Sage 8.9.beta4 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 

1537fe511b (tag: 8.9.beta4, trac/develop) Updated SageMath version to 8.9.beta4

096a9755fe Trac #26425: Conversion problem between Laurent polynomial ring and its field of fractions

a9515bf16f Trac #25070: coRSK algorithm

73219b94d2 Trac #24877: bug in polyhedron over RDF

d3f5091abb Trac #22809: Pass number of variables to polygens

d73b8508d9 Trac #28176: Fix few bugs in ClusterAlgebra discovered while working on #26771

46ec2926af Trac #28167: Py3: Fix combinat.root_system.root_lattice_realisations doctests

a8ba7e4858 Trac #27271: spkg-configure.m4 for ecm

363beb1f49 Trac #27270: spkg-configure.m4 for arb

8e66f62ab6 Trac #17895: Computing all roots is faster than computing a single one

5cf466d2c0 Trac #28216: small cleanup in polynomial_element.pyx

104da148b5 Trac #28203: pep8 cleanup of toy_buchberger

4789281747 Trac #28202: pyflakes and pep8 cleanup of latin squares

1af4a25f6c Trac #28169: Option for adjugate in conjugation function with normalize_coordinates

894956f430 Trac #28161: Make _number_field_from_algebraics() return a dynamical system

67d3b8fc89 Trac #28135: Implement plot()/ascii_art/unicode_art for parallelogram polyominoes

edaaeed81f Trac #27562: Fix string formatting in ValueError in sql_db.py

7bab17d73f Trac #25786: Fix introspection with ? when doc source is not available

8c6e4d5fca Trac #25237: normlize_coordinates for projective morphism not fully normalized

83cc4f547e Trac #11217: Bad displays for Macaulay2

ed5f50991e Trac #28210: py3 perfect_matching

df355bcda0 Trac #28206: py3: fix str vs bytes issues with gurobi

140589bd4f Trac #28205: remove deprecated alias in posets

fb1a565fe1 Trac #28198: Add method is_bipartite to BipartiteGraph

ce65edcfe6 Trac #28192: more py3 passing folders

73025683eb Trac #28191: simplify root refinement in qqbar.py

9d13e115af Trac #28149: Implement compact real forms for matrix Lie algebras

10b03c4baa Trac #28142: Girth of directed graphs

7087c0314f Trac #27958: enhance the integration by using also giac and sympy

f20bcce449 Trac #27435: py3: failing doctest in graph_database.py with interactive_query

13ca2608e4 Trac #27232: is_isomorphic broken with keyword edge_labels=True

5392aef444 Trac #25603: Signed tensor product for graded algebras, coalgebras, etc.

e116fdfb30 Trac #24128: A framework for finite dynamics (permutations of finite sets)

dedfcbe5bf Trac #27907: Building gcc 7.2.0 fails on CentOS / Red Hat / Fedora Linux systems because of crti.o

a1e1a8f721 (tag: 8.9.beta3) Updated SageMath version to 8.9.beta3

[Thierry]

Hi,

Sage fails to download gap-4.10.2.tar.bz2 because the URLs it tries look
like

http://www-ftp.lip6.fr/pub/math/sagemath/spkg/upstream/gap_packages/gap-4.10.2.tar.bz2

instead of

http://www-ftp.lip6.fr/pub/math/sagemath/spkg/upstream/gap/gap-4.10.2.tar.bz2

Where could this come from ?

Ciao,
Thierry

> --
> 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/399a98a7-73f4-4678-ac6e-b8220775881d%40googlegroups.com.

[Emmanuel Charpentier]

FWIW, on Debian testing running on core i7 + 16 GB RAM, upgrades from 8.9.beta3 (using python 3) ; ptest-python3 has to permanent failures :

charpent@zen-book-flip:/usr/local/sage-P3$ sage -t --long --warn-long 159.6 src/sage/algebras/quantum_groups/quantum_group_gap.py

Running doctests with ID 2019-07-29-17-08-22-c25ce133.

