Mpmath 1.4 alpha update, setup.py and meson compatible by kiwifb · Pull Request #40669 · sagemath/sage

kiwifb · 2025-08-23T03:34:04Z

This branch takes over from #38565 and include fix for the meson build.
It also update to the latest mpmath 1.4.0 alpha release. Releases beyond 1.4.0.a1 have some fix for sagemath but some test may still be broken.

📝 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.

… for speed'

…dle mpmath types without Sage backend

…ion_doctest.py: Update class name in doctest output

github-actions · 2025-08-23T04:47:35Z

Documentation preview for this PR (built with commit e7f5fb5; changes) is ready! 🎉
This preview will update shortly after each push to this PR.

orlitzky · 2025-08-23T15:16:12Z

The big problem here is that no other distros are going to switch to an alpha release without a very good reason. I think for there to be any hope of it getting merged before the official 1.4.0, we would need to support 1.3.x as well. (Debian et al. may remain on 1.3.x for a while anyway).

It doesn't look like there are too many tests to worry about. Just code that needs to be enabled/disabled. With meson, we could check for the version being used,

>>> import mpmath
>>> mpmath.__version__.startswith("1.4.")
True

and then add e.g. CPPFLAGS="-DMPMATH14". The pyx files that are removed in the PR could be selectively disabled, and the functions that are deleted could be turned off by checking #ifdef MPMATH14.

orlitzky · 2025-08-23T15:29:08Z

Actually for the deleted functions like bitcount and normalize, we wouldn't even need -DMPMATH14. We could just check for mpmath.__version__ directly, and return None rather than whatever the function does now. The meson check is still necessary to exclude certain pyx/pxd files though.

antonio-rojas · 2026-02-23T20:52:09Z

1.4 is out, this needs rebasing

antonio-rojas · 2026-02-24T07:46:16Z

There are still a few issues with 1.4:

python3 -m sage.doctest --long /usr/lib/python3.14/site-packages/sage/functions/error.py  # 1 doctest failed
python3 -m sage.doctest --long /usr/lib/python3.14/site-packages/sage/functions/exp_integral.py  # 1 doctest failed
python3 -m sage.doctest --long /usr/lib/python3.14/site-packages/sage/matrix/matrix2.pyx  # 3 doctests failed
python3 -m sage.doctest --long /usr/lib/python3.14/site-packages/sage/symbolic/constants.py  # 18 doctests failed

The first two are just numerical noise. The last one is due to some constants (kinchin, twinprime, glaisher, mertens) no longer being evaluated. The third one is a coercion error:

**********************************************************************
File "/usr/lib/python3.14/site-packages/sage/matrix/matrix2.pyx", line 7152, in sage.matrix.matrix2.Matrix.Sequence
Failed example:
    m.eigenvalues(algorithm="mpmath")  # abs tol 1e-14
Exception raised:
    Traceback (most recent call last):
      File "sage/rings/complex_mpfr.pyx", line 993, in sage.rings.complex_mpfr.ComplexNumber.__init__
        rr = R(real)
      File "sage/structure/parent.pyx", line 899, in sage.structure.parent.Parent.__call__
        return mor._call_(x)
      File "sage/structure/coerce_maps.pyx", line 92, in sage.structure.coerce_maps.DefaultConvertMap_unique._call_
        raise
      File "sage/structure/coerce_maps.pyx", line 87, in sage.structure.coerce_maps.DefaultConvertMap_unique._call_
        return C._element_constructor(x)
      File "sage/rings/real_mpfr.pyx", line 700, in sage.rings.real_mpfr.RealField_class._element_constructor_
        z._set(x, base)
      File "sage/rings/real_mpfr.pyx", line 1574, in sage.rings.real_mpfr.RealNumber._set
        raise TypeError("unable to convert {!r} to a real number".format(s))
    TypeError: unable to convert '(0.0-1.41421356237309j)' to a real number

    During handling of the above exception, another exception occurred:

