fix(Mpolynomial_libsingular): support division over transcendental ex… #40076

polykernel · 2025-05-09T04:41:20Z

…tension

Currently, dividing a Singular backed multivariate polynomial by a fractional field element(of the same type) is unsupported and triggers a conversion error on the Singular end indicating the denominator is nonconstant. This commit extracts the commit from Singular trunk which adds support for the use case¹ to a patch to be applied on top of release 4.4.1.

fixes #39801

Relevant discussions on sage-devel: https://groups.google.com/g/sage-devel/c/Npn48ofmHF4

I have opted to not add doctests due to anticipated difficulty in coordinating adoption of the patch across different distribution channels. Notwithstanding this, I still think doctests should be added at some point to guard against regression. If this is important enough to warrant adding the relevant tests now, please let me know. Otherwise, I think the tests can be added after the next Singular release whenever that occurs.

This is my first time contributing to sage. It is possible I missed some details in the developer documentation and contributing guidelines. Please advise me if I am doing anything unidiomatic or incorrect.

📝 Checklist

The title is concise and informative.
The description explains in detail what this PR is about.
I have linked a relevant issue or discussion.
I have created tests covering the changes.
I have updated the documentation and checked the documentation preview.

⌛ Dependencies

#40033 - potential compatibility issues/breakage with Singular release 4.4.1 should be sorted out first

https://github.com/Singular/Singular/commit/0a4bc0e2888e660d1a5229a7b4842539a46c912a ↩

fchapoton · 2025-05-09T05:45:56Z

you need to add a doctest showing what has been fixed

dimpase · 2025-05-09T15:39:01Z

CI fails (e.g. https://github.com/sagemath/sage/actions/runs/14929279618/job/41946713381?pr=40076) with

sage -t --warn-long 5.0 --random-seed=12863317776592486881340390273080966829 src/sage/rings/polynomial/multi_polynomial_libsingular.pyx
**********************************************************************
Error: Failed example:: Exception raised:
Traceback (most recent call last):
  File "/Users/runner/miniconda3/envs/sage-dev/lib/python3.11/site-packages/sage/doctest/forker.py", line 730, in _run
    self.compile_and_execute(example, compiler, test.globs)
  File "/Users/runner/miniconda3/envs/sage-dev/lib/python3.11/site-packages/sage/doctest/forker.py", line 1154, in compile_and_execute
    exec(compiled, globs)
  File "<doctest sage.rings.polynomial.multi_polynomial_libsingular.MPolynomial_libsingular._div_[29]>", line 1, in <module>
    f / d
    ~~^~~
  File "element.pyx", line 1736, in sage.structure.element.Element.__truediv__
  File "coerce.pyx", line 1238, in sage.structure.coerce.CoercionModel.bin_op
  File "element.pyx", line 1734, in sage.structure.element.Element.__truediv__
  File "multi_polynomial_libsingular.pyx", line 2364, in sage.rings.polynomial.multi_polynomial_libsingular.MPolynomial_libsingular._div_
  File "parent.pyx", line 902, in sage.structure.parent.Parent.__call__
  File "coerce_maps.pyx", line 183, in sage.structure.coerce_maps.DefaultConvertMap_unique._call_with_args
  File "coerce_maps.pyx", line 176, in sage.structure.coerce_maps.DefaultConvertMap_unique._call_with_args
  File "/Users/runner/miniconda3/envs/sage-dev/lib/python3.11/site-packages/sage/rings/fraction_field.py", line 783, in _element_constructor_
    return self._element_class(self, x, y, coerce=coerce)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "fraction_field_element.pyx", line 130, in sage.rings.fraction_field_element.FractionFieldElement.__init__
ZeroDivisionError: fraction field element division by zero

    f / d
Exception raised:
    Traceback (most recent call last):
      File "/Users/runner/miniconda3/envs/sage-dev/lib/python3.11/site-packages/sage/doctest/forker.py", line 730, in _run
        self.compile_and_execute(example, compiler, test.globs)
      File "/Users/runner/miniconda3/envs/sage-dev/lib/python3.11/site-packages/sage/doctest/forker.py", line 1154, in compile_and_execute
        exec(compiled, globs)
      File "<doctest sage.rings.polynomial.multi_polynomial_libsingular.MPolynomial_libsingular._div_[29]>", line 1, in <module>
        f / d
        ~~^~~
      File "element.pyx", line 1736, in sage.structure.element.Element.__truediv__
      File "coerce.pyx", line 1238, in sage.structure.coerce.CoercionModel.bin_op
      File "element.pyx", line 1734, in sage.structure.element.Element.__truediv__
      File "multi_polynomial_libsingular.pyx", line 2364, in sage.rings.polynomial.multi_polynomial_libsingular.MPolynomial_libsingular._div_
      File "parent.pyx", line 902, in sage.structure.parent.Parent.__call__
      File "coerce_maps.pyx", line 183, in sage.structure.coerce_maps.DefaultConvertMap_unique._call_with_args
      File "coerce_maps.pyx", line 176, in sage.structure.coerce_maps.DefaultConvertMap_unique._call_with_args
      File "/Users/runner/miniconda3/envs/sage-dev/lib/python3.11/site-packages/sage/rings/fraction_field.py", line 783, in _element_constructor_
        return self._element_class(self, x, y, coerce=coerce)
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
      File "fraction_field_element.pyx", line 130, in sage.rings.fraction_field_element.FractionFieldElement.__init__
    ZeroDivisionError: fraction field element division by zero
**********************************************************************
Error: Failed example:: Got: 0
                                                                                                                                                                                                                                                                                                                                                                                                                                                                                          
