Make sage-conf optional #37138

tobiasdiez · 2024-01-22T13:26:06Z

As noted in #37024 and in #36489, sage-conf is actually not needed on a few systems now. For this reason, we make the installation of it optional (as there are no optional install requires as far as I know, this means we remove it from pyproject.toml and setup.cfg). We also remove it from "install all dependencies via conda" as its not needed there.

📝 Checklist

The title is concise, informative, and self-explanatory.
The description explains in detail what this PR is about.
I have linked a relevant issue or discussion.
I have created tests covering the changes.
I have updated the documentation accordingly.

⌛ Dependencies

Revert PR with disputed dependencies #37796

tobiasdiez

This is ready for review.

@orlitzky @tornaria another PR you might be interested in it.

mkoeppe · 2024-01-22T17:01:51Z

That sage-conf is a required dependency at build and runtime is a crucial part of the design. It supports the PyPI distribution strategy. This is all explained in the README of sage-conf.

tornaria · 2024-01-22T17:27:23Z

That sage-conf is a required dependency at build and runtime is a crucial part of the design. It supports the PyPI distribution strategy. This is all explained in the README of sage-conf.

No, it's not. python -m build -nxw on unpatched sagemath-standard works just fine, thanks.

mkoeppe · 2024-01-22T17:32:59Z

I'll rephrase. That sage-conf is declared a required dependency at build and runtime is a crucial part of the design.

tornaria · 2024-01-22T17:56:31Z

I'll rephrase. That sage-conf is declared a required dependency at build and runtime is a crucial part of the design.

It's incorrectly declared a required dependency.

mkoeppe · 2024-01-22T18:07:04Z

It is not meaningful to declare that it is "correct" or "incorrect". We declare this particular dependency as part of a design that supports multiple use cases, not just yours. For installing directly from PyPI, it is crucial that sage-conf is pulled in at build and run time; again, see the README.

tornaria · 2024-01-22T19:00:22Z

It is not meaningful to declare that it is "correct" or "incorrect". We declare this particular dependency as part of a design that supports multiple use cases, not just yours. For installing directly from PyPI, it is crucial that sage-conf is pulled in at build and run time; again, see the README.

I install directly from pypi. I don't claim you have to not-install sage_conf, you are still free to do that. Of course I'm still free to use -x and falsify the assertion that sage_conf is a requirement to build or run sagemath.

Maybe there should be a separate distribution on pypi that gets the kitchen sink.

mkoeppe · 2024-01-22T19:44:08Z

falsify the assertion that sage_conf is a requirement

Once again, the declaration is there by design; it is not an "assertion", so it cannot be "falsified".

tornaria · 2024-01-22T20:24:49Z

falsify the assertion that sage_conf is a requirement

Once again, the declaration is there by design; it is not an "assertion", so it cannot be "falsified".

🤷

tornaria · 2024-01-22T21:11:46Z

falsify the assertion that sage_conf is a requirement to build or run sagemath.

Just for the record

$ du -h sage*.tar.gz
85K     sage-setup-10.3b6.tar.gz
19M     sagemath-standard-10.3b6.tar.gz
$ tar xf sagemath-standard-10.3b6.tar.gz 
$ tar -C sagemath-standard-10.3b6 -xf sage-setup-10.3b6.tar.gz sage-setup-10.3b6/sage_setup/ --strip-components=1
$ time SAGE_NUM_THREADS=32 python -m build -nxw sagemath-standard-10.3b6 -o dist 2>&1 | tail -1
Successfully built sagemath_standard-10.3b6-cp312-cp312-linux_x86_64.whl

real    1m25.147s
user    9m40.425s
sys     0m53.292s
$ du -h dist/sagemath_standard-10.3b6-cp312-cp312-linux_x86_64.whl 
307M    dist/sagemath_standard-10.3b6-cp312-cp312-linux_x86_64.whl

This wheel works perfectly fine if installed in /usr. To work from a venv you need to manually set SAGE_LOCAL=/usr because it defaults to SAGE_VENV (that should be improved):