    Traceback (most recent call last):
      File "/usr/lib/python3.14/site-packages/sage/doctest/forker.py", line 734, in _run
        self.compile_and_execute(example, compiler, test.globs)
        ~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
      File "/usr/lib/python3.14/site-packages/sage/doctest/forker.py", line 1158, in compile_and_execute
        exec(compiled, globs)
        ~~~~^^^^^^^^^^^^^^^^^
      File "<doctest sage.matrix.matrix2.Matrix.Sequence[23]>", line 1, in <module>
        m.eigenvalues(algorithm="mpmath")  # abs tol 1e-14
        ~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^
      File "sage/matrix/matrix2.pyx", line 7178, in sage.matrix.matrix2.Matrix.eigenvalues
        self._eigenvectors_left(
      File "sage/matrix/matrix2.pyx", line 7439, in sage.matrix.matrix2.Matrix._eigenvectors_left
        return self._fix_eigenvectors_extend(self._eigenvectors_left_mpmath(), extend)
      File "sage/matrix/matrix2.pyx", line 7624, in sage.matrix.matrix2.Matrix._eigenvectors_left_mpmath
        (C(e), [vector(C, V)], 1)
      File "sage/rings/complex_mpfr.pyx", line 478, in sage.rings.complex_mpfr.ComplexField_class.__call__
        return Parent.__call__(self, x)
      File "sage/structure/parent.pyx", line 899, in sage.structure.parent.Parent.__call__
        return mor._call_(x)
      File "sage/structure/coerce_maps.pyx", line 92, in sage.structure.coerce_maps.DefaultConvertMap_unique._call_
        raise
      File "sage/structure/coerce_maps.pyx", line 87, in sage.structure.coerce_maps.DefaultConvertMap_unique._call_
        return C._element_constructor(x)
      File "sage/rings/complex_mpfr.pyx", line 564, in sage.rings.complex_mpfr.ComplexField_class._element_constructor_
        return ComplexNumber(self, x)
      File "sage/rings/complex_mpfr.pyx", line 998, in sage.rings.complex_mpfr.ComplexNumber.__init__
        raise TypeError("unable to coerce to a ComplexNumber: %s" % type(real))
    TypeError: unable to coerce to a ComplexNumber: <class 'mpmath.ctx_mp_python.mpc'>
**********************************************************************
File "/usr/lib/python3.14/site-packages/sage/matrix/matrix2.pyx", line 7612, in sage.matrix.matrix2.Matrix.list
Failed example:
    m.eigenvectors_left(algorithm="mpmath")
Exception raised:
    Traceback (most recent call last):
      File "sage/rings/complex_mpfr.pyx", line 993, in sage.rings.complex_mpfr.ComplexNumber.__init__
        rr = R(real)
      File "sage/structure/parent.pyx", line 899, in sage.structure.parent.Parent.__call__
        return mor._call_(x)
      File "sage/structure/coerce_maps.pyx", line 92, in sage.structure.coerce_maps.DefaultConvertMap_unique._call_
        raise
      File "sage/structure/coerce_maps.pyx", line 87, in sage.structure.coerce_maps.DefaultConvertMap_unique._call_
        return C._element_constructor(x)
      File "sage/rings/real_mpfr.pyx", line 700, in sage.rings.real_mpfr.RealField_class._element_constructor_
        z._set(x, base)
      File "sage/rings/real_mpfr.pyx", line 1574, in sage.rings.real_mpfr.RealNumber._set
        raise TypeError("unable to convert {!r} to a real number".format(s))
    TypeError: unable to convert '(0.0-1.41421356237309j)' to a real number

    During handling of the above exception, another exception occurred:

