Sage 2.10.alpha3 released

[mabshoff]

Hi,

Sage 10.2.alpha3 is out. The combinatorics update has been
sorted out and doctests should pass. There were also a
whole bunch of small updates. The big push for alpha4 [at
least on my end] will be the package update and the merging
of the 64 bit MacIntel fixes.

Tarball [199MB] is the usual place at

http://sage.math.washington.edu/home/mabshoff/release-cycles-2.10/sage-2.10.alpha3.tar

Cheers,

Michael

Known Problems:

#1744: FLINT 1.05 "make check" failure on Linux/Itanium

I had a single, non-reproducible doctest failure in
sage/rings/qqbar.py on OSX 10.5. I reran that doctest
about 10 times and couldn't reproduce it.

Merged in alpha3:

#1428: Mike Hansen: add SVD method to matrix_complex_double_dense
#1570: Francis Clarke: typo in rings/number_field/number_field.py
#1571: Minh Nguyen, Michael Abshoff: corrections for tut.tex
#1657: Ted Kosan: make a build-from-source version of jmol spkg
[the spkg has been put in the optional spkg repo]
#1668: Burcin Erocal: fix polybori's variable names [BSD issue]
#1679: Mike Hansen: two documentation typos
#1682: Robert Bradshaw: make enabling implicit multiplication
easy
#1691: Craig Citro: old bug in pari.gen __setitem__ code
#1716: Paul Zimmermann: add function log1p in mpfr interface
#1732: Robert Bradshaw: block matrix construction
#1735: Michael Abshoff: do not mark a spkg as installed if
sage-check fails
#1743: Michael Abshoff: implausibly old time stamp in optional
package gap_packages-4.4.10_2.spkg
#1759: Michael Abshoff: Various files still mention GPL V2 only
#1769: Craig Citro: Fast Pari <--> Sage p-adic capped relative
element conversions
#1771: Peter Jipsen: latex bug with symbolics
#1776: William Stein: symbolic function preparser bug
#1778: Bobby Moretti, William Stein: plot() does not follow the
same interval range conventions as plot3d()
#1779: Bobby Moretti: setup.py computes the cache of some
irrelevent files
#1780: Michael Abshoff: add 64 bit MacIntel build support to
mpfr, fix stack smashing issue

Merged in alpha2:

#511: Michael Abshoff: consider changing python to UCS4
#1091: Willem Jan Palenstijn, Michael Abshoff: small memleaks
exposed by doctesting ntl_mat_GF2E.pyx
#1685: restructuring symmetric functions and misc. combinatorics
updates.
#1712: Burcin Erocal: logic error in coercion
#1739: Robert Miller, Michael Abshoff: Use of uninitialised value
of size 8" in binary_code.pyx
#1749: Michael Abshoff: rings/polynomial/polynomial_element.pyx
numerical noise doctest failure on FC7/x86
#1752: Michael Abshoff: sage make install bug

Merged in alpha1:

#975: Jason Grout: work around Library incompatibilities when
launching external applications
#1073: Martin Albrecht: print out some useful information right
after "sage -pkg" creates a package
#1669: Josh Kantor: remove bogus recommendation to set SAGE_ATLAS
when numpy fails
#1693: Martin Albrecht: jmol stubles over browser caching
#1709: Robert Miller: Make experimental jmol graph plotting work
#1717: Alex Ghitza, David Harvey: trivial fixes in the
programming guide (mostly typos)
#1720: Josh Kantor, Michael Abshoff: update numpy to 1.0.4, also
fix silent build problems with gfortran
#1725: Jaap Spies: Experimental package for wxPython
[linux only for now]
#1726: Jaap Spies: Experimental package for Enthought Mayavi2
[linux only for now]
#1728: Michael Abshoff: numpy 1.0.4 causes various numerical noise
doctest failures
#1729: Robert Bradshaw, Tom Boothby: disable password prompt on
initial startup
#1740: Paul Zimmermann: Fix Pentium M detection for ATLAS BLAS
#1741: Michael Abshoff: Build zlib with debug info
#1742: Michael Abshoff: Build ntl with debug info

Merged in alpha0:

