Sage 9.8 released

[Volker Braun]

The "master" git branch has been updated to Sage-9.8. 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

And as you are undoubtedly aware, the git repository is now hosted at https://github.com/sagemath/sage, so update your git remote.

To commemorate the move (and because we are getting really close to .10), I'm proposing to target Sage version 10 as the next release. 

The changes over 9.8.rc0 are mainly housekeeping for the new repository location. 

I'll try to finish up merging remaining positively reviewed trac issues, and then move on to github pull requests.
 

c000c953eb6 (HEAD -> develop, github/master, github/develop, tag: 9.8) Updated SageMath version to 9.8
104dde9eaa8 Merge pull request #35022 from tobiasdiez/pr-template-title
17d8a73296f Merge pull request #35018 from yyyyx4/public/35017
6a4667bb5c5 Merge pull request #34983 from tobiasdiez/add-issue-templates
66b319baffb Merge pull request #35011 from dimpase/develop
872b32b0c20 Merge pull request #34987 from mezzarobba/missing_long_time
d57804e6dec Merge pull request #34964 from yyyyx4/public/linter-20230206
698001b0e6a remove prohibition, promise details
0c5f7baf6c2 remove trac references, promise details soon
200557e00c3 (tag: 9.8.rc1) Updated SageMath version to 9.8.rc1

[Nasser M. Abbasi]

Hello;

When I go to  http://www.sagemath.org/download-latest.html  it says  sage-9.8.rc1.tar.gz as latest.  I tried few mirrors and I do not see 9.8

Is 9.8 supposed to be at one of these mirror sites to download?

thanks

--Nasser

[Dima Pasechnik]

On Sat, Feb 11, 2023 at 4:18 PM 'Nasser M. Abbasi' via sage-release <sage-r...@googlegroups.com> wrote:

Hello;

When I go to  http://www.sagemath.org/download-latest.html  it says  sage-9.8.rc1.tar.gz as latest.  I tried few mirrors and I do not see 9.8

Is 9.8 supposed to be at one of these mirror sites to download?

https://github.com/sagemath/sage/releases

(or just use git)

 

thanks

--Nasser

On Saturday, February 11, 2023 at 7:47:09 AM UTC-6 Volker Braun wrote:

The "master" git branch has been updated to Sage-9.8. 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

And as you are undoubtedly aware, the git repository is now hosted at https://github.com/sagemath/sage, so update your git remote.

To commemorate the move (and because we are getting really close to .10), I'm proposing to target Sage version 10 as the next release. 

The changes over 9.8.rc0 are mainly housekeeping for the new repository location. 

I'll try to finish up merging remaining positively reviewed trac issues, and then move on to github pull requests.
 

c000c953eb6 (HEAD -> develop, github/master, github/develop, tag: 9.8) Updated SageMath version to 9.8
104dde9eaa8 Merge pull request #35022 from tobiasdiez/pr-template-title
17d8a73296f Merge pull request #35018 from yyyyx4/public/35017
6a4667bb5c5 Merge pull request #34983 from tobiasdiez/add-issue-templates
66b319baffb Merge pull request #35011 from dimpase/develop
872b32b0c20 Merge pull request #34987 from mezzarobba/missing_long_time
d57804e6dec Merge pull request #34964 from yyyyx4/public/linter-20230206
698001b0e6a remove prohibition, promise details
0c5f7baf6c2 remove trac references, promise details soon
200557e00c3 (tag: 9.8.rc1) Updated SageMath version to 9.8.rc1

--
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/be983746-2404-4087-ae3c-525f395ad8e5n%40googlegroups.com.

[Volker Braun]

Apologies, the download link for the release is the "stable" version, not develop: https://www.sagemath.org/download-source.html

At least one mirror already has it: https://mirrors.mit.edu/sage/src/index.html

[François Bissey]

When will we have packages up on pypi? I now rely on these for stable
release in sage-on-gentoo.

François

> --
> 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

> <mailto:sage-release...@googlegroups.com>.

