MetaWindow won't have been > created in mutter yet otherwise). Calling it in realize seems totally fine - it is where we do this under X. > > Regarding how well it works, testing random applications trying to get the > application id first from the GtkApplication id, then g_get_prgname, and > then gdk_get_program_class (): > > Correct (desktop file base name and application id): > gnome-maps (org.gnome.Maps) > gnome-contacts (org.gnome.Contacts) > glade (glade) > gnome-font-viewer (org.gnome.font-viewer) > evince (evince) > > Half correct (desktop file name vs resulting application id): > gucharmap (gucharmap.desktop vs org.gnome.Charmap) > eog (eog.desktop vs org.gnome.eog.ApplicationService) > ibus-setup (ibus-setup.desktop vs main.py) > > Using gdk_get_program_class () instead of g_get_prgname () makes the result > slightly worse (glade and evince will no longer be correct) One thing that plays into this is that the identity desktop file basename == busname is only mandatory for applications which have DBusActivatable=true in the desktop files, while desktop file basename == application id should always hold. Another thing to notice is that under X, the application id is not actually needed on the bus, from what I can see, since we also set _GTK_UNIQUE_BUS_NAME as a window property, and that seems to be what the shell uses (e.g. gucharmap does not take org.gnome.Charmap as a well-known name, yet its appmenu works fine).</pre> </div> <hr> You are receiving this mail because: <ul> <li>You are on the CC list for the bug.</li> </ul> </body> </html>