Sage 9.2.beta1 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 

e2dcdeeabb (tag: 9.2.beta1) Updated SageMath version to 9.2.beta1

2fb1ae046c Trac #29773: Remove a few unnessary future imports

e979e37c84 Trac #29747: Add tangent_line() and related methods to curves and ambient spaces

7bf295ef72 Trac #29745: Clean up SPKG.rst for packages

3f7b31b829 Trac #29641: spring cleanup in finite state machine

7be3b1099a Trac #29781: properly catch exceptions in Dijkstra_Boost

e5adcbff10 Trac #29776: fix tab-completion for methods injected by MorphismMethods

16ec3e5890 Trac #29774: Ignore jedi deprecation warnings

8776e70918 Trac #29770: pyflakes cleanup for sage/misc

aec8ffc155 Trac #29757: .github/workflows/tox-gcc_spkg.yml: Remove python2

79e11f963e Trac #29698: Implement the cohomology ring of a right-angled Artin group

9504cf2845 Trac #29696: Implement the blob algebra

dea0eeb3fd Trac #26532: sparse matrix multiplication is slow

a0da5926af Trac #29680: Fix checksums: print both package name and tarball name

01d9c1deff Trac #29523: Sample points exponentially when plotting log-scale graphs

48416a457b Trac #29750: Fix a bug in singular points computation of projective curves

6932618e02 Trac #29749: incorrect package name for python37-urllib3 on Cygwin in installation guide

5fc5fbf287 Trac #29643: Typo ticket (2020-05)

209bc8b4a4 Trac #29741: Bender-Knuth involutions

6d0af7f30f Trac #29737: Add L-polynomial to integral curves

85bb17039f Trac #29684: fix some details in the doc of zinbiel algebras

d5381d1531 Trac #29679: Cleanup of 28757

c6db6429d8 Trac #29768: remove very old deprecated code in fast_eval and fast_callable

41c4fea464 Trac #29637: some details in Lyndon words

e2849f066a Trac #29765: get rid of _cmp_ for polynomials

2705552296 Trac #29762: fix some details in doc of Hasse diagrams

9179459a97 Trac #29761: fix one doc glitch in spanning tree

4a38c69fe5 Trac #29727: dihedral type reflection groups

04305d29cd Trac #27880: Update Kenzo and its interface

06435c8cb3 Trac #29566: CombinatorialPolyhedron: Store incidence matrix on initialization

5467a6680b Trac #29217: extract code from OEIS

c1f445290f Trac #19010: Can't multiply an element of a group with an element of a subgroup

81e8fe699c Trac #10930: principal and exponential specializations for symmetric functions

a3766ee0f5 Trac #29759: fix Macaulay2 interface broken by #28000

1ae21f9cdd Trac #29734: Defect in `shortest_path_lengths` method in `generic_graph.py`

02e7e53465 Trac #29732: little enhancements for STL 3D export

5e17cb6851 Trac #29685: details about dilog and polylog

32c92da9b6 Trac #29682: spring cleanup in pushout.py

11642cd301 Trac #29634: some details in combinat about partitions

0c5cf79598 Trac #29595: Improve iteration of multivariate polynomials

56610a29dc Trac #29480: Upgrade Cython to 0.29.17

3742f64428 Trac #29726: hash for reflection groups

69c1c28151 Trac #29722: some care for Coxeter and Reflection groups

c0a8b8286d Trac #29713: Broken conversion from FractionField over PolynomialRing over Field back to Field

5f42795e02 Trac #29712: some cleanup for core.py

b66d9e3ed2 Trac #29700: ValueError raised when computing numerical approx of a matrix on QQbar

0e59f415b7 Trac #29699: spring cleanup in partition tuples

d689a20375 Trac #29697: src/setup.py, src/sage/env.py (sage_include_directories): Do not add another copy of SAGE_INC, SAGE_LOCAL/lib to include dirs, library dirs

86269e9422 Trac #29695: Improve speed of diagram algebras by avoiding checks

7b21a9a7da Trac #29692: Braid closure gives wrong link

990407a73a Trac #29689: little fix in shuffle algebras

fb8b44c95f Trac #29669: For Sage 9.2: Remove Python 2 support from the build system and CI scripts

709340cd85 Trac #29666: Serious precision problems with heights of elliptic curve points over number fields

8b1e5bcbf1 Trac #29633: src/doc/bootstrap: Generate list of packages (with info and links) for ref manual from build/pkgs

a5c6f3465e Trac #29583: Obtain product with both Vrep and Hrep (if backend supports it)

24590bae92 Trac #28457: Build binder compatible docker images from Trac

27aae327ef Trac #29748: MR42: Make docker versions compatible in GitLab CI

795593fce0 Trac #29547: Update to matplotlib 3

860e4dc988 (tag: 9.2.beta0, trac/develop) Updated SageMath version to 9.2.beta0

[Dima Pasechnik]

the develop branch is still on beta0.

