Submit News Alternative Tip Form

GNOME+Mono hackfest day 2+3

Notice: This post is more than a year old. It may be outdated.

A double whammy update, mostly since day 2+3 were largly just expanding on the work everyone started on day one.

Alan and Oliviers work on the bindings of GTK3 is now down from 150 errors and 25 warnings to around 5 errors. However Alan’s current approach basically turns the gir output into output the bindings generator we use already can understand and this has turned out to be limiting and hard to debug. So he started writing a new generator meaning there has been little progress in getting a full set of bindings. He will reuse a lot of code from the current gapi generator as I understand but this means we are unlikely to reach a stage where we can demo a “Hello world!” example by the end of the hackfest.

Jérémie and Bertrand have been doing a lot of work on DBus-sharp and according to Jérémie Banshee now runs without any of the workarounds we had in place for our current buggy ndesk-dbus implementation. DBus-sharp is compatible with ndesk-dbus so application developers will only have to make minor changes and can then remove a lot of workarounds. The end result should be nicer code and more stable applications. The unit testing for dbus-sharp has also been expanded greatly over ndesk-dbus.

Tim has been working hard on removing the last bit of C code from F-Spot, mainly since this is a limitation when porting F-Spot to non-Linux platforms but also because it currently makes developing F-Spot in MonoDevelop cumbersome. The code in question covers the image effects so Tim’s initial approach was an attempt to reuse the effects from Pinta (Olivier is a Pinta developer and was helpful in this effort). This however turned out to be suboptimal and instead he has opted for porting the C code directly to C#. This work is now nearly at a point where libfspot can be removed, even if it has cost many sleepless nights for him.

Jonathan started writing a specification for handling unsupported media players, which I will continue to work on today. Alan has volunteered to implement the resulting data collector application once we have the specification finished.

All in all we are very pleased with our effort so far and have enjoyed the long days in our, relatively, cold room here in lovely Brussels.