Hi,

sympy 0.6.4 beta3 is out. Tarballs available here:

http://code.google.com/p/sympy/downloads/list

or from our front page.

This is the last beta and unless there show up some release critical
bugs, I expect to release the final 0.6.4 on Tuesday or Wednesday.

Changes since the last beta2 attached. Besides lots of bugfixes, sympy
now has a robust multivariate factorization. In some cases it's even
faster than Maxima via Sage:

sympy:

In [1]: time factor(2*x*y*z + x*y**2 + x*z**2 + y*x**2 + y*z**2 +
z*x**2 + z*y**2)
CPU times: user 0.04 s, sys: 0.00 s, total: 0.04 s
Wall time: 0.04 s
Out[2]: (x + y)⋅(y + z)⋅(x + z)

Sage:

sage: var("x y z")
(x, y, z)
sage: time factor(y*z^2 + x*z^2 + y^2*z + 2*x*y*z + x^2*z + x*y^2 + x^2*y)
CPU times: user 0.09 s, sys: 0.03 s, total: 0.12 s
Wall time: 2.12 s
(y + x)*(z + x)*(z + y)
sage: time factor(y*z^2 + x*z^2 + y^2*z + 2*x*y*z + x^2*z + x*y^2 + x^2*y)
CPU times: user 0.03 s, sys: 0.01 s, total: 0.04 s
Wall time: 0.21 s
(y + x)*(z + x)*(z + y)
sage: time factor(y*z^2 + x*z^2 + y^2*z + 2*x*y*z + x^2*z + x*y^2 + x^2*y)
CPU times: user 0.02 s, sys: 0.01 s, total: 0.04 s
Wall time: 0.19 s
(y + x)*(z + x)*(z + y)
sage: time factor(y*z^2 + x*z^2 + y^2*z + 2*x*y*z + x^2*z + x*y^2 + x^2*y)
CPU times: user 0.04 s, sys: 0.00 s, total: 0.05 s
Wall time: 0.21 s
(y + x)*(z + x)*(z + y)

That was implemented by Mateusz Paprocki. Great job again Mateusz!

In some other cases it takes about the same, in some more complex
cases it takes longer than maxima, but it always returns, and I think
it's a great start. There are still possible rooms for improvements,
as Mateusz says here:

http://code.google.com/p/sympy/issues/detail?id=500#c7


Ondrej

----------------------------

Ondrej Certik (31):
      0.6.4.beta2
      Remove the print statement from tests_latex.py
      Slow tests from core moved to sympy/slow_tests
      solvers.py: typo in documentation fixed
      Fixes the A*f(x).diff(x)*A problem (#1327)
      Make Matrix.jacobian() work with any vectors
      curvilinear_coordinates example added
      recursive trigsimp() implemented, thanks to Alan Bromborsky
      Fixes the trigsimp(2.0*sin(x)**2+2.0*cos(x)**2) problem (#1274)
      trigsimp() two hacks removed and done right
      Removing the hack from fraction() as it is not necessary anymore
      Added a new test for roots (#1305)
      solve(Eq(cos(x), sin(x)), x) raises a NotImplementedError
      ./setup.py test_doc now runs our testsuite
      Integral._eval_subs() implemented
      Make all examples run again (#1297)
      Examples documentation improved with regards to the PYTHONPATH issues
      Use systemwide sphinx (e.g. don't download it)
      Fixes the test_secondquant tests to use symbols instead of var
      guess_solve_strategy() improved to handle numbers as well
      solve() extended to handle equations of the form 4*x*(1 - a*x**(S(1)/2))
      solve( x**Rational(1,2) - 2, x) fixed
      Printing precision fixed all over sympy (#1052)
      Make PrettyPrinter use the profile mechanism
      Akshay and Aaron added to credits
      Integral.subs() fixed for integration variables
      Make Integral(exp(-x**2), (3, -oo, oo)) raise a ValueError
      Make lambdify() work with Integral
      setup.py: sympy.slow_tests added to the tests list
      setup.py: modules list refactored
      v0.6.4.beta3

Mateusz Paprocki (9):
      Implemented recursive dense multivariate polynomials
      Implemented factorization of x**n - 1 and x**n + 1 polynomials
      Improved sparse polynomial GCD in Z[X] by reducing polynomials
      Implemented more useful tools for polynomials in Z[X]
      Fixed corner case in heuristic GCD algorithm
      Implemented Eisenstein's criterion, improved factorization in Z[x]
      Implemented Wang's algorithm for multivariate factorization
      Use Wang's multivariate factorization algorithm in factor()
      Use sqrt() from mpmath in HEUGCD, if gmpy is not available

Aaron Meurer (2):
      Solvers: dsolve - Added Bernoulli case
      Fixed typo in docstring

Freddie Witherden (2):
      LaTeX: support multiple styles of inverse trig functions.
      Add unit tests for the new inverse trig styles.

Akshay Srinivasan (1):
      Fixed printing of 1/cos(x)*sin(x) to sin(x)/cos(x)

Andrew Docherty (1):
      Fixes the (1/(1+(a+b)*x)).series(x, 0) problem (#1342)

Fabian Seoane (1):
      roots: Just added a docstring reminding that we don't check the
correct result

Priit Laes (1):
      Added tests for issues 580 and 689.

Robert Kern (1):
      BUG: Fixes the issue 1263.

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"sympy" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [email protected]
For more options, visit this group at http://groups.google.com/group/sympy?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to