Planning a deep refactoring
I’m planning a deep refactoring.
The more I read the codebase the mode it’s obvious it requires a deep refactoring. I won’t elaborate on the reasons. I’ve already talked enough about why for weeks in the mailing list.
This will happen as soon as the next stable v6.5.7 will be out.
I won’t merge anymore features during the coming cycle. If you intend to implement one, it’s not the good time to start it now.
As far as I can tell today, here are some of the planned tasks:
- better names for variables, objects, etc;
- improve comments, especially about the logic;
- better policy on objects;
- revamp inheritance of objects;
- revamp the factorization accross objects;
- better namespacing;
- don’t blindly inherit from library objects;
- usual patterns more strictly applied;
- seriously improve our debugging/hacking sessions.
For details, see TODO.rst
in the source tree.
Anyone interested might like to follow the work in progress with git.
Follow up: imapfw is made public.