Sage 7.5.beta6 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

608cfc2 Updated SageMath version to 7.5.beta6

98f6946 Trac #22039: Docstring on SageNB error with moebius algebra

34d61b9 Trac #22011: Make giac a standard package

33ba206 Trac #21793: Document whitespace policy, provide editor configurations for non-Python source files

54934ba Trac #22031: PEP8 compliance: replace "== None" by "is None" in graphs package

d0e3225 Trac #21625: Strange linking order for omalloc due to pkgconfig bug

f9c7b2e Trac #21613: Don't use make for autogenerated modules in sagelib

214e1d2 Trac #21608: Univariate Polynomial Ring does not support lc, lm, lt

763da96 Trac #21718: Add pictures to polygon.py

1af2b15 Trac #22023: fixing typos and and & the the

36c1fa0 Trac #21986: LaTeX and digraph of crystals.infinity.Tableaux never finish

e5d3852 Trac #21906: Bug in bessel_K

1a4df33 Trac #22020: py3 get rid of cmp() in fraction field element pyx

209c15d Trac #22012: Py3 : Handle cmp() in pyx files in misc and modules

bd1c6f1 Trac #21989: Remove _richcmp_ from FastDoubleFunc

7c74580 Trac #21778: correct refinement in cvxopt sdp backend

94a6e9d Trac #21669: Segmentation fault when saving certain empty matrices

b07da46 Trac #21587: Implement ribbon graphs

efe93f9 Trac #21428: AssertionError plotting real part of complex function with float coefficients

99f6548 Trac #22015: Document use of SAGE_EXTCODE

ea8c03a Trac #22014: Move Three.js template to src/ext

0cd788e Trac #22004: Allow algorithm='sympy' in symbolic_sum function

c24e811 Trac #21910: py3: our own integer cmp and permutation groups

79ca171 Trac #21393: Partial bindings for Arb polynomials with complex coefficients

34d779d Trac #20680: Some optimizations for addition in combinatorial free modules and dict_* methods

0e02b51 Trac #22009: sage -t src/sage/misc/latex.py opens 2 sage.pdf file

53a3308 Trac #21903: LatticePoset: add neutral_elements

d5bb829 Trac #16457: Rename xrange() to range()

df0e68f8 Trac #21998: any_root() sometimes fails over finite fields

fa39630 Trac #21997: Fix the check for Littlewood-Richardson tableaux

f6ba228 Trac #21983: Remaining py3 syntax level errors

163489e Updated SageMath version to 7.5.beta5

[Paul Masson]

There are still two uncommitted changes on both master and develop with files that start with

# This file is auto-generated by sage_setup/autogen/pari/generator.py

I known that I can just delete them both, but it must be confusing to new developers to see these changes. Please remove them from the repository.

These uncommitted changes must have been around for some time, since I've seen at least one branch on Trac from a seasoned developer that has them attached to 7.5.beta2.

[Volker Braun]

We can't remove uncommitted files from the repository because by definition they aren't being tracked by the repo.

The only workaround would be to put them into a .gitignore to leave the files in the worktree but not show them as untracked.

[Dima Pasechnik]

On Saturday, December 10, 2016 at 9:14:14 AM UTC, Volker Braun wrote:

We can't remove uncommitted files from the repository because by definition they aren't being tracked by the repo.

The only workaround would be to put them into a .gitignore to leave the files in the worktree but not show them as untracked.

if these files are indeed generated during the (normal) Sage installation then by right they should be in .gitignore.

[Francois Bissey]

> On 10/12/2016, at 22:53, Dima Pasechnik <dim...@gmail.com> wrote:
>
> On Saturday, December 10, 2016 at 9:14:14 AM UTC, Volker Braun wrote:
> We can't remove uncommitted files from the repository because by definition they aren't being tracked by the repo.
>
> The only workaround would be to put them into a .gitignore to leave the files in the worktree but not show them as untracked.
>
> if these files are indeed generated during the (normal) Sage installation then by right they should be in .gitignore.

+1
This email may be confidential and subject to legal privilege, it may
not reflect the views of the University of Canterbury, and it is not
guaranteed to be virus free. If you are not an intended recipient,
please notify the sender immediately and erase all copies of the message
and any attachments.

Please refer to http://www.canterbury.ac.nz/emaildisclaimer for more
information.

[Jeroen Demeyer]

On 2016-12-10 02:28, Paul Masson wrote:
> There are still two uncommitted changes on both master and develop with
> files that start with
>
> # This file is auto-generated by sage_setup/autogen/pari/generator.py
>
> I known that I can just delete them both, but it must be confusing to
> new developers to see these changes. Please remove them from the repository.

