Sage 10.9.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

565f7eced5f (tag: 10.9.beta2, github/develop) Updated SageMath version to 10.9.beta2
1b5a5dd78b5 gh-41395: Fix test failures with pyparsing 3.3
34ba69bafc5 gh-41394: Fix one more doctest failure with Python 3.14
9837e722d76 gh-41392: remove some deprecated stuff in modules
d0466ce5767 gh-41391: Fix inline literal interpreted as LaTeX math
b1796e26135 gh-41385: remove some deprecated material in interfaces
c5e75433344 gh-41384: remove deprecated stuff in homology
242e1fb1209 gh-41383: remove some deprecated stuff in groups
3f91b4b1c73 gh-41382: remove deprecated stuff in geometry
0ca70d0aa6a gh-41381: remove deprecated stuff in data_structures
9d81532bfa6 gh-41380: remove deprecated functions in algebras/
abdbdeab044 gh-41379: remove some deprecated things in combinat/
f76a06b5e86 gh-41378: a bunch of typos
b8c381fc984 gh-41373: The bases for the free Lie algebras are graded bases
e975db81354 gh-41371: Improve printing of elements of free nilpotent Lie algebra
deafe32bdf0 gh-41369: various small changes in combinat/
d42459e5a2f gh-41360: bump memory_allocator to 0.2.0
af67015a2fd gh-41358: Remove unused variables (mostly SAGE_VENV)
449dbde5d79 gh-41350: some cleanup in lie_conformal_algebras/
1eee112c9db gh-41347: bump cddlib to 0.94n
380949e6eed gh-41346: Fix execution_count for newer ipython version
56f67c507ef gh-41345: fix Laplace transform doctests
5fb0dee7ab4 gh-41341: bump primecount to 8.0 and primecountpy to 0.2.1
14ca933f6c3 gh-41340: RealSet: Implement simplest_rational for RealSet
87627f3172c gh-41323: some type annotations in categories/examples
be619561422 gh-41309: RDF[] for libsingular, delegate ideal comparison and `.divides()` to singular
e79d621a09d gh-41306: Implement Polynomial_real_arb
eb86861049d gh-41247: Update FLINT package version and checksums to 3.4.0
e46eb0f6271 gh-41142: Docs: unconditionally show plain Python example
1a14a417aed gh-40999: implement the lattices of sashes and pellytopes
38c57e0c680 gh-40945: introduce the category of graded lattices
5b824675148 gh-40465: using pari for elliptic and Eisenstein L-series
845d8d42edc gh-39909: Remove some confusion in gap interface conversion method
3bfcc2bde13 gh-39383: use decomposition into bi/tri-connected components and atoms for iterating over minimal separators
4015f9189b6 (tag: 10.9.beta1) Updated SageMath version to 10.9.beta1

[Jaap Spies]

On Fedora 42 upgraded from beta1, with  CC=gcc-14 CXX=g++-14 FC=gfortran-14

----------------------------------------------------------------------
All tests passed!
----------------------------------------------------------------------
Total time for all tests: 739.7 seconds
   cpu time: 5181.2 seconds
   cumulative wall time: 4718.6 seconds

--
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/3766307b-78db-497d-a9bc-a9d884738708n%40googlegroups.com.

[John H Palmieri]

With Python3.14 on OS X, I am seeing some doctests fail because of the warning message

"DeprecationWarning: codecs.open() is deprecated. Use open() instead."

(At least I think this is due to using Python3.14.) For example I see failures in sage/misc/sageinspect.py, sage/doctest/forker.py, and others. Perhaps take care of this for now by removing the warning text in sage/doctest/parsing.py?

[Dima Pasechnik]

On Sat, Jan 10, 2026 at 9:11 PM John H Palmieri <jhpalm...@gmail.com> wrote:
>
> With Python3.14 on OS X, I am seeing some doctests fail because of the warning message
>
> "DeprecationWarning: codecs.open() is deprecated. Use open() instead."
>
> (At least I think this is due to using Python3.14.) For example I see failures in sage/misc/sageinspect.py, sage/doctest/forker.py, and others. Perhaps take care of this for now by removing the warning text in sage/doctest/parsing.py?

These come from cython.

https://github.com/sagemath/sage/pull/41409

takes care of this by updating cython to the version where these are fixed.

Dima

> To view this discussion visit https://groups.google.com/d/msgid/sage-release/a5102c5a-aab0-4457-8ce5-15354caa4dcen%40googlegroups.com.

[John H Palmieri]

Got it, thanks, Dima.

