Sage 9.3.beta7 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 

8453ffb849 (tag: 9.3.beta7, trac/develop) Updated SageMath version to 9.3.beta7

6096110e06 Trac #30517: MemoryError in doctesting combinat/designs/gen_quadrangles_with_spread.pyx

e67fc51749 Trac #31148: MR51: Allow Matplotlib to use system qhull >= 7.2.0

f3740e8add Trac #31147: Add missing curly braces in latex of Kodaira symbol

d856033dd6 Trac #31144: refresh the file polyhedron/plot.py

eb9bae531a Trac #31143: shorten range(0, *) in combinat folder

9d4f2d1525 Trac #31140: use ⨂ for unicode of tensor symbol

5a4ad12275 Trac #31138: Document WSL Installation

c8a7960598 Trac #31134: Update setuptools and setuptools_scm

45e7ada314 Trac #24459: Segfault in matrix_integer_dense

9bdf1d149f Trac #24317: Doctest: Improve conversion of inexact symbolic values

85f81e86d3 Trac #21907: Bug in Maxima interface wrt polylog

913b025ac5 Trac #31283: 1000× speedup of conjugate of double dense matrices

bfe6fd9654 Trac #31273: Remove deprecated `form` input for bundle connections

00cacf3803 Trac #31272: Improve the documentation for tensor()

50c1a5d655 Trac #31270: Remove sage-location's "sage-force-relocate" mechanism, fix script to work without SAGE_ROOT

3f5faf58b6 Trac #31266: Signed tensor products does not allow tensor products with non-signed modules

1d8b95def5 Trac #31265: changes in kschur to get rid of specific product definition there

9c39188e60 Trac #31262: Implement non zero chunks for sparse bitsets

35ad55518c Trac #31261: Update pplpy to 0.8.6

d5581ba0ea Trac #31258: Upgrade cmake to 3.19.3

57211cf752 Trac #31257: quo_rem gives wrong answers for LaurentPolynomial_mpair

b282cf6d2d Trac #31256: more care for pbori

e8c83f5b70 Trac #31255: Follow Up: Mutability of Sections and Tensor Fields

573df518ba Trac #31254: Do not perform unnecesssary subdivisions in matrices

09f5b07182 Trac #31239: Add ore_algebra optional package

6a5ad3f3eb Trac #31216: tox.ini (local): Add environment variables to skip system package installs and other steps, add mechanism for a local interactive shell

1ad5cb3ec9 Trac #31197: Use bitsets/binary matrix for edges of dense graphs

3925e96e0a Trac #31196: Code Improvements for Mutability module

01ad25fc3f Trac #31194: Make Class with Mutability compatible with require_mutable wrapper

99d4307f61 Trac #31121: Further refactoring of eta products file

6dda91cd3b Trac #31035: Remove mathjax configuration/symlink from jupyter notebook

2c32dab63a Trac #27103: Enable SIMD-instructions for Bitsets

6b64832a66 Trac #21783: QEPCAD: get rid of qepcad bits in src/bin/sage-location

5e36f5d72b Trac #31271: MR52: upgrade ubuntu system in docker image

7823deee62 Trac #31263: Broken `sage -b`

f0ca7dfb31 Trac #30725: macOS: spkg-install scripts that force use of clang conflict with '-march=native'

aadda8f345 Trac #30731: Replace use of build/bin/sage-python23 by just python3

70291edd23 Trac #14821: Weird error in exponential integral

583e83e86d Trac #7423: plot3d can't handle log(0)

d7c2285977 Trac #31243: Stronger Categories for Manifold Examples: Connectedness

6ef10b86c1 Trac #31240: sage-flatsurf pip package

513046f42d Trac #31238: Add admcycles as an optional package

a33370064e Trac #31237: Add slabbe as an optional package

712ca6e170 Trac #31234: multiplication of matrix with zero columns or rows fails

8c3804cc20 Trac #31233: Categories for Homset of Manifolds do not meet

e1bd5e3fe5 Trac #31229: more typing annotations in Tamari interval posets

02d0fc8821 Trac #31226: Generalize face iterator of combinatorial polyhedron to locally branched lattices

d9745ca3e1 Trac #31224: Add surface-dynamics as a pip package

8101d78056 Trac #31221: Alllow arbitrary folder as output directory of docbuild

96bac10d15 Trac #31220: Remove test for safe directory while doctesting

7eebf61583 Trac #31219: Fix escape of docstring for derivative method

a704ef5b17 Trac #31215: Documentation of Item Assignment Behavior for Bundle Connections

c4f0516700 Trac #31212: fix submonoid method when the generating set include the identity

bf9205aebc Trac #31210: Make it possible to run github workflows manually

5477ab510c Trac #31208: Use popcnt and tzcnt to speed up bitsets on Intel and AMD