$ python -m venv /tmp/sage-from-pypi --system-site-packages --without-pip
$ source /tmp/sage-from-pypi/bin/activate
(sage-from-pypi) $ time python -m installer dist/sagemath_standard-10.3b6-cp312-cp312-linux_x86_64.whl

real    0m20.339s
user    0m19.484s
sys     0m0.769s
(sage-from-pypi) $ python -m installer dist/conway_polynomials-0.9-py3-none-any.whl
(sage-from-pypi) $ SAGE_LOCAL=/usr sage -tp 32 --all
...
----------------------------------------------------------------------
All tests passed!
----------------------------------------------------------------------
Total time for all tests: 212.2 seconds
    cpu time: 5378.1 seconds
    cumulative wall time: 5103.6 seconds
Features detected for doctesting: conway_polynomials,cvxopt,database_ellcurves,database_graphs,fpylll,gap_package_atlasrep,gap_package_polycyclic,gfan,imagemagick,lrcalc_python,mpmath,nauty,networkx,numpy,palp,pandoc,pexpect,pillow,polytopes_db,pplpy,primecountpy,ptyprocess,pyparsing,sage.combinat,sage.geometry.polyhedron,sage.graphs,sage.groups,sage.libs.braiding,sage.libs.ecl,sage.libs.flint,sage.libs.gap,sage.libs.linbox,sage.libs.m4ri,sage.libs.ntl,sage.libs.pari,sage.libs.singular,sage.misc.cython,sage.modular,sage.modules,sage.numerical.mip,sage.plot,sage.rings.complex_double,sage.rings.finite_rings,sage.rings.function_field,sage.rings.number_field,sage.rings.padics,sage.rings.polynomial.pbori,sage.rings.real_double,sage.rings.real_mpfr,sage.schemes,sage.symbolic,scipy,singular,sympy,threejs

Appart from installing conway_polynomials-0.9 by hand in the venv, I didn't have to prepare the system in any way, of course I already had all the dependencies installed, all from distro packages.

Note that a lot of mess can happen if you have a system sage installed because .../site-packages/sage/__init__.py is missing, so it's a good idea to touch /tmp/sage-from-pypi/lib/python3.12/site-packages/sage/__init__.py.

IMNSHO, sagemath-standard distribution should include sage/__init__.py, and also sage_setup.

mkoeppe · 2024-01-22T22:04:45Z

To work from a venv you need to manually set SAGE_LOCAL=/usr

Do you know what functionality still depends on SAGE_LOCAL?

tornaria · 2024-01-22T22:31:50Z

To work from a venv you need to manually set SAGE_LOCAL=/usr

Do you know what functionality still depends on SAGE_LOCAL?

It's mainly used as PREFIX, so lots of things: gap search paths, database search paths, jmol, threejs, mathjax. Mostly via SAGE_SHARE=SAGE_LOCAL/share but also SAGE_LOCAL/lib (for SAGE_LOCAL/lib/gap/sysinfo.gap).

A problem is that variables are not clearly defined, and most of them were created as part of sage-the-distro (when that was the only thing) and then used for sage-the-lib.

For example: is SAGE_SHARE the place where sage-the-distro install "share" stuff, or is it the place where sage-the-lib goes to search for "share" stuff. Maybe there is an installation where some database lives in /usr/share (used from system) but another database was installed by sage in its own SAGE_LOCAL. Or maybe the user installed it in DOT_SAGE/db, etc.

It seems maybe we need to think a little bit about search paths to make something coherent for different search paths that can accomodate different situations and it's easy to reason about.

Using sys.prefix seems the right way to go, except when I'm on a venv the sys.prefix corresponds to the venv, but I want to use stuff from the system prefix (/usr) at least as a fallback (kind of like --system-site-packages, but for other stuff).

mkoeppe · 2024-01-22T22:58:39Z

It seems maybe we need to think a little bit about search paths to make something coherent for different search paths that can accommodate different situations and it's easy to reason about.

+1

Using sys.prefix seems the right way to go, except when I'm on a venv the sys.prefix corresponds to the venv, but I want to use stuff from the system prefix (/usr) at least as a fallback (kind of like --system-site-packages, but for other stuff).

In a venv, of course there's also sys.base_prefix; but it's not clear to me that this would be the right thing to use at all.
For example, suppose I install a python in the prefix /opt/python3.12; then probably we do not expect to find stuff other than Python in that prefix. We would likely still want to find things in default search paths in prefixes such as /usr, /usr/local.

mkoeppe · 2024-01-22T23:03:08Z

It's mainly used as PREFIX, so lots of things: gap search paths, database search paths, jmol, threejs, mathjax. Mostly via SAGE_SHARE=SAGE_LOCAL/share but also SAGE_LOCAL/lib (for SAGE_LOCAL/lib/gap/sysinfo.gap).

A problem is that variables are not clearly defined, and most of them were created as part of sage-the-distro (when that was the only thing) and then used for sage-the-lib.

