Hello, Marc Leeman, le mar. 14 oct. 2025 09:15:45 +0200, a ecrit: > There were a couple packaging changes in master wrt gi that were > released in unstable. I was holding off applying those until there was > a new upstream experimental release, but I'll push it today.
I'm not sure to understand: do you mean changes in unstable that fix this particular issue, which were not applied to experimental, thus showing the issue only in experimental, and it won't show up once the new upstream release is uploaded to unstable? Samuel > On Mon, 13 Oct 2025 at 21:17, Samuel Thibault <[email protected]> wrote: > > > > Control: reassign -1 gstreamer1.0 > > Control: retitle -1 1.27.2-1 broke Gst.init(None), making applications > > startup break > > Control: found -1 1.27.2-1 > > Control: affects -1 + orca > > Control: severity -1 grave > > > > Hello, > > > > Huey Chen, le dim. 12 oct. 2025 15:55:49 -0700, a ecrit: > > > File "/usr/lib/python3/dist-packages/orca/sound.py", line 118, in init > > > Gst.init(None) > > > ~~~~~~~~^^^^^^ > > > TypeError: Argument 1 does not allow None as a value > > > > Oops, API change on the gstreamer side: > > > > > ii gir1.2-gstreamer-1.0 1.27.2-1 > > > > in the experimental version. Simply running > > > > import gi > > from gi.repository import Gst > > Gst.init(None) > > > > breaks. > > > > https://codesearch.debian.net/search?q=Gst.init%28None%29&literal=1 > > > > seems to be showing a lot of Gst.init(None) calls, so perhaps the > > gstreamer1.0 should rather fix the API break? Or else a MBR should be > > done to let packages fix their call before breaking all of them. > > > > With regards, > > Samuel
