upgrade normaliz to 3.7.1, pynormaliz to 2.1 and package e-antic

[videlec]

New release of Normaliz is available.

It depends on an additional library e-antic that is packaged in this ticket.

Tarballs

• https://github.com/Normaliz/Normaliz/releases/download/v3.7.1/normaliz-3.7.1.tar.gz
• http://www.labri.fr/perso/vdelecro/e-antic/e-antic-0.1.3b0.tar.gz
• https://files.pythonhosted.org/packages/69/74/a32187f49a2524978d9bc6f704049d8021b555abe3b71aa74b687a958217/PyNormaliz-2.1.tar.gz

follow up tickets: #25091, #25097

(see also the meta-ticket #22420)

---

Step by step instructions if you want to try it out

1. Go to the Sage source tree and pull the branch associated to this ticket

2. Get the e-antic and normaliz tarballs and put them in the upstream repository

$ wget -P upstream https://github.com/Normaliz/Normaliz/releases/download/v3.7.1/normaliz-3.7.1.tar.gz
$ wget -P upstream http://www.labri.fr/perso/vdelecro/e-antic/e-antic-0.1.3b0.tar.gz
$ wget -P upstream https://files.pythonhosted.org/packages/69/74/a32187f49a2524978d9bc6f704049d8021b555abe3b71aa74b687a958217/PyNormaliz-2.1.tar.gz

3. Compile and install everything

$ sage -i pynormaliz

4. Run make

$ make build

CC: @w-bruns @jplab @sebasguts @mkoeppe

Component: packages: optional

Author: Vincent Delecroix, Matthias Koeppe

Branch/Commit: 64f1f19

Reviewer: Jean-Philippe Labbé, Travis Scrimshaw

Issue created by migration from https://trac.sagemath.org/ticket/27682

[videlec]

Branch: u/vdelecroix/27682

[videlec]

Commit: 6e96a5a

[sagetrac-git]

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

7618ea9

package e-antic

[sagetrac-git]

Changed commit from 6e96a5a to 7618ea9

[sagetrac-git]

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

d752708

upgrade normaliz

[sagetrac-git]

Changed commit from 7618ea9 to d752708

[jplab]

comment:7

So far, both packages install without issues on debian stretch with gcc 6.3.0. Waiting for the tarball for the new version of pynormaliz.

[sagetrac-git]

Changed commit from d752708 to 4eb02f8

[sagetrac-git]

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

4eb02f8

fix doctest in backend_normaliz.py

[jplab]

comment:12

After a make distclean and proper reinstall as the ticket description, I got everything back in order and all tests pass in the folder polyhedron.

[videlec]

comment:13

Replying to @jplab:

After a make distclean and proper reinstall as the ticket description, I got everything back in order and all tests pass in the folder polyhedron.

Then could you try to reproduce #22, #23, #26, #27? (and report back on the github issues)

[jplab]

comment:14

Replying to @videlec:

Replying to @jplab:

After a make distclean and proper reinstall as the ticket description, I got everything back in order and all tests pass in the folder polyhedron.

Then could you try to reproduce #22, #23, #26, #27? (and report back on the github issues)

Just to confirm: all 4 issues are now solved in sage (some of them had different traces inside of sage...).

and

$ sage -t *.py --optional=dochtml,e_antic,gfortran,memlimit,mpir,normaliz,python2,sage,pynormaliz
Using --optional=dochtml,e_antic,gfortran,memlimit,mpir,normaliz,pynormaliz,python2,sage
Doctesting 25 files.
----------------------------------------------------------------------
All tests passed!
----------------------------------------------------------------------

[sagetrac-git]

Changed commit from 4eb02f8 to e5c669d

[sagetrac-git]

Branch pushed to git repo; I updated commit sha1. This was a forced push. New commits:

eaa4584	package e-antic 0.1.3b0
e5c669d	upgrade normaliz

[mkoeppe]

comment:18

Please no force pushs... I'm merging this ticket in #25097...

[videlec]

comment:19

sorry... won't do anymore.

[videlec]

comment:20

On the other hand, it does not make much sense to merge anywhere. Normaliz and e-antic needs upgrade and PyNormaliz needs to be fixed and released.

[mkoeppe]

comment:21

Yes, of course I'm waiting for these upgrades.

[mkoeppe]

Changed branch from u/vdelecroix/27682 to u/mkoeppe/27682

[mkoeppe]

Changed author from Vincent Delecroix to Vincent Delecroix, Matthias Koeppe

[mkoeppe]

comment:26

Updated. Builds and tests OK on macOS Mojave

---

New commits:

14d6c83	Update normaliz to 3.7.1, pynormaliz to 2.1
64f1f19	fix doctest in backend_normaliz.py

[mkoeppe]

Changed commit from e5c669d to 64f1f19

[jplab]

comment:27

Builds and tests (polyhedron and thematic_tutorial folders) Oll Korrekt on debian stretch.

[tscrim]

comment:28

It passes all tests for me (in particular, in the folder I care about: combinat/rigged_configurations ;)). I am also getting around a 5-10% speedup in my tests:

sage: RC = RiggedConfigurations(['E',8,1], [[4,1]])
sage: %time len(RC.kleber_tree())
CPU times: user 2.29 s, sys: 8.24 ms, total: 2.3 s
Wall time: 352 ms
87
sage: RC = RiggedConfigurations(['E',8,1], [[4,1],[2,1]])
sage: %time len(RC.kleber_tree())
CPU times: user 53.4 s, sys: 52.1 ms, total: 53.5 s
Wall time: 7.68 s
2137
sage: RC = RiggedConfigurations(['F',4,1], [[3,2],[2,2]])
sage: %time len(RC.kleber_tree())
CPU times: user 9.18 s, sys: 20.3 ms, total: 9.2 s
Wall time: 1.85 s
1257

vs before

sage: RC = RiggedConfigurations(['E',8,1], [[4,1]])
sage: %time len(RC.kleber_tree())
CPU times: user 2.29 s, sys: 7.74 ms, total: 2.29 s
Wall time: 360 ms
87
sage: RC = RiggedConfigurations(['E',8,1], [[4,1],[2,1]])
sage: %time len(RC.kleber_tree())
CPU times: user 55.7 s, sys: 59.4 ms, total: 55.8 s
Wall time: 8.23 s
2137
sage: RC = RiggedConfigurations(['F',4,1], [[3,2],[2,2]])
sage: %time len(RC.kleber_tree())
CPU times: user 9.43 s, sys: 24 ms, total: 9.45 s
Wall time: 1.92 s
1257

[tscrim]

Reviewer: Jean-Philippe Labbé, Travis Scrimshaw

[vbraun]

Changed branch from u/mkoeppe/27682 to 64f1f19