7edddbd058 Trac #31207: Implement copy for binary matrix

3994d91f70 Trac #31200: Small improvements to binary matrix

db29ef0dc9 Trac #31195: yet another cleanup for pbori

3e3dcb7ce8 Trac #31192: Installation manual: Fix documentation of python prerequisite (sage-bootstrap-python)

6b14dc7113 Trac #31190: add Farey maps as examples of simplicial surfaces

0c64e7a577 Trac #31189: local quicklisp install interfers with Maxima package building

97c19345d5 Trac #31184: refresh one givaro file

09b4dd2854 Trac #31181: Mutability class does not return is_immutable

4a1d0f8c95 Trac #31084: Makefile: Add "ptest" targets that do not depend on the docbuild

01c9f0b244 Trac #3889: extend parameter for number field sqrt method

c28601e7f6 Trac #31218: bootstrap: Fix "sage-venv-config: /doesnotexist/python3: bad interpreter"

e3dc2c8982 Trac #31183: Fix scipy for MacOS 11 (Big Sur)

5854f7a9fc Trac #31146: cygwin-standard: fpylll build fails

77d9d52abf Trac #31132: homebrew: Unused packages (singular, pari, ...) in /usr/local leak into build when using homebrew's python3

aac2decc62 Trac #31097: build/pkgs/gcc/spkg-configure.m4: Fix SAGE_BROKEN_GCC test

b426f95131 Trac #31062: tox / GH Actions: Disable testsuites of packages depending on pip packages (pytest, ...)

e88216c277 Trac #30944: tox: Improve local-sudo-ubuntu-standard

a549452a4b Trac #30589: Upgrade Python to 3.9.1, pip to 20.3.3

20fbba25e5 Trac #29124: Add script packages build/pkgs/_prereq, build/pkgs/_bootstrap

13b40902a9 (tag: 9.3.beta6) Updated SageMath version to 9.3.beta6

[Matthias Köppe]

On Sunday, February 7, 2021 at 2:04:42 PM UTC-8 Volker Braun 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 

Thanks, Volker, for preparing the new release.

 

8453ffb849 (tag: 9.3.beta7, trac/develop) Updated SageMath version to 9.3.beta7 [...] 

a549452a4b Trac #30589: Upgrade Python to 3.9.1, pip to 20.3.3

All - if you use python installed by Sage, note that this update contains the Python 3.9 upgrade, which will trigger rebuilds of all Python packages.

Developers who use homebrew on macOS -- I recommend that you update homebrew packages to latest; in particular python@3.9 to the latest build (3.9.1_8). Further fixes for known build errors on homebrew are on https://trac.sagemath.org/ticket/31335 (which needs testing).

 

[Matthias Köppe]

On Sunday, February 7, 2021 at 2:04:42 PM UTC-8 Volker Braun 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 

8453ffb849 (tag: 9.3.beta7, trac/develop) Updated SageMath version to 9.3.beta7

First test results from https://github.com/sagemath/sage/actions/runs/545994902:

ubuntu-bionic-standard: (python 3.6) still https://trac.sagemath.org/ticket/31191

debian-bullseye-standard: New(?) numerics issues:

sage -t --random-seed=0 src/sage/finance/time_series.pyx  # 8 doctests failed

sage -t --random-seed=0 src/sage/geometry/polyhedron/library.py  # 1 doctest failed

sage -t --random-seed=0 src/sage/geometry/polyhedron/base.py  # 1 doctest failed

sage -t --random-seed=0 src/sage/matrix/matrix_double_dense.pyx  # 2 doctests failed

sage -t --random-seed=0 src/sage/matrix/matrix_integer_dense.pyx  # 1 doctest failed

sage -t --random-seed=0 src/sage/numerical/sdp.pyx  # 2 doctests failed

sage -t --random-seed=0 src/sage/tests/books/computational-mathematics-with-sagemath/float_doctest.py  # 1 doctest failed

 

[Samuel Lelièvre]

2021-02-07 22:04 UTC, Volker Braun:

>
> 8453ffb849 (tag: 9.3.beta7, trac/develop) Updated SageMath version to 9.3.beta7

Thanks for this new release.