For example: is SAGE_SHARE the place where sage-the-distro install "share" stuff, or is it the place where sage-the-lib goes to search for "share" stuff. Maybe there is an installation where some database lives in /usr/share (used from system) but another database was installed by sage in its own SAGE_LOCAL. Or maybe the user installed it in DOT_SAGE/db, etc.

I agree that this needs a sharper definition. In my view, we should move toward a model where in a typical distro installation (also conda), SAGE_LOCAL (and likewise SAGE_SHARE as its subdirectory) is not defined. Rather, SAGE_LOCAL, when defined, would be exclusively something managed by sage-the-distro (or whatever subset incarnation of it).

mkoeppe · 2024-01-22T23:07:51Z

On removing this overloading of SAGE_LOCAL, see also my previous comment #36489 (comment)

tornaria · 2024-01-22T23:14:20Z

Also: it's NOT OK that sage-env is doing this: export PATH="$SAGE_VENV/bin:$PATH".

It's not ok, it's not ok, it's not ok.

Say SAGE_VENV=/usr, and PATH=~/bin:/usr/local/bin:/usr/bin, then you just broke my whole system (this has nothing to do with sage: it means that if I subshell from sage my environment is screwed up wrt what I expect, which is that binaries run from ~/bin first, then from /usr/local, then from /usr).

Broken, broken, broken.

I hack my way out of that by doing unset SAGE_VENV in sage-env-conf but I shouldn't have to (worse: nobody tells me to do that, so I learn the hard way).

mkoeppe · 2024-01-22T23:16:31Z

I know, that's #33625

tornaria · 2024-01-22T23:22:16Z

I agree that this needs a sharper definition. In my view, we should move toward a model where in a typical distro installation (also conda), SAGE_LOCAL (and likewise SAGE_SHARE as its subdirectory) is not defined. Rather, SAGE_LOCAL, when defined, would be exclusively something managed by sage-the-distro (or whatever subset incarnation of it).

This sounds reasonable. We still need some kind of PREFIX, and for some stuff this maybe needs to be a search list of possible prefixes, I don't know.

mkoeppe · 2024-01-22T23:27:24Z

yes, some SAGE_PREFIX_PATH or some such

orlitzky · 2024-01-23T00:20:28Z

yes, some SAGE_PREFIX_PATH or some such

Can we please not reinvent the wheel?

$ meson setup --help
usage: meson setup [-h] [--prefix PREFIX]

mkoeppe · 2024-01-23T01:05:26Z

Hm? My suggestion was for the search list of possible prefixes.

orlitzky · 2024-01-23T01:10:15Z

Hm? My suggestion was for the search list of possible prefixes.

You don't have to guess though, the user is going to tell you the one right answer when he builds sagelib.

Searching will always get the edge cases wrong. For example, if you're searching for libdir and if you try /usr/lib before /usr/lib64, then on multilib systems where /usr/lib contains 32-bit libraries, you'll pick the wrong one. If I want to use /home/mjo/usr instead of /usr, you aren't going to guess that, etc.

tornaria · 2024-01-23T01:25:39Z

yes, some SAGE_PREFIX_PATH or some such

Can we please not reinvent the wheel?
$ meson setup --help
usage: meson setup [-h] [--prefix PREFIX]

👍

But how is PREFIX used at runtime ? This seems ok as a build time prefix to search for build time stuff (pkgconfig, headers, libs, etc).

But it seems to me that python wheels are kind of "relocatable". To paraphrase you:

$ python -m installer -h
usage: python -m installer [-h] [--destdir path] [--prefix path] [--compile-bytecode level] [--no-compile-bytecode] wheel

That is the install --prefix (which defaults to sys.prefix). But you are not allowed to run meson or anything at install time. In any case, the issue here is what to do when sys.prefix is different from the prefix where other packages (gap, databases, etc) are installed.

The question is how much diversity we really need / want to handle in this regard.

orlitzky · 2024-01-23T03:31:45Z

But how is PREFIX used at runtime ? This seems ok as a build time prefix to search for build time stuff (pkgconfig, headers, libs, etc).

It's actually only used at install/runtime. At build time, you're searching the standard system locations or whatever pkg-config tells you. (And at install time, you're really installing to a tree under --destdir that includes the prefix.)

