Sage 10.7.beta6 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

With this version I can now build Sage with gcc 15 (Fedora 42), test pass except for one segfault in src/sage/schemes/curves/point.py

11baeed4600 (origin/develop, tag: 10.7.beta6) Updated SageMath version to 10.7.beta6
10c19f1b79e gh-40242: cleaning coding/ using cython-lint
63b5554e3df gh-40240: Fix a bug in khuri-makdisi small model
c81c2ab5566 gh-40235: Annoate methods with `Self`
eabf4d95333 gh-40234: some details in ternary quadratic forms
281c31a7673 gh-40231: Remove compilation dependency of `rings/convert/mpfi` on cypari2
e1547099d9d gh-40230: Port ECL to C23
616376e7232 gh-40229: Upgrade to palp 2.21
26ee21e1a3b gh-40228: Apply gap build fix for gcc-15
b8cedec1ea4 gh-40227: Upgrade to symmetrica 3.1.0
be4f8ac4e62 gh-40226: glpk build fix for gcc-15 (remove duplicate bool typedef)
436948e9c52 gh-40225: Complete function prototype gmp for C23 compatibility
ff0b6149f79 gh-40222: pep8 cleanup for KR tableaux
a1f9934211f gh-40221: Fix caching in Khuri-Makdisi Jacobian implementation
7179e533078 gh-40220: some cleanup for ell_field.py
f453378a134 gh-40218: some cleanup in combinat/sf
693eac4313b gh-40208: add checks to the linter once again
394db56903c gh-40206: fix some typos in modular
7e5f0fbf764 gh-40203: Fix misspelt word `intalling` -> `installing` in suitesparse documentation
259e2cdb701 gh-40202: fix docstring in btquotient.py
703182a50c6 gh-40201: Hochschild lattices and fans
d775c8cd7cc gh-40200: faster listing of oriented triangles in `src/sage/combinat/cluster_algebra_quiver/mutation_type.py`
9ecd4070b0a gh-40199: details about ruff code RET in modular folder
f33495e5b26 gh-40197: simplify some details in posets
836bba381d5 gh-40196: further typing annotations for methods has_something
dcc0f652cd3 gh-40195: some fixes for ruff,pyflakes,pycodestyle
fd1a51f89b2 gh-40192: Fix synchronize labels GitHub action
d79515447b9 gh-40190: cddlib[tools] needed for cddlib on gentoo
c8c30f57b5e gh-40185: Drop sage-cleaner interface
a2b0b263e8a gh-40184: Fixes #40134: correct signs in inner_product_matrix docstring
014d67a7ead gh-40183: Fpylll cython3.1 fix
86d4e5ab42e gh-40181: Increase tolerance for agm
e0762e2439c gh-40154: Add python-dateutil to documentation dependencies
2732afff962 gh-40153: update planaritry to 4.0.0.0
bcdc712fb01 gh-40141: Adding the polylogarithm as lazy series
ecef2daa303 gh-40137: Rewriting Lie subalgebras and ideals to always do computations in the ambient Lie algebra
eb1bc9f96bd gh-40117: Implement crt for multivariate polynomials
54fc233caf4 gh-40086: some fixes for C4 (set comprehensions)
42c9b1ac989 gh-40082: simplify a little bit "_an_element_" in parent_old.pyx
9fc76607dd9 gh-40011: remove bzip2 spkg
ef0a63ddfeb gh-39741: Implementation of Specht and simple modules for signed permutation group
c8d0afbba95 gh-39639: sketch of svg display for tableaux
85df9ff4ede gh-39598: adding has_subgraph_decomposition method to GenericGraph
9352a3258c0 (tag: 10.7.beta5) Updated SageMath version to 10.7.beta5

[Volker Braun]

I've created https://github.com/sagemath/sage/issues/40253 for the segfault

$ ./sage -gdb
sage: A.<x,y> = AffineSpace(QQ, 2) ## line 304 ##
sage: C = A.curve([x^5 - x^3*y^2 + 5*x^4 - x^3*y - 3*x^2*y^2 +
....: x*y^3 + 10*x^3 - 3*x^2*y - 3*x*y^2 + y^3 + 10*x^2 - 3*x*y - y^2 +
....: 5*x - y + 1]) ## line 305 ##
[Thread 0x7fff997496c0 (LWP 1881254) exited]