I believe these are automatically deleted when running "make", so I
consider it not an issue.

[fchap...@gmail.com]

Hello

it seems that once again somebody broke the python3 compatibility of the scripts:

[sagelib-7.5.beta6]   File "/home/chapoton/sage3/src/sage_setup/autogen/__init__.py", line 9, in autogen_all

[sagelib-7.5.beta6]     pari.rebuild()

[sagelib-7.5.beta6]   File "/home/chapoton/sage3/src/sage_setup/autogen/pari/__init__.py", line 11, in rebuild

[sagelib-7.5.beta6]     src_files = [join(pari_share(), 'pari.desc')] + \

[sagelib-7.5.beta6]   File "/home/chapoton/sage3/local/lib/python3.5/posixpath.py", line 89, in join

[sagelib-7.5.beta6]     genericpath._check_arg_types('join', a, *p)

[sagelib-7.5.beta6]   File "/home/chapoton/sage3/local/lib/python3.5/genericpath.py", line 145, in _check_arg_types

[sagelib-7.5.beta6]     raise TypeError("Can't mix strings and bytes in path components") from None

[sagelib-7.5.beta6] TypeError: Can't mix strings and bytes in path components

I have not searched for the culprit, but please take care of that.

In the future, the scripts should be *TESTED* for python3 compatibility.

Frederic

[Dima Pasechnik]

On Saturday, December 10, 2016 at 10:13:58 AM UTC, fchap...@gmail.com wrote:

Hello

it seems that once again somebody broke the python3 compatibility of the scripts:

[sagelib-7.5.beta6]   File "/home/chapoton/sage3/src/sage_setup/autogen/__init__.py", line 9, in autogen_all

[sagelib-7.5.beta6]     pari.rebuild()

[sagelib-7.5.beta6]   File "/home/chapoton/sage3/src/sage_setup/autogen/pari/__init__.py", line 11, in rebuild

[sagelib-7.5.beta6]     src_files = [join(pari_share(), 'pari.desc')] + \

[sagelib-7.5.beta6]   File "/home/chapoton/sage3/local/lib/python3.5/posixpath.py", line 89, in join

[sagelib-7.5.beta6]     genericpath._check_arg_types('join', a, *p)

[sagelib-7.5.beta6]   File "/home/chapoton/sage3/local/lib/python3.5/genericpath.py", line 145, in _check_arg_types

[sagelib-7.5.beta6]     raise TypeError("Can't mix strings and bytes in path components") from None

[sagelib-7.5.beta6] TypeError: Can't mix strings and bytes in path components

I have not searched for the culprit, but please take care of that.

In the future, the scripts should be *TESTED* for python3 compatibility.

Scripts? Do you mean the places where the system Python is Python 3?

[fchap...@gmail.com]

I mean that "export SAGE_PYTHON3=yes & make build" breaks at the point displayed in my message, and it was not doing that before 7.5.b6.

I monitor on every new release where is the first py3 problem. This is a regression.

Frederic

[fchap...@gmail.com]

culprit ticket is https://trac.sagemath.org/ticket/21613#comment:63

[Volker Braun]

Maybe we can add a "make test-python3" target (or so) to test things that should work already on py3. Then set up a buildbot to monitor regressions....

[Dima Pasechnik]

it is an issue in the sense of these files being created at the wrong place, IMHO.

Aren't we trying to separate build space from source?

 

[Jeroen Demeyer]

On 2016-12-10 13:54, Dima Pasechnik wrote:
> Aren't we trying to separate build space from source?

Yes, we are in the process of doing that. Although this is about
auto-generated sources, so it's not necessarily wrong to put those in
the sources.

[Sébastien Labbé]

On OS X 10.10.2, make ptestlong has only one already reported failing test ( `sage126` is not defined line 514 in singular.py).

Also, I get a strange log file for sqlite which contains a little bit more than the SPKG.txt file :

$ cat logs/pkgs/sqlite.log
Found local metadata for sqlite-autoconf-3130000
= SQLite =

== Description ==

SQLite is a software library that implements a self-contained,
serverless, zero-configuration, transactional SQL database engine.

