On a fresh build on 64-bit Fedora 10 (Opteron), only one doctest
fails:
sage -t "devel/sage/sage/matrix/matrix2.pyx"
due to numerical noise as follows:
**********************************************************************
File "/scratch/sage-4.0.rc0/devel/sage/sage/matrix/matrix2.pyx", line
4964:
sage: m.change_ring(CDF).eigenvalues() # again for display
purposes
Expected:
[10.463115298 - 4.98835989975e-16*I, 7.42365754809 -
5.05298436852e-16*I, 3.36964641458 + 5.59670199836e-17*I,
1.25904669699 + 4.34508443713e-17*I, 0.00689184179485 -
5.75358699674e-17*I, 0.330700789655 + 5.1816322259e-16*I]
Got:
[10.463115298 - 3.57021884933e-16*I, 7.42365754809 +
2.91121322571e-17*I, 3.36964641458 - 1.54788874696e-16*I,
1.25904669699 - 5.01646753379e-17*I, 0.00689184179485 -
3.97030024181e-16*I, 0.330700789655 + 1.32245788594e-16*I]
**********************************************************************
Kiran
On May 21, 10:14 am, mabshoff <mabsh...@googlemail.com> wrote:
> Hi,
>
> I gotta run to the airport, but sources and upgrade bits of 4.0.rc0
> are in
>
> http://sage.math.washington.edu/home/mabshoff/release-cycles-4.0/rc0/
>
> Once I get online I will post more info, but I am sure this will take
> a couple hours. Sorry.
>
> Cheers,
>
> Michael
>
> Merged in Sage 4.0.rc0:
>
> #6104: Michael Abshoff: Fix Solaris specific build issue for
> libfplll.spkg [Needs review]
>
> #3932: Ryan Dingman: generate size n trees in linear time [Reviewed by
> Robert Miller]
> #4875: Marshall Hampton: Polyhedron.show(fill=True) fails [Reviewed by
> David Perkison]
> #5076: David Roe: bug with (coercion in?) p-adic extension fields
> [Reviewed by Robert Bradshaw]
> #5120: Nicolas Thiery: An alternative implementation of the Unique
> Representation design pattern [Reviewed by David Roe]
> #5236: David Roe: x^(-pm) in ramified extensions of Qp [Reviewed by
> Kiran Kedlaya]
> #5554: Nick Alexander: implement cholesky_decomposition for matrices
> other than RDF [Reviewed by Rob Beezer, Michael Abshoff]
> #5581: Marshall Hampton: Face lattices and f-vectors for polytopes;
> bug fixes for polyhedra.py [Reviewed by David Perkison]
> #5619: Minh Van Nguyen: doctests for error messages in groebner_fan.py
> [Reviewed by Marshall Hampton]
> #5621: Minh Van Nguyen, Mike Hansen: doctests for calculus.py
> [Reviewed by John Palmieri]
> #5753: Burcin Erocal: update to pynac 0.1.4 [Reviewed by Mike Hansen]
> #5777: Burcin Erocal: update to pynac 0.1.6 [Reviewed by Mike Hansen]
> #5783: Nicolas Thiery: Lazy attribute fixes and improvements [Reviewed
> by David Roe]
> #5582: Robert Bradshaw: Coercion from float to QQ is missing [Reviewed
> by Jason Bandlow, Mike Hansen]
> #5837: Luiz Berlioz: bug in rational_diagonal_form() from
> QuadraticForm class [Reviewed by John Cremona, Gonzalo Tornaria]
> #5879: Dan Bump: Added crystal of letters for type E [Reviewed by Anne
> Schilling, Michael Abshoff]
> #5907: Gonzalo Tornaria, Carl Witty: incorrect fast_float evaluation
> of constant polynomials [Reviewed by Mike Hansen]
> #5930: Mike Hansen, William Stein, Carl Witty, Robert Bradshaw: switch
> from maxima to pynac for core symbolic manipulation system [Reviewed
> by Carl Witty, Robert Bradshaw, William Stein, Mike Hansen, Tom
> Boothby, Michael Abshoff]
> #5940: Emily Kirkman: graph theory -- bug in is_circular_planar
> [Reviewed by Robert Miller]
> #5944: Burcin Erocal: fix a nasty pynac bug in exp [Reviewed by Mike
> Hansen]
> #5954: Jon Hanke: Added documentation/doctests for all quadratic form
> genus symbol routines [Reviewed by Gonzalo Tornaria]
> #6088: Nicolas Thiery: LatticePolytope: Removed a try/catch which
> could involuntarily hide exceptions from lower code [Reviewed by
> Andrey Novoseltsev]
> #6029: Robert Bradshaw: bug in floor() for python types [Reviewed by
> Mike Hansen]
> #6035: Wilfried Huss: Don't draw vertical asympotes [Reviewed by
> Marshall Hampton]
> #6036: Kwankyu Lee: a bug in polynomial() for PolynomialRing(GF(5),2)
> [Reviewed by Mike Hansen]
> #6037: Jon Hanke: Major Upgrade to QuadraticForm Local Density
> Routines [Reviewed by Gonzalo Tornaria]
> #6040: Jon Hanke: Added Doctests for QuadraticForms methods [Reviewed
> by Gonzalo Tornaria]
> #6042: John Cremona: Bring doctests of modular/modsym/ambient.py up to
> 100% [Reviewed by David Loeffler, Alex Ghitza]
> #6048: John Palmieri: fix non-ASCII character in mpolynomialsystem
> [Reviewed by Martin Albrecht]
> #6049: Jason Grout: bitset complement zeroes out last word if the
> bitset is a multiple of the word size [Reviewed by Carl Witty]
> #6052: Gonzalo Tornaria: partial_fraction_decomposition is broken for
> irreducible denominators [Reviewed by Mike Hansen]
> #6054: Mike Hansen: fix display2d with ecl on OSX [Reviewed by Michael
> Abshoff]
> #6059: Gonzalo Tornaria: speed regresion in hilbert_symbol after #5834
> [Reviewed by John Cremona]
> #6060: Fredrik Johansson: divmod fails when one argument is Integer
> and the other is a Python int [Reviewed by Mike Hansen]
> #6061: William Stein: refresh the pickle jar [Reviewed by Michael
> Abshoff]
> #6062: Gonzalo Tornaria: QQ method is_padic_square fails when argument
> is python int [Reviewed John Cremona]
> #6064: Gonzalo Tornaria: allow rationals in kronecker_symbol and
> legendre_symbol [Reviewed John Cremona]
> #6065: Carl Witty: implement some missing trig methods on
> RealIntervalFieldElement [Reviewed by Mike Hansen]
> #6066: Radoslav Kirov: constructing Sage graphs is slow compared to
> NetworkX graphs [Reviewed by Robert Miller]
> #6070: David Loeffler: Get doctest coverage in sage/modular to 100%
> [Reviewed by John Cremona]
> #6077: Marshall Hampton: simplicial complex method for polytopes
> [Reviewed by John Palmieri]
> #6080: Fredrik Johansson: Add utility methods to Integer (needed for
> mpmath) [Reviewed by Mike Hansen]
> #6082: Robert Bradshaw: realloc called too often for Integer
> construction/deconstruction [Reviewed by William Stein]
> #6086: David Perkinson: fixed laplacian_matrix in graph.py [Reviewed
> by Emily Kirkman]
> #6107: Mike Hansen: misc. fixes for symbolics for 4.0.rc0 [Reviewed by
> William Stein]
--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~----------~----~----~----~------~----~------~--~---