> --
> 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/c8a96fe3-9723-4d27-b06f-52c0f0856d2ao%40googlegroups.com.

[Justin C. Walker]

> On Jun 13, 2020, at 15:28 , 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

Pulled fresh clones of the develop tree on the usual three macOS systems:
10.13.6 (2017 iMac Pro, 18-core Xeon W)
10.14.6 (2017 MBP Pro, Quad-core Core i7)
10.15.5 (2019 MBP Pro, 9-core Core i9)

In each case, the build completed without complaint, and all tests (‘ptestlong’) passed!

Justin

--
Justin C. Walker, Curmudgeon at Large
Director
Institute for the Enhancement of the Director's Income
--
In mathematics you don't understand
things. You just get used to them.
--John von Neumann
--

[kcrisman]

On Mac OS X 10.11.  Upgrading via git pull trac develop and make from somewhere in the 9.0 betas, I get a lot of good things, until:

mkdir ../../../../library/tools/libs

installing 'sysdata.rda'

dyld: Library not loaded: /Users/.../sage/local/lib/libopenblas_haswellp-r0.3.6.dylib

  Referenced from: /Users/.../sage/local/lib/R/lib/libR.dylib

  Reason: image not found

/bin/bash: line 1: 24759 Done                    echo "tools:::sysdata2LazyLoadDB(\"./R/sysdata.rda\",\"../../../library/tools/R\")"

     24760 Trace/BPT trap: 5       | R_DEFAULT_PACKAGES=NULL LC_ALL=C R_ENABLE_JIT=0 TZ=UTC ../../../bin/R --vanilla --slave

make[8]: *** [sysdata] Error 133

make[7]: *** [all] Error 2

make[6]: *** [R] Error 1

make[5]: *** [R] Error 1

make[4]: *** [R] Error 1

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

Error building r-3.6.2.p0

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

real    2m5.131s

user    0m47.882s

sys     0m46.001s

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

Error installing package r-3.6.2.p0

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

$ ls local/lib/libopenblas[tab]

libopenblas.0.dylib                libopenblas.dylib                  libopenblas_haswellp-r0.3.9.dylib  

So I'm not sure why it's asking for an older version of libopenblas.  Full R log at http://www.math.gordon.edu/~kcrisman/r-3.6.2.p0.log

[kcrisman]

mkdir ../../../../library/tools/libs

installing 'sysdata.rda'

dyld: Library not loaded: /Users/.../sage/local/lib/libopenblas_haswellp-r0.3.6.dylib

  Referenced from: /Users/.../sage/local/lib/R/lib/libR.dylib

For reference, doing

$ rm -rf local/lib/R/

ended up solving the problem.  But that seems like a bug - shouldn't the previous R install properly be deleted, in our current paradigm of first uninstalling (if I understand it correctly)?

[Dima Pasechnik]

yes, it appears to be a bug (macOS-specific?). Probably the list of
files installed by R spkg is incomplete, and so what's in the list is
removed, but
this one is not.
You can see the file list in local/var/lib/sage/installed/
and check whether libR.dylib
is there.

HTH
Dima

>
> --
> 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/f50588d7-2094-498a-870c-b63716e105d3o%40googlegroups.com.

[Matthias Köppe]

A possibly related ticket: https://trac.sagemath.org/ticket/28008

[Justin C. Walker]

> On Jun 17, 2020, at 13:22 , Dima Pasechnik <dim...@gmail.com> wrote:
>
> On Wed, Jun 17, 2020 at 6:38 PM kcrisman <kcri...@gmail.com> wrote:
>>
>>
>>
>>
>>>
>>> mkdir ../../../../library/tools/libs
>>>
>>> installing 'sysdata.rda'
>>>
>>> dyld: Library not loaded: /Users/.../sage/local/lib/libopenblas_haswellp-r0.3.6.dylib
>>>
>>> Referenced from: /Users/.../sage/local/lib/R/lib/libR.dylib
>>
>>
>> For reference, doing
>>
>> $ rm -rf local/lib/R/
>>
>> ended up solving the problem. But that seems like a bug - shouldn't the previous R install properly be deleted, in our current paradigm of first uninstalling (if I understand it correctly)?
>
> yes, it appears to be a bug (macOS-specific?). Probably the list of
> files installed by R spkg is incomplete, and so what's in the list is
> removed, but
> this one is not.
> You can see the file list in local/var/lib/sage/installed/
> and check whether libR.dylib
> is there.