On macOS 10.14.6 with lots of Homebrew packages,
I failed to report on beta6 for which cypari2 failed to build;
with beta7, cypari2 now builds but not linbox. Logs:

https://www.math.u-psud.fr/~lelievre/t/slel-2021-02-08-s93b7-brew.zip

I'll try #31355.

[Steven Trogdon]

This beta does not pick up on my system (Gentoo) python. From config.log

configure:39140: result: python3-3.9.1:                               no suitable system package; will be installed as an SPKG

And my system python

$ python

Python 3.9.1 (default, Jan 26 2021, 00:24:17) 

[GCC 9.3.0] on linux

Type "help", "copyright", "credits" or "license" for more information.

>>> 

$ python3

Python 3.9.1 (default, Jan 26 2021, 00:24:17) 

[GCC 9.3.0] on linux

Type "help", "copyright", "credits" or "license" for more information.

>>> 

[Steven Trogdon]

OK, here is the reason

g++ -std=gnu++11 -fPIC -I/local/sage-git/sage/conftest_venv/include -I/usr/include/python3.9 -c conftest.cpp -o conftest.dir/temp.linux-x86_64-3.9/conftest.o -std=c++11

creating conftest.dir/lib.linux-x86_64-3.9

x86_64-pc-linux-gnu-g++ -pthread -shared -Wl,-O1 -Wl,--as-needed conftest.dir/temp.linux-x86_64-3.9/conftest.o -L/usr/lib64 -o conftest.dir/lib.linux-x86_64-3.9/config_check_distutils_cxx.cpython-39-x86_64-linux-gnu.so

        LDFLAGS = "-Wl,-O1 -Wl,--as-needed -L."

configure:32030: result: no, this is a misconfigured Python whose sysconfig compiler/linker flags contain -I or -L options, which may cause wrong versions of libraries to leak into the build of Python packages - see https://trac.sagemath.org/ticket/31132; to use it anyway, use ./configure --with-python=/usr/bin/python3

I have 

$ echo $LDFLAGS

-Wl,-O1 -Wl,--as-needed

and I've had this set for some time (several years).

[Steven Trogdon]

The LDFLAGS is set only when building Sage. I have set nothing explicitly when building system python.

[Matthias Köppe]

These LDFLAGS are read from `python3 -m sysconfig`. The "-L." flag there is causing this python3 to be rejected by the new checks from https://trac.sagemath.org/ticket/31132 .

Looks like using this flag is a gentoo specialty. 

[Matthias Köppe]

Fixed in https://trac.sagemath.org/ticket/31358

[Kenji Iohara]

On my mac with OS 11.1, it compiled except for some problems with python. make ptestlong has a similar problem.

[Emmanuel Charpentier]

On Debian testing running on core i7 + 16 GB RAM, `ptestlong` gives two permanent failures, both GAP related, both already reported :

```

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

sage -t --long --warn-long 169.5 --random-seed=0 src/sage/tests/gap_packages.py  # 1 doctest failed

sage -t --long --warn-long 169.5 --random-seed=0 src/sage/groups/abelian_gps/abelian_group_gap.py  # 1 doctest failed

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

```

HTH,

[Sébastien Labbé]

On Ubuntu 18.04, running tests with various optional/external packages, I get:

Using --optional=4ti2,cbc,ccache,cryptominisat,database_jones_numfield,debian,dot2tex,e_antic,external,fricas,glucose,latte_int,libnauty,lidia,lrslib,memlimit,normaliz,notedown,pandoc_attributes,pip,pycosat,pynormaliz,rst2ipynb,sage,sage_numerical_backends_coin,sage_spkg

----------------------------------------------------------------------
sage -t --long --warn-long 72.7 --random-seed=0 src/sage/tests/cmdline.py  # 1 doctest failed
sage -t --long --warn-long 72.7 --random-seed=0 src/sage/geometry/cone.py  # 2 doctests failed
----------------------------------------------------------------------
External software detected for doctesting: ffmpeg,graphviz,imagemagick,internet,latex,octave,pandoc

Rerunning failed tests, I get:

----------------------------------------------------------------------
sage -t --long --warn-long 72.0 --random-seed=0 src/sage/tests/cmdline.py  # 1 doctest failed
----------------------------------------------------------------------
External software detected for doctesting: internet,pandoc