But it seems to me that python wheels are kind of "relocatable". To paraphrase you:
$ python -m installer -h
usage: python -m installer [-h] [--destdir path] [--prefix path] [--compile-bytecode level] [--no-compile-bytecode] wheel
That is the install --prefix (which defaults to sys.prefix). But you are not allowed to run meson or anything at install time. In any case, the issue here is what to do when sys.prefix is different from the prefix where other packages (gap, databases, etc) are installed.

The main thing that changed for us with meson-python is exactly that: there's now a real build system that can be invoked from python to handle each of the configuration, build, and install phases. When you pip install a meson-python package, it runs meson for you to configure, build, and install the thing. Though you can also run meson yourself, which is much simpler once you get used to it.

The "configure" phase of that would find, say, GAP, via $PATH at build time, regardless of the meson --prefix. That works because $PATH and/or -lgap are standard interfaces.

The database packages are trickier, because they're just data files with no build system and that supply no information about themselves via standard interfaces. That's why I packaged conway-polynomials and started on the MOLS data. The end goal is to make them available via some standard interface, such as

In python, via a module in the standard python search path
In a library, via -ldatabase
On the CLI, from a tool in PATH that can print the data
Or via a bare pkg-config file that points to the database

All of those require the databases to be separately packaged with their own build systems. But once they are, they can be found by meson setup (or pip install), completely independent of where sagelib is going to be installed.

In short, I would stop worrying about the databases because they're just going to mess up your mental model. There's only like three of them left that need to be packaged, and when they're done, it will be a lot easier to see how building sagelib should work.

mkoeppe · 2024-01-23T03:54:41Z

The main thing that changed for us with meson-python is exactly that: there's now a real build system that can be invoked from python to handle each of the configuration, build, and install phases.
When you pip install a meson-python package, it runs meson for you to configure, build, and install the thing.

meson-python only installs into a temporary directory and builds a wheel from it. The "prefix" has no meaning, it's just the temporary directory. The wheel is always installable in an arbitrary venv.

orlitzky · 2024-01-23T04:35:21Z

meson-python only installs into a temporary directory and builds a wheel from it. The "prefix" has no meaning, it's just the temporary directory. The wheel is always installable in an arbitrary venv.

it still invokes meson, though, which uses the --prefix to whatever extent you decide. In the context of dependency paths, I think the relevant cases are,

If it's not a database package, the sage --prefix is irrelevant, because they can all be found via standard interfaces
A few databases still can't be found automatically but that's easy to fix by packaging them with a build system

With (2) done, there's no need to coordinate any paths, and SAGE_SHARE, SAGE_LOCAL, etc. all become irrelevant. We can then use whatever --prefix, --datadir, etc. the user asks meson to use. And that's one less sage-ism to carry around.

mkoeppe · 2024-01-23T20:15:03Z

We can then use whatever --prefix, --datadir, etc. the user asks meson to use.

No, we can't. As I just explained, in meson-python, the prefix is temporary and has no meaning at the runtime of the package.

meson-python is just another PEP-517/660 build backend. If used right, it can have benefits over using setuptools, in particular regarding dependency tracking.

If you are excited about switching to meson because it would allow us to depend on the installation prefix and to do arbitrary things at install time... let me just point out that the packages that have led the development of meson-python (scipy, numpy) don't do that.

tobiasdiez · 2024-01-24T16:02:29Z

falsify the assertion that sage_conf is a requirement

Once again, the declaration is there by design; it is not an "assertion", so it cannot be "falsified".

Well, this is nothing for us to design. PEP518 specifies the meaning of the requires key as "This key must have a value of a list of strings representing PEP 508 dependencies required to execute the build system (currently that means what dependencies are required to execute a setup.py file)." And it's pretty clear that sage_conf is not required in this sense.

Moreover, nobody prevents you or anyone else to install sage_conf before installing sagelib/sagemath-standard.
I can make this more explicit in the docs if you want. Is the Readme under pkgs the correct place, or is there a better one?

kwankyu · 2024-03-29T04:15:45Z

@kwankyu: Precisely, the proposal is to create a separate package with a different name that will be a standalone package. Major goals from the onset:

all external dependencies would be provided by the environment and it's out of the scope of this package how those are installed,

other than that, everything else would be provided in a single self-contained package.

IOW, assuming all the required dependencies are installed in standard locations, it would only take a standard build/install step to have sagemath working on the system (identical to e.g. numpy, scipy, sympy, ipython, etc)

One way it could be implemented is to create it as a new subdirectory inside pkgs in the monorepo. This subdirectoory would only contain the build scripts and configuration, but no sagelib code; the latter would be taken from src/sage (just like most packages inside pkgs do.)

