Hello folks,
Sage 2.10.1 has been released on February 2nd, 2008. It is available at
http://sagemath.org/download.html
* About Sage (http://www.sagemath.org)
Sage is developed by volunteers and combines over 75 open source
packages. It is available for download from sagemath.org and its
mirrors in source or binary form. If you have any questions and/or
problems please report them to the google groups sage-devel,
sage-support, sage-forum or sage-newbie. You can also drop by in
#sage-devel on irc.freenode.net.
---------------------------------------------------------------------
At least the following 27 developers contributed to this release:
* Michael Abshoff
* Martin Albrecht
* Nick Alexander
* Robert Bradshaw
* Michael Brickenstein
* Craig Citro
* Timothy Clemans
* John Cremona
* Burcin Erocal
* Alex Ghitza
* Jason Grout
* Mike Hansen
* David Harvey
* David Joyner
* Josh Kantor
* Simon King
* David Kohel
* Robert Miller
* Joel Mohler
* Willem Jan Palenstijn
* Yi Qiang
* Denis Simone
* William Stein
* Steve Vonn
* Ralf-Philip Weinmann
* Carl Witty
* Paul Zimmermann
Cheers,
Michael Abshoff (release chair), William Stein
* Major Features, New Spkgs and Bugfixes
* Updated to eclib.spkg from cremona.spkg.
* Update to FLINT 1.0.6
* Updated to MPFR 2.3.1 which contains an important fix that
avoids smashing the stack with large precisions
* Improved ATLAS with pre-tuned settings for more CPUs. This
greatly reduced the compile time on Pentium M and Athlon CPUS.
* Massive improvements on the doctesting coverage.
* Fix security issue with the notebook (found by T. Clemans).
It is *highly* recommended that you upgrade.
* Fixed various memleaks in PolyBoRi. There are still some
issues in the PolyBoRi wrapper.
* Progress on the Solaris port as well as the 64-bit OS X port.
Neither port is yet ready for prime time.
* Known Issues with 2.10.1:
* GNUTLS might fail to build on OS X if guile is present. For a
workaround use the updated GNUTLS spkg linked from ticket #2003
* You need to update the optional valgrind.spkg to the 3.3.0
release because the interface has been changed in 2.10.1 and it
isn't backward compatible with earlier releases.
* Bug Statistics
We closed 186 tickets. For details see
http://trac.sagemath.org/sage_trac/milestone/sage-2.10.1
* Upcoming Releases
The next release will be 2.10.2, chaired by Michael Abshoff. The release
is planned post Sage Days 7, but details are sketchy at the moment.
Potential features include:
* David Roe's unramified and Eisenstein extensions of Qp and Zp
* John Voight's code for enumerating totally real fields [with
generalizations for not-totally real fields]
* OS X 10.5 support in 64-bit mode [this will depend on fixing
a vexing segfault in libSingular and getting disutils to
play nice with numpy]
* GCC 4.3 support out of the box
* FreeBSD support out of the box
* Michael Abshoff's improved experimental Solaris support
* William Stein's rewrite of the mwrank Cython interface
* Doctesting Coverage
For 2.10 we had:
Overall weighted coverage score: 35.4%
Total number of functions: 18296
We increased coverage by 2.9% in 2.10.1 while removing 6 functions:
Overall weighted coverage score: 38.3%
Total number of functions: 18290
* Closed Tickets:
Merged in rc5:
#1547: Michael Abshoff: Add pre-tuned settings for ATLAS for
certain CPUs
#1886: Michael Abshoff: add ATLAS pretuning for AMD Athlon MP
processors
#1977: Michael Abshoff: valgrind 3.3.0 no longer appends $PID per
default to the logs
#1986: William Stein: Numerical noise in fast evaluation code.
#1999: Michael Abshoff, William Stein: delete mpl3d code from sage
#2019: David Joyner: problem with gap_packages-4.4.10_3
#2021: Jason Grout: Only start one maxima instance
#2026: William Stein, Jason Grout: matrix.eigenspaces doctest
description is misleading
#2027: Ralf-Philip Weinmann, Michael Brickenstein: update PolyBoRi
to current CVS: fix memleak and remove joke patent claim
#2029: Michael Abshoff: adjust TIMEOUT for long and valgrinded
doctests
#2035: William Stein: transform.pyx is now broken (probably caused
by other fixes)
Merged in rc4:
#1926: Burcin Erocal: fixes for the maple interface
[this is another patch to fix some more doctest issues]
#1960: Martin Albrecht: bug when reducing Gröbner basis
#1967: William Stein: fix matplotlib locale-related bugs once and for
all
#1968: Timothy Clemans and William Stein: notebook -- remove
capability of using live tutorial for users not signed in to
the notebook server; this was a potential security risk.
