sage-5.0.beta4 released
Jeroen Demeyer
We're releasing Sage 5.0.beta4.
Source archive:
http://boxen.math.washington.edu/home/release/sage-5.0.beta4/sage-5.0.beta4.tar
Upgrade path:
http://boxen.math.washington.edu/home/release/sage-5.0.beta4/sage-5.0.beta4/
The source and upgrade path can also be found on the mirror network
(you might need to wait a while before the mirrors are synchronized):
http://www.sagemath.org/download-latest.html
Please build, test, and report! We'd love to hear about your
experiences with this release.
== Tickets ==
* We closed 168 tickets in this release. For details, see
http://boxen.math.washington.edu/home/release/sage-5.0.beta4/tickets.html
Merged in sage-5.0.beta4:
#4498: Karen T. Kohl, Burcin Erocal: Implement a symbolic version of the
arg function [Reviewed by Karl-Dieter Crisman, Burcin Erocal]
#6581: John Perry: Groebner basis not working over symbolic ring
[Reviewed by Martin Albrecht]
#10807: David Loeffler: Modular Symbols trouble restoring saved objects
[Reviewed by Marco Streng]
#10998: Frédéric Chapoton, Christian Stump, Nicolas M. Thiéry:
Categories for posets [Reviewed by Franco Saliola, Christian Stump,
Nicolas M. Thiéry, Florent Hivert]
#11003: Mike Hansen: Improve support for lazy_import inside classes
[Reviewed by Nicolas M. Thiéry, Florent Hivert]
#11118: Florent Hivert: Add a cache for .list() method in
FiniteEnumeratedSet [Reviewed by Nicolas M. Thiéry]
#11250: Florent Hivert: Fix link in coercion.rst [Reviewed by Nicolas M.
Thiéry]
#11257: Nicolas Borie: Avoid a coercion when computing an element to the
power of 0 [Reviewed by Nathann Cohen]
#11967: John Palmieri: os x 10.7 Lion -- Sage segfaults on startup when
initializing GiNaC [Reviewed by Jeroen Demeyer]
#12006: william stein: sage-4.8 and magma-2.18 not playing well together
[Reviewed by Marco Streng]
#12034: William Stein: improve doctest coverage of matrix/benchmark.py
from 0% to 100% [Reviewed by Martin Albrecht, Karl-Dieter Crisman]
#12135: Lukáš Lánský: Multiple loops handling in c_graph [Reviewed by
Nathann Cohen]
#12318: Lukáš Lánský: **kwds in the Graph constructor is confusing
[Reviewed by Nathann Cohen]
#12336: Martin Albrecht: The optional doctests in
sage/databases/symbolic_data.py are misformatted so they are never run
[Reviewed by Michael Orlitzky]
#12344: Martin Albrecht: Interface to GINV [Reviewed by Dmitrii Pasechnik]
#12353: David Roe: wrong comparison between RealIntervalField and
RealField [Reviewed by Paul Zimmermann]
#12363: Jeroen Demeyer: Upgrade PARI to 2.5.1 [Reviewed by John Palmieri]
#12366: Jeroen Demeyer: In mpfr, delete old libraries *after* build
[Reviewed by Michael Orlitzky]
#12367: Jeroen Demeyer: Add a script to check whether the compiler
supports a certain flag [Reviewed by R. Andrew Ohana]
#12373: Sonseeahray Rucker, William Stein: Possible typo in Units Module
Avogadro constant [Reviewed by William Stein]
#12385: Keshav Kini: Questionable semantics of
DiGraph().all_simple_paths [Reviewed by Nathann Cohen]
#12416: Jeroen Demeyer: Mercurial assumes that any system with XCode
*installed* will use XCode to *build* Mercurial [Reviewed by R. Andrew
Ohana]
#12430: R. Andrew Ohana, Jeroen Demeyer: Add -fwrapv cflag when building
Python [Reviewed by R. Andrew Ohana]
#12464: Florent Hivert, Nicolas M. Thiéry: Improved uniqueness for
combinatorial free modules [Reviewed by Nicolas M. Thiéry, Florent Hivert]
#12474: Jean-Pierre Flori: Doc fix for UniqueRepresentation class
[Reviewed by Nicolas M. Thiéry]
#12478: Andrey Novoseltsev: Typo in an error message of Fan constructor
[Reviewed by Volker Braun]
#12481: Jeroen Demeyer: Fix branch_current_hg() [Reviewed by André Apitzsch]
#12483: Nicolas M. Thiéry: Workaround #12482 causing history-dependent
behaviour of Family [Reviewed by Viviane Pons]
#12490: Florent Hivert: Create a role :trac: to link to trac ticket in
sphinx [Reviewed by Nathann Cohen, Jeroen Demeyer]
#12492: Keshav Kini: Polyhedron().radius() is nonsensical [Reviewed by
Punarbasu Purkayastha]
#12500: John Palmieri: sagetex: fix self-tests on OpenSolaris [Reviewed
by Jeroen Demeyer]
Florent Hivert
> Dear Sage lovers,
>
> We're releasing Sage 5.0.beta4.