Failures are copied below:

sage -t --long --warn-long 72.0 --random-seed=0 src/sage/tests/cmdline.py
**********************************************************************
File "src/sage/tests/cmdline.py", line 606, in sage.tests.cmdline.test_executable
Failed example:
    err  # optional - internet
Expected:
    ''
Got:
    '/home/slabbe/GitBox/sage/local/lib/python3.8/site-packages/sage/misc/package.py:114: UserWarning: failed to fetch the version of pkg=\'ore_algebra\' at https://pypi.org/pypi/ore_algebra/json\n  warnings.warn("failed to fetch the version of pkg={!r} at {}".format(pkg, url))\n'
**********************************************************************
1 item had failures:
   1 of 246 in sage.tests.cmdline.test_executable
    14 build tests not run
    3 gdb tests not run
    6 py2 tests not run
    6 sagenb tests not run
    0 tests not run because we ran out of time
    [245 tests, 1 failure, 29.78 s]



sage -t --long --warn-long 72.7 --random-seed=0 src/sage/geometry/cone.py
**********************************************************************
File "src/sage/geometry/cone.py", line 6189, in sage.geometry.cone.?
Failed example:
    K = random_cone(min_ambient_dim=5, min_rays=2,
                    strictly_convex=False)
Exception raised:
    Traceback (most recent call last):
      File "/home/slabbe/GitBox/sage/local/lib/python3.8/site-packages/sage/doctest/forker.py", line 714, in _run
        self.compile_and_execute(example, compiler, test.globs)
      File "/home/slabbe/GitBox/sage/local/lib/python3.8/site-packages/sage/doctest/forker.py", line 1133, in compile_and_execute
        exec(compiled, globs)
      File "<doctest sage.geometry.cone.?[16]>", line 1, in <module>
        K = random_cone(min_ambient_dim=Integer(5), min_rays=Integer(2),
      File "/home/slabbe/GitBox/sage/local/lib/python3.8/site-packages/sage/geometry/cone.py", line 6617, in random_cone
        K = Cone(rays, lattice=L)
      File "/home/slabbe/GitBox/sage/local/lib/python3.8/site-packages/sage/geometry/cone.py", line 462, in Cone
        return _Cone_from_PPL(cone, lattice, rays)
      File "/home/slabbe/GitBox/sage/local/lib/python3.8/site-packages/sage/geometry/cone.py", line 500, in _Cone_from_PPL
        for g in cone.minimized_generators():
      File "ppl/polyhedron.pyx", line 335, in ppl.polyhedron.Polyhedron.minimized_generators
    RuntimeError: Aborted
**********************************************************************
File "src/sage/geometry/cone.py", line 6191, in sage.geometry.cone.?
Failed example:
    K.is_strictly_convex()
Expected:
    False
Got:
    True
**********************************************************************
1 item had failures:
   2 of  77 in sage.geometry.cone.?
    0 tests not run because we ran out of time
    [1257 tests, 2 failures, 1458.43 s]

[fchap...@gmail.com]

This release breaks (again) our automated docker build on gitlab : 

see https://gitlab.com/sagemath/sage/-/pipelines

Le dimanche 7 février 2021 à 23:04:42 UTC+1, Volker Braun a écrit :

[Matthias Köppe]

On Tuesday, February 9, 2021 at 11:38:29 AM UTC-8 fchap...@gmail.com wrote:

This release breaks (again) our automated docker build on gitlab : 

see https://gitlab.com/sagemath/sage/-/pipelines

From the log:

Error building Sage.

The following package(s) may have failed to build (not necessarily

during this run of 'make pari_jupyter'):

* package:         pari_jupyter-1.3.2

  last build time: Feb 8 16:16

  log file:        /home/sage/sage/logs/pkgs/pari_jupyter-1.3.2.log

  build directory: /home/sage/sage/local/var/tmp/sage/build/pari_jupyter-1.3.2

It is safe to delete any log files and build directories, but they

This is an optional package.

 

[Matthias Köppe]

[pari_jupyter-1.3.2]   gcc -Wno-unused-result -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wno-unused -O2 -g -O2 -g -fPIC -I/home/sage/sage/local/include/python3.9 -c PARIKernel/io.c -o build/temp.linux-x86_64-3.9/PARIKernel/io.o

[pari_jupyter-1.3.2]   PARIKernel/io.c: In function 'PyInit_io':

[pari_jupyter-1.3.2]   PARIKernel/io.c:2905:43: error: 'PyTypeObject' {aka 'struct _typeobject'} has no member named 'tp_print'

[pari_jupyter-1.3.2]    2905 |   __pyx_type_10PARIKernel_2io_PARIKernelIO.tp_print = 0;

this comes from the Python 3.9 upgrade.

[Matthias Köppe]

I've opened https://trac.sagemath.org/ticket/31372 for this.

