Migrate from old assumptions to new assumptions

[mrocklin]

What is necessary to remove the old assumptions system?  

This issue is here to collect other issues.  It serves as root of an issue tree.  Search for blocked on issues.

Original issue for #6730: http://code.google.com/p/sympy/issues/detail?id=3631
Original author: https://code.google.com/u/109882876523836932473/
Referenced issues: #6786, #6731, #6732, #6733, #6734, #6738, #6735, #5820, #6783

[mrocklin]

New assumptions should be fast

**Blockedon:** sympy:3632  

Original comment: http://code.google.com/p/sympy/issues/detail?id=3631#c1
Original author: https://code.google.com/u/109882876523836932473/

[mrocklin]

Contradictory assumptions should raise an error

**Blockedon:** sympy:3633  

Original comment: http://code.google.com/p/sympy/issues/detail?id=3631#c2
Original author: https://code.google.com/u/109882876523836932473/

[mrocklin]

`is_attribute` syntax in core

**Blockedon:** sympy:3634  

Original comment: http://code.google.com/p/sympy/issues/detail?id=3631#c3
Original author: https://code.google.com/u/109882876523836932473/

[mrocklin]

Relational assumptions

**Blockedon:** sympy:2721  

Original comment: http://code.google.com/p/sympy/issues/detail?id=3631#c4
Original author: https://code.google.com/u/109882876523836932473/

[mrocklin]

Backwards compatibility with old assumptions

**Blockedon:** sympy:3635  

Original comment: http://code.google.com/p/sympy/issues/detail?id=3631#c5
Original author: https://code.google.com/u/109882876523836932473/

[mrocklin]

Several modules depend on the old assumptions structure

**Blockedon:** sympy:3636  

Original comment: http://code.google.com/p/sympy/issues/detail?id=3631#c6
Original author: https://code.google.com/u/109882876523836932473/

[mrocklin]

Discussion should not occur on this issue.  

Please add your comment to the relevant sub-issue.

If the appropriate sub-issue for your comment does not exist please do the following.

1. Create a new sub- issue
2 . List that issue here by adding a new comment
  a. Add the new issue number to the "Blocked on" list.  
  b. Add title of the new issue in the comment (this is not done automatically)

Original comment: http://code.google.com/p/sympy/issues/detail?id=3631#c7
Original author: https://code.google.com/u/109882876523836932473/

[asmeurer]

A very important thing is issue 5295 , "Document assumptions." We need to be very precise about what each assumption means, so that we don't have any logical inconsistencies like the one from issue 5976 (see also all the other blocking issues from issue 5295 ).  Also, people just will just use some assumption assuming that it means a certain thing, but it's not clear if it really does, and sometimes it actually doesn't (especially with esoteric assumptions like infinitesimal, which IMHO should not make its way into the new system, at least in its current form).

**Blockedon:** 5295  

Referenced issues: #5976, #5295
Original comment: http://code.google.com/p/sympy/issues/detail?id=3631#c8
Original author: https://code.google.com/u/[email protected]/

[asmeurer]

Issue 1047 has been merged into this issue.

**Cc:** [email protected]  

Original comment: http://code.google.com/p/sympy/issues/detail?id=3631#c9
Original author: https://code.google.com/u/[email protected]/

[asmeurer]

The old issue for the new assumptions was issue 4146 .  There's a lot of discussion there, and a lot of the things discussed are either bad ideas, or are already implemented, but I'm noting it here for completeness.  But if you're looking to work on assumptions, focus your energies on the blocking issues of this one.

Referenced issues: #4146
Original comment: http://code.google.com/p/sympy/issues/detail?id=3631#c10
Original author: https://code.google.com/u/[email protected]/

[asmeurer]

Just stumbled across an old assumptions issue that brings up another point: what to do about user-defined assumptions.  See issue 3447 "User-defined assumptions".

**Blockedon:** 3447  

Referenced issues: #3447
Original comment: http://code.google.com/p/sympy/issues/detail?id=3631#c11
Original author: https://code.google.com/u/[email protected]/

[asmeurer]

By the way, who wants to do a scan of all issues tagged "Assumptions", and see how they relate to the new assumptions?  As far as I am concerned, the old assumptions are deprecated, and we shouldn't waste energy on them.

Original comment: http://code.google.com/p/sympy/issues/detail?id=3631#c12
Original author: https://code.google.com/u/[email protected]/

[rlamy]

I think it would be a serious mistake to throw away a working system (the old assumptions) for the sake of an untested one with known deficiencies (the new assumptions). It would probably be more productive to view the new assumptions as complementary to the old ones, and treat it as a given that they are intrinsically more powerful, but slower and more cumbersome to use than the old ones.

Original comment: http://code.google.com/p/sympy/issues/detail?id=3631#c13
Original author: https://code.google.com/u/101272611947379421629/

[mrocklin]

I've created an issue for that discussion.  Again, I'd like for this issue to be mostly a table of contents.

Should we remove old assumptions?

**Blockedon:** sympy:3639  

Original comment: http://code.google.com/p/sympy/issues/detail?id=3631#c14
Original author: https://code.google.com/u/109882876523836932473/

[asmeurer]

Get rid of confusing assumptions

**Blockedon:** sympy:3684  

Original comment: http://code.google.com/p/sympy/issues/detail?id=3631#c15
Original author: https://code.google.com/u/[email protected]/

[asmeurer]

Redundant assumptions.  

This is a very important principle in SymPy in general, and we need to make sure it is considered when designing things.

**Blockedon:** sympy:3687  

Original comment: http://code.google.com/p/sympy/issues/detail?id=3631#c16
Original author: https://code.google.com/u/[email protected]/

[asmeurer]

An interesting discussion about assumptions of explicit objects (like is_zero, and generalizations of it) on issue 6105 and issue 6835 .

**Blockedon:** 6105 sympy:3736  

Referenced issues: #6105, #6835
Original comment: http://code.google.com/p/sympy/issues/detail?id=3631#c17
Original author: https://code.google.com/u/[email protected]/