>
> Source archive:
>
> http://boxen.math.washington.edu/home/release/sage-5.0.beta4/sage-5.0.beta4.tar
Compiled from scratch on an openSUSE 11.3 (x86_64). It seems to work fine
however there is a problem with the documentation:
spkg/pipestatus "./sage --docbuild --no-pdf-links all html 2>&1" "tee -a dochtml.log"
sphinx-build -b html -d /home/data/Sage-Install/sage-5.0.beta2/devel/sage/doc/output/doctrees/en/reference -A hide_pdf_links=1 /home/data/Sage-Install/sage-5.0.beta2/devel/sage/doc/en/reference /home/data/Sage-Install/sage-5.0.beta2/devel/sage/doc/output/html/en/reference
Running Sphinx v1.1.2
Configuration error:
There is a syntax error in your configuration file: invalid syntax (conf.py, line 131)
Build finished. The built documents can be found in /home/data/Sage-Install/sage-5.0.beta2/devel/sage/doc/output/html/en/reference
The same is repeated for all documents...
Anyone has already seen something similar ?
Cheers,
Florent
Jeroen Demeyer
> Anyone has already seen something similar ?
Florent Hivert
Ok ! I got the problem. Recently (I think with sage_4_8) something changed in
the way sage honor the $SAGE_ROOT environment variable. I have a $SAGE_ROOT
set-up in my .profile script. Now, when I install a new sage, I first have to
edit the new installed sage script to override the $SAGE_ROOT, otherwise sage
will use the ancient one for several stuff including compiling the doc: for
example here since my SAGE_ROOT point to
/home/data/Sage-Install/sage-5.0.beta2/, while I was compiling a new sage in
/home/data/Sage-Install/sage-5.0.beta4/ sage tried to compile a broken doc in
/home/data/Sage-Install/sage-5.0.beta2/ I think its a new behavior of sage,
which I find bad.
Am I the only one affected ? Should I remove the assignment to $SAGE_ROOT in
my .profile ?
Cheers,
Florent
Jeroen Demeyer
> Should I remove the assignment to $SAGE_ROOT in my .profile ?
better, so I understand what you want to do.
Florent Hivert
Nothing special. I wan't to have a default sage which start when I type sage,
and to have several non default where I have to type
/path/to/sage-version..../sage
Currently what I do is to have in my ~/bin/ a symlink to
/path/to/sage-default-version/sage
Sa probably setting $SAGE_ROOT is redundant and confuse my sage.
Cheers,
Florent
Jeroen Demeyer
> Sa probably setting $SAGE_ROOT is redundant and confuse my sage.
SAGE_ROOT, no matter what. I.e. your SAGE_ROOT from .profile should
override whatever Sage you are trying to run. So probably, you don't
want to set SAGE_ROOT in your .profile.
Justin C. Walker
On Feb 15, 2012, at 08:25 , Jeroen Demeyer wrote:
> Dear Sage lovers,
>
> We're releasing Sage 5.0.beta4.
>
> Source archive:
>
> http://boxen.math.washington.edu/home/release/sage-5.0.beta4/sage-5.0.beta4.tar
Full build on Mac OS X, 10.6.8, Dual 6-core Xeon: build completed w/o problems; dochtml build completed w/o problems; tests ('ptestlong') all passed successfully.
Justin
--
Justin C. Walker, Curmudgeon at Large
Institute for the Absorption of Federal Funds
-----------
Like the ski resort full of girls hunting for husbands
and husbands hunting for girls, the situation is not
as symmetrical as it might seem.
- Alan MacKay
--
Kelvin Li
> Dear Sage lovers,
>
> We're releasing Sage 5.0.beta4.