[Matthias Köppe]

On Sunday, February 7, 2021 at 2:04:42 PM UTC-8 Volker Braun 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 

8453ffb849 (tag: 9.3.beta7, trac/develop) Updated SageMath version to 9.3.beta7

The runs on GH Actions show a very strange failure on fedora-34 and archlinux-latest (https://github.com/sagemath/sage/runs/1850248457):

+ ./bootstrap

rm -rf config configure build/make/Makefile-auto.in

rm -f src/doc/en/installation/*.txt

rm -rf src/doc/en/reference/spkg/*.rst

rm -f src/doc/en/reference/repl/*.txt

rm -f environment.yml

rm -f src/environment.yml

rm -f environment-optional.yml

rm -f src/environment-optional.yml

./sage: no Sage installation found in $SAGE_ROOT=/sage

./sage: no Sage installation found in $SAGE_ROOT=/sage

./sage: no Sage installation found in $SAGE_ROOT=/sage

This can be reproduced locally using "tox -e docker-fedora-34-standard".

It turns out that the built-in command "test" of bash is broken:

[root@9d0e0458cde5 sage]# test -r VERSION.txt && echo yes

[root@9d0e0458cde5 sage]# bash --version
GNU bash, version 5.1.0(1)-release (x86_64-redhat-linux-gnu)

Compare with /usr/bin/test

[root@9d0e0458cde5 sage]# /usr/bin/test -r VERSION.txt && echo yes
yes

 

[Justin C. Walker]

> On Feb 7, 2021, at 14:04, 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

I’ve been off on a tangent for a while, and am just getting back to this.

I am using macOS 10.15.7, and have had the same problem, starting with 9.3-beta2:

“make configure” gives me this for each:

gettext and the gettextize program must be installed and be in
your PATH. E.g. Homebrew installs them in /usr/local/opt/gettext/bin.
Bootstrap failed, downloading required files instead.
Using cached file /Users/Sage/sage-9.3.beta1/upstream/configure-ee1d038453af9bb563b79ab9bdbb08c18540207a.tar.gz
tar: Error opening archive: Failed to open 'upstream/configure-ee1d038453af9bb563b79ab9bdbb08c18540207a.tar.gz'
make: *** [configure] Error 1

(Differing name fo the tarball). This does not happen on macOS 10.13., where I build successfully for each.

FWIW, there is no “upstream” directory in any of the failing SAGE_ROOTs.

Have not seen similar problems reported.

I do not have Homebrew (or other package manager) installed.

Thoughts? Pointers?

Thanks!

Justin

[Dima Pasechnik]

a proper sequence of commands is

./bootstrap -d

./configure

make

You get an error from ./bootstrap being

invoked without -d.

hth

Dima

Thanks!

Justin

--
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/02AAFF37-36C6-403B-A7E2-813AEC112FC6%40mac.com.

[Justin C. Walker]

> On Feb 10, 2021, at 13:55, Dima Pasechnik <dim...@gmail.com> wrote:
>
>
>
> On Wed, 10 Feb 2021, 21:29 'Justin C. Walker' via sage-release, <sage-r...@googlegroups.com> wrote:
>
>
> > On Feb 7, 2021, at 14:04, 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

> a proper sequence of commands is
>
> ./bootstrap -d
> ./configure
> make

Thanks for your response.

Is this new? I’ve been starting with “make configure” (which worked through Beta1).

In any case, this doesn’t seem to help. I started from scratch, pulling a new Beta7 tree, and then ran bootstrap, as above. This ended up looking just like the result of running “make configure” earlier.

Not sure what I am missing

Justin

[Dima Pasechnik]

could you show the exact message?

If it says "using cached configure tarball" on a clean directory it is weird^2

Not sure what I am missing

Justin

--
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/D325FE54-753D-4365-AB19-D64F82CB7E34%40mac.com.

[Samuel Lelievre]

2021-02-08 02:11:22 UTC, Samuel Lelievre:

>
> On macOS 10.14.6 with lots of Homebrew packages,

> [failed to build linbox]. Logs:

>
> https://www.math.u-psud.fr/~lelievre/t/slel-2021-02-08-s93b7-brew.zip
>
> I'll try #31355.

Same failure to build linbox after applying #31355.
I'm sticking to Sage 9.3.beta5 for now. I should maybe
check if I can still build Sage 9.3.beta5. Or maybe try
removing a few Homebrew packages in case that helps.

[Dima Pasechnik]

probably you have Macaulay2 installed via the Homebrew tap.
And it includes macaulay2/tap/fflas-ffpack - so that's causing an error here.

