sage-2.9.txt

Hello folks,

Sage 2.9 has been released. It is available at

           http://sagemath.org/download.html

The following people contributed this release:

 * Michael Abshoff
 * Martin Albrecht
 * Robert Bradshaw
 * Tom Boothby
 * Ondrej Certik
 * Craig Citro
 * F. Clark
 * John Cremona
 * Alyson Deines
 * Burcin Erocal
 * Jan Groenewald
 * David Harvey
 * Mike Hansen
 * Josh Kantor
 * Robert Miller
 * Bobby Moretti
 * Rich Morin
 * Bill Page
 * Willem Jan Palenstijn
 * Yi Qiang
 * David Roe
 * William Stein
 * Carl Witty
 * Paul Zimmermann

Cheers,

Michael Abshoff (release chair), William Stein

* Major Features, New Spkgs and Bugfixes

We achieved most of the goals we set ourselves for 2.9. The major
changes compared to 2.8.15 are:

 * Integration of PolyBoRi
 * Integration of ATLAS BLAS/Lapack on non-OSX
 * Integration of R 2.6.1 and rpy
 * Update to FLINT 1.02
 * Update to Symmetrica 2.0
 * Much faster permutation arithmetic
 * Very fast rubik's cube solvers
 * Many build fixes applied and many bugs squashed

We closed a total of 110 tickets, for details see below or our 
tracker at

http://www.sagetrac.org/sage_trac/query?status=closed&milestone=sage-2.9

* Known Issues with 2.9

#1137: Import gd fails to import on OSX 10.4 ppc. Any attempt 
       to fix this breaks other applications on different OSX 
       flavors. We hope to fix this in 2.9.1.
#1497: ATLAS fails to build on FC7, Dual core [known ATLAS 
       issue with workaround: disable power management] 

* Upcoming Releases

The following releases are planned for the rest of the year:

 * 2.9.1: chaired by Robert Miller, bug fixes, release 
          planned about a week from now
 * 2.9.2: chaired by Michael Abshoff, bug fixes, release 
          planned about two weeks from now

2.9.2 will be the release we plan to distribute on DVD at the AMS 
meeting, so we are shooting hard for an excellent release.

* Doctesting Coverage 

For 2.9:

Overall weighted coverage score:  35.3%
Total number of functions:  17947

Compared to 2.8.15 this is an increase by 0.5%.

* About Sage (http://www.sagemath.org)

Sage is developed by volunteers and combines 71 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 in
freenode.

* Closed Tickets:

Merged in rc3:

#1529: William Stein: update install_scripts to also install R script.
#1530: William Stein: building tut.tex is currently broken
#1531: Michael Abshoff: fix doctest failure sage/calculus/calculus.py
#1532: Josh Kantor; Michael Abshoff: Error out with intelligent
       message if ATLAS tune failed

Without tickets:

William Stein: update libpng.spkg
William Stein: don't let "sage -ba" start sage
William Stein: link libcsage.so against atlas on non-OSX
William Stein: link IML against atlas on non-OSX
Willaim Stein, Michael Abshoff: Various doctest fixes
Michael Abshoff: link LinBox against atlas on non-OSX
Michael Abshoff: remove forced check from FLINT.spkg

Merged in rc1/2:

#1258: Willem Jan Palenstijn, Robert Miller: additions and changes 
       to linear_codes
#1457: William Stein: "BUG: Rational.__pow__ called on a non-Rational"
#1464: Robert Miller: binary code canonical labels & automorphism
       group generators
#1503: William Stein: formal function calls don't coerce correctly to
       Mathematica
#1511: Robert Bradshaw: Export 3d objects in jmol format
#1527: William Stein: doctest R with "import rpy" 
#1528: William Stein: make "sage -R" run R.