Thread 1 "python" received signal SIGSEGV, Segmentation fault.
0x00007fffa3315f4a in fmpz_poly_divides (q=q@entry=0x7fffffffacb0, a=0x555557c5bef8, b=0xffffffffffffffe8) at src/fmpz_poly/divides.c:69
69     if (fmpz_poly_is_zero(b))
(gdb) bt
#0  0x00007fffa3315f4a in fmpz_poly_divides (q=q@entry=0x7fffffffacb0, a=0x555557c5bef8, b=0xffffffffffffffe8) at src/fmpz_poly/divides.c:69
#1  0x00007fffa35a4c54 in fmpz_bpoly_divides (Q=Q@entry=0x7fffffffae10, A=A@entry=0x7fffffffb290, B=B@entry=0x7fffffffae50)
    at src/fmpz_mpoly_factor/bpoly_factor.c:169
#2  0x00007fffa35a610e in _recombine_naive (F=0x7fffffffb2c0, B=0x7fffffffb290, alpha=0x7fffffffadb0, I=0x7fffffffaf00)
    at src/fmpz_mpoly_factor/bpoly_factor.c:744
#3  fmpz_bpoly_factor (c=c@entry=0x7fffffffb100, F=F@entry=0x7fffffffb2c0, B=B@entry=0x7fffffffb290) at src/fmpz_mpoly_factor/bpoly_factor.c:918
#4  0x00007fffa35a7f92 in _factor_irred_compressed (Af=Af@entry=0x7fffffffb3b0, A=0x555558111c30, ctx=ctx@entry=0x7fffffffb970, algo=algo@entry=7)
    at src/fmpz_mpoly_factor/factor.c:186
#5  0x00007fffa35a95ec in _compressed_content_to_irred (g=g@entry=0x7fffffffb4c0, f=f@entry=0x555557b4a8f8, e=<optimized out>,
    ctx=ctx@entry=0x7fffffffb970, algo=algo@entry=7) at src/fmpz_mpoly_factor/factor.c:650
#6  0x00007fffa35a9818 in _factor (f=0x7fffffffb940, ctx=0x7fffffffb970, algo=7, A=<optimized out>) at src/fmpz_mpoly_factor/factor.c:727
#7  0x00007fffa1a4db45 in factorize (f=..., issqrfree=issqrfree@entry=false) at cf_factor.cc:710
#8  0x00007fffa1d2d378 in singclap_factorize (f=<optimized out>, v=v@entry=0x7fffffffbe98, with_exps=with_exps@entry=0, r=r@entry=0x7fffa1825030)
    at clapsing.cc:1077
#9  0x00007fffa2217e3f in __pyx_pf_4sage_5rings_10polynomial_28multi_polynomial_libsingular_23MPolynomial_libsingular_104factor (

[Kazuyoshi Furutaka]

Hi,

Unfortunately, the build failed on my fedora-42 machine due to maxima (the log attached).

By the way, it seems that there's no rpm package named maxima-runtime-ecl:

$ dnf list maxima-runtime\*
Updating and loading repositories:
Repositories loaded.
Installed packages
maxima-runtime-sbcl.x86_64  5.47.0-8.fc42 fedora
Available packages
maxima-runtime-clisp.x86_64 5.47.0-8.fc42 fedora
maxima-runtime-gcl.x86_64   5.47.0-8.fc42 fedora

Kazuyoshi

2025年6月15日(日) 7:21 Volker Braun <vbrau...@gmail.com>:

--
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 visit https://groups.google.com/d/msgid/sage-release/bafb42fb-b71c-4905-9688-507ab5561374n%40googlegroups.com.

--

Kazuyoshi Furutaka

[Volker Braun]

You are picking up /usr/include/ecl/object.h, which isn't patched to work with gcc-15. Where did you get the rpm from? I don't see a f42 build on https://koji.fedoraproject.org/koji/packageinfo?packageID=1640. 

It should work if you uninstall the ecl rpm so we can build our own.

[Kazuyoshi Furutaka]

Hi again...

$ rpm -qif /usr/include/ecl/object.hName        : ecl
Version     : 23.9.9
Release     : 3.fc40
Architecture: x86_64
Install Date: Wed 24 Apr 2024 06:47:51 PM JST
Group       : Unspecified
Size        : 19869547
License     : LGPL-2.0-or-later AND BSD-2-Clause AND MIT AND X11 AND HPND AND LOOP AND LicenseRef-Fedora-Public-Domain
Signature   : RSA/SHA256, Wed 24 Jan 2024 07:30:07 PM JST, Key ID 0727707ea15b79cc
Source RPM  : ecl-23.9.9-3.fc40.src.rpm
Build Date  : Wed 24 Jan 2024 07:14:19 PM JST
Build Host  : buildvm-x86-28.iad2.fedoraproject.org
Packager    : Fedora Project
Vendor      : Fedora Project
URL         : https://common-lisp.net/project/ecl/
Bug URL     : https://bugz.fedoraproject.org/ecl
Summary     : Embeddable Common-Lisp
Description :
ECL (Embeddable Common Lisp) is an implementation of the Common Lisp
language as defined by the ANSI X3J13 specification.  ECL features a
bytecode compiler and interpreter, the ability to build standalone
executables and libraries, and extensions such as ASDF, sockets, and
Gray streams.

So it seems to be a remnant of F40 rpms...

The build after doing `dnf erase ecl && dnf reconfigure` finished successfully!

Tnx...

Kazuyoshi  

2025年6月15日(日) 20:34 Volker Braun <vbrau...@gmail.com>:

To view this discussion visit https://groups.google.com/d/msgid/sage-release/3a1dac88-5c08-4dd0-b1ce-98aa44f4a8b6n%40googlegroups.com.

--

Kazuyoshi Furutaka