>
> --
> 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/515d3489-4c1b-47fd-b801-73c7c9676a07n%40googlegroups.com.

[Justin C. Walker]

> On Feb 10, 2021, at 14:41, Dima Pasechnik <dim...@gmail.com> wrote:
>
>
>
> On Wed, 10 Feb 2021, 22:29 'Justin C. Walker' via sage-release, <sage-r...@googlegroups.com> wrote:
>
>
> > On Feb 10, 2021, at 13:55, Dima Pasechnik <dim...@gmail.com> wrote:
> >
> >
> >
> > On Wed, 10 Feb 2021, 21:29 'Justin C. Walker' via sage-release, <sage-r...@googlegroups.com> wrote:
> >
> >
> > > On Feb 7, 2021, at 14:04, 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
>
> > a proper sequence of commands is
> >
> > ./bootstrap -d
> > ./configure
> > make
>
> Thanks for your response.
>
> Is this new? I’ve been starting with “make configure” (which worked through Beta1).
>
> In any case, this doesn’t seem to help. I started from scratch, pulling a new Beta7 tree, and then ran bootstrap, as above. This ended up looking just like the result of running “make configure” earlier.
>
> could you show the exact message?
> If it says "using cached configure tarball" on a clean directory it is weird^2

Here’s the full output from “bootstrap

rm -rf config configure build/make/Makefile-auto.in
rm -f src/doc/en/installation/*.txt
rm -rf src/doc/en/reference/spkg/*.rst
rm -f src/doc/en/reference/repl/*.txt
rm -f environment.yml
rm -f src/environment.yml
rm -f environment-optional.yml
rm -f src/environment-optional.yml

src/doc/bootstrap:92: installing src/doc/en/installation/arch.txt and src/doc/en/installation/arch-optional.txt
src/doc/bootstrap:92: installing src/doc/en/installation/debian.txt and src/doc/en/installation/debian-optional.txt
src/doc/bootstrap:92: installing src/doc/en/installation/fedora.txt and src/doc/en/installation/fedora-optional.txt
src/doc/bootstrap:92: installing src/doc/en/installation/cygwin.txt and src/doc/en/installation/cygwin-optional.txt
src/doc/bootstrap:92: installing src/doc/en/installation/homebrew.txt and src/doc/en/installation/homebrew-optional.txt
src/doc/bootstrap:66: installing environment.yml, environment-optional.yml, src/environment.yml and src/environment-optional.yml
src/doc/bootstrap:103: installing src/doc/en/reference/spkg/*.rst
src/doc/bootstrap:182: installing src/doc/en/reference/repl/options.txt

gettext and the gettextize program must be installed and be in
your PATH. E.g. Homebrew installs them in /usr/local/opt/gettext/bin.
Bootstrap failed, downloading required files instead.

Using cached file /Users/Sage/sage-9.3.beta1/upstream/configure-6096110e060e441df3219fd8caef1c4525145983.tar.gz
tar: Error opening archive: Failed to open 'upstream/configure-6096110e060e441df3219fd8caef1c4525145983.tar.gz’

The “cached” line is always there for this failure. The line about gettext and gettextsize has been present as well. Both are new starting with Beta2. I don’t see them, e.g., on 10.13.6.

Thanks, as always!

Justin

[Dima Pasechnik]

something is not clean in the repo

maybe run

      make bootstrap-clean

--
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/15A6FB09-7EB3-4045-88DD-83D6F78BBB39%40mac.com.

[Justin C. Walker]

> On Feb 11, 2021, at 08:52, Dima Pasechnik <dim...@gmail.com> wrote:
>
> something is not clean in the repo
>
> maybe run
>
> make bootstrap-clean

I ran this, “make distclean” and “make clean”. All three had the same output as for “bootstrap -d”, which I guess is not too surprising.

I’m at the edge of my understanding of the build process now…

Justin

[John H Palmieri]

I just noticed something which has been happening for all of the 9.3 beta releases, maybe before that. In the dochtml.log file, I see the message

    Help file /Applications/usr/share/giac/doc/fr/aide_cas not found

(This is on OS X.) That's a strange path — there is no /Applications/usr — so it's not surprising that it's not found. What is producing this message? The path should be fixed.