#1092: Willem Jan Palenstijn, Michael Abshoff: small memleaks exposed
by doctesting ntl_ZZ_pE.pyx
#1093: Michael Abshoff, Willem Jan Palenstijn: small memleaks exposed
by doctesting ntl_ZZ_pX.pyx
#1541: Burcin Erocal: improve PolyBoRi integration
#1544: David Joyner, Rich Morin: SAGE Tutorial nits #2
#1553: Kiran Kedlaya, Michael Abshoff: SCons related build failure
of PolyBoRi on 64-bit RHEL5
#1598: Martin Albrecht: fix SIGSEGV in libSINGULAR interface on
Solaris
#1694: Michael Abshoff, Bill Hart: Update FLINT to 1.05 release
#1701: Martin Albrecht: attempt to clean up currRing if deallocated
#1702: Martin Albrecht: fix memleak in fplll.pyx
#1703: Michael Abshoff: memleak in Singular: one mpz is leaked in
longrat.cc

[Jaap Spies]

mabshoff wrote:
> Hi,
>
> Sage 10.2.alpha3 is out. The combinatorics update has been
> sorted out and doctests should pass. There were also a
> whole bunch of small updates. The big push for alpha4 [at
> least on my end] will be the package update and the merging
> of the 64 bit MacIntel fixes.
>
> Tarball [199MB] is the usual place at
>
> http://sage.math.washington.edu/home/mabshoff/release-cycles-2.10/sage-2.10.alpha3.tar
>

No surprises here. On Fedora 32 bits all tests passed.

Jaap

[Kate]

Michael,

sage-2.10.alpha3 fails one test (polynomial_element.pyx) on
both
x86-Linux (pentium4-fc6)
x86_64-Linux (Opteron-fc6)

Both were built using gcc-4.2.2

Details follow.

Kate

******
x86-Linux
******

sage -t devel/sage-main/sage/rings/polynomial/
polynomial_element.pyx**********************************************************************
File "polynomial_element.pyx", line 2549:
sage: f.roots(multiplicities=False)
Expected:
[...1.00000000000000...*I, 1.00000000000000...*I]
Got:
[1.00000000000000 + 1.11022302462516e-16*I, 1.12045416424138e-16 +
1.00000000000000*I]
**********************************************************************


*****
x86_64-Linux
*****

sage -t devel/sage-main/sage/rings/polynomial/
polynomial_element.pyx**********************************************************************
File "polynomial_element.pyx", line 2644:
sage: p.roots(ring=ComplexIntervalField(200))
Expected:

[([1.1673039782614186842560458998548421807205603715254890391400816 ..
1.1673039782614186842560458998548421807205603715254890391400829], 1),
([0.18123244446987538390180023778112063996871646618462304743773153 ..
0.18123244446987538390180023778112063996871646618462304743773341] +
[1.0839541013177106684303444929807665742736402431551156543011306 ..
1.0839541013177106684303444929807665742736402431551156543011344]*I,
1),
([0.18123244446987538390180023778112063996871646618462304743773153 ..
0.18123244446987538390180023778112063996871646618462304743773341] -
[1.0839541013177106684303444929807665742736402431551156543011306 ..
1.0839541013177106684303444929807665742736402431551156543011344]*I,
1),
([-0.76488443360058472602982318770854173032899665194736756700777454 ..
-0.76488443360058472602982318770854173032899665194736756700777...] +
[0.35247154603172624931794709140258105439420648082424733283769... ..
0.35247154603172624931794709140258105439420648082424733283769...]*I,
1),
([-0.76488443360058472602982318770854173032899665194736756700777454 ..
-0.764884433600584726029823187708541730328996651947367567007772...] -
[0.352471546031726249317947091402581054394206480824247332837691... ..
0.35247154603172624931794709140258105439420648082424733283769341]*I,
1)]
Got:

