Sage 10.5.beta2 released

[Volker Braun]

As always, you can get the latest beta version from the "develop" git branch. Alternatively, the self-contained source tarball is at http://www.sagemath.org/download-latest.html

e042294b5ed (HEAD -> develop, tag: 10.5.beta2, github/develop) Updated SageMath version to 10.5.beta2

b178c10a740 (origin/develop) gh-38492: expose fplll enumeration routines in IntegralLattice
8587c1351b7 gh-38479: Choose special name to avoid '@c' by singular's absFactorize
95a580db85e gh-38476: Include modular composition for polynomial rings over finite fields
cac6e29f1a8 gh-38470: `build/pkgs/furo`: Set lower bound
6522292ad72 gh-38466: introduce half-angle transformations
80ba1d5955c gh-38458: simplify conversion code in cubic Hecke
7c41ea58284 gh-38444: Moving ideal methods to category of rings
b20628a403f gh-38442: `build/pkgs/gcc`: Update to 13.3
a58fd635e62 gh-38433: Make # abs tol compare over the complex numbers
896eee58941 gh-38432: fix memory leak in is_planar
cee79e3abe2 gh-38414: Properly check dimension in curve constructor
ce92b938f24 gh-38409: `xgcd` for multiple arguments
d83f90cfafc gh-38395: get rid of some sage-eval in gap3-related code
f369f6576f5 gh-38347: support passing two base points to .log() for elliptic-curve points
8abb61cab64 gh-38331: use pkg-config to find msolve, otherwise try to run it
66f679d6f29 gh-38330: Lattes to curve function code commits for enhancement #38329
8d9da51214c gh-38323: some fixes for ruff code C41 (about loops and iteration)
5be4da7b396 gh-38296: Deprecate `is_SchemeMorphism`, `is_SchemeTopologicalPoint`
095e6ee6579 gh-38273: Update the deprecation policy
a1deb75ea13 gh-38185: Simplify doc build process by removing empty directories locally
27c9b02a90e gh-37372: Strength 2 Covering Array constructions
0f3477394fb (tag: 10.5.beta1) Updated SageMath version to 10.5.beta1

[Jaap Spies]

On Fedora 40 I get

----------------------------------------------------------------------
sage -t --long --warn-long 29.7 --random-seed=312965183184317696093439207698161157083 src/sage/rings/pol
ynomial/multi_polynomial_ideal.py  # 1 doctest failed
sage -t --long --warn-long 29.7 --random-seed=312965183184317696093439207698161157083 src/sage/rings/pol
ynomial/groebner_fan.py  # 47 doctests failed
----------------------------------------------------------------------
Those failures were already known in sage-10.4.rcx

Doctesting 1 file.
sage -t --long --warn-long 29.7 --random-seed=312965183184317696093439207698161157083 src/sage/rings/pol
ynomial/multi_polynomial_ideal.py
**********************************************************************
File "src/sage/rings/polynomial/multi_polynomial_ideal.py", line 4181, in sage.rings.polynomial.multi_po
lynomial_ideal.NCPolynomialIdeal.groebner_fan
Failed example:
   g.reduced_groebner_bases()
Exception raised:
   Traceback (most recent call last):
     File "/home/jaap/sagemath/sage-10.5.beta2/src/sage/rings/polynomial/groebner_fan.py", line 1057, i
n reduced_groebner_bases
       return self.__reduced_groebner_bases
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
   AttributeError: 'GroebnerFan' object has no attribute '_GroebnerFan__reduced_groebner_bases'. Did yo
u mean: '_GroebnerFan__is_groebner_basis'?

   During handling of the above exception, another exception occurred:

   Traceback (most recent call last):
     File "/home/jaap/sagemath/sage-10.5.beta2/src/sage/rings/polynomial/groebner_fan.py", line 1008, i
n _gfan_reduced_groebner_bases
       return self.__gfan_reduced_groebner_bases
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
   AttributeError: 'GroebnerFan' object has no attribute '_GroebnerFan__gfan_reduced_groebner_bases'

   During handling of the above exception, another exception occurred:

   Traceback (most recent call last):
     File "/home/jaap/sagemath/sage-10.5.beta2/src/sage/doctest/forker.py", line 715, in _run
       self.compile_and_execute(example, compiler, test.globs)
     File "/home/jaap/sagemath/sage-10.5.beta2/src/sage/doctest/forker.py", line 1145, in compile_and_e
xecute
       exec(compiled, globs)
     File "<doctest sage.rings.polynomial.multi_polynomial_ideal.NCPolynomialIdeal.groebner_fan[3]>", l
ine 1, in <module>
       g.reduced_groebner_bases()
     File "/home/jaap/sagemath/sage-10.5.beta2/src/sage/rings/polynomial/groebner_fan.py", line 1059, i
n reduced_groebner_bases
       G = self._gfan_reduced_groebner_bases()
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
     File "/home/jaap/sagemath/sage-10.5.beta2/src/sage/rings/polynomial/groebner_fan.py", line 1010, i
n _gfan_reduced_groebner_bases
       B = self.gfan(cmd='bases')
           ^^^^^^^^^^^^^^^^^^^^^^
     File "/home/jaap/sagemath/sage-10.5.beta2/src/sage/rings/polynomial/groebner_fan.py", line 1132, i
n gfan
       s = gfan(I, cmd, verbose=self.__verbose)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
     File "/home/jaap/sagemath/sage-10.5.beta2/src/sage/misc/decorators.py", line 656, in wrapper
       return func(*args, **kwds)
              ^^^^^^^^^^^^^^^^^^^
     File "/home/jaap/sagemath/sage-10.5.beta2/src/sage/interfaces/gfan.py", line 118, in __call__
       raise RuntimeError(err)
   RuntimeError: Falling back on CddLib
   Falling back on CddLib

