Martin Vidner
Martin Vidner
Status: some stalled attempts to solve this, from Jaunary 2011, are in https://github.com/mvidner/ruby-dbus/tree/multithreading
OK, I've had a look at the library that will be first to use this, cockpit.js for D-Installer. [DBusClient.watch](https://cockpit-project.org/guide/latest/cockpit-dbus.html#cockpit-dbus-watch) says that `ObjectManager` is not necessary: > Internally these watches work...
Indeed. To reproduce it, ``` bash (unset DBUS_SESSION_BUS_ADDRESS DISPLAY; ruby -r dbus -e 'p DBus.session_bus') ``` I see, a bug fix broke a method that is not covered by tests:...
@imobachgs I see [the-installer is using EventMachine](https://github.com/yast/the-installer/commit/3098e80b59d9b768dcbb36952ed36760bea9dab0), which I have no experience with. Would it help to add proper EM support? Something like this PR?
The library currently does not care about closing the connection. That is bad form but I wonder whether it actually hurts in practice. You can get at the socket used...
Notes to self: Things to consider when adding `DBus::Connection#close`: - what about non-empty `@buffer`, do we care? - what about the singleton instances `DBus.system_bus`, `DBus.session_bus`? Should we allow reopening? Or...
Thanks for the report! In what context are you getting this error? Do you have a simple test case for it? BTW lps-it.fr seems to be down, but I copied...
Hi, thanks for the report! In ruby-dbus, thread safety is more of an accident than a goal. That said, I think I have found a workaround for your use case:...
Good to hear that! I'll document it and then close this myself.
You're right, UNIX_FD is not actually implemented. A major reason is that the spec is vague about the file descriptors are actually passed, and I was not determined enough to...