== License ==

Public Domain

== Upstream contact ==

 * http://www.sqlite.org

== Dependencies ==

* readline

== Special Update/Build Instructions ==

* Use the autoconf version of sqlite.
== Type ==

standard

[Jeroen Demeyer]

On 2016-12-10 01:16, Volker Braun wrote:
> d0e3225 Trac #21625: Strange linking order for omalloc due to pkgconfig bug

This breaks building Sage with ATLAS: https://trac.sagemath.org/ticket/22047

[Eric Gourgoulhon]

On Ubuntu 16.04 x86_64 Xeon E5-2623 + 16 GB RAM, after a fresh git clone + pull develop, make -j8 failled with the following error:

===========================
cd ../.. && sage-logger -p './sage --docbuild --no-pdf-links all html ' logs/dochtml.log
[dochtml] /home/eric/sage/7.5.beta6/local/bin/python: No module named interpreters.wrapper_rdf; 'sage_setup.docbuild' is a package and cannot be directly executed
Makefile:1063 : la recette pour la cible « doc-html » a échouée
make[2]: *** [doc-html] Erreur 1
make[2] : on quitte le répertoire « /home/eric/sage/7.5.beta6/build/make »
Makefile:886 : la recette pour la cible « all » a échouée
make[1]: *** [all] Erreur 2
make[1] : on quitte le répertoire « /home/eric/sage/7.5.beta6/build/make »

real    28m48.747s
user    182m38.168s
sys    22m4.972s
***************************************************************
Error building Sage.

The following package(s) may have failed to build (not necessarily
during this run of 'make all'):

The build directory may contain configuration files and other potentially
helpful information. WARNING: if you now run 'make' again, the build
directory will, by default, be deleted. Set the environment variable
SAGE_KEEP_BUILT_SPKGS to 'yes' to prevent this.

Makefile:16 : la recette pour la cible « all » a échouée
make: *** [all] Erreur 1
==========================================

Running make again (simply 'make', no parallelization) resulted in a successful build. 

Then make ptestlong gave "All tests passed!" but with a slow doctest warning:

==========================================
sage -t --long --warn-long 54.1 src/sage/symbolic/series.pyx
**********************************************************************
File "src/sage/symbolic/series.pyx", line 97, in sage.symbolic.series
Warning, slow doctest:
    sin(x*sin(x*sin(x*sin(x)))).series(x,12)
Test ran for 75.18 s
    [51 tests, 79.47 s]
==========================================

This seems to be sporadic: running
./sage -t --long --warn-long 54.1 src/sage/symbolic/series.pyx
yields

==========================================
Running doctests with ID 2016-12-12-15-13-48-d0fd7083.
Git branch: develop
Using --optional=mpir,python2,sage
Doctesting 1 file.
sage -t --long --warn-long 54.1 src/sage/symbolic/series.pyx
    [51 tests, 0.32 s]
----------------------------------------------------------------------
All tests passed!
----------------------------------------------------------------------
Total time for all tests: 0.4 seconds
    cpu time: 0.3 seconds
    cumulative wall time: 0.3 seconds
==========================================

Running it again gave the warning, then no warning, then the same warning again...

Eric.

[Daniel Krenn]

On 2016-12-12 15:22, Eric Gourgoulhon wrote:
> On Ubuntu 16.04 x86_64 Xeon E5-2623 + 16 GB RAM, after a fresh git clone
> + pull develop, make -j8 failled with the following error:

I get the same error on Linux Mint 17.3 (which is similar to Ubuntu). A
fresh git clone as well, compiled with
MAKE='make -j4' make

[Sébastien Labbé]

Also, I get a strange log file for sqlite which contains a little bit more than the SPKG.txt file :

$ cat logs/pkgs/sqlite.log

I found out that this junk file is created by sage -t src/sage/tests/cmdline.py, I created:

https://trac.sagemath.org/ticket/22062

[Justin C. Walker]

> On Dec 9, 2016, at 16:16 , 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 from tarsal successfully (macOS 10.11.6, with recent update of command-line tools), but the documentation build quits with only one line in dochtml.log:

/Users/Sage/sage-7.5.beta6/local/bin/python: No module named interpreters.wrapper_rdf; 'sage_setup.docbuild' is a package and cannot be directly executed

Eek!

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

[Emmanuel Charpentier]

I've seen analogous messages in about two out of three makes. The cuse is, as always "make doc-clean && make", which eats more and more time, now that 3D plots adorn our online help...

HTH,

--
Emmanuel Charpentier

[Jeroen Demeyer]

On 2016-12-20 06:34, Justin C. Walker wrote:
> /Users/Sage/sage-7.5.beta6/local/bin/python: No module named interpreters.wrapper_rdf; 'sage_setup.docbuild' is a package and cannot be directly executed

If you know how to reproduce this: please report at
https://trac.sagemath.org/ticket/22094

[Jeroen Demeyer]

On 2016-12-12 15:22, Eric Gourgoulhon wrote:

> [dochtml] /home/eric/sage/7.5.beta6/local/bin/python: No module named
> interpreters.wrapper_rdf; 'sage_setup.docbuild' is a package and cannot
> be directly executed