forked from sympy/sympy
-
Notifications
You must be signed in to change notification settings - Fork 0
Plan for SymPy 1.0
Aaron Meurer edited this page Mar 12, 2011
·
2 revisions
Draft of our roadmap.
The items in each block represent what we should work on in each version and what needs to be finished in order to bump the version.
- good numpy and mpmath integration
- good Sage integration + tests (so that things like , or never occur again)
- All useful SymPy modules and all important features should be documented at http://docs.sympy.org/, so that anyone who tries to find if SymPy can do something, can easily find it in docs.
- assumptions
The next release after 0.6.6 should be a 0.7 release.
- Merge in new assumptions (hopefully completely replacing the old assumptions).
- Merge in the new Polys module.
- Merge in the 2010 Google Summer of Code projects (Risch integration, SAT solving, quantum mechanics, code generation)
- At some point in this release cycle, we will drop Python 2.4 support.
- Python 3 support.
- merge our experimental core sympyx, that basically should make sympy as fast as sympycore (or nearly as fast)
- speed sympy to the level of sympycore, merge all important things
- maybe add some (optional!) C/Cython module for speed
- all important things working, general polishing of things
- full test coverage (using tools like figleaf)
Well tested product. This is not the end, rather the opposite. Just a milestone on the way to be a good CAS in Python.
All other improvements will just bump the third number, i.e. 0.6.1, 0.6.2, 0.6.3, ... The priority is to merge all useful (and stable) things that anyone contributes immediatelly and also that sympy is usable and robust right now. The above plan is just the general direction, so that we know where we are standing.