my current plans for wajig

Some stuff I might work on in future:

  • I want to start using argcomplete at some point, a more dynamic shell completion tool, replacing the current hand-rolled solution. I find shell programming hard, and argcomplete, which is inspired by optcomplete is a brilliant workaround.
  • Someone asked me to restore gjig, the GUI tool that I removed from wajig some releases ago. This will be an opportunity to learn GTK+ again. I did a bit of it, even contributing patches (Tracker project), but was never quite confident. It's very likely that I won't work on this... too much work.

But ultimately, I would love for the standard Debian packaging tools to get the the point where they make wajig irrelvant. After all, wajig was created to be a wrapper that helps ease the mess.

Here's the work I did in the past working on wajig, have a look at its tag. Off the top of my head:

  • improved, simplied packaging
  • port to Python 3
  • port to argparse (from getopt), which also meant proper handling of subcommands
  • increased usage of python-apt, instead of hand-crafted (and ugly) code
  • a few feature additions
  • (lots of) modernising the code
  • (lots of) cleaning up

And over 800 commits later, I am glad.