> To view this discussion on the web visit

> https://groups.google.com/d/msgid/sage-release/c9f7a210-5377-4cf1-a176-d693fce12c86n%40googlegroups.com <https://groups.google.com/d/msgid/sage-release/c9f7a210-5377-4cf1-a176-d693fce12c86n%40googlegroups.com?utm_medium=email&utm_source=footer>.

[Matthias Köppe]

Neither the "dist.yml" nor the "ci-..." workflows have run on the push to the 9.8 tag on sagemath/sage.

I'll investigate.

[Dima Pasechnik]

On Sat, Feb 11, 2023 at 10:31 PM Matthias Köppe <matthia...@gmail.com> wrote:

Neither the "dist.yml" nor the "ci-..." workflows have run on the push to the 9.8 tag on sagemath/sage.

I'll investigate.

I've also created the GitHub release 9.8 manually. (The tag was created, the release was not)

 

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/0386f230-5ea3-44b8-9a98-4e60b988cf1fn%40googlegroups.com.

[Matthias Köppe]

I've created a new PyPI deployment token and made it available to our repos.

I've started https://github.com/sagemath/sage/actions/workflows/dist.yml manually.

[Matthias Köppe]

The sdists are now up on PyPI.

[François Bissey]

Thanks!

> <https://github.com/sagemath/sage>, so update your git

> https://groups.google.com/d/msgid/sage-release/c9f7a210-5377-4cf1-a176-d693fce12c86n%40googlegroups.com <https://groups.google.com/d/msgid/sage-release/c9f7a210-5377-4cf1-a176-d693fce12c86n%40googlegroups.com> <https://groups.google.com/d/msgid/sage-release/c9f7a210-5377-4cf1-a176-d693fce12c86n%40googlegroups.com?utm_medium=email&utm_source=footer <https://groups.google.com/d/msgid/sage-release/c9f7a210-5377-4cf1-a176-d693fce12c86n%40googlegroups.com?utm_medium=email&utm_source=footer>>.

>
> --
> 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/0386f230-5ea3-44b8-9a98-4e60b988cf1fn%40googlegroups.com <https://groups.google.com/d/msgid/sage-release/0386f230-5ea3-44b8-9a98-4e60b988cf1fn%40googlegroups.com?utm_medium=email&utm_source=footer>.

>
> --
> 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
> <mailto:sage-release...@googlegroups.com>.
> To view this discussion on the web visit

> https://groups.google.com/d/msgid/sage-release/7d555461-a37d-47ac-a2a5-ecd12451b60cn%40googlegroups.com <https://groups.google.com/d/msgid/sage-release/7d555461-a37d-47ac-a2a5-ecd12451b60cn%40googlegroups.com?utm_medium=email&utm_source=footer>.

[Nasser M. Abbasi]

Thanks. I downloaded 9.8 and build it OK. No errors building it.

But unlike 9.8 rc1, I see that 9.8 is back to using system python which is 3.10.9 and not sagemath python 3.11.1 like 9.8 rc1 did, even though I used the same exact configuration commands. These are the commands I used

unset SAGE_ROOT
unset SAGE_LOCAL
./configure --with-system-python=no
make
make install


And now from a new terminal I type

>which python
/usr/bin/python

>python --version
Python 3.10.9

>sage --version
SageMath version 9.8, Release Date: 2023-02-11

>sage --python --version
Python 3.10.9

>sage
──────────────┐
│ SageMath version 9.8, Release Date: 2023-02-11                     │
│ Using Python 3.10.9. Type "help()" for help.                       │
└──────────────────────────────────────────────────────

Any idea why 9.8 did not use sagemath python which is supposed to be 3.11.1 like 9.8 rc1 did? This is all on same Linux virtual box

>lsb_release -a
LSB Version:    n/a
Distributor ID:    ManjaroLinux
Description:    Manjaro Linux
Release:    22.0.2
Codename:    Sikaris

>uname -r
6.1.9-1-MANJARO

I put the config.log and the build.log in this [folder](https://12000.org/tmp/sagemath_98_logs/)

[Thierry Dumont]

Restarting from a fresh clone of 9.8 (from github):

./configure --without-system-singular (is it necessary now? it was with
9.8rc*)

I got a problem building the documentation, which seems to be a segfault.
Logfile is joined.