--

John

On Sunday, February 7, 2021 at 2:04:42 PM UTC-8 Volker Braun 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 

8453ffb849 (tag: 9.3.beta7, trac/develop) Updated SageMath version to 9.3.beta7

[Matthias Köppe]

On Thursday, February 11, 2021 at 3:49:44 PM UTC-8 John H Palmieri wrote:

I just noticed something which has been happening for all of the 9.3 beta releases, maybe before that. In the dochtml.log file, I see the message

    Help file /Applications/usr/share/giac/doc/fr/aide_cas not found

(This is on OS X.) That's a strange path — there is no /Applications/usr — so it's not surprising that it's not found. What is producing this message? The path should be fixed.

See https://trac.sagemath.org/ticket/29552, https://trac.sagemath.org/ticket/30537

 

[Samuel Lelièvre]

2021-02-11 15:35 UTC, Dima:
>
> On Thu, Feb 11, 2021 at 5:45 AM Samuel Lelievre:

> >
> > 2021-02-08 02:11:22 UTC, Samuel Lelievre:
> > >
> > > On macOS 10.14.6 with lots of Homebrew packages,
> > > [failed to build linbox]. Logs:
> > >
> > > https://www.math.u-psud.fr/~lelievre/t/slel-2021-02-08-s93b7-brew.zip
> > >
> > > I'll try #31355.
> >
> > Same failure to build linbox after applying #31355.
> > I'm sticking to Sage 9.3.beta5 for now. I should maybe
> > check if I can still build Sage 9.3.beta5. Or maybe try
> > removing a few Homebrew packages in case that helps.
>
> probably you have Macaulay2 installed via the Homebrew tap.
> And it includes macaulay2/tap/fflas-ffpack - so that's causing an error here.

Thanks Dima, well spotted! I uninstalled every `macaulay2/tap/*`
package and now sagelib builds. It still fails to build the html
documentation though. Zipped `config.log` and `dochtml.log`:

https://www.math.u-psud.fr/~lelievre/t/slel-mac-brew/slel-2021-02-13-s93b7-brew.zip

[Thierry]

Hi,

On Sun, Feb 07, 2021 at 02:04:42PM -0800, Volker Braun 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

Thanks Volker and all devs for this release.

Since this 9.3.beta7 release, all the patchbots i am running fail, even
after `make distclean`, here is a typical log:

https://patchbot.sagemath.org/log/0/Linux/1_SMP_Debian_4.19.171-2_(2021-01-30)/x86_64/4.19.0-14-amd64/tmonteil-lxc/2021-02-13%2023:01:46

If some of you understands the issue, it would be helpful (there are
currently only 2 patchbots running).

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/5c66e760-31f5-4b92-8490-6a50bd902da8n%40googlegroups.com.

[jonatha...@googlemail.com]

Here is what I think is relevant

in control.py

 431     def _init_warn_long(self):
 432 +---- 23 lines: """------------------------------------------------------------------------------------------------------------------------------------------------------------
 455         if self.options.warn_long is not None:     # Specified on the command line
 456             return
 457         try:
 458             self.options.warn_long = 60.0 * self.second_on_modern_computer()
 459         except RuntimeError as err:
 460             if not sage.doctest.DOCTEST_MODE:
 461                 print(err)   # No usable timing information

It appears there is a RuntimeError but it is not printed. You can edit the line to see what is going on.

At least I'm guessing that this is the problem.

[jonatha...@googlemail.com]

Oh yes, you probably have too few sucessful tests

I would propose to change this (init_warn_long) to set

self.optioins.warn_long = 60

in case of this RuntimeError

[Samuel Lelièvre]

2021-02-17 12:05 UTC, Samuel Lelièvre:

Dima, I reported this to the Macaulay Homebrew tap repo:

https://github.com/Macaulay2/homebrew-tap/issues/68

Can you say how you spotted what was causing the problem?
Is it something about using fflas-ffpack with vs without openmp?

[Dima Pasechnik]

I spotted include paths in the log that pointed to Macaulay2 tap.

> Is it something about using fflas-ffpack with vs without openmp?

I don't recall, but it could be.

>
> --
> 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/CAEcArF2PHhNiQzvkGu_kw8Y5sXZMbi2Th-Tz3Y4nAQjeqFf1zg%40mail.gmail.com.