    Traceback (most recent call last):
      File "/usr/lib/python3.14/site-packages/sage/doctest/forker.py", line 734, in _run
        self.compile_and_execute(example, compiler, test.globs)
        ~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
      File "/usr/lib/python3.14/site-packages/sage/doctest/forker.py", line 1158, in compile_and_execute
        exec(compiled, globs)
        ~~~~^^^^^^^^^^^^^^^^^
      File "<doctest sage.matrix.matrix2.Matrix.list[1]>", line 1, in <module>
        m.eigenvectors_left(algorithm="mpmath")
        ~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^
      File "sage/matrix/matrix2.pyx", line 7417, in sage.matrix.matrix2.Matrix.eigenvectors_left
        return self._eigenvectors_left(
      File "sage/matrix/matrix2.pyx", line 7439, in sage.matrix.matrix2.Matrix._eigenvectors_left
        return self._fix_eigenvectors_extend(self._eigenvectors_left_mpmath(), extend)
      File "sage/matrix/matrix2.pyx", line 7624, in sage.matrix.matrix2.Matrix._eigenvectors_left_mpmath
        (C(e), [vector(C, V)], 1)
      File "sage/rings/complex_mpfr.pyx", line 478, in sage.rings.complex_mpfr.ComplexField_class.__call__
        return Parent.__call__(self, x)
      File "sage/structure/parent.pyx", line 899, in sage.structure.parent.Parent.__call__
        return mor._call_(x)
      File "sage/structure/coerce_maps.pyx", line 92, in sage.structure.coerce_maps.DefaultConvertMap_unique._call_
        raise
      File "sage/structure/coerce_maps.pyx", line 87, in sage.structure.coerce_maps.DefaultConvertMap_unique._call_
        return C._element_constructor(x)
      File "sage/rings/complex_mpfr.pyx", line 564, in sage.rings.complex_mpfr.ComplexField_class._element_constructor_
        return ComplexNumber(self, x)
      File "sage/rings/complex_mpfr.pyx", line 998, in sage.rings.complex_mpfr.ComplexNumber.__init__
        raise TypeError("unable to coerce to a ComplexNumber: %s" % type(real))
    TypeError: unable to coerce to a ComplexNumber: <class 'mpmath.ctx_mp_python.mpc'>
**********************************************************************
File "/usr/lib/python3.14/site-packages/sage/matrix/matrix2.pyx", line 7615, in sage.matrix.matrix2.Matrix.list
Failed example:
    m.eigenvectors_left(algorithm="mpmath", extend=False)
Exception raised:
    Traceback (most recent call last):
      File "sage/rings/complex_mpfr.pyx", line 993, in sage.rings.complex_mpfr.ComplexNumber.__init__
        rr = R(real)
      File "sage/structure/parent.pyx", line 899, in sage.structure.parent.Parent.__call__
        return mor._call_(x)
      File "sage/structure/coerce_maps.pyx", line 92, in sage.structure.coerce_maps.DefaultConvertMap_unique._call_
        raise
      File "sage/structure/coerce_maps.pyx", line 87, in sage.structure.coerce_maps.DefaultConvertMap_unique._call_
        return C._element_constructor(x)
      File "sage/rings/real_mpfr.pyx", line 700, in sage.rings.real_mpfr.RealField_class._element_constructor_
        z._set(x, base)
      File "sage/rings/real_mpfr.pyx", line 1574, in sage.rings.real_mpfr.RealNumber._set
        raise TypeError("unable to convert {!r} to a real number".format(s))
    TypeError: unable to convert '(0.0-1.41421356237309j)' to a real number

    During handling of the above exception, another exception occurred:

    Traceback (most recent call last):
      File "/usr/lib/python3.14/site-packages/sage/doctest/forker.py", line 734, in _run
        self.compile_and_execute(example, compiler, test.globs)
        ~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
      File "/usr/lib/python3.14/site-packages/sage/doctest/forker.py", line 1158, in compile_and_execute
        exec(compiled, globs)
        ~~~~^^^^^^^^^^^^^^^^^
      File "<doctest sage.matrix.matrix2.Matrix.list[2]>", line 1, in <module>
        m.eigenvectors_left(algorithm="mpmath", extend=False)
        ~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
      File "sage/matrix/matrix2.pyx", line 7417, in sage.matrix.matrix2.Matrix.eigenvectors_left
        return self._eigenvectors_left(
      File "sage/matrix/matrix2.pyx", line 7439, in sage.matrix.matrix2.Matrix._eigenvectors_left
        return self._fix_eigenvectors_extend(self._eigenvectors_left_mpmath(), extend)
      File "sage/matrix/matrix2.pyx", line 7624, in sage.matrix.matrix2.Matrix._eigenvectors_left_mpmath
        (C(e), [vector(C, V)], 1)
      File "sage/rings/complex_mpfr.pyx", line 478, in sage.rings.complex_mpfr.ComplexField_class.__call__
        return Parent.__call__(self, x)
      File "sage/structure/parent.pyx", line 899, in sage.structure.parent.Parent.__call__
        return mor._call_(x)
      File "sage/structure/coerce_maps.pyx", line 92, in sage.structure.coerce_maps.DefaultConvertMap_unique._call_
        raise
      File "sage/structure/coerce_maps.pyx", line 87, in sage.structure.coerce_maps.DefaultConvertMap_unique._call_
        return C._element_constructor(x)
      File "sage/rings/complex_mpfr.pyx", line 564, in sage.rings.complex_mpfr.ComplexField_class._element_constructor_
        return ComplexNumber(self, x)
      File "sage/rings/complex_mpfr.pyx", line 998, in sage.rings.complex_mpfr.ComplexNumber.__init__
        raise TypeError("unable to coerce to a ComplexNumber: %s" % type(real))
    TypeError: unable to coerce to a ComplexNumber: <class 'mpmath.ctx_mp_python.mpc'>
**********************************************************************