Git branch: develop

Using --optional=build,dochtml,dot2tex,fricas,gap_packages,giacpy_sage,memlimit,mpir,python2,sage

Doctesting 1 file.

sage -t --long --warn-long 159.6 src/sage/algebras/quantum_groups/quantum_group_gap.py

**********************************************************************

File "src/sage/algebras/quantum_groups/quantum_group_gap.py", line 1553, in sage.algebras.quantum_groups.quantum_group_gap.QuantumGroupModule._latex_

Failed example:

    latex(S)  # optional - gap_packages  # random (depends on dot2tex)

Exception raised:

    Traceback (most recent call last):

      File "/usr/local/sage-P3/local/lib/python3.7/site-packages/sage/doctest/forker.py", line 681, in _run

        self.compile_and_execute(example, compiler, test.globs)

      File "/usr/local/sage-P3/local/lib/python3.7/site-packages/sage/doctest/forker.py", line 1105, in compile_and_execute

        exec(compiled, globs)

      File "<doctest sage.algebras.quantum_groups.quantum_group_gap.QuantumGroupModule._latex_[4]>", line 1, in <module>

        latex(S)  # optional - gap_packages  # random (depends on dot2tex)

      File "/usr/local/sage-P3/local/lib/python3.7/site-packages/sage/misc/latex.py", line 925, in __call__

        return LatexExpr(x._latex_())

      File "/usr/local/sage-P3/local/lib/python3.7/site-packages/sage/algebras/quantum_groups/quantum_group_gap.py", line 1559, in _latex_

        return latex(self.crystal_graph())

      File "/usr/local/sage-P3/local/lib/python3.7/site-packages/sage/algebras/quantum_groups/quantum_group_gap.py", line 2097, in crystal_graph

        vertices = [CrystalGraphVertex(self, d[repr(p)[1:-1]]) for p in G[bytes('points')]]

    TypeError: string argument without an encoding

**********************************************************************

File "src/sage/algebras/quantum_groups/quantum_group_gap.py", line 1658, in sage.algebras.quantum_groups.quantum_group_gap.QuantumGroupModule.crystal_graph

Failed example:

    G = V.crystal_graph(); G            # optional - gap_packages

Exception raised:

    Traceback (most recent call last):

      File "/usr/local/sage-P3/local/lib/python3.7/site-packages/sage/doctest/forker.py", line 681, in _run

        self.compile_and_execute(example, compiler, test.globs)

      File "/usr/local/sage-P3/local/lib/python3.7/site-packages/sage/doctest/forker.py", line 1105, in compile_and_execute

        exec(compiled, globs)

      File "<doctest sage.algebras.quantum_groups.quantum_group_gap.QuantumGroupModule.crystal_graph[2]>", line 1, in <module>

        G = V.crystal_graph(); G            # optional - gap_packages

      File "/usr/local/sage-P3/local/lib/python3.7/site-packages/sage/algebras/quantum_groups/quantum_group_gap.py", line 1666, in crystal_graph

        vertices = [CrystalGraphVertex(self, repr(p)) for p in G[bytes('points')]]

    TypeError: string argument without an encoding

**********************************************************************

File "src/sage/algebras/quantum_groups/quantum_group_gap.py", line 1662, in sage.algebras.quantum_groups.quantum_group_gap.QuantumGroupModule.crystal_graph

Failed example:

    G.is_isomorphic(B.digraph(), edge_labels=True)  # optional - gap_packages

Exception raised:

    Traceback (most recent call last):

      File "/usr/local/sage-P3/local/lib/python3.7/site-packages/sage/doctest/forker.py", line 681, in _run

        self.compile_and_execute(example, compiler, test.globs)

      File "/usr/local/sage-P3/local/lib/python3.7/site-packages/sage/doctest/forker.py", line 1105, in compile_and_execute

        exec(compiled, globs)

      File "<doctest sage.algebras.quantum_groups.quantum_group_gap.QuantumGroupModule.crystal_graph[4]>", line 1, in <module>

