Upgrade to pynac-0.7.13

Question

Upgrade to pynac-0.7.13

rwst opened this issue 8 years ago · 30 comments

rwst commented 8 years ago

There is:

fix ex.coefficients() bugs (#23545, #24147)
fix interface to Singular GCD (#23845)
fix compiler warnings
asin/atan enhancements (#24211)
set_of_all() function (#24176)
upgrade ax_cxx_compile_stdcxx_11.m4

https://github.com/pynac/pynac/releases/download/pynac-0.7.13/pynac-0.7.13.tar.bz2

CC: @kiwifb

Component: packages: standard

Author: Ralf Stephan

Branch/Commit: u/rws/upgrade_to_pynac_0_7_13 @ 438816e

Reviewer: Emmanuel Charpentier

Issue created by migration from https://trac.sagemath.org/ticket/24262

Answer 1 · 2017-11-21T16:36:20.000Z

Branch: u/rws/upgrade_to_pynac_0_7_13

Answer 2 · 2017-11-21T17:54:00.000Z

New commits:

`46a7ebc`	`24262: pkg version/chksum`

Answer 3 · 2017-11-21T17:54:00.000Z

Commit: 46a7ebc

Answer 4 · 2017-11-22T06:56:10.000Z

comment:3

No doctests need to be adapted for this release. Please review.

Answer 5 · 2017-11-22T22:11:10.000Z

comment:4

On Debian 8.1.rc2 + #24240, compiles without difficulty.

fix ex.coefficients() bugs (#23545, #24147) : yes
fix interface to Singular GCD (#23845) : yes. The planned doctest don't seem to have been added.
fix compiler warnings : I still get :

charpent@asus16-ec:~$ cat /usr/local/sage-8/logs/pkgs/pynac-0.7.13.log  | grep warn
/usr/local/sage-8/local/include/flint/fmpz.h:428:23: warning: declaration of 'mpz_ptr' shadows a global declaration [-Wshadow]
/usr/local/sage-8/local/include/flint/fmpz.h:428:23: warning: declaration of 'mpz_ptr' shadows a global declaration [-Wshadow]
/usr/local/sage-8/local/include/flint/fmpz.h:428:23: warning: declaration of 'mpz_ptr' shadows a global declaration [-Wshadow]

asin/atan enhancements (#24211) : partial fix :

sage: [arcsin(t) for t in [0, 1/2, sqrt(2)/2, sqrt(3)/2, 1]] # Fixed
[0, 1/6*pi, 1/4*pi, 1/3*pi, 1/2*pi]
sage: [arctan(t) for t in [0, 1/sqrt(3), 1, sqrt(3)]] # Fixed
[0, 1/6*pi, 1/4*pi, 1/3*pi]
sage: [arccos(t) for t in [0, 1/2, sqrt(2)/2, sqrt(3)/2, 1]] # Partially fixed
[1/2*pi, 1/3*pi, arccos(1/2*sqrt(2)), arccos(1/2*sqrt(3)), 0]
sage: [arccot(t) for t in [0, 1/sqrt(3), 1, sqrt(3)]] # Partially fixed
[1/2*pi, arccot(1/3*sqrt(3)), 1/4*pi, arccot(sqrt(3))]

The simplifications are not fixed either (but I'm not sure they could be by pynac's upgrade...) :

sage: assume(-pi/2<x, x<pi/2)
sage: arctan(sin(x)/cos(x)).simplify() # Not fixed
arctan(sin(x)/cos(x))
sage: arctan(sin(x)/cos(x)).simplify_full() # Not fixed
arctan(sin(x)/cos(x))

set_of_all() function (#24176) : can't test directly. Doctest ?
upgrade ax_cxx_compile_stdcxx_11.m4 : I dont't know how to test. Doctest ?
Your name needs to go in the Authors field.