Platform: Debian Wheezy/Testing in QEMU-KVM guest on Debian Wheezy/
Testing host, everything 32-bit.
Host:
$ grep name /proc/cpuinfo
model name : Intel(R) Core(TM)2 Duo CPU P8400 @ 2.26GHz
model name : Intel(R) Core(TM)2 Duo CPU P8400 @ 2.26GHz
$ uname -a
Linux qu1 3.2.0-1-686-pae #1 SMP Sun Feb 5 23:52:49 UTC 2012 i686
GNU/Linux
Guest:
$ uname -a
Linux crap 3.2.0-1-686-pae #1 SMP Sun Feb 5 23:52:49 UTC 2012 i686
GNU/Linux
During building and testing, the guest was suspended/resumed multiple
times (this is a laptop I carry around). "make" finished without any
errors, but then "make ptestlong" gave several failures (sorry, I lost
the details here...). However, doing "sage -ba" then "make ptestlong"
gave an all-tests-pass success.
-- Kelvin
Dima Pasechnik
Jeroen Demeyer
> $ ./sage
> ----------------------------------------------------------------------
> | Sage Version 5.0.beta4, Release Date: 2012-02-14 |
> | Type notebook() for the GUI, and license() for information. |
> ----------------------------------------------------------------------
> **********************************************************************
> * *
> * Warning: this is a prerelease version, and it may be unstable. *
> * *
> **********************************************************************
> sage: ModularSymbols(12,8).character()
> /usr/local/src/sage/sage-5.0.beta4/local/lib/libcsage.so(print_backtrace+0x31)[0x7f307d623986]
> /usr/local/src/sage/sage-5.0.beta4/local/lib/libcsage.so(sigdie+0x14)[0x7f307d6239b8]
> /usr/local/src/sage/sage-5.0.beta4/local/lib/libcsage.so(sage_signal_handler+0x20c)[0x7f307d623606]
> /lib/x86_64-linux-gnu/libpthread.so.0(+0xf030)[0x7f3082c3f030]
> /usr/local/src/sage/sage-5.0.beta4/local/lib/libatlas.so(ATL_dJIK56x56x56TN56x56x0_a1_b0+0xac)[0x7f307796f7bc]
Did you try the same command on the same system on an earlier Sage version?
Justin C. Walker
On Feb 19, 2012, at 23:46 , Jeroen Demeyer wrote:
> On 2012-02-19 15:44, Dima Pasechnik wrote:
>> $ ./sage
>> ----------------------------------------------------------------------
>> | Sage Version 5.0.beta4, Release Date: 2012-02-14 |
>> | Type notebook() for the GUI, and license() for information. |
>> ----------------------------------------------------------------------
>> **********************************************************************
>> * *
>> * Warning: this is a prerelease version, and it may be unstable. *
>> * *
>> **********************************************************************
>> sage: ModularSymbols(12,8).character()
FWIW, it seems to work on 5.0-b3+gcc.
Justin
--
Justin C. Walker, Curmudgeon at Large
Institute for the Absorption of Federal Funds
--
Democracy is two wolves and a lamb
voting on what to have for lunch.
Liberty is a well-armed lamb contesting
the vote.
Jeroen Demeyer
>
> On Feb 19, 2012, at 23:46 , Jeroen Demeyer wrote:
>
>> On 2012-02-19 15:44, Dima Pasechnik wrote:
>>> $ ./sage
>>> ----------------------------------------------------------------------
>>> | Sage Version 5.0.beta4, Release Date: 2012-02-14 |
>>> | Type notebook() for the GUI, and license() for information. |
>>> ----------------------------------------------------------------------
>>> **********************************************************************
>>> * *
>>> * Warning: this is a prerelease version, and it may be unstable. *
>>> * *
>>> **********************************************************************
>>> sage: ModularSymbols(12,8).character()
>
> FWIW, it seems to work on 5.0-b3+gcc.
$SAGE_ROOT/spkg/installed/gcc-4.6.2 and is there $SAGE_ROOT/local/bin/gcc)
Justin C. Walker
What I actually did, since I seem to be stuck in a massive mismatch of components, is to grab a tar ball from William that was built from the "beta3+gcc" release you made earlier. He built it from your source tarball.