Not sure this is related to the above problem, but I checked the results of the builds in each of my macOS systems (10.13, 10.14, 10.15), and there are no “libR.*” files in either SAGE_ROOT/local/var/lib/sage/installed or SAGE_ROOT/local/lib.

Is one of the Mac package gizmos (macports/homebrew/etc) involved?

Justin

--
Justin C. Walker
Curmudgeon-at-large
--
Network, n., Difference between work
charged for and work done

[Dima Pasechnik]

Sage is able to use R package of homebrew, yes. If this happens on a
system then obviously there will be no data on R installed by Sage.
(look at config.log to see what is used)
Otherwise it's a bug.

On Linux I see
$ grep libR local/var/lib/sage/installed/r-3.6.2.p0
"lib/R/lib/libR.so",
"lib/pkgconfig/libR.pc",

Do you actually get local/var/lib/sage/installed/r-3.6.2.p0 on macOS
if R is istalled by Sage?

[kcrisman]

Otherwise it's a bug.

I was not using a system package for R.  While Matthias' link may conceivably be relevant as well, I'm currently leaning toward the other option.  On the old run (upgrading R) I have in the log the first time (unsuccessful):

Warning: No files to uninstall for 'r'
Uninstalling existing 'r'

While on the second (unsuccessful) and most recent (successful) run I see:

Package 'r' is currently not installed

No legacy uninstaller found for 'r'; nothing to do

Perhaps that helps someone who understands the uninstaller logic - that came after I had to stop keeping track of build system developments.

 

On Linux I see
$ grep libR local/var/lib/sage/installed/r-3.6.2.p0
        "lib/R/lib/libR.so",
        "lib/pkgconfig/libR.pc",

It's the Mac equivalent.  (I don't think any system packages are used for my install, or very few.)  Unfortunately I don't have the state before this upgrade any more.

$ grep libR local/var/lib/sage/installed/r-3.6.2.p0 

        "lib/R/lib/libR.dylib",

        "lib/pkgconfig/libR.pc", 

[Dima Pasechnik]

could you try to re-install R, and check what you see in the log?
(i.e. "Package 'r' is currently not installed" would be wrong message,
I suppose, I'd rather see something about "uninstalling")

>
> --
> 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/161201ad-43ba-4f12-b917-6617d111b89bo%40googlegroups.com.

[Justin C. Walker]

> On Jun 18, 2020, at 02:23 , Dima Pasechnik <dim...@gmail.com> wrote:
>
> On Thu, Jun 18, 2020 at 3:45 AM 'Justin C. Walker' via sage-release
> <sage-r...@googlegroups.com> wrote:
>>
>>
>>
>>> On Jun 17, 2020, at 13:22 , Dima Pasechnik <dim...@gmail.com> wrote:
>>>
>>> On Wed, Jun 17, 2020 at 6:38 PM kcrisman <kcri...@gmail.com> wrote:

[snip]

>> Is one of the Mac package gizmos (macports/homebrew/etc) involved?
>
> Sage is able to use R package of homebrew, yes. If this happens on a
> system then obviously there will be no data on R installed by Sage.
> (look at config.log to see what is used)
> Otherwise it's a bug.
>
> On Linux I see
> $ grep libR local/var/lib/sage/installed/r-3.6.2.p0
> "lib/R/lib/libR.so",
> "lib/pkgconfig/libR.pc",
>
> Do you actually get local/var/lib/sage/installed/r-3.6.2.p0 on macOS
> if R is istalled by Sage?

Yep! I do have that on my system. To be clear, I do not use any of the macOS packaging schemes (I have Mac Ports, but not accessible unless I want to use an app that this packager provides [or I make a mistake]).

HTH

Justin

--
Justin C. Walker, Curmudgeon-at-Large
() The ASCII Ribbon Campaign
/\ Help Cure HTML Email

[kcrisman]

could you try to re-install R, and check what you see in the log?
(i.e. "Package 'r' is currently not installed" would be wrong message,
I suppose, I'd rather see something about "uninstalling")

I would too, but:

[r-3.6.2.p0] Package 'r' is currently not installed

[r-3.6.2.p0] No legacy uninstaller found for 'r'; nothing to do

It otherwise installs fine, though. 