[([1.1673039782614186842560458998548421807205603715254890391400816 ..
1.1673039782614186842560458998548421807205603715254890391400829], 1),
([0.18123244446987538390180023778112063996871646618462304743773153 ..
0.18123244446987538390180023778112063996871646618462304743773341] +
[1.0839541013177106684303444929807665742736402431551156543011306 ..
1.0839541013177106684303444929807665742736402431551156543011344]*I,
1),
([0.18123244446987538390180023778112063996871646618462304743773153 ..
0.18123244446987538390180023778112063996871646618462304743773341] -
[1.0839541013177106684303444929807665742736402431551156543011306 ..
1.0839541013177106684303444929807665742736402431551156543011344]*I,
1),
([-0.76488443360058472602982318770854173032899665194736756700777516 ..
-0.76488443360058472602982318770854173032899665194736756700777204] +
[0.35247154603172624931794709140258105439420648082424733283769091 ..
0.35247154603172624931794709140258105439420648082424733283769372]*I,
1),
([-0.76488443360058472602982318770854173032899665194736756700777454 ..
-0.76488443360058472602982318770854173032899665194736756700777204] -
[0.35247154603172624931794709140258105439420648082424733283769091 ..
0.35247154603172624931794709140258105439420648082424733283769372]*I,
1)]
**********************************************************************
File "polynomial_element.pyx", line 2666:
sage: p.roots(ring=CIF)
Expected:
[([-1.4142135623730952 .. -1.4142135623730949], 1),
([1.4142135623730949 .. 1.4142135623730952], 1),
([-1.2146389322441827 .. -1.2146389322441821] -
[0.1414250525823937... .. 0.1414250525823939...]*I, 2),
([-0.141425052582393... .. -0.14142505258239376] +
[1.2146389322441821 .. 1.2146389322441827]*I, 2),
([0.141425052582393... .. 0.141425052582393...] -
[1.2146389322441821 .. 1.2146389322441827]*I, 2),
([1.2146389322441821 .. 1.2146389322441827] + [0.14142505258239376 ..
0.14142505258239399]*I, 2)]
Got:
[([-1.4142135623730952 .. -1.4142135623730949], 1),
([1.4142135623730949 .. 1.4142135623730952], 1),
([-1.2146389322441827 .. -1.2146389322441821] -
[0.14142505258239373 .. 0.14142505258239397]*I, 2),
([-0.14142505258239397 .. -0.14142505258239373] +
[1.2146389322441821 .. 1.2146389322441827]*I, 2),
([0.14142505258239376 .. 0.14142505258239397] - [1.2146389322441821 ..
1.2146389322441827]*I, 2), ([1.2146389322441821 .. 1.2146389322441827]
+ [0.14142505258239376 .. 0.14142505258239399]*I, 2)]
**********************************************************************

[Justin C. Walker]

One error in testing (see below):

On Jan 15, 2008, at 12:05 AM, mabshoff wrote:

> Sage 10.2.alpha3 is out. The combinatorics update has been
> sorted out and doctests should pass. There were also a
> whole bunch of small updates. The big push for alpha4 [at
> least on my end] will be the package update and the merging
> of the 64 bit MacIntel fixes.
>
> Tarball [199MB] is the usual place at
>
> http://sage.math.washington.edu/home/mabshoff/release-cycles-2.10/
> sage-2.10.alpha3.tar

Mac OS X, 10.4.11 (Dual Quad Xeon):

Install succeeded:

real 76m49.526s
user 63m16.093s
sys 31m32.407s

Testing had one glitch:

sage -t devel/sage-main/sage/schemes/generic/projective_space.py
A mysterious error (perphaps a memory error?) occurred, which
may have crashed d
octest.
[11.9 s]

----------------------------------------------------------------------
The following tests failed:

sage -t devel/sage-main/sage/schemes/generic/
projective_space.py
Total time for all tests: 2743.9 seconds

There was a python crash during the testing, but I'm not at home to
determine where that might have occurred. I've pushed that crash
dump to sage.math.washington.edu:~justin/logs/python.log, in case it
contains interesting stuff. I won't be home to do any more checking
until late tonight.

Justin

--
Justin C. Walker, Curmudgeon-At-Large
Institute for the Absorption of Federal Funds
--------
If you're not confused,
You're not paying attention
--------

[mabshoff]

On Jan 15, 10:32 pm, Kate <kate01...@gmail.com> wrote:
> Michael,

Hi Kate,

> sage-2.10.alpha3 fails one test (polynomial_element.pyx) on
> both
> x86-Linux (pentium4-fc6)
> x86_64-Linux (Opteron-fc6)
>
> Both were built using gcc-4.2.2
>
> Details follow.
>
> Kate
>
> ******
> x86-Linux
> ******
>
> sage -t devel/sage-main/sage/rings/polynomial/
> polynomial_element.pyx**********************************************************************
> File "polynomial_element.pyx", line 2549:
> sage: f.roots(multiplicities=False)
> Expected:
> [...1.00000000000000...*I, 1.00000000000000...*I]
> Got:
> [1.00000000000000 + 1.11022302462516e-16*I, 1.12045416424138e-16 +
> 1.00000000000000*I]
> **********************************************************************

This is now #1791 - but it looks like there is no way for this
particular example to really doctest anything. I might make this
#random until we can come up with a numerically more stable doctest.

This is now #1790 - and easy enough to fix. I should have a patch in a
couple minutes. Thanks for the reports.