    f // g
Expected:
    1/a
Got:
    0
**********************************************************************
Error: Failed example:: Got: doctest:warning
  File "<doctest sage.rings.polynomial.multi_polynomial_libsingular.MPolynomial_libsingular.factor[2]>", line 1, in <module>
    F = f.factor(); F
  File "/Users/runner/miniconda3/envs/sage-dev/lib/python3.11/warnings.py", line 110, in _showwarnmsg
    sw(msg.message, msg.category, msg.filename, msg.lineno,
:
UserWarning: error in Singular ignored: 
x * (x^2 + x + 1) * (x^2 + 2*y^2)
                                                            
    F = f.factor(); F
Expected:
    x * (x^2 + x + 1) * (x^2 + 2*y^2)
Got:
    doctest:warning
      File "<doctest sage.rings.polynomial.multi_polynomial_libsingular.MPolynomial_libsingular.factor[2]>", line 1, in <module>
        F = f.factor(); F
      File "/Users/runner/miniconda3/envs/sage-dev/lib/python3.11/warnings.py", line 110, in _showwarnmsg
        sw(msg.message, msg.category, msg.filename, msg.lineno,
    :
    UserWarning: error in Singular ignored: 
    x * (x^2 + x + 1) * (x^2 + 2*y^2)
**********************************************************************
3 items had failures:
   1 of  31 in sage.rings.polynomial.multi_polynomial_libsingular.MPolynomial_libsingular._div_
   1 of  36 in sage.rings.polynomial.multi_polynomial_libsingular.MPolynomial_libsingular._floordiv_
   1 of 135 in sage.rings.polynomial.multi_polynomial_libsingular.MPolynomial_libsingular.factor
    [1352 tests, 3 failures, 30.77s wall]

polykernel · 2025-05-09T16:14:49Z

The first two CI failures (for _div_ and _floordiv_) appears to be caused by meson using Singular 4.4.1 from conda-forge instead of building from source with the patch. I don't know how to fix this part as it doesn't seem like the meson build system is hooked into build/pkgs at all and there doesn't seem to be any precedence with configuring subprojects and wraps.

I am not sure about the cause of the third error on factor.

dimpase · 2025-05-12T01:38:46Z

We now have Singular 4.4.1 in the develop branch, so you'll need to update this.

polykernel · 2025-05-12T13:19:59Z

Thanks for the heads-up, I have rebased on top of the develop branch and squashed the fixup commit.

…tension Currently, dividing a Singular backed multivariate polynomial by a fractional field element(of the same type) is unsupported and triggers a conversion error on the Singular end indicating the denominator is nonconstant. This commit extracts the commit from Singular trunk which adds support for the use case[^1] to a patch to be applied on top of release 4.4.1. We force an in-source build of Singular if the system Singular package does not contain the patch. [^1]: Singular/Singular@0a4bc0e

polykernel · 2025-05-23T15:14:04Z

There still seems to be some mismatches with sonames. I am not very familiar with the sage build system, any pointers on how to resolve these would be greatly appreciated.

dimpase · 2025-05-23T21:03:58Z

do you mean, in CI logs?
That's a current "feature" of our CI : 😢

polykernel · 2025-05-26T17:11:48Z

I see. I have tested the changes locally and addressed the previous comments. There is one doctest failure I got when testing the changes to multi_polynomial_libsingular.pyx.

File "src/sage/rings/polynomial/multi_polynomial_libsingular.pyx", line 4202, in sage.rings.polynomial.multi_polynomial_libsingular.MPolynomial_libsingular._floordiv_
Failed example:
    ((x+y)^3+x+z)//(x+y)
Expected:
    x^2 + 2*x*y + y^2
Got:
    x^2 + 2*x*y + y^2 + 1
**********************************************************************
1 item had failures:
   1 of  36 in sage.rings.polynomial.multi_polynomial_libsingular.MPolynomial_libsingular._floordiv_
    [1352 tests, 1 failure, 19.72s wall]

I am not sure why the expected result is $x^2 + 2xy + y^2$. Under the default graded reverse lexicographic ordering, the quotient is indeed $x^2 + 2xy + y^2 + 1$ (verified with Macaulay2). I did a quick ./sage -br against the develop branch and the testcase also fails there so I don't think this is a regression caused by this PR. In any case, it would be great if someone else can verify this is indeed the case.

