Completely untested modules in sympy.polys

[rlamy]

The modules densepolys, sparsearith, sparsebasic, sparsepolys and sparsetools have 0% coverage by the test suite, which means that they aren't even imported. They don't have any doctest either.

Quite unsurprisingly, they don't seem to work correctly:

In [1]: from sympy.polys.densepolys import DensePoly

In [2]: DensePoly(x, x)
---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)

/media/sda2/Boulot/Projets/sympy-git/<ipython console> in <module>()

/media/sda2/Boulot/Projets/sympy-git/sympy/polys/densepolys.pyc in __init__(self, rep, dom, lev)
     10     def __init__(self, rep, dom, lev=None):
     11         if lev is None:
---> 12             rep, lev = dmp_validate(rep)
     13 
     14         self.rep = rep

NameError: global name 'dmp_validate' is not defined

In [3]: from sympy.polys.sparsepolys import SparsePoly

In [4]: SparsePoly(x, x)
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)

/media/sda2/Boulot/Projets/sympy-git/<ipython console> in <module>()

TypeError: __init__() takes at least 4 arguments (3 given)


So they should either be fixed, documented and tested before the release, or be removed entirely (since they aren't used at all, they're probably not that useful).

Original issue for #5470: http://code.google.com/p/sympy/issues/detail?id=2371
Original author: https://code.google.com/u/101272611947379421629/
Original owner: https://code.google.com/u/101069955704897915480/

[asmeurer]

I think at least some of this (like the sparse stuff) is WIP code that isn't finished yet.

**Cc:** matt...@gmail.com  

Original comment: http://code.google.com/p/sympy/issues/detail?id=2371#c1
Original author: https://code.google.com/u/asmeurer@gmail.com/

[asmeurer]

**Labels:** Polynomial  

Original comment: http://code.google.com/p/sympy/issues/detail?id=2371#c2
Original author: https://code.google.com/u/asmeurer@gmail.com/

[asmeurer]

These are not imported into the global namespace with "from sympy import *", so I don't see too much danger in leaving them.  I am pretty sure that it is just WIP code.

Mateusz, did you want to work on this?  I am postponing otherwise.

**Labels:** -Milestone-Release0.7.0 Milestone-Release0.7.1  

Original comment: http://code.google.com/p/sympy/issues/detail?id=2371#c3
Original author: https://code.google.com/u/asmeurer@gmail.com/

[mattpap]

This code will replace polyclasses.py (there is even preliminary support for this in polyoptions.py). I'm fine with postponing, because this will require still some work and it will be convenient to work on this in parallel with merge of lpoly1.

**Owner:** matt...@gmail.com  
**Cc:** -matt...@gmail.com  

Original comment: http://code.google.com/p/sympy/issues/detail?id=2371#c4
Original author: https://code.google.com/u/101069955704897915480/

[rlamy]

This is public code, since it's accessible via public names, and the Repr option is actually imported but can't work. I really think we shouldn't release broken code if we can avoid it. It's easy to remove this now and it'll be easy to add it back in: https://github.com/sympy/sympy/pull/372

**Labels:** NeedsReview  

Original comment: http://code.google.com/p/sympy/issues/detail?id=2371#c5
Original author: https://code.google.com/u/101272611947379421629/

[asmeurer]

I'd like to do a small 0.7.1 release with IPython 0.11 support, so these will be postponed until 0.7.2.

**Labels:** Milestone-Release0.7.2  

Original comment: http://code.google.com/p/sympy/issues/detail?id=2371#c6
Original author: https://code.google.com/u/asmeurer@gmail.com/

[rlamy]

Well we released with these broken modules, so this is moot.

**Status:** WontFix  

Original comment: http://code.google.com/p/sympy/issues/detail?id=2371#c7
Original author: https://code.google.com/u/101272611947379421629/

[vks]

I think we should still fix them and add tests, or remove them.

**Status:** Verified  

Original comment: http://code.google.com/p/sympy/issues/detail?id=2371#c8
Original author: https://code.google.com/u/Vinzent.Steinberg@gmail.com/

[rlamy]

I think you mean "Accepted", not "Verified". The last time, Aaron and Mateusz thought it was useful to release broken code, but I guess we could reopen the discussion for the upcoming release. I still don't see what we lose in removing them (isn't it the whole point of using a VCS that it's easy to go back to a previous version of the code?) and they do cause problems, such as causing a ton of warnings in pyflakes (e.g. when running './setup.py audit').

**Status:** NeedsDecision  
**Labels:** -NeedsReview  

Original comment: http://code.google.com/p/sympy/issues/detail?id=2371#c9
Original author: https://code.google.com/u/101272611947379421629/

[rlamy]

**Blocking:** 4555  

Referenced issues: #4555
Original comment: http://code.google.com/p/sympy/issues/detail?id=2371#c10
Original author: https://code.google.com/u/101272611947379421629/

[asmeurer]

**Blocking:** 5958  

Referenced issues: #5958
Original comment: http://code.google.com/p/sympy/issues/detail?id=2371#c11
Original author: https://code.google.com/u/asmeurer@gmail.com/

[sympy-issue-migrator]

I'm running into exactly the same kind of problem as Ronan: I'm getting lots of distracting unhelpful diagnostics about code that isn't finished enough that it's even supposed to work. densepolys.py alone is spamming me with over 100 errors (missing dmp_ functions) and another 100 warnings (first parameter of class functions should be named "self").

I second removing them from the 'master' branch, until they have matured enough to be reintegrated again.
If we really want this code in the official codebase, create an 'untested' branch and move the code there.

Original comment: http://code.google.com/p/sympy/issues/detail?id=2371#c12
Original author: https://code.google.com/u/103802103633477003048/

[rlamy]

Second attempt: https://github.com/sympy/sympy/pull/1283

**Labels:** NeedsReview  

Original comment: http://code.google.com/p/sympy/issues/detail?id=2371#c13
Original author: https://code.google.com/u/101272611947379421629/

[smichr]

OK, Ronan, I have now read here the rational for removing the things that you removed. I am in agreement that until they are mature enough to include they should not be included.

Original comment: http://code.google.com/p/sympy/issues/detail?id=2371#c14
Original author: https://code.google.com/u/117933771799683895267/

[smichr]

They are now removed.

**Status:** Fixed  

Original comment: http://code.google.com/p/sympy/issues/detail?id=2371#c15
Original author: https://code.google.com/u/117933771799683895267/

[jrioux]

**Blocking:** -sympy:1456 -sympy:2859 sympy:1456 sympy:2859 sympy:1456  

Original comment: http://code.google.com/p/sympy/issues/detail?id=2371#c16
Original author: https://code.google.com/u/102137482174297837682/