Cheers,

Michael

[mabshoff]

Hi Justin,

I have been seeing non-reproducible doctest failures on OSX for quite
a while now, but so far nobody has had any luck tracking them down and
more importantly fixing them. The failures themselves always seem to
be caused on exit, i.e. there are never any "wrong" results being
printed. I suspect that it is caused by #1337, i.e. "tp_new leads to
munmap_chunk(): invalid pointer segfault". It causes a segfault on
every exit on Linux/PPC. There was a patch on cython-dev about a week
ago named "Re: [Cython-dev] fix for tp_clear()" [sorry, no link since
the archives are list members only - I don't know why] - which seems
at least related. The ticket was never updated by me with the latest
info, but that segfault is caused by line 2200 in structure/
element.pyx:

# This is pretty nasty low level stuff. The idea is to speed up
construction
# of EuclideanDomainElements (in particular Integers) by skipping some
tp_new
# calls up the inheritance tree.
PY_SET_TP_NEW(EuclideanDomainElement, Element)

in my opinion. I haven't had time to fix it yet and I didn't feel like
bothering Robert about it during the last couple weeks.

>
> --
> Justin C. Walker, Curmudgeon-At-Large
> Institute for the Absorption of Federal Funds

Cheers,

Michael

[Kate]

Michael,

The patch for #1790 is ok, but after applying the patch for #1791
I now get

./sage -t devel/sage-main/sage/rings/polynomial/polynomial_element.pyx

sage -t devel/sage-main/sage/rings/polynomial/
polynomial_element.pyx**********************************************************************
File "polynomial_element.pyx", line 2644:
sage: p.roots(ring=ComplexIntervalField(200))
Expected:

[([1.1673039782614186842560458998548421807205603715254890391400816 ..
1.1673039782614186842560458998548421807205603715254890391400829], 1),
([0.18123244446987538390180023778112063996871646618462304743773153 ..
0.18123244446987538390180023778112063996871646618462304743773341] +
[1.0839541013177106684303444929807665742736402431551156543011306 ..
1.0839541013177106684303444929807665742736402431551156543011344]*I,
1),
([0.18123244446987538390180023778112063996871646618462304743773153 ..
0.18123244446987538390180023778112063996871646618462304743773341] -
[1.0839541013177106684303444929807665742736402431551156543011306 ..
1.0839541013177106684303444929807665742736402431551156543011344]*I,
1),

([-0.76488443360058472602982318770854173032899665194736756700777... ..

-0.76488443360058472602982318770854173032899665194736756700777...] +
[0.35247154603172624931794709140258105439420648082424733283769... ..
0.35247154603172624931794709140258105439420648082424733283769...]*I,
1),
([-0.76488443360058472602982318770854173032899665194736756700777454 ..
-0.764884433600584726029823187708541730328996651947367567007772...] -

[0.35247154603172624931794709140258105439420648082424733283769... ..
0.3524715460317262493179470914025810543942064808242473328376934...]*I,

1 items had failures:
1 of 109 in __main__.example_58
***Test Failed*** 1 failures.
For whitespace errors, see the file .doctest_polynomial_element.pyx
[6.6 s]
exit code: 256

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

Kate

[mabshoff]

On Jan 16, 5:08 pm, Kate <kate01...@gmail.com> wrote:
> Michael,

Hi Kate,

> The patch for #1790 is ok, but after applying the patch for #1791
> I now get

I see that I missed one single digit being different. Please try

http://trac.sagemath.org/sage_trac/attachment/ticket/1790/Sage-2.10.alpha3-%231790-leftovers.patch

which should fix the issue hopefully for good.

> ./sage -t devel/sage-main/sage/rings/polynomial/polynomial_element.pyx
> sage -t devel/sage-main/sage/rings/polynomial/
> polynomial_element.pyx**********************************************************************
> File "polynomial_element.pyx", line 2644:
> sage: p.roots(ring=ComplexIntervalField(200))
> Expected:
>

<SNIP>
> ----------------------------------------------------------------------
>
> Kate

Cheers,

Michael

[Kate]

Michael,

This fixes the issue. Thanks!

Kate

On Jan 16, 11:22 am, mabshoff <Michael.Absh...@fsmath.mathematik.uni-

dortmund.de> wrote:
> On Jan 16, 5:08 pm, Kate <kate01...@gmail.com> wrote:
>
> > Michael,
>
> Hi Kate,
>
> > The patch for #1790 is ok, but after applying the patch for #1791
> > I now get
>
> I see that I missed one single digit being different. Please try
>

> http://trac.sagemath.org/sage_trac/attachment/ticket/1790/Sage-2.10.a...