Gonzalo, do you understand the bullet points and agree? I am curious.

The proposal is not trying to solve "user packages", so the bullet points you refer to are not relevant. Note that we would use a different name to avoid any type of conflict with that. Supporting user packages is a worthwhile goal; having this separate standalone package is also a way to remove some pressure from the current packages so they can move faster towards that goal.

Thanks. So this new distribution package with whatever name xxx has zero chance to conflict with the current modularization effort. All would happen inside the directory pkgs/xxx. This distribution package would exist solely for the software sage deployment to linux platforms, or to environments like conda or possibly homebrew.

Matthias may regard it as a duplicate since in his opinion, sagemath-standard serves for the same purpose. But if some downstream packagers would prefer xxx to sagemath-standard for whatever reasons, we may keep both distribution packages side by side for users to choose. Matthias?

kwankyu · 2024-03-29T04:18:23Z

@culler: I use the shorthand "distro" to mean "software distribution" (as in "a distribution of software packages"). I try to stress the shorthand in an attempt to avoid the confusion with the python term.

The terms "distribution" and "package" are too much overloaded. We should qualify the terms when we use them.

Please do not use the terms for general concepts.

Sage-the-distro is special in this regard since it aims to work in many different environments which makes the task more difficult. Other distributions have it easier. ...

Please do not use the term "sage-the-distribution" as something equal to a linux distribution.

One of the packages available in some "distros" is a python package which in sage-the-distro is called sagelib (see build/pkgs/sagelib). ...

That is a confusing statement. Yes, there is build/pkgs/sagelib for the sage installation procedure. But we should not think sage-the-library as part of sage-the-distribution. The main purpose of sage-the-distribution is to provide external dependencies of sage-the-library.

As for the python term "distribution", it's a shortcut for "source distribution package" aka "sdist" which means an individual "python package".

For that meaning, let's use only the terms "source distribution" ("sdist" ) and "binary distribution" ("bdist").

Flat directory menans "build/pkgs" has 400+ subdirectories, one for each package. As opposed to a tree-like structure. This is not so important IMO, although handling a directory that is too large leads to some inconveniences (the void packages repo has a directory with 13000+ entries and this is really painful sometimes).

I don't think a tree structure would be useful for "build/pkgs".

dimpase · 2024-03-29T10:16:16Z

Flat directory menans "build/pkgs" has 400+ subdirectories, one for each package. As opposed to a tree-like structure. This is not so important IMO, although handling a directory that is too large leads to some inconveniences (the void packages repo has a directory with 13000+ entries and this is really painful sometimes).

well, mentioning how Void does things as an excuse is a typical whataboutism.
Void is a bit of a joke in this sense - all it seems to offer, if one wants a particular package, is keywords search facility. :-(
Our story is rather different, and Void or any other Linux distro are not an excuse for the status quo.

Unlike Void (or other Linux distros), majority of spkgs can't be installed at will
(only optional and experimental can). Also, there are only very few spkg which must be installed, and not substituted by an external analog, already installed on the system - it's sagelib, sage-conf, sagelib doctesting tools, and few other, bootstrap-related, things.
In this sense, the rest of standard spkgs is optional.
Let's have a closer look at these standard packages. The rough classification is as follows, and it IMHO should make its way into the docs.

standard package classification (sans sagelib+sage-conf+bootstrap)

python (an - outdated - Python source) and its toolchain - for building wheels etc.
compilers (C, C++, Fortran, Lisp, Cython); other build tools, such as patch
sagelib deps - what one needs to build and use sagelib - other than 1. and 2.
- build+runtime
- runtime-only
"Schroedinger" sagelib deps, such as rpy2 (they are optional by right, but not quite, and so marked at standard)
sage preparser+terminal IO - to support sage's "language" (^ instead of **, etc) + ipython
notebook+jupyter(lab) - that's (G)UI. One can build and use sagelib without it
sage's Jupyter kernel - needed for 6. to be used
sphinx & Co. - tools to build docs (they do use a small part of 6.)
... something else?

PS. Note that a chunk of 5 is actually in sagelib. (Same for 7.?)