make ptestlong gives me one (non-transient) failure :

charpent@asus16-ec:/usr/local/sage-8$ sage -t --long src/sage/rings/asymptotic/asymptotics_multivariate_generating_functions.py
too many failed tests, not using stored timings
Running doctests with ID 2017-11-22-23-04-04-f8ba9318.
Git branch: tst24240_n_24262
Using --optional=database_gap,dot2tex,fricas,giacpy_sage,igraph,mpir,python2,python_igraph,sage
Doctesting 1 file.
sage -t --long src/sage/rings/asymptotic/asymptotics_multivariate_generating_functions.py
**********************************************************************
File "src/sage/rings/asymptotic/asymptotics_multivariate_generating_functions.py", line 1652, in sage.rings.asymptotic.asymptotics_multivariate_generating_functions.FractionWithFactoredDenominator.?
Failed example:
    asy # long time
Expected:
    (4/3*sqrt(3)*sqrt(r)/sqrt(pi) + 47/216*sqrt(3)/(sqrt(pi)*sqrt(r)),
     1, 4/3*sqrt(3)*sqrt(r)/sqrt(pi) + 47/216*sqrt(3)/(sqrt(pi)*sqrt(r)))
Got:
    (4/3*sqrt(3)*sqrt(r)/sqrt(pi) + 191/216*sqrt(3)/(sqrt(pi)*sqrt(r)),
     1,
     4/3*sqrt(3)*sqrt(r)/sqrt(pi) + 191/216*sqrt(3)/(sqrt(pi)*sqrt(r)))
**********************************************************************
File "src/sage/rings/asymptotic/asymptotics_multivariate_generating_functions.py", line 1655, in sage.rings.asymptotic.asymptotics_multivariate_generating_functions.FractionWithFactoredDenominator.?
Failed example:
    F.relative_error(asy[0], alpha, [1, 2, 4, 8], asy[1]) # long time
Expected:
    [((3, 3, 2), 0.9812164307, [1.515572606], [-0.54458543...]),
     ((6, 6, 4), 1.576181132, [1.992989399], [-0.26444185...]),
     ((12, 12, 8), 2.485286378, [2.712196351], [-0.091301338...]),
     ((24, 24, 16), 3.700576827, [3.760447895], [-0.016178847...])]
Got:
    [((3, 3, 2), 0.9812164307, [2.167042622], [-1.208526635]),
     ((6, 6, 4), 1.576181132, [2.453648265], [-0.5567045026]),
     ((12, 12, 8), 2.485286378, [3.037931359], [-0.2223667203]),
     ((24, 24, 16), 3.700576827, [3.990777328], [-0.07842034220])]
**********************************************************************
1 item had failures:
   2 of 104 in sage.rings.asymptotic.asymptotics_multivariate_generating_functions.FractionWithFactoredDenominator.?
    [805 tests, 2 failures, 27.28 s]
----------------------------------------------------------------------
sage -t --long src/sage/rings/asymptotic/asymptotics_multivariate_generating_functions.py  # 2 doctests failed
----------------------------------------------------------------------
Total time for all tests: 27.4 seconds
    cpu time: 27.3 seconds
    cumulative wall time: 27.3 seconds

==>needs_work.

Answer 6 · 2017-11-22T22:11:10.000Z

Reviewer: Emmanuel Charpentier

Answer 7 · 2017-11-23T07:00:12.000Z

comment:5

Replying to @EmmanuelCharpentier:

fix interface to Singular GCD (#23845) : yes. The planned doctest don't seem to have been added.

They are always added in the reporting tickets.

fix compiler warnings : I still get :

charpent@asus16-ec:~$ cat /usr/local/sage-8/logs/pkgs/pynac-0.7.13.log  | grep warn
/usr/local/sage-8/local/include/flint/fmpz.h:428:23: warning: declaration of 'mpz_ptr' shadows a global declaration [-Wshadow]
/usr/local/sage-8/local/include/flint/fmpz.h:428:23: warning: declaration of 'mpz_ptr' shadows a global declaration [-Wshadow]
/usr/local/sage-8/local/include/flint/fmpz.h:428:23: warning: declaration of 'mpz_ptr' shadows a global declaration [-Wshadow]

They need to be fixed by updating Flint, as I have submitted the fix to Flint some time ago.

asin/atan enhancements (#24211) : partial fix :

sage: [arccot(t) for t in [0, 1/sqrt(3), 1, sqrt(3)]] # Partially fixed
[1/2*pi, arccot(1/3*sqrt(3)), 1/4*pi, arccot(sqrt(3))]

Sorry but you didn't report that. Of course I'll add it soon.

The simplifications are not fixed either (but I'm not sure they could be by pynac's upgrade...) :

I have pointed out already that simplify omissions are basically Maxima's.

set_of_all() function (#24176) : can't test directly. Doctest ?

Well, see #24176.

upgrade ax_cxx_compile_stdcxx_11.m4 : I dont't know how to test. Doctest ?

This is a Pynac build enhancement.

Your name needs to go in the Authors field.