**********************************************************************
1 item had failures:
  1 of   5 in sage.rings.polynomial.multi_polynomial_ideal.NCPolynomialIdeal.groebner_fan
   [986 tests, 1 failure, 3.74 s]
----------------------------------------------------------------------
sage -t --long --warn-long 29.7 --random-seed=312965183184317696093439207698161157083 src/sage/rings/pol
ynomial/multi_polynomial_ideal.py  # 1 doctest failed
----------------------------------------------------------------------
Total time for all tests: 4.3 seconds
   cpu time: 3.6 seconds
   cumulative wall time: 3.7 seconds

Jaap

--
You received this message because you are subscribed to the Google Groups "sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sage-release...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sage-release/07e44a5e-b281-45b4-b225-f62bdde24f72n%40googlegroups.com.

[Enrique Artal]

I had similar errors with system cddlib also in Fedora 40 and with a previous version, see https://github.com/sagemath/sage/pull/38358;  forcing the installation of the package, it works. I do not know why.

[Matthias Köppe]

I think it would be worth reporting it as a bug to Fedora.

[John H Palmieri]

This looks like a new failure on OS X 14.5:

sage -t --long --warn-long 24.7 --random-seed=330717456732602079039915918344955119792 src/sage/modules/free_quadratic_module_integer_symmetric.py
**********************************************************************
File "src/sage/modules/free_quadratic_module_integer_symmetric.py", line 1544, in sage.modules.free_quadratic_module_integer_symmetric.FreeQuadraticModule_integer_symmetric._fplll_enumerate
Failed example:
    [next(close) for _ in range(10)]
Expected:
    [(1, 0, 6, -9), (1, -1, 5, -9), (2, 0, 6, -9), (1, 0, 5, -9), (1, -1, 5, -10),
     (2, 1, 6, -9), (1, 0, 5, -10), (2, 0, 5, -9), (1, 0, 6, -8), (1, -1, 6, -9)]
Got:
    [(1, 0, 6, -9),
     (1, -1, 5, -9),
     (2, 0, 6, -9),
     (1, 0, 5, -9),
     (1, -1, 5, -10),
     (2, 1, 6, -9),
     (1, 0, 5, -10),
     (1, 0, 6, -8),
     (2, 0, 5, -9),
     (1, -1, 6, -9)]
**********************************************************************
File "src/sage/modules/free_quadratic_module_integer_symmetric.py", line 1601, in sage.modules.free_quadratic_module_integer_symmetric.FreeQuadraticModule_integer_symmetric.enumerate_short_vectors
Failed example:
    [next(short) for _ in range(20)]
Expected:
    [(1, -1, 1, 0), (2, -2, 2, 0), (3, -3, 3, 0), (0, 3, 2, 4), (1, 2, 3, 4),
     (4, 4, 1, 0), (3, 2, -2, -4), (3, 5, 0, 0), (4, 1, -1, -4), (-1, 4, 1, 4),
     (2, 1, 4, 4), (5, 3, 2, 0), (2, 3, -3, -4), (2, 6, -1, 0), (5, 0, 0, -4),
     (-2, 5, 0, 4), (4, -4, 4, 0), (6, 2, 3, 0), (1, 4, -4, -4), (3, 0, 5, 4)]
Got:
    [(1, -1, 1, 0),
     (2, -2, 2, 0),
     (3, -3, 3, 0),
     (0, 3, 2, 4),
     (1, 2, 3, 4),
     (4, 4, 1, 0),
     (3, 2, -2, -4),
     (3, 5, 0, 0),
     (4, 1, -1, -4),
     (-1, 4, 1, 4),
     (2, 1, 4, 4),
     (5, 3, 2, 0),
     (2, 3, -3, -4),
     (5, 0, 0, -4),
     (2, 6, -1, 0),
     (-2, 5, 0, 4),
     (4, -4, 4, 0),
     (6, 2, 3, 0),
     (1, 4, -4, -4),
     (3, 0, 5, 4)]
**********************************************************************
2 items had failures:
   1 of   7 in sage.modules.free_quadratic_module_integer_symmetric.FreeQuadraticModule_integer_symmetric._fplll_enumerate
   1 of   4 in sage.modules.free_quadratic_module_integer_symmetric.FreeQuadraticModule_integer_symmetric.enumerate_short_vectors
    [239 tests, 2 failures, 4.37 s]
----------------------------------------------------------------------
sage -t --long --warn-long 24.7 --random-seed=330717456732602079039915918344955119792 src/sage/modules/free_quadratic_module_integer_symmetric.py  # 2 doctests failed