Merged in rc0:

#444: Robert Bradshaw: solve the rubik's cube fast!
#553: Mike Hansen, William Stein: calling of symbolic expressions is
      sometimes ridiculous
#1077: Yi Qiang: DSage restarts two workers after timeout, various 
       other fixes and improvements
#1119: Martin Albrecht, Robert Bradshaw: EllipticCurve.random_element
       for char=2
#1137: Robert Miller: matrix visualize_structure is completely broken
       on OSX [later reverted]
#1162: Paul Zimmermann, Carl Witty: fix issues in RealField <-> RQDF
       conversions
#1183: David Roe, William Stein, Alyson Deines: Residue fields are
       broken. This patch set also fixes #1242 and #1185
#1232: Craig Citro: bug in modular symbols over GF(2)
#1237: David Roe: E.torsion_order() fails on curves with big
       coefficients!
#1239: Robert Bradshaw, John Cremona: Wrap Simon's new gp two 
       descent code
#1393: Robert Miller: is_integral_domain may return incorrect answer
#1395: Mike Hansen: notebook(directory="foo/") misbehaves
#1401: Mike Hansen, William Stein: deprecate A[n] for n a matrix
#1404: F. Clark, William Stein: bug in %latex feature in the notebook)
#1407: William Stein: deciding that generators don't generate an order
       in some extensions can be way way too slow.
#1423: William Stein: bug/issue in python.eval from sage
#1425: Mike Hansen, William Stein: wrong automatic simplification of
       pow
#1442: Mike Hansen: fix wrong statement in calculus.py
#1447: Burcin Erocal: Polybori permissions issues [fixed in 
       polybori-0.1-r5.spkg]
#1454: William Stein: Change the prompt for sage -sh
#1460: William Stein, Carl Witty: bug in float( ... ) conversion in
       calculus
#1461: William Stein: implement general subs_expr for calculus
       expressions
#1472: Josh Kantor: gnuplotpy optional package doesn't work with numpy
       [updated optional spkg]
#1473: Robert Bradschaw: make Java3D usable from command line as well
       as notebook 
#1480: William Stein: implement P.show() for mathematica elements --
       nice mathematica graphics in the sage notebook!
#1491: William Stein: improve conversion from interface to sage
       objects (i.e., the dot sage method)
#1494: William Stein: bug coercing from maximal order of cyclotomic
       field into cyclotomic field
#1502: Mike Hansen: calculus -- bug in argument ordering for formal
       functions
#1506: William Stein: NTL.spkg - dumb intentional error during the 
       build
#1507: William Stein: document how to use pylab / matlab style
       plotting from sage
#1513: Michael Abshoff: FLINT install uses make -B, which isn't an
       option on (slightly) older make versions
#1515: Robert Bradshaw: fix ParametricSurface bug
#1518: William Stein: cache computing order of elliptic curve over
       finite field
#1519: William Stein: hg problem applying tickets via trac url: remove 
       characters after "?" from name
#1520: Michael Abshoff: prebuilt 2.9-alpha7 osx ppc binary has
       hardcoded paths

Merged in alpha7:

#1340: Robert Bradshaw: %cython broken if directory name contains
       spaces
#1398: Mike Hansen: Functions on integers which are not 
       object-oriented
#1462: William Stein: make sure that not to update the timestamp 
       cache if compilation fails
#1487: Michael Abshoff: ATLAS breaks with parallel make
#1488: William Stein: fix output of symbolic vectors
#1489: William Stein: serious bug in calculus maxima parsing
#1490: Michael Abshoff: fix numerical noise doctest failure in
       numerical/test.py
#1492: Yi Qiang: update sqlite.spkg to 3.5.3.p1 [this also closes
       #1152, #1202]
#1493  Burcin Erocal: polybori doesn't free m4ri data on exit
#1497: Josh Kantor: Configure ATLAS properly for 32 bit user space 
       on 64 bit CPU
#1499: Michael Abshoff: fix import issues related to ATLAS BLAS on
       non-OSX
#1500: William Stein: solve_mod -- implement solving modulo n in sage
#1501: Josh Kantor: udpate cvxopt to use ATLAS on Linux

Merged in alpha6:

#469: Burcin Erocal: Add BooleanPolynomialRing to global namespace. 
      With this patch PolyBoRi's integration should be finished
#823: Josh Kantor, Michael Abshoff: make atlas standard in Sage
#1430: Josh Kantor: sparse eigenvalues and splines [scipy sandbox]
#1434: Josh Kantor: cut sandbox directory out of scipy spkg
#1462: William Stein: speed up "sage -b" -- don't do anything cython
       stuff if no .pyx, .pxd, or .pxi file changes 
#1463: William Stein: polymake package has wrong cddlib,gmp versions
#1479: William Stein: Fix doctest failure fallout from #1296

Merged in alpha5:

#469: Burcin Erocal: partial merge of the PolyBoRi wrapper, updated 
      PolyBoRi.spkg [still some code to merge, so disable doctests 
      for now, BooleanRing is missing]
#1189: Ondrej Certik: SymPy <-> SAGE automatic conversion
#1235: William Stein: doctest fixes for 32 bit Linux
#1296: Robert Bradshaw: Fast permutation arithmetic [with bugfixes
       and debugging by William Stein, Michael Abshoff, Carl Witty 
       and Mike Hansen - this was a real team effort :)]
#1394: Robert Miller: make plot documentation readable in graph.py
#1450: William Stein: update the error message for using maple/
       mathematica from sage
#1453: William Stein: fix cython dependency computation for new files
       [later reverted, due to issues with #1366]
#1455: Michael Abshoff: Fix OSX doctest failures due to numerical
       noise

Merged in alpha4:

No tickets, but build fixes for gfan and R by William Stein

Merged in alpha3:

#1366: Bobby Moretti, William Stein: speed up "sage -br" -- make it
       cache the dependency diagram instead of computing it every 
       time [later reverted]
#1411: William Stein: pdate the README.txt file included in Sage
#1416: William Stein, Mike Hansen, Josh Kantor: make R standard 
       component
#1427: Josh Kantor: rpy doesn't build on osx 10.4
#1433: William Stein: Fix some documentation typos pointed out by 
       Rich Morin
#1436: Michael Abshoff, Bill Hart: Update FLINT.spkg to 1.02 relase
#1437: Michael Abshoff: fix spkg-check in FLINT to work "stand alone"
#1444: William Stein: fix some serious hard-coding issues that break
       all binary installs
#1445: Mike Hansen: symmetrica's longints need to be converted

Merged in alpha2:

#469: Burcin Erocal: add PolyBoRi.spkg [integration patch still 
      missing]
#1164: Mike Hansen: Error in tutorial documentation about %hist
#1167: Mike Hansen: Spelling errors in documentation
#1182: Mike Hansen: update freetype.spkg to 2.3.5
#1235: William Stein: bug solving equations using maxima
#1243: Mike Hansen: Minor spelling mistakes in documentation
#1267: Mike Hansen: documentation for piecewise does not show up in
       notebook
#1288: Mike Hansen: misformating of some of the reference manual in
       live version; also out of date
#1324: Carl Witty: 2.8.14: doctest failure in sage/rings/real_rqdf.pyx
       on FC6, x86-64
#1348: Mike Hansen: Incorporate Jan Groenewald's documentation fixes
#1378: Mike Hansen: add ssh-client (i.e., ssh-keygen) as a required
       package in various places in the docs
#1412: Tom Boothby: creating directory in notebook cell #x makes
       evaluation of cell #x hang
#1417: Mike Hansen: update symmetrica.spkg to 2.0
#1420: Mike Hansen: documentation typo
#1435: Michael Abshoff: Update FLINT.spkg to 1.01

Merged in alpha1:

#491: Michael Abshoff: gcc 4.3: fix givaro build due to ::memcpy
      failure
#492: Craig Citro, David Harvey: gcc 4.3: fix ntl_wrap.cc in   
      sage_c_lib [this was some time ago]
#1260: Michael Abshoff, Bill Hart: update to FLINT 1.0 and clean 
       up spkg-install
#1403: John Cremona: Analog fixes for termination issues for 
       the other mwrank binaries
#1406: Mike Hansen, William Stein: bug in coercion of multivariate
       polynomials 
#1410: Michael Abshoff: fix leftovers from 1256, make mwrank 
       binaries link dynamically, fix cython doctest issue

Merged in alpha0:

#1002: Bill Page: Update clisp to the 2.43 release [later reverted]
#1256: Michael Abshoff: mwrank*.spkg now redundant, included in
       cremona*.spkg
#1337: Carl Witty: technically incorrect code in integer_mod.pyx
#1403: John Cremona: mwrank has termination issues
#1405: Michael Abshoff, Carl Witty: gcc 4.3-snapshot: cremona 
       compile fixes. 

Reviewers: Carl Witty, William Stein, Mike Hansen, Robert Miller,
           Craig Citro, Michael Abshoff; apologies to anybody I
           forgot.

Feedback: Jaap Spies, apologies to anybody I forgot.