        G.is_isomorphic(B.digraph(), edge_labels=True)  # optional - gap_packages

    NameError: name 'G' is not defined

**********************************************************************

File "src/sage/algebras/quantum_groups/quantum_group_gap.py", line 2082, in sage.algebras.quantum_groups.quantum_group_gap.HighestWeightSubmodule.crystal_graph

Failed example:

    G = S.crystal_graph()                    # optional - gap_packages

Exception raised:

    Traceback (most recent call last):

      File "/usr/local/sage-P3/local/lib/python3.7/site-packages/sage/doctest/forker.py", line 681, in _run

        self.compile_and_execute(example, compiler, test.globs)

      File "/usr/local/sage-P3/local/lib/python3.7/site-packages/sage/doctest/forker.py", line 1105, in compile_and_execute

        exec(compiled, globs)

      File "<doctest sage.algebras.quantum_groups.quantum_group_gap.HighestWeightSubmodule.crystal_graph[4]>", line 1, in <module>

        G = S.crystal_graph()                    # optional - gap_packages

      File "/usr/local/sage-P3/local/lib/python3.7/site-packages/sage/algebras/quantum_groups/quantum_group_gap.py", line 2097, in crystal_graph

        vertices = [CrystalGraphVertex(self, d[repr(p)[1:-1]]) for p in G[bytes('points')]]

    TypeError: string argument without an encoding

**********************************************************************

File "src/sage/algebras/quantum_groups/quantum_group_gap.py", line 2083, in sage.algebras.quantum_groups.quantum_group_gap.HighestWeightSubmodule.crystal_graph

Failed example:

    sorted(G.vertices(), key=str)            # optional - gap_packages

Exception raised:

    Traceback (most recent call last):

      File "/usr/local/sage-P3/local/lib/python3.7/site-packages/sage/doctest/forker.py", line 681, in _run

        self.compile_and_execute(example, compiler, test.globs)

      File "/usr/local/sage-P3/local/lib/python3.7/site-packages/sage/doctest/forker.py", line 1105, in compile_and_execute

        exec(compiled, globs)

      File "<doctest sage.algebras.quantum_groups.quantum_group_gap.HighestWeightSubmodule.crystal_graph[5]>", line 1, in <module>

        sorted(G.vertices(), key=str)            # optional - gap_packages

    NameError: name 'G' is not defined

**********************************************************************

File "src/sage/algebras/quantum_groups/quantum_group_gap.py", line 2087, in sage.algebras.quantum_groups.quantum_group_gap.HighestWeightSubmodule.crystal_graph

Failed example:

    sorted(S.crystal_graph(False).vertices(), key=str)  # optional - gap_packages

Exception raised:

    Traceback (most recent call last):

      File "/usr/local/sage-P3/local/lib/python3.7/site-packages/sage/doctest/forker.py", line 681, in _run

        self.compile_and_execute(example, compiler, test.globs)

      File "/usr/local/sage-P3/local/lib/python3.7/site-packages/sage/doctest/forker.py", line 1105, in compile_and_execute

        exec(compiled, globs)

      File "<doctest sage.algebras.quantum_groups.quantum_group_gap.HighestWeightSubmodule.crystal_graph[6]>", line 1, in <module>

        sorted(S.crystal_graph(False).vertices(), key=str)  # optional - gap_packages

      File "/usr/local/sage-P3/local/lib/python3.7/site-packages/sage/algebras/quantum_groups/quantum_group_gap.py", line 2092, in crystal_graph

        return QuantumGroupModule.crystal_graph(self)

      File "/usr/local/sage-P3/local/lib/python3.7/site-packages/sage/algebras/quantum_groups/quantum_group_gap.py", line 1666, in crystal_graph

        vertices = [CrystalGraphVertex(self, repr(p)) for p in G[bytes('points')]]

    TypeError: string argument without an encoding

**********************************************************************

3 items had failures:

   3 of   8 in sage.algebras.quantum_groups.quantum_group_gap.HighestWeightSubmodule.crystal_graph