Done.

make ptestlong gives me one (non-transient) failure :

What do you mean with "non-transient"? "non-transient using make ptestlong" or "non-transient using sage -t manually"? I cannot reproduce using sage -t on OpenSuSE, so this might be system-specific, or it might be make ptestlong-specific.

Answer 8 · 2017-11-23T07:01:07.000Z

Author: Ralf Stephan

Answer 9 · 2017-11-23T15:49:44.000Z

comment:7

Replying to @rwst:

Replying to @EmmanuelCharpentier:

[ Snip... ]}

asin/atan enhancements (#24211) : partial fix :
sage: [arccot(t) for t in [0, 1/sqrt(3), 1, sqrt(3)]] # Partially fixed
[1/2*pi, arccot(1/3*sqrt(3)), 1/4*pi, arccot(sqrt(3))]
Sorry but you didn't report that. Of course I'll add it soon.

I just did that.

Answer 10 · 2017-11-23T15:56:00.000Z

comment:8

Replying to @rwst:

Replying to @EmmanuelCharpentier:

[ Snip... ]

make ptestlong gives me one (non-transient) failure :

What do you mean with "non-transient"? "non-transient using make ptestlong" or "non-transient using sage -t manually"? I cannot reproduce using sage -t on OpenSuSE, so this might be system-specific, or it might be make ptestlong-specific.

I get the failure during ptestlong, and I can reproduce at will by doing :

sage -t --long src/sage/rings/asymptotic/asymptotics_multivariate_generating_functions.py

Answer 11 · 2017-11-23T16:19:10.000Z

comment:9

Confirmed with --long.

Answer 12 · 2017-11-24T08:03:31.000Z

comment:10

Okay the reason is a regression in subs, minimal case: sage: (1+1/x).subs({x:1/x}) ... 1/x + 1. We try to add a patch here with doctest.

Answer 13 · 2017-11-24T08:15:57.000Z

comment:11

Replying to @rwst:

Okay the reason is a regression in subs, minimal case: sage: (1+1/x).subs({x:1/x}) ... 1/x + 1. We try to add a patch here with doctest.

Hmmm... This a catastrophic one : subs is a basic method. IMHO, this deserves its own ticket, marked as blocker.

Answer 14 · 2017-11-24T08:44:20.000Z

comment:12

A blocker is a ticket that reports/fixes an existing bug in Sage. This is a ticket that potentially introduces a severe bug, but with the right patch, it becomes a normal ticket. The sky does not fall down.

Answer 15 · 2017-11-24T09:33:36.000Z

comment:13

Replying to @rwst:

A blocker is a ticket that reports/fixes an existing bug in Sage. This is a ticket that potentially introduces a severe bug, but with the right patch, it becomes a normal ticket. The sky does not fall down.

Ah ! You meant that this regression was introduced by the upgrade to pynac ? That was not obvious (at least to me...) ; I thought that you have found a pre-existing regression. And that was deserving a blocker ticket.

Answer 16 · 2017-11-26T13:42:45.000Z

Changed commit from 46a7ebc to 1782f91

Answer 17 · 2017-11-26T13:42:45.000Z

Branch pushed to git repo; I updated commit sha1. New commits:

`10a108d`	`24262: last-minute fix`
`1782f91`	`24262: doctest last-minute-fix`

Answer 18 · 2017-11-26T13:48:17.000Z

comment:15

This fixes the issue. My ptestlong passes but you might want to test again too.

Note that the behaviour of this added doctest:

sage: (sqrt(x) + 1/sqrt(x)).subs({x: 1/x})
1/sqrt(x) + 1/1/sqrt(x)

is unchanged with respect to previous Sage versions but I have opened a Pynac issue: pynac/pynac#287

Answer 19 · 2017-11-26T19:32:01.000Z

comment:16

Doesn't smell good.. On top of 8.0.rc3+#24240, I get :