[Emmanuel Charpentier]

FWIW :

On Debian testing running on core i7 + 16 GB RAM, upgrading 9.1.beta0 (configured to use systewide packages whenever possible) and running ptestlong gives 1 (transient) timeout and 11 (permanent) failures :

File	Result	P/T
src/sage/schemes/elliptic_curves/ell_number_field.py	3 doctests failed	P
src/sage/tests/parigp.py	Timed out	T
src/sage/rings/number_field/number_field.py	7 doctests failed	P
src/sage/rings/number_field/number_field_element.pyx	1 doctest failed	P
src/sage/rings/number_field/number_field_ideal.py	2 doctests failed	P
src/sage/tests/gap_packages.py	1 doctest failed	P
src/sage/numerical/backends/glpk_backend.pyx	1 doctest failed	P
src/sage/lfunctions/dokchitser.py	2 doctests failed	P
src/sage/rings/polynomial/polynomial_quotient_ring.py	2 doctests failed	P
src/sage/lfunctions/pari.py	1 doctest failed	P
src/sage/libs/glpk/error.pyx	1 doctest failed	P
src/sage/rings/number_field/unit_group.py	1 doctest failed	P

AFAICT, all these failures have been already reported and explained away.

Attached files : the result of re-running the failed doctests is in chkerrs.txt, configuration can be deduced from config.log.

HTH,

[Sébastien Labbé]

Testing various optional and external packages (on June 15, but I am reporting only now sorry for the delay), I get

Using --optional=4ti2,cbc,ccache,cryptominisat,dot2tex,e_antic,external,fricas,glucose,latte_int,lidia,lrslib,memlimit,normaliz,notedown,openssl,pandoc_attributes,pycosat,pynormaliz,rst2ipynb,sage,sage_numerical_backends_coin,sage_numerical_backends_cplex,sage_numerical_backends_gurobi

----------------------------------------------------------------------
sage -t --long src/sage/combinat/matrices/hadamard_matrix.py  # 1 doctest failed
sage -t --long src/sage/combinat/quickref.py  # 1 doctest failed
sage -t --long src/sage/combinat/species/library.py  # 1 doctest failed
sage -t --long src/sage/combinat/tutorial.py  # 1 doctest failed
sage -t --long src/sage/databases/findstat.py  # 17 doctests failed
sage -t --long src/sage/databases/oeis.py  # 1 doctest failed
sage -t --long src/sage/geometry/polyhedron/base.py  # Bad exit: 1
sage -t --long src/sage/graphs/generators/smallgraphs.py  # 2 doctests failed
sage -t --long src/sage/sat/boolean_polynomials.py  # 1 doctest failed
----------------------------------------------------------------------
External software detected for doctesting: cplex,ffmpeg,graphviz,imagemagick,internet,latex,pandoc

Rerunning failed tests, I get

----------------------------------------------------------------------
sage -t --long src/sage/combinat/quickref.py  # 1 doctest failed
sage -t --long src/sage/combinat/species/library.py  # 1 doctest failed
sage -t --long src/sage/combinat/tutorial.py  # 1 doctest failed
sage -t --long src/sage/databases/findstat.py  # 17 doctests failed
sage -t --long src/sage/databases/oeis.py  # 1 doctest failed
sage -t --long src/sage/geometry/polyhedron/base.py  # Bad exit: 1
sage -t --long src/sage/graphs/generators/smallgraphs.py  # 2 doctests failed
----------------------------------------------------------------------
External software detected for doctesting: internet


Many of them are related to recent changes in oeis, see below.