   1 of   6 in sage.algebras.quantum_groups.quantum_group_gap.QuantumGroupModule._latex_

   2 of   6 in sage.algebras.quantum_groups.quantum_group_gap.QuantumGroupModule.crystal_graph

    [529 tests, 6 failures, 8.25 s]

----------------------------------------------------------------------

sage -t --long --warn-long 159.6 src/sage/algebras/quantum_groups/quantum_group_gap.py  # 6 doctests failed

----------------------------------------------------------------------

Total time for all tests: 8.4 seconds

    cpu time: 8.3 seconds

    cumulative wall time: 8.3 seconds

charpent@zen-book-flip:/usr/local/sage-P3$ sage -t --long --warn-long 159.6 src/sage/rings/polynomial/polynomial_rational_flint.pyx

Running doctests with ID 2019-07-29-17-08-54-16852f02.

Git branch: develop

Using --optional=build,dochtml,dot2tex,fricas,gap_packages,giacpy_sage,memlimit,mpir,python2,sage

Doctesting 1 file.

sage -t --long --warn-long 159.6 src/sage/rings/polynomial/polynomial_rational_flint.pyx

**********************************************************************

File "src/sage/rings/polynomial/polynomial_rational_flint.pyx", line 2055, in sage.rings.polynomial.polynomial_rational_flint.Polynomial_rational_flint.galois_group

Failed example:

    G = f.galois_group(); G

Expected:

    Transitive group number 5 of degree 4

Got:

    Exception (FLINT memory_manager). Unable to allocate memory.

    Transitive group number 5 of degree 4

**********************************************************************

1 item had failures:

   1 of  16 in sage.rings.polynomial.polynomial_rational_flint.Polynomial_rational_flint.galois_group

    [397 tests, 1 failure, 3.86 s]

----------------------------------------------------------------------

sage -t --long --warn-long 159.6 src/sage/rings/polynomial/polynomial_rational_flint.pyx  # 1 doctest failed

----------------------------------------------------------------------

Total time for all tests: 3.9 seconds

    cpu time: 3.5 seconds

    cumulative wall time: 3.9 seconds

Both have been already reported.

HTH,

[Justin C. Walker]

> On Jul 28, 2019, at 15:54 , Volker Braun <vbrau...@gmail.com> wrote:
>
> 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

Built w/o problems, from a fresh clone of the develop/8.9-b4 tree, on three macOS systems. Testing (‘ptestlong’ in each case) yielded:

10.11.6 (mid-2015 MBP, Quad-core Core i7): all tests passed!

10.13.6 (2017 iMac Pro, 18-core Xeon W): one test failed:
./sage -t --long --warn-long 70.3 src/sage/libs/singular/polynomial.pyx
The usual issue with many leaks. This test intermittently passes and fails when run by itself.

10.14.6 (2017 MBP, Quad-core Core i7): all tests passed!

Justin

--
Justin C. Walker
Director
Institute for the Enhancement of the Director's Income
--
Fame is fleeting, but obscurity
just drags on and on. F&E

[Emmanuel Charpentier]

On a smaller (core i5, 8 GB RAM) machine with a similar setup, I get only the second failure :

charpent@p-202-021:/usr/local/sage-python3$ sage -t --long src/sage/rings/polynomial/polynomial_rational_flint.pyx

too many failed tests, not using stored timings

Running doctests with ID 2019-07-30-14-28-30-306c6b55.

Git branch: develop

Using --optional=build,dochtml,dot2tex,fricas,giacpy_sage,memlimit,python2,sage

Doctesting 1 file.

sage -t --long src/sage/rings/polynomial/polynomial_rational_flint.pyx

**********************************************************************

File "src/sage/rings/polynomial/polynomial_rational_flint.pyx", line 2055, in sage.rings.polynomial.polynomial_rational_flint.Polynomial_rational_flint.galois_group

Failed example:

    G = f.galois_group(); G

Expected:

    Transitive group number 5 of degree 4

Got:

    Exception (FLINT memory_manager). Unable to allocate memory.

