Aaron Meurer

Roots of large numbers

3

``` In [18]: a = solve(S("1768292677839237920489538677417507171630859375*x^109-2734577732179183863586489182929671773182898498218854181690460140337930774573792597743853652058046464"))[0] In [19]: print a 2*4213284111726743293784894879025636320185854716489383140024880463872**(1/109)*66741943359375**(6/109)/15 In [20]: b = S("2*2**(22/109)*3**(42/109)*5**(90/109)*7**(71/109)/15") In [24]: factorint(4213284111726743293784894879025636320185854716489383140024880463872) Out[24]: {2: 22, 7: 71} The a above should have been...

``` See the end of https://github.com/ipython/ipython/issues/2234 . ``` Original issue for #7050: http://code.google.com/p/sympy/issues/detail?id=3951 Original author: https://code.google.com/u/[email protected]/

Issues with assumptions

2

``` Just some things I am noticing about assumptions as I go through it. - The handlers are staticmethods. This is silly. They should be regular methods. - You should...

See https://github.com/sympy/sympy/pull/2256/files#L4R232 . This could be factored out of the fabfile so that it can be tested automatically on Travis. Original issue for #7033: http://code.google.com/p/sympy/issues/detail?id=3934 Original author: https://code.google.com/u/[email protected]/

sift

2

I just saw this: https://coderwall.com/p/fun3vq?utm_campaign=weekly_digest&utm_content=A+simple%2C+recruiter+free%2C&utm_medium=email How does this compare to our sift? ``` Original issue for #6997: http://code.google.com/p/sympy/issues/detail?id=3898 Original author: https://code.google.com/u/[email protected]/ ```

One of the promises of the plot_implicit algorithm was the ability to plot the graph as it is made. This sort of thing could also be done just for simple...

Doctest the cheatsheet

1

``` Thanks to a GCI student, we now have a cheatsheet in doc/cheatsheet (run make cheatsheet to build it). We should restructure it so that it can be doctested. Perhaps...

degree((x+1)**10000) takes too long

24

degree() should be able to return almost instantly, but something like degree((x+1)**10000) takes forever to run, because it expands the polynomial first. The function should be rewritten to compute the...

Apply 2to3 "idioms" fixer to code base

9

See http://docs.python.org/2/library/2to3.html#2to3fixer-idioms . If you run "2to3 -f idioms sympy", it will spit out a diff file with all the changes (note, make sure your 2to3 is the one that...

Implement the formulas from the matrix cookbook

19

See http://orion.uwaterloo.ca/~hwolkowi/matrixcookbook.pdf . These should all be implemented, e.g., with MatrixExpr. In particular, currently differentiation with respect to matrices is wrong because it doesn't handle transposes correctly. For example, compare...