antonio-rojas · 2026-02-24T22:01:15Z

This fixes the remaining issues

diff --git a/src/sage/functions/error.py b/src/sage/functions/error.py
index 78eb4992aa6..8822b3ed584 100644
--- a/src/sage/functions/error.py
+++ b/src/sage/functions/error.py
@@ -257,7 +257,7 @@ class Function_erf(BuiltinFunction):
             sage: erf(2).n(200)                                                         # needs sage.symbolic
             0.99532226501895273416206925636725292861089179704006007673835
             sage: erf(pi - 1/2*I).n(100)                                                # needs sage.symbolic
-            1.0000111669099367825726058952 + 1.6332655417638522934072124547e-6*I
+            1.0000111669099367825726058952 + 1.633265541763852293407212454...e-6*I
 
         TESTS:
 
diff --git a/src/sage/functions/exp_integral.py b/src/sage/functions/exp_integral.py
index e074c430808..09d2bbda71b 100644
--- a/src/sage/functions/exp_integral.py
+++ b/src/sage/functions/exp_integral.py
@@ -993,7 +993,7 @@ class Function_cos_integral(BuiltinFunction):
             sage: N(cos_integral(10^-10), digits=30)                                    # needs sage.symbolic
             -22.4486352650389239795759024568
             sage: cos_integral(ComplexField(100)(I))                                    # needs sage.symbolic
-            0.83786694098020824089467857943 + 1.5707963267948966192313216916*I
+            0.8378669409802082408946785794... + 1.5707963267948966192313216916*I
         """
         return _mpmath_utils_call(_mpmath_ci, z, parent=parent)
 
diff --git a/src/sage/libs/mpmath/all.py b/src/sage/libs/mpmath/all.py
index ea35faeb721..14a4be971e6 100644
--- a/src/sage/libs/mpmath/all.py
+++ b/src/sage/libs/mpmath/all.py
@@ -22,5 +22,5 @@ _constants_funcs = {
 
 
 def eval_constant(name, ring):
-    prec = ring.precision() + 20
+    prec = int(ring.precision() + 20)
     return ring(_constants_funcs[name](prec)) >> prec
diff --git a/src/sage/rings/complex_mpfr.pyx b/src/sage/rings/complex_mpfr.pyx
index 11f58008fcf..ba26d194b1f 100644
--- a/src/sage/rings/complex_mpfr.pyx
+++ b/src/sage/rings/complex_mpfr.pyx
@@ -54,6 +54,7 @@ from sage.rings.integer_ring import ZZ
 
 cimport gmpy2
 gmpy2.import_gmpy2()
+from mpmath import mp
 
 try:
     from cypari2.gen import Gen as pari_gen
@@ -984,7 +985,7 @@ cdef class ComplexNumber(sage.structure.element.FieldElement):
                 real, imag = real
             elif isinstance(real, complex):
                 real, imag = real.real, real.imag
-            elif type(real) is gmpy2.mpc:
+            elif isinstance(real, (gmpy2.mpc, mp.mpc)):
                 real, imag = (<gmpy2.mpc>real).real, (<gmpy2.mpc>real).imag
             else:
                 imag = 0

kiwifb · 2026-02-24T22:24:27Z

Thanks for the extra work @antonio-rojas I am not sure when I will have time to act on this. I am bit swamped in my real job and life in general.