charpent@asus16-ec:/usr/local/sage-8$ sage -t --long src/sage/rings/asymptotic/asymptotics_multivariate_generating_functions.py
too many failed tests, not using stored timings
Running doctests with ID 2017-11-26-20-26-24-fdb78fa1.
Git branch: tst24240_n_24262
Using --optional=database_gap,dot2tex,fricas,giacpy_sage,igraph,mpir,python2,python_igraph,sage
Doctesting 1 file.
sage -t --long src/sage/rings/asymptotic/asymptotics_multivariate_generating_functions.py
**********************************************************************
File "src/sage/rings/asymptotic/asymptotics_multivariate_generating_functions.py", line 1652, in sage.rings.asymptotic.asymptotics_multivariate_generating_functions.FractionWithFactoredDenominator.?
Failed example:
    asy # long time
Expected:
    (4/3*sqrt(3)*sqrt(r)/sqrt(pi) + 47/216*sqrt(3)/(sqrt(pi)*sqrt(r)),
     1, 4/3*sqrt(3)*sqrt(r)/sqrt(pi) + 47/216*sqrt(3)/(sqrt(pi)*sqrt(r)))
Got:
    (4/3*sqrt(3)*sqrt(r)/sqrt(pi) + 191/216*sqrt(3)/(sqrt(pi)*sqrt(r)),
     1,
     4/3*sqrt(3)*sqrt(r)/sqrt(pi) + 191/216*sqrt(3)/(sqrt(pi)*sqrt(r)))
**********************************************************************
File "src/sage/rings/asymptotic/asymptotics_multivariate_generating_functions.py", line 1655, in sage.rings.asymptotic.asymptotics_multivariate_generating_functions.FractionWithFactoredDenominator.?
Failed example:
    F.relative_error(asy[0], alpha, [1, 2, 4, 8], asy[1]) # long time
Expected:
    [((3, 3, 2), 0.9812164307, [1.515572606], [-0.54458543...]),
     ((6, 6, 4), 1.576181132, [1.992989399], [-0.26444185...]),
     ((12, 12, 8), 2.485286378, [2.712196351], [-0.091301338...]),
     ((24, 24, 16), 3.700576827, [3.760447895], [-0.016178847...])]
Got:
    [((3, 3, 2), 0.9812164307, [2.167042622], [-1.208526635]),
     ((6, 6, 4), 1.576181132, [2.453648265], [-0.5567045026]),
     ((12, 12, 8), 2.485286378, [3.037931359], [-0.2223667203]),
     ((24, 24, 16), 3.700576827, [3.990777328], [-0.07842034220])]
**********************************************************************
1 item had failures:
   2 of 104 in sage.rings.asymptotic.asymptotics_multivariate_generating_functions.FractionWithFactoredDenominator.?
    [805 tests, 2 failures, 27.18 s]
----------------------------------------------------------------------
sage -t --long src/sage/rings/asymptotic/asymptotics_multivariate_generating_functions.py  # 2 doctests failed
----------------------------------------------------------------------
Total time for all tests: 27.3 seconds
    cpu time: 27.2 seconds
    cumulative wall time: 27.2 seconds

These errors don't seem to be cosmetic (i. e. space, printing order, etc...).

ptestlong underway to find other snags... In the interim, needs_work

Sorry...

Answer 20 · 2017-11-26T21:05:53.000Z

comment:17

Replying to @EmmanuelCharpentier:

Doesn't smell good.. On top of 8.0.rc3+#24240, I get :

[ Snip... ]

ptestlong gave another error, which is reproducible standalone :

sage -t --long src/sage/symbolic/expression.pyx
**********************************************************************
File "src/sage/symbolic/expression.pyx", line 5153, in sage.symbolic.expression.
Expression.substitute
Failed example:
    (1 + 1/x).subs({x: 1/x})
Expected:
    x + 1
Got:
    1/x + 1
**********************************************************************
1 item had failures:
   1 of  68 in sage.symbolic.expression.Expression.substitute
    [2776 tests, 1 failure, 85.34 s]

Definitely needs_work.

Answer 21 · 2017-11-26T21:40:33.000Z

comment:18

It doesn't look like you updated your copy of the ticket branch, e.g.. git trac pull.24262.

Answer 22 · 2017-11-26T22:10:01.000Z

comment:19

Replying to @rwst:

It doesn't look like you updated your copy of the ticket branch, e.g.. git trac pull.24262.

