[gedit-list] gedit is unmaintained, some thoughts


gedit is no longer maintained, I've added it to this wiki page:

Any developer interested to take over the maintenance of gedit?

Here are my thoughts:

I think a high-priority issue is that there are no checks to see if a
plugin is compatible with the gedit version. Currently enabling a plugin
can make gedit to crash. See my comment on this libpeas feature request:

Once the above is done, what I would do in gedit is to port the core to
Tepl (and continuing developing Tepl and making the gedit core features
re-usable). See:
By making the code re-usable and putting it in a library, the code is
shared among several applications, which means that it has a greater
chance to be better maintained, and during a longer time. Also by
developing a library, the code is cleaner, less entangled, better
documented, and it's possible to (unit-) test each component
individually. Of course it's possible to do that for an application as
well, but it's not really how gedit is implemented.

Note that I still develop Tepl, so over time it has more and more
features. If gedit is not developed during several years, maybe it'll be
possible to remove a big amount of code from gedit by porting it to
Tepl. By doing so, it'll maybe be difficult to keep the gedit API
backward-compatible, hence the need to check if a plugin is compatible.

BTW while the gedit core is written in C (with a bit of Objective-C for
Mac OS X support), some plugins are written in Vala or Python. If you
take over gedit maintenance, you'll need to deal with 4 programming
languages (without counting the build system). The Python code is not
compiled, so when doing refactorings in gedit core, good luck to port
all the plugins (the Python code is also less "greppable" than C). At
least with Vala there is a compiler, even if I would not recommend Vala.

Also by contributing to gedit (probably for free), you help this guy
selling gedit on Mac:
If you read this, please don't buy gedit there, there is a free (but
older) version here:

Still interested?

Note that with GTK+ 3 being now stable, gedit will probably still
continue to work fine during a long time, even if not perfect.

If gedit dies, I think there is a more general lesson for all GTK+
applications: it is important to write more libraries, sharing the code
and maintenance among several similar applications. GtkSourceView is
still maintained, but gedit has still more code than GtkSourceView.
Before I contribute to GtkSourceView, there was 8000 lines of code in
gedit for the file loading and saving (just the backend part, not the
frontend). Do you seriously think that only gedit needed to load and
save files with GtkSourceView? There are other text editors out there…
See for example Anjuta (also a big codebase no longer maintained), and
now gnome-builder doing the same mistake (developing in its corner a lot
of text editor features; do you really think that the Vim mode is useful
only to gnome-builder?!).

And this applies not just to text editors. How many times the wheel has
been reinvented for music players? Photo managers? IRC/chat clients?
Showing the wheather? Etc.

/me back to Tepl development.


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]