[Enrique Artal]

Reported: https://bugzilla.redhat.com/show_bug.cgi?id=2305344

Already got a reply (and I do not know what to say):

%%%%%%

The only important difference between the Fedora 39 and 40 cddlib package builds is the addition of this upstream commit: https://github.com/cddlib/cddlib/commit/1210faf9ee9cf02e29fc67e11a8ea4e847c58ddd Does sagemath access cddlib from multiple threads and assume they will share data? %%%%%%

[Dima Pasechnik]

line 113 in this file, before the error is caught,
is
ans, err = gfan_processes.communicate(input=input)

What `ans` and `err` do you get while running this test?
(add a print and see, please)

>>>>> <https://groups.google.com/d/msgid/sage-release/07e44a5e-b281-45b4-b225-f62bdde24f72n%40googlegroups.com?utm_medium=email&utm_source=footer>
>>>>> .
>>>>>
>>>>
>

[Enrique Artal]

For the test in line 967 in src/sage/rings/polynomial/groebner_basis.py, after adding print(ans, err) to weight_vectors it gives

spkg: [(3, 7, 1), (5, 1, 2), (7, 1, 4), (5, 1, 4), (1, 1, 1), (1, 4, 8), (1, 4, 10)]

system: b'{\n(3,7,1),\n(5,1,2),\n(7,1,4),\n(5,1,4),\n(1,1,1),\n(1,4,8),\n(1,4,10)}\n' b'List of linked LP solvers:\n SoPlexCddGmp\n cddgmp\n cdd\n'
[(3, 7, 1), (5, 1, 2), (7, 1, 4), (5, 1, 4), (1, 1, 1), (1, 4, 8), (1, 4, 10)]

[Enrique Artal]

Sorry, with the spkg it is; 

b'{\n(3,7,1),\n(5,1,2),\n(7,1,4),\n(5,1,4),\n(1,1,1),\n(1,4,8),\n(1,4,10)}\n' b'List of linked LP solvers:\n cddgmp\n cdd\n'

[(3, 7, 1), (5, 1, 2), (7, 1, 4), (5, 1, 4), (1, 1, 1), (1, 4, 8), (1, 4, 10)]

[Dima Pasechnik]

On 17 August 2024 10:39:35 BST, Enrique Artal <enriqu...@gmail.com> wrote:
>Sorry, with the spkg it is;
>b'{\n(3,7,1),\n(5,1,2),\n(7,1,4),\n(5,1,4),\n(1,1,1),\n(1,4,8),\n(1,4,10)}\n'
>b'List of linked LP solvers:\n cddgmp\n cdd\n'
>[(3, 7, 1), (5, 1, 2), (7, 1, 4), (5, 1, 4), (1, 1, 1), (1, 4, 8), (1, 4,
>10)]

That is, no runtime error raised for you?
Or it comes later?

[Enrique Artal]

Sorry, I misunderstood and put the print in rings/polynomial/groebner_fan.py and put the output of the test of the changed function.  I put now 

       print("ans = ", ans)
       print('err = ', err)
in interfaces/gfan.py and got

ans =  Q[x,y]
{{
x^2+1-y^2}
,
{
y^2-1-x^2}
}

err =  Falling back on CddLib
Falling back on CddLib

[John H Palmieri]

I think this failure is due to https://github.com/sagemath/sage/pull/38492

[G. M.-S.]

FWIW, here is what I get on Apple Silicon.

macOS 12.7.6:  All tests passed!

macOS 13.6.9:  This failure (among others).

macOS 14.6.1:  This failure (among others).

Guillermo

[Enrique Artal]

There is a new cddlib rpm for fedora which tries to solve the problem. It doesn't for me but I am not sure if it is because I do something wrong, since I get the same errors with former packages. Can someone else with access to Fedora 40 make the test? 

Anyway, I do not know how to proceed when a system package is updated. Must sage be rebuilt? The exact steps would be helpful. Best, Enrique.

[Dima Pasechnik]

On 20 August 2024 09:06:24 BST, Enrique Artal <enriqu...@gmail.com> wrote:
>There is a new cddlib rpm for fedora which tries to solve the problem. It
>doesn't for me but I am not sure if it is because I do something wrong,
>since I get the same errors with former packages. Can someone else with
>access to Fedora 40 make the test?

I can do this next week. Afk now.

>Anyway, I do not know how to proceed when a system package is updated.

when a system library is updated, and is not compatible with the previous version (some updates might be e.g. just adding new functionality)
then the Python modules using it, and excutables/libraries using it, might need a rebuild.

In most cases it's just sagelib which needs a rebuild:

make sagelib-clean sagelib-uninstall && make build

will do the job.
However in case of cddlib, there are more packages which depend on it, e.g. singular.
(although you might get Singular from OS, too, then you don't need to (re)build it, obviously.)

git grep cdd build/pkgs/*/dep*

should give you an idea what needs a rebuild.



Dima

[Enrique Artal]

Thanks! Rebuilding gfan rpm and singular spkg (also sagelib), the tests pass.