From the top of my git log :

commit 7d9462cc8b91bc25f35559a17b2fce8a4efc8d5e
Merge: e885a67024 1782f910de
Author: Emmanuel Charpentier <emm.charpentier@free.fr>
Date:   Sun Nov 26 20:18:24 2017 +0100

    Merge branch 'u/rws/upgrade_to_pynac_0_7_13' of trac.sagemath.org:sage into tst24240_n_24262

commit 1782f910ded326252088dcd2c8169e0ffc471653
Author: Ralf Stephan <ralf@ark.in-berlin.de>
Date:   Sun Nov 26 08:49:40 2017 +0100

    24262: doctest last-minute-fix

commit 10a108dc41f6400dca0215e17cc5b57cb0d81a53
Author: Ralf Stephan <ralf@ark.in-berlin.de>
Date:   Sun Nov 26 08:48:53 2017 +0100

    24262: last-minute fix

commit e885a6702409e1bc1bdc1f5988057f2856655568
Merge: 7cde1424f3 46a7ebcbaa
Author: Emmanuel Charpentier <emm.charpentier@free.fr>
Date:   Wed Nov 22 21:58:13 2017 +0100

    Merge branch 'u/rws/upgrade_to_pynac_0_7_13' of trac.sagemath.org:sage into tst24240_n_24262

commit 46a7ebcbaa98ba7bcb61f1f5c8698489b0c9b9ca
Author: Ralf Stephan <ralf@ark.in-berlin.de>
Date:   Tue Nov 21 17:35:03 2017 +0100

    24262: pkg version/chksum

commit 7cde1424f39327192dfd72e7ca9ca0cb1db21895
Author: Emmanuel Charpentier <emm.charpentier@free.fr>
Date:   Sun Nov 19 18:46:59 2017 +0100

    24240 : Updated freetype to 2.8.1.

commit 92f95cef93000042532acbf8d84e4bede8e699ea
Author: Volker Braun <vbraun.name@gmail.com>
Date:   Fri Nov 17 14:21:52 2017 +0100

    Updated SageMath version to 8.1.rc2

commit 50a34e5f4378b9b4aba1ea972c57c4dd1e0eb8fb
Merge: e47ca9ea99 520fdab4f6
Author: Release Manager <release@sagemath.org>

I don't understand it, either...

Answer 23 · 2017-11-27T06:27:22.000Z

comment:20

The usual gotcha: I forgot to bump the patchlevel, so Pynac was not recompiled when you started make.

Answer 24 · 2017-11-27T06:29:39.000Z

Branch pushed to git repo; I updated commit sha1. New commits:

`438816e`	`24262: patchlevel`

Answer 25 · 2017-11-27T06:29:39.000Z

Changed commit from 1782f91 to 438816e

Answer 26 · 2017-11-27T08:29:54.000Z

comment:23

Smells better :

charpent@asus16-ec:/usr/local/sage-8$ sage -t --long src/sage/symbolic/expression.pyx
too many failed tests, not using stored timings
Running doctests with ID 2017-11-27-09-25-00-2de1e0c6.
Git branch: tst24240_n_24262
Using --optional=database_gap,dot2tex,fricas,giacpy_sage,igraph,mpir,python2,python_igraph,sage
Doctesting 1 file.
sage -t --long src/sage/symbolic/expression.pyx
    [2776 tests, 21.99 s]
----------------------------------------------------------------------
All tests passed!
----------------------------------------------------------------------
Total time for all tests: 22.4 seconds
    cpu time: 21.0 seconds
    cumulative wall time: 22.0 seconds
charpent@asus16-ec:/usr/local/sage-8$ sage -t --long src/sage/rings/asymptotic/asymptotics_multivariate_generating_functions.py
too many failed tests, not using stored timings
Running doctests with ID 2017-11-27-09-26-14-fa0e402c.
Git branch: tst24240_n_24262
Using --optional=database_gap,dot2tex,fricas,giacpy_sage,igraph,mpir,python2,python_igraph,sage
Doctesting 1 file.
sage -t --long src/sage/rings/asymptotic/asymptotics_multivariate_generating_functions.py
    [805 tests, 26.29 s]
----------------------------------------------------------------------
All tests passed!
----------------------------------------------------------------------
Total time for all tests: 26.4 seconds
    cpu time: 26.3 seconds
    cumulative wall time: 26.3 seconds

ptestlong underway...

Answer 27 · 2017-11-27T09:48:48.000Z

comment:24

On top of 8.0.rc3+#24240, passes ptestlong with no error whatsoever.

==>positive_review

Answer 28 · 2017-12-11T10:43:04.000Z

comment:25

Merge conflict

Answer 29 · 2017-12-11T13:51:45.000Z

comment:26

We might as well upgrade to 0.7.14 in one go. I'll prepare #24329 as soon as 8.2.beta1 is out.

Answer 30 · 2017-12-11T13:56:21.000Z

comment:27

Duplicate of #24329.