#1973: Jason Grout, Mike Hansen: native partition_associated function
#1983: William Stein: change 0^0, for 0 a Sage integer, to evaluate
to 1 for consistency with Python, PARI, Magma, Maple, MPFR,
GMP, etc.
#1985: Steve Vonn, Michael Abshoff: is_pseudoprime docstring
doesn't wrap
#1988: Craig Citro: doctest fallout -- osx ppc 10.4 overflow error
#1989: Craig Citro: doctest fallout -- osx ppc 10.4 matrix index out
of range
#1996: Craig Citro: doctest failure on 2.10.1.rc3: libecm.pyx
#2002: Jason Grout: creating a fresh new notebook in sage-2.10.1.rc3
is broken
#2005: William Stein: The .sage directory -- adjust permissions
#2006: Michael Abshoff: crap -- gap contains a binary
#2008: Michael Abshoff: crap -- cython contains a stupid OS X file
#2009: Michael Abshoff: crap -- networkx spkg in sage-2.10.1.rc3
contains a bunch of .svn directories
#2010: Michael Abshoff: crap -- libpng contains lots and lots of
weird (OS X?) temp or meta files
#2011: Michael Abshoff: crap -- mpfi ships with two precompiled
binaries
#2012: Michael Abshoff: crap -- jmol contains some os x files
that shouldn't be there
#2013: Michael Abshoff: crap -- f2c spkg contains precompiled binary
#2014: Michael Abshoff: crap -- matplotlib contains an annoying OS
X temp file
#2015: Michael Abshoff: crap -- twisted contains a bunch of ._ osx
crap
#2016: Michael Abshoff: crap -- there is a precompiled object
file in the SAGE_ROOT/examples directory
#2018: Michael Abshoff: crap -- scipy sandbox has a bunch of .svn
directories
#2022: William Stein: bug with load and loads when data is not
compressed during save
Merged in rc3:
#740: John Cremona: Implement standard transformations for elliptic
curves & points
#1328: Michael Abshoff: 2.8.14/Solaris: partitions_c.h compile fix
#1329: Michael Abshoff: 2.8.14/Solaris: real_rqdf.pyx compile fixes
#1554: Mike Hansen: issues with calling symbolic expressions
#1637: Michael Abshoff: Update to mpfr 2.3.1 - fix small issues in
the spkg
#1887: Robert Miller: graph isomorphism for multiple edges
#1923: Jason Grout: Make an obvious checkbox or menu to switch on
pretty printing in the notebook
#1938: Robert Bradshaw: Fast (double) function evaluation for
plotting, etc.
#1958: Craig Citro: fix problems with ANSI codes in sage0.py
#1966: Josh Kantor: inline_fortran broke on OS X
#1970: Willaim Stein, Martin Albrecht: gnutls should not be needed
if you're running the notebook insecurely, so don't require it
#1972: William Stein: programming guide -- fix a confusing typo
#1978: Yi Qiang, Michael Abshoff: update python-gnutls
#1979: John Cremona, Michael Abshoff: doctest fall out from #740
in tut.tex
Merged in rc2:
#571: John Cremona, Denis Simone, William Stein: weird doctest
failure in simon_two_descent [to quote William: "John and
Denis Simone in fact did greatly update this code and it
is now in Sage. And this particularly doctest runs almost
instantly now, so remove #long"]
#1058: John Cremona, William Stein: the mwrank interface barfs on
bad input
#1649: John Cremona, William Stein, Michael Abshoff: Updated
eclib.spkg to eclib-20080127.spkg
#1650: John Cremona: eclib.spkg: fix tsat test failure
#1823: Robert Bradshaw, Carl Witty: RDF/CDF coverage, consistent
hashing
#1922: Jason Grout: change and update lprint
#1940: Martin Albrecht: Segmentation fault when comparing two empty
ideals
#1953: Carl Witty: fix problems in calculus.py found by Jason while
reviewing #1945
#1954: Mike Hansen: sage/modules/free_module_element.pyx computing
abs(vector(...))