But the compiled sage seems to run correctly.

I got the same problem when trying to upgrade from the last 9.8rc to 9.8.

My Debian testing is uptodate.

t.d.

[Emmanuel Charpentier]

On Debian testing running on core i7 + 16 GB RAM, upgrading `9.8.rc1` to `9.8` and running `ptestlong` results in three permanent failures :

| Doctest                                | Result            |
|----------------------------------------|-------------------|
| src/sage/interfaces/fricas.py          | 3 doctests failed |
| src/sage/graphs/generators/families.py | 2 doctests failed |
| src/sage/graphs/graph_generators.py    | 1 doctest failed  |

All these failures were already reported for previous betas.

HTH,

[Dima Pasechnik]

On Sun, Feb 12, 2023 at 9:10 AM 'Nasser M. Abbasi' via sage-release <sage-r...@googlegroups.com> wrote:

Thanks. I downloaded 9.8 and build it OK. No errors building it.

But unlike 9.8 rc1, I see that 9.8 is back to using system python which is 3.10.9 and not sagemath python 3.11.1 like 9.8 rc1 did, even though I used the same exact configuration commands.

there is absolutely nothing in the diff between 9.8.rc1 and 9.8 that would explain this.

So it's the environment you built that was different - I don't know exactly how.

But --with-system-python3 appears to be a noop. I could not find a place where it is processed.

(it should be in build/pkgs/python3/spkg-configure.m4, IMHO)

I opened https://github.com/sagemath/sage/issues/35089

So it was a change in the build enviroment that made Sage accept python3

By the way, I noticed a bug in Sage along the way.

From your config.log (and install) I see that you don't have pkg-config installed.

It's actually a sort of a bug in configure; your system is similar to Arch linux, and

the system package pkg-config is needed on Arch (and on Manjaro,  I suppose)

for the build to function correctly.

If you ran ./bootstrap (something should be a mandatory thing to do on Linux for installing from source,

but let's not get into this here)

then it would force you to install pkg-config, as it is mentioned in

build/pkgs/_bootstrap/distros/arch.txt

- but you didn't, and as build/pkgs/_prereq/distros/arch.txt does not mention pkg-config,

./configure didn't complain 

I've opened https://github.com/sagemath/sage/issues/35088 for the latter problem.

 

--

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/75d380a7-ded5-4925-aa9b-88a00752d69an%40googlegroups.com.

[Dima Pasechnik]

On Sun, Feb 12, 2023 at 9:19 AM Thierry Dumont <tdu...@math.univ-lyon1.fr> wrote:

Restarting from a fresh clone of 9.8 (from github):

./configure --without-system-singular (is it necessary now? it was with
9.8rc*)

I got a problem building the documentation, which seems to be a segfault.
Logfile is joined.

But the compiled sage seems to run correctly.

do doctests involving Singular pass?

E.g.

./sage -tp src/sage/libs/singular/

?

 

I got the same problem when trying to upgrade from the last 9.8rc to 9.8.

My Debian testing is uptodate.

t.d.

--
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/fd45ed00-7b8c-0a23-957e-571c607315e8%40math.univ-lyon1.fr.

[Thierry Dumont]

Le 12/02/2023 à 13:33, Dima Pasechnik a écrit :
> /sage -tp src/sage/libs/singular/

Ok.... this gives a segfault and may be other problems.
(file joined)
Thanks a lot.

[Dima Pasechnik]

it would be interesting to have a look at logs/pkgs/singular*.log

--
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/52337dc1-6e43-c372-03be-4dd4140d6c5e%40math.univ-lyon1.fr.

[Matthias Köppe]

From config.log, one can see that you typed "
"./configure --with-system-python=no"
but the correct option is "--with-system-python3=no".