    Transitive group number 5 of degree 4

**********************************************************************

1 item had failures:

   1 of  16 in sage.rings.polynomial.polynomial_rational_flint.Polynomial_rational_flint.galois_group

    [397 tests, 1 failure, 3.32 s]

----------------------------------------------------------------------

sage -t --long src/sage/rings/polynomial/polynomial_rational_flint.pyx  # 1 doctest failed

----------------------------------------------------------------------

Total time for all tests: 3.4 seconds

    cpu time: 3.2 seconds

    cumulative wall time: 3.3 seconds

Furthemore, having absentmindedly executed ptestlong on the same machine, I've got "only" 7 permanent failures. Is this worth pursuing ?

HTH,

[Eric Gourgoulhon]

Le lundi 29 juillet 2019 07:01:10 UTC+2, Thierry a écrit :

Hi,

Sage fails to download gap-4.10.2.tar.bz2 because the URLs it tries look
like

http://www-ftp.lip6.fr/pub/math/sagemath/spkg/upstream/gap_packages/gap-4.10.2.tar.bz2

instead of

http://www-ftp.lip6.fr/pub/math/sagemath/spkg/upstream/gap/gap-4.10.2.tar.bz2

Where could this come from ?

I confirm the issue: when building Sage 8.9.beta4 from a fresh git clone, one gets

[gap-4.10.2.p0] Attempting to download package gap-4.10.2.tar.bz2 from mirrors
[gap-4.10.2.p0] http://www-ftp.lip6.fr/pub/math/sagemath/spkg/upstream/gap_packages/gap-4.10.2.tar.bz2
[gap-4.10.2.p0] ERROR [transfer|run:135]: [Errno 404] Not Found: '//www-ftp.lip6.fr/pub/math/sagemath/spkg/upstream/gap_packages/gap-4.10.2.tar.bz2'
[gap-4.10.2.p0] http://www.mirrorservice.org/sites/www.sagemath.org/spkg/upstream/gap_packages/gap-4.10.2.tar.bz2
[gap-4.10.2.p0] ERROR [transfer|run:135]: [Errno 404] Not Found: '//www.mirrorservice.org/sites/www.sagemath.org/spkg/upstream/gap_packages/gap-4.10.2.tar.bz2'
[gap-4.10.2.p0] https://mirror.koddos.net/sagemath/spkg/upstream/gap_packages/gap-4.10.2.tar.bz2
[gap-4.10.2.p0] ERROR [transfer|run:135]: [Errno 404] Not Found: '//mirror.koddos.net/sagemath/spkg/upstream/gap_packages/gap-4.10.2.tar.bz2'
[gap-4.10.2.p0] https://mirrors.up.pt/pub/sage/spkg/upstream/gap_packages/gap-4.10.2.tar.bz2
[gap-4.10.2.p0] ERROR [transfer|run:135]: [Errno 404] Not Found: '//mirrors.up.pt/pub/sage/spkg/upstream/gap_packages/gap-4.10.2.tar.bz2'

and similar errors for many mirrors.

It seems that only the mirror  http://sagepad.org/ is successfull in providing gap-4.10.2.tar.bz2, because in the log we have

[gap-4.10.2.p0] http://sagepad.org/spkg/upstream/gap_packages/gap-4.10.2.tar.bz2

without any 404 error associated to it and after this the Sage build proceeds smoothly. 

Eric.