#1957: Michael Abshoff: Solaris 9 fixes for libfplll.spkg
#1959: Michael Abshoff: Solaris 9 fixes for rubiks.spkg
Merged in rc1:
#384: Alex Ghitza, Mike Hansen: latex formatting issues with
symbolic expressions
#652: Nick Alexander: .sage files destroy existing .py files
#743: Robert Miller, Carl Witty: graphs: simplify interface for
accessing vertex objects
#980: Martin Albrecht, Didier Deshommes: random_element()
for multivariate polynomials
#1083: Craig Citro: Bug in degree 1 number fields; manifests itself
in element.matrix(), .trace(), .norm(), etc
#1155: Mike Hansen: PermutationGroup coercion bug
#1209: Mike Hansen: make it so maple.gcd?? shows source code
[later reopened due to more review]
#1456: Mike Hansen: gaussian_binomial bug
#1575: Jason Grout: plotting -- fix vector plotting
#1845: Josh Kantor, Carl Witty: list_plot3d should be able to accept
lists of points in arbitrary positions
#1875: William Stein: increase idle timeout and save interval
on the notebook to 6 minutes (shorter times cause serious
performance problems, esp on large notebook servers)
#1929: Alex Ghitza: add more examples to qqbar.py
#1943: Michael Abshoff: rpy fails to build on OS X 10.4
#1945: Carl Witty: improve reference manual formatting
Merged in rc0:
#1600: Craig Citro: another weird coercion bug
#1911: William Stein, Craig Citro: elliptic curves -- make
heegner_index command return index instead of square
of index; clarify why sometimes result is not an
integer (it's not a bug, it's part of the algorithm)
#1930: Michael Abshoff: R is broken for parallel make, rpy is
broken for fresh installs, remove recommended tar.gz
#1931: Michael Abshoff: sage.spkg need to depend on ecm.spkg
#1933: Michael Abshoff: libgpg_error-1.6.spkg broken make install
for parallel make
#1934: Michael Abshoff: sage-check for ATLAS broken on OS X
#1935: Craig Citro: legendre_symbol currently quite slow
#1937: Martin Albrecht: fix LLL documentation
#1939: Michael Abshoff: Fix OS X rpy import issues, i.e.
DYLD_LIBRARY_PATH fix
Merged in alpha2:
#1028: Michael Abshoff: fix omega support for new valgrind
release (3.3.0 once it is out)
#1050: Michael Abshoff: update optional valgrind.spkg to 3.3.0
release
#1147: William Stein: change location of valgrind output files
to something less obnoxious
#1550: Robert Miller: use libecm instead of pexpect+ecm binary
#1593: Martin Albrecht: m4ri -- the documentation of the echelon
command only lists 1 algorithm but >= 2 algorithms are
supported
#1617: William Stein: speed problem when factoring polynoms
#1623: Michael Abshoff: update gsl to 1.10
#1724: Craig Citro: Creating ModularForms from q-expansions
#1817: Martin Albrecht: crypto.mq.sr doctest coverage at 100%
#1818: Martin Albrecht: include AES polynomial system generator
in reference manual
#1852: Michael Abshoff: Configure R to use Atlas, other build
improvements to R package
#1869: Robert Miller: Implement show(list(graphs(n)))
#1873: Mike Hansen: elementary function expansion returns result
in the wrong ring
#1889: Martin Albrecht: 2.10.1.alpha2 doctest failure in
crypto/mq/sr.py
#1890: Robert Miller: interfaces/libecm.pyx doctest failure
#1894: Martin Albrecht: toy_buchberger failures
#1895: Martin Albrecht: deal with passed through keyword arguments
for Gröbner basis computation
#1898: Michael Abshoff, William Stein: numerical doctest failure
for polynomial_element.pyx on Linux/Itanium
#1899: Robert Miller: Making graphs with a matrix and loops=True
blows up
#1900: Robert Miller, Jason Grout: Clean up adjacency matrix
functions for graphs
#1905: Alex Ghitza: both E.Lseries and E.L_series are defined.
Just define one of them, say E.Lseries.