[Samuel Lelièvre]

2021-02-17 23:06 UTC, Dima Pasechnik:
>
> On Wed, Feb 17, 2021 at 9:51 PM Samuel Lelièvre:

> >
> > 2021-02-17 12:05 UTC, Samuel Lelièvre:
> > >
> > > 2021-02-11 15:35 UTC, Dima:
> > > >
> > > > On Thu, Feb 11, 2021 at 5:45 AM Samuel Lelievre:
> > > > >
> > > > > 2021-02-08 02:11:22 UTC, Samuel Lelievre:
> > > > > >
> > > > > > On macOS 10.14.6 with lots of Homebrew packages,
> > > > > > [failed to build linbox]. Logs:
> > > > > >
> > > > > > https://www.math.u-psud.fr/~lelievre/t/slel-2021-02-08-s93b7-brew.zip
> > > > > >
> > > > > > I'll try #31355.
> > > > >
> > > > > Same failure to build linbox after applying #31355.
> > > > > I'm sticking to Sage 9.3.beta5 for now. I should maybe
> > > > > check if I can still build Sage 9.3.beta5. Or maybe try
> > > > > removing a few Homebrew packages in case that helps.
> > > >
> > > > probably you have Macaulay2 installed via the Homebrew tap.
> > > > And it includes macaulay2/tap/fflas-ffpack - so that's causing an error here.
> > >
> > > Thanks Dima, well spotted! I uninstalled every `macaulay2/tap/*`
> > > package and now sagelib builds. It still fails to build the html
> > > documentation though. Zipped `config.log` and `dochtml.log`:
> > >
> > > https://www.math.u-psud.fr/~lelievre/t/slel-mac-brew/slel-2021-02-13-s93b7-brew.zip
> >
> > Dima, I reported this to the Macaulay Homebrew tap repo:
> >
> > https://github.com/Macaulay2/homebrew-tap/issues/68
> >
> > Can you say how you spotted what was causing the problem?
>
> I spotted include paths in the log that pointed to Macaulay2 tap.
>
> > Is it something about using fflas-ffpack with vs without openmp?
>
> I don't recall, but it could be.

Thanks. The problem I reported about the html documentation
not building is solved by #31344. With that, `make` succeeds
and `make testlong` gives the joyous "All tests passed!".

[Matthias Köppe]

On Tuesday, February 9, 2021 at 9:41:54 PM UTC-8 Matthias Köppe wrote:

The runs on GH Actions show a very strange failure on fedora-34 and archlinux-latest (https://github.com/sagemath/sage/runs/1850248457):

+ ./bootstrap

....

./sage: no Sage installation found in $SAGE_ROOT=/sage

This can be reproduced locally using "tox -e docker-fedora-34-standard".

It turns out that the built-in command "test" of bash is broken:

[root@9d0e0458cde5 sage]# test -r VERSION.txt && echo yes

Compare with /usr/bin/test

[root@9d0e0458cde5 sage]# /usr/bin/test -r VERSION.txt && echo yes
yes

Also on ubuntu-hirsute-standard.

This appears to be a problem with recent glibc versions, related to the faccessat2 system call - see for example https://bugzilla.redhat.com/show_bug.cgi?id=1900021

 

[John H Palmieri]

Sorry to resurrect this old thread, but I'm seeing the same numerical doctest failures on an OS X machine with Sage 9.5.beta1. Any ideas where these could be coming from? Some miscompiled component coming from the system, or something from within Sage itself? I started seeing these after upgrading to OS X 11.5.2 (it was on 10.?? before). I tried reinstalling all of homebrew's packages, and that didn't help.

On Sunday, February 7, 2021 at 5:12:40 PM UTC-8 matthia...@gmail.com wrote:

On Sunday, February 7, 2021 at 2:04:42 PM UTC-8 Volker Braun 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 

8453ffb849 (tag: 9.3.beta7, trac/develop) Updated SageMath version to 9.3.beta7

First test results from https://github.com/sagemath/sage/actions/runs/545994902:

ubuntu-bionic-standard: (python 3.6) still https://trac.sagemath.org/ticket/31191

debian-bullseye-standard: New(?) numerics issues:

sage -t --random-seed=0 src/sage/finance/time_series.pyx  # 8 doctests failed

sage -t --random-seed=0 src/sage/geometry/polyhedron/library.py  # 1 doctest failed

sage -t --random-seed=0 src/sage/geometry/polyhedron/base.py  # 1 doctest failed

sage -t --random-seed=0 src/sage/matrix/matrix_double_dense.pyx  # 2 doctests failed

sage -t --random-seed=0 src/sage/matrix/matrix_integer_dense.pyx  # 1 doctest failed

sage -t --random-seed=0 src/sage/numerical/sdp.pyx  # 2 doctests failed

sage -t --random-seed=0 src/sage/tests/books/computational-mathematics-with-sagemath/float_doctest.py  # 1 doctest failed

 

[Matthias Köppe]

https://trac.sagemath.org/ticket/31621 tracks this issue.