antonio-rojas · 2026-02-28T08:31:11Z

split and rebased the sagelib changes in #41728. The mpmath update requires patching sympy, and I personally no longer want to spend any time on sage-the-distro (and, in general, I think it's good to separate the sagelib changes from the sage-the-distro version upgrades whenever possible)

sagemathgh-41728: Make sagelib compatible with mpmath 1.4 pick sagelib changes from sagemath#40669 and sagemath#38565 for mpmath 1.4 compatibility URL: sagemath#41728 Reported by: Antonio Rojas Reviewer(s):

orlitzky · 2026-03-06T23:59:04Z

Looks like the most recent PR will make it in

sagemathgh-41728: Make sagelib compatible with mpmath 1.4 pick sagelib changes from sagemath#40669 and sagemath#38565 for mpmath 1.4 compatibility URL: sagemath#41728 Reported by: Antonio Rojas Reviewer(s):

Matthias Koeppe and others added 16 commits August 10, 2024 21:02

src/sage/env.py: Set MPMATH_NOSAGE

07c4f98

src/sage/libs/mpmath/utils.pyx: Remove 'patches some mpmath functions…

0f6e4b0

… for speed'

src/sage/libs/mpmath/ext_libmp.pyx: Remove

3ed5cca

src/sage/libs/mpmath/ext_libmp.p*: Remove

d52b7ec

src/sage/libs/mpmath/ext_main.p*: Remove

ec767ec

src/sage/libs/mpmath/utils.pyx: Use gmpy2

cf0a7a1

src/sage/rings/real_mpfr.pyx: Remove duplicate cimport

b57e0a4

src/sage/structure/coerce.pyx (is_mpmath_type, py_scalar_parent): Han…

125ae4e

…dle mpmath types without Sage backend

src/sage/tests/books/computational-mathematics-with-sagemath/integrat…

b713521

…ion_doctest.py: Update class name in doctest output

src/sage/libs/mpmath/utils.pyx: Update doctest output for 'ei'

84f280c

build/pkgs/mpmath/version_requirements.txt: Reject 1.4

ed6d0c2

Merge branch 'no_mpmath_sage' into mpmath-1.4

12c6b9f

build/pkgs/mpmath: Update to 1.4.0a1

81d3cce

Merge branch 'develop' into mpmath-1.4

e5cb262

bump required version to 1.4.0a5 which has many fixes

7714d6c

update meson build for new layout

e7f5fb5

github-actions bot added the s: needs review label Aug 23, 2025

kiwifb mentioned this pull request Aug 23, 2025

Add meson-based CI run for Gentoo #40663

Open

5 tasks

Merge branch 'develop' into mpmath-1.4

65a3fcb

orlitzky mentioned this pull request Oct 8, 2025

Update for mpmath-1.4 while remaining backwards-compatible #41014

Closed

antonio-rojas added s: needs work and removed s: needs review labels Feb 23, 2026

antonio-rojas mentioned this pull request Feb 27, 2026

Make sagelib compatible with mpmath 1.4 #41728

Merged

orlitzky closed this Mar 6, 2026

github-actions bot removed the s: needs work label Mar 6, 2026

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Mpmath 1.4 alpha update, setup.py and meson compatible#40669

Mpmath 1.4 alpha update, setup.py and meson compatible#40669
kiwifb wants to merge 17 commits intosagemath:developfrom
kiwifb:mpmath-1.4

kiwifb commented Aug 23, 2025

Uh oh!

github-actions bot commented Aug 23, 2025

Uh oh!

orlitzky commented Aug 23, 2025

Uh oh!

orlitzky commented Aug 23, 2025

Uh oh!

antonio-rojas commented Feb 23, 2026

Uh oh!

antonio-rojas commented Feb 24, 2026

Uh oh!

antonio-rojas commented Feb 24, 2026

Uh oh!

kiwifb commented Feb 24, 2026

Uh oh!

antonio-rojas commented Feb 28, 2026 •

edited

Loading

Uh oh!

orlitzky commented Mar 6, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

Uh oh!

Conversation

kiwifb commented Aug 23, 2025

📝 Checklist

Uh oh!

github-actions bot commented Aug 23, 2025

Uh oh!

orlitzky commented Aug 23, 2025

Uh oh!

orlitzky commented Aug 23, 2025

Uh oh!

antonio-rojas commented Feb 23, 2026

Uh oh!

antonio-rojas commented Feb 24, 2026

Uh oh!

antonio-rojas commented Feb 24, 2026

Uh oh!

kiwifb commented Feb 24, 2026

Uh oh!

antonio-rojas commented Feb 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

orlitzky commented Mar 6, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

antonio-rojas commented Feb 28, 2026 •

edited

Loading