#1906: Alex Ghitza: eisenstein_series_qexp does not pay attention
to the field parameter
#1913: Burcin Erocal: poles of gamma
#1917: William Stein: get rid of crazy OS X charpoly workaround,
since linbox isn't broken anymore
#1924: Robert Bradshaw, William Stein, Clement Pernet: Optimize
matrix multiply cache friendliness
Merged in alpha1:
#1166: William Stein: 2D terminal output is inconsistent and
corrupted
#1217: Willem Jan Palenstijn: libfplll error codes
#1274: Martin Albrecht: modify singular interface to allow for
verbose output
#1294: Mike Hansen: v.n() function blows up when v is a vector
#1306: Robert Miller: Bundles of graphs
#1418: Joel Mohler: magma element __floordiv__
#1508: William Stein: axes_label in plot() broken
#1577: Martin Albrecht: .coefficients() and .monomials() differ
in order in multivariate polynomial rings
#1583: Craig Citro: simple modules of modular symbols over finite
fields
#1604: William Stein: more locale problems with python exposed
by matplotlib
#1606: William Stein: plotting -- add aspect_ratio option to show
command
#1652: Mike Hansen: length of DAGs with loops calculation runs
infinite
#1704: David Harvey: replace _DivPolyContext by _multiply_point
#1713: Martin Albrecht: fix SageElement._sage_
#1737: William Stein: ctl-c doesn't exit job in parametric_plot3d
#1718: William Stein: bug in parametric_plot
#1721: Josh Kantor, Michael Abshoff: Introduce SAGE_ATLAS_LIB to
skip building of ATLAS
#1787: Michael Abshoff: Apply row-major GEMM major ATLAS errata
#1794: Willem Jan Palenstijn: Gram-Schmidt typo
#1804: Nick Alexander: Factorization.__mul__ assumes that the
primes in the factorization commute, which may not be a
valid assumption
#1812: Martin Albrecht: doctest coverage for finite_field_givaro up
to 100%
#1814: Martin Albrecht: docstrings for stream_cipher.py
#1820: Mike Hansen: update documentation for skew_partition.py
#1829: Martin Albrecht: small improvements for term ordering module
#1831: David Joyner: 3d docs: add more examples of parametric plots
#1832: Mike Hansen: change how real(...) is defined in Sage
#1833: William Stein: plot3d and parametric_plot3d can be very slow
on some inputs
#1835: Martin Albrecht: some minor changes to the documentation
of basis_is_groebner
#1836: Martin Albrecht: return reduced Groebner bases by default
#1837: Martin Albrecht: pass through options from groebner_basis
[hunk 2 seems to have snuck in]
#1840: David Kohel: Hill cipher addition to crypto package
#1843: William Stein: cputime for gp interface
#1844: Craig Citro: Get doctest coverage in sage/modular/modform
up to 100%
#1854: David Joyner: add more 3d plots to the reference manual
#1856: William Stein: 3d graphics -- bug in setting options via
the show command
#1858: William Stein: plot.py coverage is crap -- improve it
#1872: William Stein: serious bug in pickling elliptic curves
#1874: Robert Miller: implement bipartite graphs
#1878: Paul Zimmermann: add new function from mpfr-2.3.0
#1880: Michael Abshoff: qqbar numerical doctest failure
#1881: Michael Abshoff: tut.tex doctes failure with factorize()
#1883: Michael Abshoff: Sage distribution includes Windows
executables
Merged in alpha0:
#712: William Stein: SAGE/Maxima is sometimes unable to solve a
linear system [solved by #1827 merged in 2.10]
#827: Nick Alexander: nfrootsof1 from Pari
#1391: Martin Albrecht, William Stein : bug in unit part of
factorizations of multivariate polynomials
#1485: David Joyner, Simon King: wrapper for invariant_ring and
invariant_algebra_reynolds in Singular
#1525: Robert Miller: NetworkX upgraded spkg, Bipartite graphs
#1616: Mike Hansen, David Joyner: Change asin to arcsin, etc.
#1622: Michael Abshoff: update gnutls to 2.2.1
#1627: Michael Abshoff: update libgcrypt to 1.4
#1630: Michael Abshoff: update opencdk to 0.6.6
#1653: David Harvey: Bug raising an integer to a float
#1710: William Stein: switch default location of trac server
away from http://www.sagemath.org:9002/
#1715: Martin Albrecht: PolyBoRi pickling/hashing
#1746: Alex Ghitza: add p-norm as a method to vectors
#1772: Mike Hansen, Robert Miller: bug somewhere in the symbolics
#1789: William Stein: standalone Sage scripts don't work on Linux,
they work fine on OS X
#1800: Carl Witty: bug with RealIntervalField/MPFI
#1801: Robert Bradshaw: Update cython package to 0.9.6.11
#1813: Martin Albrecht: misc enhancements for ref.pdf
#1821: Michael Abshoff, Bill Hart: Update FLINT to 1.0.6
#1826: Robert Bradshaw, Robert Miller: cdefs.pxi and stdsage.pxi
were missing some function declarations
#1850: Robert Bradshaw: serious bug in parametric plotting of curves
#1851: Robert Miller: add documentation, doctests for
graphs/bruhat_sn.pyx
#1853: Michael Abshoff: update libgpg_error to 1.6
#1855: William Stein: trivial but visibly very painful bug in plotting
#1859: Robert Bradshaw: 3d and 2d graphics -- some unification