Is there anything else that I need to work on to get the PR to a mergeable state?

dimpase · 2025-05-26T17:27:02Z

OK, so this doctest should be fixed. Please don't just correct the result, but add some text explaining why this is true.

polykernel · 2025-05-26T18:06:24Z

Understood, should I open another PR for the doctest fix?

dimpase · 2025-05-26T18:09:02Z

why - this one is still open, do it here

…d output This commit updates the expected output for one failing doctest under the _floordiv_ function, namely the test for floor division over large prime characteristic fields. The offending test computes ((x+y)^3+x+z)//(x+y) with the expected answer x^2 + 2*x*y + y^2 under the default grevlex order. However the actual quotient should be x^2 + 2*x*y + y^2 + 1 as computed by Macaulay2 and verified by hand calculation.

polykernel · 2025-05-27T01:56:29Z

Ok, I thought it would be beneficial to track these changes separately since they are unrelated. I have just appended a commit fixing the doctest.

dimpase · 2025-05-27T02:12:10Z

if this test result changes as a result of this PR changes, then it should be updated here. Otherwise it's up to you.

github-actions · 2025-06-12T16:07:57Z

Documentation preview for this PR (built with commit d6eb1e0; changes) is ready! 🎉
This preview will update shortly after each push to this PR.

github-actions bot added the s: needs review label May 9, 2025

polykernel force-pushed the fix/mpolynomial_libsingular-division-over-transext branch from 43fadd2 to c1cc3c7 Compare May 9, 2025 04:42

polykernel force-pushed the fix/mpolynomial_libsingular-division-over-transext branch from c1cc3c7 to 161c2a9 Compare May 9, 2025 12:47

polykernel force-pushed the fix/mpolynomial_libsingular-division-over-transext branch from 10698a2 to 55259bf Compare May 9, 2025 21:33

fchapoton added the c: commutative algebra label May 11, 2025

polykernel force-pushed the fix/mpolynomial_libsingular-division-over-transext branch from 55259bf to d866b47 Compare May 12, 2025 13:13

polykernel force-pushed the fix/mpolynomial_libsingular-division-over-transext branch from d866b47 to fb8d1c5 Compare May 23, 2025 01:28

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Uh oh!

fix(Mpolynomial_libsingular): support division over transcendental ex… #40076

fix(Mpolynomial_libsingular): support division over transcendental ex… #40076

Uh oh!

polykernel commented May 9, 2025 •

edited

Loading

Uh oh!

fchapoton commented May 9, 2025

Uh oh!

dimpase commented May 9, 2025 •

edited

Loading

Uh oh!

polykernel commented May 9, 2025

Uh oh!

dimpase commented May 12, 2025

Uh oh!

polykernel commented May 12, 2025 •

edited

Loading

Uh oh!

polykernel commented May 23, 2025

Uh oh!

dimpase commented May 23, 2025

Uh oh!

polykernel commented May 26, 2025

Uh oh!

dimpase commented May 26, 2025

Uh oh!

polykernel commented May 26, 2025

Uh oh!

dimpase commented May 26, 2025

Uh oh!

polykernel commented May 27, 2025

Uh oh!

dimpase commented May 27, 2025

Uh oh!

github-actions bot commented Jun 12, 2025

Uh oh!

Uh oh!

Uh oh!

fix(Mpolynomial_libsingular): support division over transcendental ex… #40076

Are you sure you want to change the base?

fix(Mpolynomial_libsingular): support division over transcendental ex… #40076

Uh oh!

Conversation

polykernel commented May 9, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

📝 Checklist

⌛ Dependencies

Footnotes

Uh oh!

fchapoton commented May 9, 2025

Uh oh!

dimpase commented May 9, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

polykernel commented May 9, 2025

Uh oh!

dimpase commented May 12, 2025

Uh oh!

polykernel commented May 12, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

polykernel commented May 23, 2025

Uh oh!

dimpase commented May 23, 2025

Uh oh!

polykernel commented May 26, 2025

Uh oh!

dimpase commented May 26, 2025

Uh oh!

polykernel commented May 26, 2025

Uh oh!

dimpase commented May 26, 2025

Uh oh!

polykernel commented May 27, 2025

Uh oh!

dimpase commented May 27, 2025

Uh oh!

github-actions bot commented Jun 12, 2025

Uh oh!

Uh oh!

polykernel commented May 9, 2025 •

edited

Loading

dimpase commented May 9, 2025 •

edited

Loading

polykernel commented May 12, 2025 •

edited

Loading