[Nasser M. Abbasi]

Thanks. That was it. I did not notice the `3` was missing when I typed it. Now it works:

>sage
│ SageMath version 9.8, Release Date: 2023-02-11                 

│ Using Python 3.11.1. Type "help()" for help.                    

I find it very strange that sagemath ./configure will accept an invalid or unknown option. This should have been an error.

If it had complained about it, then the user will know they have used a wrong option immediately.  

This would save time and effort in the future if this happens again.

--Nasser

[Dima Pasechnik]

On Sun, Feb 12, 2023 at 5:04 PM 'Nasser M. Abbasi' via sage-release <sage-r...@googlegroups.com> wrote:

Thanks. That was it. I did not notice the `3` was missing when I typed it. Now it works:

>sage
│ SageMath version 9.8, Release Date: 2023-02-11                 
│ Using Python 3.11.1. Type "help()" for help.                    

I find it very strange that sagemath ./configure will accept an invalid or unknown option. This should have been an error.

If it had complained about it, then the user will know they have used a wrong option immediately.

it did complain, at the end of ./configure run:

....

config.status:4085: creating convenience symlink prefix -> local
config.status:4100: creating convenience symlink venv -> local/var/lib/sage/venv-python3.10
configure:120940: WARNING: unrecognized options: --with-system-python

configure:

...

 

--

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/a46714d7-3b4c-4340-801f-87977e7e20fen%40googlegroups.com.

[Nasser M. Abbasi]

I see. But I think it would be much better to stop the script immediately and give an error message at the terminal, may even be with a loud beep as well to alert the user.

One can still overlook these warnings if they are in the log file.

But now I know, I will make sure to check the log file more carefully in the future. But I really think the script should have generated an error for this and stopped. But this might be a design issue. 

Thanks

--Nasser

[Matthias Köppe]

It's also displayed in the terminal (not just the log file), just before the other messages that are there for the user to read.

[Marc Culler]

I am trying to build the binary release of Sage 9.8 for macOS and I have run into a mysterious problem.  I wonder if anyone on this list has any insight into it.

There are really two separate binary releases: one for Intel and one for Apple Silicon.  These are built using the same build script but on two separate machines whose native architecture matches the target.  That is, there is no cross-compilation involved.

This is what I see on Intel:

sage: %gap LoadPackage("semigroups");
true

This is what I see on Apple M1:

sage: %gap LoadPackage("semigroups");
** Gap crashed or quit executing 'LoadPackage("semigroups");' **
Restarting Gap and trying again
** Gap crashed or quit executing 'LoadPackage("semigroups");' **
Restarting Gap and trying again
** Gap crashed or quit executing 'LoadPackage("semigroups");' **
Restarting Gap and trying again
^C---------------------------------------------------------------------------

(You have to kill Sage or it will keep restarting Gap forever.)

Does anyone have any ideas?

- Marc

On Saturday, February 11, 2023 at 7:47:09 AM UTC-6 Volker Braun wrote:

[Marc Culler]

By running the load command in sage -gap I can see that GAP is crashing with an illegal instruction error while trying to load the semigroups package.

- Marc

[Dima Pasechnik]

does

     sage: libgap.LoadPackage("semigroup")

crash, as well?

and how about the same at GAP prompt:

./sage --gap

..

   gap> LoadPackage("semigroup");

--

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/be0b9508-3a85-4219-ad3f-4fe8cd0bbb8en%40googlegroups.com.

[Dima Pasechnik]

One reason might be is that this version of the package is just too old.

It was released long before all those macOS updates, and it is quite heavy C++ stuff.

You might get better luck with using positively reviewed update with GAP 4.12.2

[Marc Culler]

Thanks for the suggestions!

I have found the problem, although I cannot begin to explain it.  Many of the libraries in sage/local targeted x86_64 even though they were built on an M1 system.  Executables and libraries in the venv targeted the correct arm64 architecture.  I can only guess that there was some very weird environment in place when I did the build.  Anyway, rebuilding after make distclean seems to be working correctly.  So this thread can now be ignored.  Thanks.

- Marc