[Eric Gourgoulhon]

On Ubuntu 18.04 running on bi-Xeon E5-2623 (8 cores) + 16 GB RAM:

- python2:

  - build (-j16) from a fresh git clone completed, with the mirror issue for gap-4.10.2.tar.bz2 reported by Thierry (cf. https://groups.google.com/d/msg/sage-release/bC7Fbk5JyKE/e9aUtIGvEgAJ)

  - all tests from ptestlong passed

- python3:

  - a first attempt to build (-j16) from a fresh git clone hanged (no CPU activity for ~ 10 minutes) without any error message after

[ipywidgets-7.4.2]     running install_scripts
[ipywidgets-7.4.2]     writing list of installed files to '/tmp/pip-record-sxim7isv/install-record.txt'
[ipywidgets-7.4.2]     Running setup.py install for ipywidgets: finished with status 'done'
[ipywidgets-7.4.2]   Removing source in /tmp/pip-req-build-0nx5ty22
[ipywidgets-7.4.2] Successfully installed ipywidgets-7.4.2
[ipywidgets-7.4.2] Cleaning up...
[ipywidgets-7.4.2] Removed build tracker '/tmp/pip-req-tracker-ehuwucce'
[ipywidgets-7.4.2]
[ipywidgets-7.4.2] real    0m1.924s
[ipywidgets-7.4.2] user    0m1.686s
[ipywidgets-7.4.2] sys    0m0.274s
[ipywidgets-7.4.2] Copying package files from temporary location /home/eric/sage/py3/local/var/tmp/sage/build/ipywidgets-7.4.2/inst to /home/eric/sage/py3/local
[ipywidgets-7.4.2] Successfully installed ipywidgets-7.4.2
[ipywidgets-7.4.2] Deleting temporary build directory
[ipywidgets-7.4.2] /home/eric/sage/py3/local/var/tmp/sage/build/ipywidgets-7.4.2
[ipywidgets-7.4.2] Finished installing ipywidgets-7.4.2.spkg

Ending the build with CTRL-C and starting again with

MAKE="make -j16" make

eventually completed the build.

  - tests from ptest-python3 passed, except for 1 permanent doctest:

sage -t --long --warn-long 51.1 src/sage/rings/polynomial/polynomial_rational_flint.pyx

**********************************************************************
File "src/sage/rings/polynomial/polynomial_rational_flint.pyx", line 2055, in sage.rings.polynomial.polynomial_rational_flint.Polynomial_rational_flint.galois_group
Failed example:
    G = f.galois_group(); G
Expected:
    Transitive group number 5 of degree 4
Got:
    Exception (FLINT memory_manager). Unable to allocate memory.
    Transitive group number 5 of degree 4

This is the same error as that reported by Emmanuel in https://groups.google.com/d/msg/sage-release/bC7Fbk5JyKE/FLN4lQG2EwAJ

Eric.

[Henri Girard]

No problem on ubuntu 19.10 amd64 8x cores 16 Go RAM

--
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/28192905-a07e-42f5-9b6d-235842ba505a%40googlegroups.com.

[Volker Braun]

I think the gap / gap_packages url confusion is because both reference the same tarball file, we don't handle that case well I think.

For the record, we had some networking issue on the fileserver. The sagepad.org machine is not part of the mirror network but my personal server and serves as fallback.

[John H Palmieri]

Do you get the same failure with Python 2? If not, any ideas why "Exception (FLINT memory_manager). Unable to allocate memory" would depend on the version of Python? Flint doesn't depend on Python, so if it's Python 3 only, is there some issue with the Flint interface?

[John H Palmieri]

Another question: does Flint pass its test suite on these machines?

[Eric Gourgoulhon]

Le vendredi 2 août 2019 00:52:07 UTC+2, John H Palmieri a écrit :

Do you get the same failure with Python 2?

In my case, no:  the test

./sage -t --long src/sage/rings/polynomial/polynomial_rational_flint.pyx

is passed with Python 2 and fails with Python3, with

Failed example:
    G = f.galois_group(); G
Expected:
    Transitive group number 5 of degree 4
Got:
    Exception (FLINT memory_manager). Unable to allocate memory.
    Transitive group number 5 of degree 4

 

If not, any ideas why "Exception (FLINT memory_manager). Unable to allocate memory" would depend on the version of Python? Flint doesn't depend on Python, so if it's Python 3 only, is there some issue with the Flint interface?

Maybe. I cannot say, being not familiar with Flint.

Eric.