Justin
--
Justin C. Walker, Curmudgeon at Large
Director
Institute for the Enhancement of the Director's Income
-----------
Nobody knows the trouble I've been
-----------
Jeroen Demeyer
> What I actually did, since I seem to be stuck in a massive mismatch of components, is to grab a tar ball from William that was built from the "beta3+gcc" release you made earlier.
unfortunately.
Dima Pasechnik
Julian Rüth
> Please build, test, and report! We'd love to hear about your
> experiences with this release.
The following code used to work with 4.8 but fails with 5.0.beta4.
"""
class C(object):
@cached_method
def __invert__(self):
raise NotImplementedError
c = C()
c.__invert__.set_cache(int())
print c.__invert__() # this works
print ~c # this raises a NotImplementedError (5.0.beta4)
print c.__invert__() # this also raises a NotImplementedError (5.0.beta4)
"""
I tried to find the source of this problem and realized that it
works if I replace "class C(object):" with "class C:". But I still don't
know why the above code fails.
Even in 4.8 ~c does something weird. It actually changes c.__invert__ if
C inherits from object:
So in 4.8 id(c.__invert__) is different before the call to ~c and after.
I put a demo of this on sagenb.org: http://sagenb.org/home/pub/4366/
Thanks for looking into this,
julian
William Stein
> * Jeroen Demeyer <jdem...@cage.ugent.be> [2012-02-15 17:25:31 +0100]:
>> Please build, test, and report! We'd love to hear about your
>> experiences with this release.
>
> The following code used to work with 4.8 but fails with 5.0.beta4.
>
> """
> class C(object):
> @cached_method
> def __invert__(self):
> raise NotImplementedError
>
> c = C()
> c.__invert__.set_cache(int())
> print c.__invert__() # this works
> print ~c # this raises a NotImplementedError (5.0.beta4)
> print c.__invert__() # this also raises a NotImplementedError (5.0.beta4)
> """
>
> I tried to find the source of this problem and realized that it
> works if I replace "class C(object):" with "class C:". But I still don't
> know why the above code fails.
The first thing I would try is copying the cached_method decorator
code to a new file with your code, and testing the above in Python 2.6
and 2.7, independent of Sage. Maybe something about semantics of
object changed in the move from Python 2.6 to 2.7.
william
>
> Even in 4.8 ~c does something weird. It actually changes c.__invert__ if
> C inherits from object:
> So in 4.8 id(c.__invert__) is different before the call to ~c and after.
> I put a demo of this on sagenb.org: http://sagenb.org/home/pub/4366/
>
> Thanks for looking into this,
> julian
>
> --
> You received this message because you are subscribed to the Google Groups "sage-release" group.
> To post to this group, send email to sage-r...@googlegroups.com.
> To unsubscribe from this group, send email to sage-release...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/sage-release?hl=en.
>
--
William Stein
Professor of Mathematics
University of Washington
http://wstein.org
Dima Pasechnik
Julian Rüth
> On Wed, Feb 22, 2012 at 10:27 AM, Julian Rüth <julian...@fsfe.org> wrote:
> > * Jeroen Demeyer <jdem...@cage.ugent.be> [2012-02-15 17:25:31 +0100]:
> >> Please build, test, and report! We'd love to hear about your
> >> experiences with this release.
> >
> > The following code used to work with 4.8 but fails with 5.0.beta4.
> >
> > """
> > class C(object):
> > @cached_method
> > def __invert__(self):
> > raise NotImplementedError
> >
> > c = C()
> > c.__invert__.set_cache(int())
> > print c.__invert__() # this works
> > print ~c # this raises a NotImplementedError (5.0.beta4)
> > print c.__invert__() # this also raises a NotImplementedError (5.0.beta4)
> > """
> >
> > I tried to find the source of this problem and realized that it
> > works if I replace "class C(object):" with "class C:". But I still don't
> > know why the above code fails.
>
> The first thing I would try is copying the cached_method decorator
> code to a new file with your code, and testing the above in Python 2.6
> and 2.7, independent of Sage. Maybe something about semantics of
> object changed in the move from Python 2.6 to 2.7.
Thanks, this actually pointed me in the right direction - it does not
seem to be related to the 2.6-2.7 move but the way python calls "special
methods" for new-style classes. I filed a bug for this:
http://trac.sagemath.org/sage_trac/ticket/12601
I just discussed this with Simon King, and the fix appears to be easy.
I'm working on it right now.
Anyway, sage does not use @cached_method for operators at any place so I
don't think it's a problem to ship 5.0 without such a patch.
julian