sage -t --long src/sage/combinat/quickref.py
**********************************************************************
File "src/sage/combinat/quickref.py", line 9, in sage.combinat.quickref
Failed example:
    s[0].programs() # optional - internet
Expected:
    0: (PARI) {a(n) = if( n<0, 0, n!^2 * 4^n * polcoeff( 1 / besselj(0, x + x * O(x^(2*n))), 2*n))}; /* _Michael Somos_, May 17 2004 */
Got:
    [('maple', 0: A000275 := proc(n) sum(z^k/k!^2, k = 0..infinity);
      1: series(%^x, z=0, n+1): n!^2*coeff(%,z,n); add(abs(coeff(%,x,k)), k=0..n) end:
      2: seq(A000275(n), n=0..17); # _Peter Luschny_, May 27 2017),
     ('mathematica',
      0: a[0] = 1; a[n_] := a[n] = Sum[(-1)^(r+n+1)*Binomial[n, r]^2 a[r], {r, 0, n-1}]; Table[a[n], {n, 0, 17}] (* _Jean-François Alcover_, Aug 05 2013 *)
      1: CoefficientList[Series[1/BesselJ[0,Sqrt[4*x]], {x, 0, 20}], x]* Range[0, 20]!^2 (* _Vaclav Kotesovec_, Mar 02 2014 *)
      2: a[ n_] := If[ n < 0, 0, (n! 2^n)^2 SeriesCoefficient[ 1 / BesselJ[ 0, x], {x, 0, 2 n}]]; (* _Michael Somos_, Aug 20 2015 *)),
     ('pari',
      0: {a(n) = if( n<0, 0, n!^2 * 4^n * polcoeff( 1 / besselj(0, x + x * O(x^(2*n))), 2*n))}; /* _Michael Somos_, May 17 2004 */)]
**********************************************************************
1 item had failures:
   1 of  22 in sage.combinat.quickref
    5 not tested tests not run
    0 tests not run because we ran out of time
    [21 tests, 1 failure, 3.06 s]


sage -t --long src/sage/combinat/species/library.py
**********************************************************************
File "src/sage/combinat/species/library.py", line 104, in sage.combinat.species.library.BinaryTreeSpecies
Failed example:
    oeis(seq)[0]                              # optional -- internet
Expected:
    A000108: Catalan numbers: C(n) = binomial(2n,n)/(n+1) = (2n)!/(n!(n+1)!). Also called Segner numbers.
Got:
    A000108: Catalan numbers: C(n) = binomial(2n,n)/(n+1) = (2n)!/(n!(n+1)!).
**********************************************************************
1 item had failures:
   1 of  10 in sage.combinat.species.library.BinaryTreeSpecies
    0 tests not run because we ran out of time
    [23 tests, 1 failure, 4.48 s]



sage -t --long src/sage/combinat/tutorial.py
**********************************************************************
File "src/sage/combinat/tutorial.py", line 224, in sage.combinat.tutorial
Failed example:
    oeis([1,1,2,5,14])                            # optional -- internet
Expected:
    0: A000108: Catalan numbers: C(n) = binomial(2n,n)/(n+1) = (2n)!/(n!(n+1)!). Also called Segner numbers.
    1: ...
    2: ...
Got:
    0: A000108: Catalan numbers: C(n) = binomial(2n,n)/(n+1) = (2n)!/(n!(n+1)!).
    1: A022562: Number of connected claw-free unlabeled graphs on n nodes.
    2: A124302: Number of set partitions with at most 3 blocks; number of Dyck paths of height at most 4; dimension of space of symmetric polynomials in 3 noncommuting variables.
**********************************************************************
1 item had failures:
   1 of 248 in sage.combinat.tutorial
    5 tests for not implemented functionality not run
    6 not tested tests not run
    1 py2 test not run
    0 tests not run because we ran out of time
    [247 tests, 1 failure, 18.41 s]



sage -t --long src/sage/databases/oeis.py
**********************************************************************
File "src/sage/databases/oeis.py", line 492, in sage.databases.oeis.OEIS.find_by_description
Failed example:
    oeis('beaver', max_results=4, first_result=2)     # optional -- internet
Expected:
    0: A131956: Busy Beaver variation: maximum number of steps for ...
    1: A141475: Number of Turing machines with n states following ...
    2: A131957: Busy Beaver sigma variation: maximum number of 1's ...
    3: A...: ...
Got:
    0: A131956: Busy Beaver variation: maximum number of steps for a 2-state, 2-symbol Turing machine running on a tape which is initialized with the number n in binary and 0's everywhere else. The machine is started at the rightmost bit in the number n.
    1: A141475: Number of Turing machines with n states following the standard formalism of the busy beaver problem where the head of a Turing machine either moves to the right or to the left, but none once halted.
    2: A333479: Busy Beaver for binary lambda calculus: the maximum normal form size of any closed lambda term of size n, or 0 if no closed term of size n exists.
    3: A131957: Busy Beaver sigma variation: maximum number of 1's on the final tape, for a 2-state, 2-symbol Turing machine running on a tape which is initialized with the number n in binary and 0's everywhere else. The machine is started at the rightmost bit in the number n.
**********************************************************************
1 item had failures:
   1 of   5 in sage.databases.oeis.OEIS.find_by_description
    5 webbrowser tests not run
    0 tests not run because we ran out of time
    [287 tests, 1 failure, 39.53 s]

[tsc...@ucdavis.edu]

This might help for some of the failing internet tests: https://trac.sagemath.org/ticket/29787