I've been trying to jump through lots of hoops to get Sage to build and for doctests to pass. Doctests are still not passing for me on OS X, in part because of old previously reported problems (#40869, #40873), but I'm seeing a new set of failures. Something is going wrong with text processing and control sequences:

src/bin/sage -t --long --warn-long 30.0 --random-seed=36948321267754839088553570681149644913 src/sage/quivers/algebra_elements.pxi
**********************************************************************
File "src/sage/quivers/algebra_elements.pxi", line 306, in sage.quivers.algebra_elements.pxi._FreeListProtector.__dealloc__
Failed example:
    s.eval("P = DiGraph({1:{1:['x','y','z']}}).path_semigroup().algebra(GF(25,'t'))")
Expected:
    ''
Got:
    '\x1b[0m\x1b[1;34mfrom\x1b[0m\x1b[8D\x1b[8C\x1b[8D\x1b[1;35'
**********************************************************************
File "src/sage/quivers/algebra_elements.pxi", line 308, in sage.quivers.algebra_elements.pxi._FreeListProtector.__dealloc__
Failed example:
    s.eval("P.inject_variables()")
Expected:
    'Defining e_1, x, y, z'
Got:
    '\x1b[0m\x1b[1;34mfrom\x1b[0m \x1b[9D\x1b[9C\x1b[9D\x1b[1;35'
**********************************************************************
File "src/sage/quivers/algebra_elements.pxi", line 310, in sage.quivers.algebra_elements.pxi._FreeListProtector.__dealloc__
Failed example:
    s.eval("x*y+y*z*x")
Expected:
    'x*y + y*z*x'
Got:
    '\x1b[0m\x1b[1;34mfrom\x1b[0m s\x1b[10D\x1b[10C\x1b[10D\x1b[1;35'
**********************************************************************
1 item had failures:

I'm also seeing failures like this after upgrading to OS X 26.2:

File "src/sage/misc/sagedoc.py", line 709, in sage.misc.sagedoc.format
Failed example:
    cython('\n'.join(cython_code))                                            # needs sage.misc.cython
Expected nothing
Got:
    ld: warning: building for macOS-15.0, but linking with dylib '/Users/palmieri/Sage/TESTING/sage-10.9.beta1/local/lib/libec.14.dylib' which was built for newer version 26.0
    ld: warning: building for macOS-15.0, but linking with dylib '/Users/palmieri/Sage/TESTING/sage-10.9.beta1/local/lib/libgsl.27.dylib' which was built for newer version 26.0

We should be able to filter out warning messages like these, but I think we should be able to set flags to specify which version of Mac OS we're building for. Is MACOSX_DEPLOYMENT_TARGET being incorrectly set somewhere?

[Dima Pasechnik]

On Sun, Jan 11, 2026 at 3:26 PM John H Palmieri <jhpalm...@gmail.com> wrote:
>
> Got it, thanks, Dima.
>
> I've been trying to jump through lots of hoops to get Sage to build and for doctests to pass. Doctests are still not passing for me on OS X, in part because of old previously reported problems (#40869, #40873), but I'm seeing a new set of failures. Something is going wrong with text processing and control sequences:
>
> src/bin/sage -t --long --warn-long 30.0 --random-seed=36948321267754839088553570681149644913 src/sage/quivers/algebra_elements.pxi
> **********************************************************************
> File "src/sage/quivers/algebra_elements.pxi", line 306, in sage.quivers.algebra_elements.pxi._FreeListProtector.__dealloc__
> Failed example:
> s.eval("P = DiGraph({1:{1:['x','y','z']}}).path_semigroup().algebra(GF(25,'t'))")
> Expected:
> ''
> Got:
> '\x1b[0m\x1b[1;34mfrom\x1b[0m\x1b[8D\x1b[8C\x1b[8D\x1b[1;35'

hmm, we got ANSI terminal control sequences here. (I see it on an
Intel mac with Python 3.14.2, too)

I ran this through https://ansi.tools
and found that it is basically printing the string "from s",
where "from" is bold blue, and then breaks down.

\x1b[0m is "reset"
\x1b[1;34m is "bold, blue foreground"
\x1b[nD is "move cursor 'n' columns back" - in your case n=8, I see n=10.
\x1b[nC is "move cursor 'n' columns forward"
\x1b[1;35 is an illegal/incomplete sequence.

Is this a beginning of a pexpect interaction, which goes wrong?
I have no idea. Why on Earth does it make it bold blue?

Dima

[Dima Pasechnik]

I've opened https://github.com/sagemath/sage/issues/41438
to deal with this

[Dima Pasechnik]

On Mon, Jan 12, 2026 at 3:06 PM Dima Pasechnik <dim...@gmail.com> wrote:
>
> I've opened https://github.com/sagemath/sage/issues/41438
> to deal with this

workaround: do

export PYTHON_BASIC_REPL=1

then everything is back to normal, it appears. E.g. ./sage -t
src/sage/quivers/algebra_elements.pxi passes