[Eric Gourgoulhon]

Le vendredi 2 août 2019 01:00:29 UTC+2, John H Palmieri a écrit :

Another question: does Flint pass its test suite on these machines?

How could I run this test suite?

Eric.

[John H Palmieri]

"sage -f -c flint"

[John H Palmieri]

Since the problem doesn't occur with Python 2, it shouldn't be a Flint problem. If you have time, can you try "git bisect" to try to track down exactly which change caused the problem?

[Eric Gourgoulhon]

Thanks for the tip.

Being out of office, I ran the test suite on another computer (Core i7-6700HQ + 16 GB RAM, running Ubuntu 18.04.2), which shows the same doctest failure with Python3 Sage 8.9.beta5.

The test suite seems passed, with many messages like

[flint-2.5.2.p4] zz_pX_to_fmpz_mod_poly....PASS
[flint-2.5.2.p4] zz_pE_to_fq....PASS
[flint-2.5.2.p4] zz_pEX_to_fq_poly....PASS
(I did not see any "FAILED") and at the end:

[flint-2.5.2.p4] Successfully installed flint-2.5.2.p4
[flint-2.5.2.p4] Deleting temporary build directory
[flint-2.5.2.p4] /home/eric/sage/py3/local/var/tmp/sage/build/flint-2.5.2.p4
[flint-2.5.2.p4] Finished installing flint-2.5.2.p4.spkg

Side note: the "./sage -f -c flint" command continued with

New packages may have been installed.
Re-running configure and make in case any dependent packages need updating.

and then proceeded by installing

arb-2.16.0.p0
eclib-20190226
linbox-1.6.3
singular-4.1.1p2.p0
pynac-0.7.24.p0
sagelib-8.9.beta5

Just in case, after "./sage -f -c flint", I run

./sage  -t --long src/sage/rings/polynomial/polynomial_rational_flint.pyx

and get the same error:

File "src/sage/rings/polynomial/polynomial_rational_flint.pyx", line 2055, in sage.rings.polynomial.polynomial_rational_flint.Polynomial_rational_flint.galois_group
Failed example:
    G = f.galois_group(); G
Expected:
    Transitive group number 5 of degree 4
Got:
    Exception (FLINT memory_manager). Unable to allocate memory.
    Transitive group number 5 of degree 4

Again, on the very same computer, this doctest is passed with Python 2 Sage 8.9.beta5.

Eric.

[Eric Gourgoulhon]

Le vendredi 2 août 2019 18:49:04 UTC+2, John H Palmieri a écrit :

Since the problem doesn't occur with Python 2, it shouldn't be a Flint problem. If you have time, can you try "git bisect" to try to track down exactly which change caused the problem?

 I had no time to really bisect, by I can tell that the issue was already here in Python 3 Sage 8.8.rc3.

Eric.

[E. Madison Bray]

It's possible this is similar to
https://trac.sagemath.org/ticket/28106 simply insofar as some test is
bumping up against the default 3300MB virtual memory limit imposed by
the test runner.

Which doesn't necessarily mean there isn't a problem: If some test is
using a lot more memory than it used to, such that it's bumping up
against that limit, there might be a leak or some other unintended
memory usage consequence to the way a test was written....

[John H Palmieri]

I get the same error within a Sage session using

sage: from sage.doctest.control import DocTestController, DocTestDefaults
sage: DC = DocTestController(DocTestDefaults(), ['/path/to/src/sage/rings/polynomial/polynomial_rational_flint.pyx'])
sage: DC.run()

I don't think this uses the memory limits discussed at #28106 — I think those are set in the script sage-runtests, not in the library doctesting framework.

--

John

[kcrisman]

On Wednesday, July 31, 2019 at 6:52:04 PM UTC-4, Volker Braun wrote:

I think the gap / gap_packages url confusion is because both reference the same tarball file, we don't handle that case well I think.

For the record, we had some networking issue on the fileserver. The sagepad.org machine is not part of the mirror network but my personal server and serves as fallback.

Just as a data point, I'm still having this issue right now.