From sagemath#37138 (comment)       ### 📝 Checklist  - [x] The title is concise and informative. - [ ] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### ⌛ Dependencies    URL: sagemath#37671 Reported by: Matthias Köppe Reviewer(s): François Bissey

tornaria · 2024-04-08T12:28:00Z

+1: @tobiasdiez @orlitzky @dimpase @tornaria
-1: @mkoeppe

Am I missing anyone?

kwankyu · 2024-04-08T13:28:11Z

src/pyproject.toml.m4

@@ -6,7 +6,6 @@ requires = [
    # Note that PEP517/518 have no notion of optional sage_spkg dependencies:
    # https://github.com/pypa/pip/issues/6144
     esyscmd(`sage-get-system-packages install-requires-toml \
-        sage_conf      \
        setuptools \


Line 5 says "Some version of sage-conf is required". But sage_conf is removed from line 9.

Then would pip install sagemath-standard work?

yes, it's still working as can be seen from the conda ci workflow.

Thanks for catching the outdated comment. I've removed it now.

commit 14a1e5c078d1684f751b9e7ef450f197e7bcf27f Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 23:01:35 2024 -0400 Update src/sage/combinat/symmetric_group_algebra.py Good idea. moved description from _dft_modular to dft docstring remove doctest with p=3, n=3 Update src/sage/combinat/symmetric_group_algebra.py Update src/sage/combinat/symmetric_group_algebra.py Co-Authored-By: Travis Scrimshaw <clfrngrown@aol.com> commit f3807cb141bae308906295f605939d4b2b205a5a Merge: 9ad2152942 8ea5214695 Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 20:12:53 2024 -0400 Merge branch 'develop' into pr/37748 commit 8ea5214695f742eccd34bd88600771b7121cdb9f Author: Release Manager <release@sagemath.org> Date: Tue Apr 9 00:29:19 2024 +0200 Updated SageMath version to 10.4.beta2 commit 9ad2152942819a55c20d80485ad8cd8925969c6c Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 13:38:43 2024 -0400 double backslash for math commit 619a48dea00a0a67b8d4e542948200dc23e9589a Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 13:33:00 2024 -0400 add examples commit 527ae2670dd3a37b25a192ab49dad7f4c16dd5aa Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 13:24:13 2024 -0400 form should default to None if form is None, check if characteristic of field divides order of group. if so, it's the modular case, and we return the projection onto idempotents change-of-basis, i.e. the MFT. if not, use semi-normal form. commit 488e931da94f4872047ff8e839edc4a0c99b163e Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 13:09:54 2024 -0400 missing parantheses commit 7e96c904b6bc47d2108930c039b95a14155d425b Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 13:00:57 2024 -0400 Update src/sage/combinat/symmetric_group_algebra.py Co-authored-by: Travis Scrimshaw <clfrngrown@aol.com> commit cf27c64dff0b77c6cca162849ec923ae14f4920f Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 13:00:14 2024 -0400 Update src/sage/combinat/symmetric_group_algebra.py I thought I could do this, don't know why I convinced myself I needed flatten. Co-authored-by: Travis Scrimshaw <clfrngrown@aol.com> commit 92699f42eba524047b09d527981dab2456802958 Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 12:59:16 2024 -0400 Update src/sage/combinat/symmetric_group_algebra.py Co-authored-by: Travis Scrimshaw <clfrngrown@aol.com> commit 68f2ef140f91356f55290d100ad8a5282f19c206 Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 12:58:30 2024 -0400 Update src/sage/combinat/symmetric_group_algebra.py Co-authored-by: Travis Scrimshaw <clfrngrown@aol.com> commit 618d5e1a8cb11006ef9fa6642e33f06d5728d6ae Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 12:57:18 2024 -0400 Update src/sage/combinat/symmetric_group_algebra.py Co-authored-by: Travis Scrimshaw <clfrngrown@aol.com> commit 04db911e3f6a08b3be1bf5c295e913a40ee8b28a Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 12:56:37 2024 -0400 Update src/sage/combinat/symmetric_group_algebra.py I see, so "None" if the user doesn't specify, and we determine which form based on whether p|n!. Sounds good. Co-authored-by: Travis Scrimshaw <clfrngrown@aol.com> commit e86f79410ebda01b8a85598a5754f9c383336ae1 Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 11:09:57 2024 -0400 add doctest to ensure issue 37751 is fixed commit f8c64d16b3d1c7b3d338d9d2339ae2e284df3c58 Merge: aedd2f3927 3079e2d897 Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 10:55:44 2024 -0400 Merge branch 'pr/37748' of https://github.com/jacksonwalters/sage into pr/37748 commit aedd2f3927aac05dceeb97bd484afa82557652a8 Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 10:55:43 2024 -0400 remove multiplication stars commit 3079e2d897396222b0f23784a4d07a40f25b3c1f Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 10:54:38 2024 -0400 Update src/sage/combinat/symmetric_group_algebra.py Co-authored-by: grhkm21 <83517584+grhkm21@users.noreply.github.com> commit b14565299a6f5a82e5fae5080de88b455ce696e5 Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 10:51:59 2024 -0400 docstring lines are wrapped at 80 characters ensure each line is less than 80 characters commit e3fcc1f59c4668f88c184649c18627480365e646 Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 10:24:51 2024 -0400 keep form argument keep form since there is a one year deprecation warning before removing arguments. show deprecation warning when default seminormal form is called, but suggest modular form which works in all cases commit 6d187b153b5b5c3e29fe07d9efe925e34bf5baa5 Merge: 551139c09f c8d260a6fe Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 23:34:58 2024 +0200 gh-37764: ECM-related tests fail after an incremental build ### TL;DR: This is because setting the default ECMBIN=ecm is only done when using the configure test, but not when skipping the configure test because the spkg is already installed. ### Steps to reproduce: * ecm is not installed in the base os * Sage make distclean && make succeeeds * Tests succeed * re-running ./bootstrap && ./configure && make works * Tests now fail with PermissionError: [Errno 13] Permission denied: '' ### Dependencies See also: * https://github.com/sagemath/sage/pull/37701 * https://github.com/sagemath/sage/pull/37011#issuecomment-2023089743 URL: https://github.com/sagemath/sage/pull/37764 Reported by: Volker Braun Reviewer(s): Matthias Köppe commit 71ad110e196d9f17934d030ce1c000e11b50fee8 Author: Jackson Walters <jacksonwalters@gmail.com> Date: Sun Apr 7 19:57:42 2024 -0400 trigger GitHub actions commit c8d260a6fe37cb0379690e57dab6fd2a3d5d8048 Author: Volker Braun <tarpit@vbraun.imap.cc> Date: Sun Apr 7 22:13:11 2024 +0200 Update build/pkgs/ecm/spkg-configure.m4 Co-authored-by: Matthias Köppe <mkoeppe@math.ucdavis.edu> commit fbf896681da3ad1fad6c164e4d5a7cf452ec2eec Author: Volker Braun <tarpit@vbraun.imap.cc> Date: Sun Apr 7 21:35:42 2024 +0200 Update build/pkgs/ecm/spkg-configure.m4 Co-authored-by: Matthias Köppe <mkoeppe@math.ucdavis.edu> commit 6531f7e415230d62fecd0a38fd804081341a9f32 Author: Volker Braun <tarpit@vbraun.imap.cc> Date: Sun Apr 7 21:35:38 2024 +0200 Update build/pkgs/ecm/spkg-configure.m4 Co-authored-by: Matthias Köppe <mkoeppe@math.ucdavis.edu> commit af0e24286a62a06d85e1baf49d0dc8f624965e5c Author: Volker Braun <vbraun.name@gmail.com> Date: Sun Apr 7 19:17:37 2024 +0200 ECM-related tests fail after an incremental build This is because setting the default ECMBIN=ecm is only done when using the configure test, but not when skipping the configure test because the spkg is already installed. Steps to reproduce: * ecm is not installed in the base os * Sage make distclean && make succeeeds * Tests succeed * re-running ./bootstrap && ./configure && make works * Tests now fail with PermissionError: [Errno 13] Permission denied: '' See also: * https://github.com/sagemath/sage/pull/37701 * https://github.com/sagemath/sage/pull/37011#issuecomment-2023089743 commit 551139c09f26a5da96b1187c3f0dd17b8d80ef84 Merge: 03be519274 547d502ed5 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:43 2024 +0200 gh-37763: Fix tolerance for scipy 1.13 After updating scipy to 1.13: ``` ********************************************************************** File "pkgs/sagemath-standard/build/lib.linux-x86_64-cpython- 312/sage/matrix/matrix_double_dense.pyx", line 3686, in sage.matrix.matrix_double_dense.Matrix_double_dense.exp Failed example: A.exp() # tol 1.1e-14 # needs sage.symbolic Expected: [-19.614602953804912 + 12.517743846762578*I 3.7949636449582176 + 28.88379930658099*I] [ -32.383580980922254 + 21.88423595789845*I 2.269633004093535 + 44.901324827684824*I] Got: [-19.61460295380496 + 12.517743846762578*I 3.7949636449581874 + 28.88379930658104*I] [-32.38358098092232 + 21.884235957898436*I 2.2696330040934853 + 44.901324827684896*I] Tolerance exceeded in 1 of 8: 2.269633004093535 vs 2.2696330040934853, tolerance 3e-14 > 1.1e-14 ********************************************************************** ``` This PR raises the tolerance to `3e-14` to fix this. ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37763 Reported by: Gonzalo Tornaría Reviewer(s): Antonio Rojas commit 03be519274f1e3aa424ea729f406180a10df1498 Merge: 4d88e898e2 cd41d1e414 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:37 2024 +0200 gh-37755: src/sage/graphs/graph_decompositions/tdlib.pyx: Use -std=c++11       This is needed with the current Xcode command line tools on macOS with boost from homebrew. Similar to https://github.com/sagemath/sage/pull/37646#issuecomment-2017108199 ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37755 Reported by: Matthias Köppe Reviewer(s): David Coudert commit 4d88e898e203d9b7297d224f40d659c213701ccb Merge: 9971cec642 6722976389 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:34 2024 +0200 gh-37753: Added rank check to doctest in `.gauss_on_polys` Fixes #37732. Also removed assignment of vectors `v` in the same doctest since they are not used. URL: https://github.com/sagemath/sage/pull/37753 Reported by: Sebastian A. Spindler Reviewer(s): Martin Rubey commit 9971cec64239679ee00ee7b22415b98ed53fbbf6 Merge: 3df3305a64 80239bc925 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:31 2024 +0200 gh-37736: Reimplementing the Witt (Sym Func) change of basis, caches, and omega involution       We reimplement the coercion from the Witt basis of symmetric functions to the $h$, $e$, and $p$ bases and their inverses by using the generating function identities and recursion for $w_n, h_n, e_n, p_n$ and the mulitplicative basis property for the general shape (in particular, we attempt to minimize the number of multiplications and maximize the use of the cache by recursively splitting the partition by its smallest part). We cache the result using `@cached_method` instead of custom classes. We also deprecate the `coerce_*` inputs. We also take advantage that $\omega w_n = w_n$ when $n$ is odd to compute the omega involution. This extends to a faster computation of the antipode. ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [x] I have created tests covering the changes. - [x] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37736 Reported by: Travis Scrimshaw Reviewer(s): Darij Grinberg, Martin Rubey commit 3df3305a6493f88ce86a1bc73e17c34ad01fbe33 Merge: 01333dfec1 ddcb3e45ca Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:27 2024 +0200 gh-37725: Fix typo/phrasing in README.md       - Removed the phrase "It assumes that you have already cloned the git repository" from the setup guide since the guide actually contains instructions on cloning the repository - Removed a random misplaced period ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37725 Reported by: Faisal Reviewer(s): gmou3 commit 01333dfec1ae7a518260c0178d4a0b254b886157 Merge: bcbb58f10a 8378fe6fdd Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:24 2024 +0200 gh-37721: Renamed "ring" argument of matrix contructor to "base_ring"       The matrix constructor used to accept ``ring`` as argument to specify a the base ring of a matrix. It now takes ``base_ring`` instead. ``ring`` is still accepted but deprecated (I added a deprecation in the code). This is for unification (``base_ring`` is what users expect from their experience with other classes). This solves @mkoeppe's issue #33380. ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [x] I have created tests covering the changes. - [x] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37721 Reported by: SandwichGouda Reviewer(s): Matthias Köppe commit bcbb58f10a840f43c1359934b85bfef9c143731c Merge: 08d5d97336 9b3ddf95fd Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:21 2024 +0200 gh-37720: pkgs/sagemath-repl/pyproject.toml.m4: Declare extra 'sphinx'       As discussed in https://github.com/sagemath/sage/pull/37589#issuecomment-2030456264, the Sage documentation system uses Sphinx to format the documentation in the terminal, for example in the help system. For the modularization project, a simple fallback for the case of Sphinx not being present was added. Here we declare the corresponding "extra" so that users can do `pip install sagemath-repl[sphinx]`. ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37720 Reported by: Matthias Köppe Reviewer(s): commit 08d5d97336d5cb87f4e9731b724b320e8cef771f Merge: 253169ff7b 5a61d3c103 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:18 2024 +0200 gh-37712: src/tox.ini (rst): Add missing Sphinx roles       Fixes #37711 ### :memo: Checklist  - [x] The title is concise and informative. - [ ] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37712 Reported by: Matthias Köppe Reviewer(s): Frédéric Chapoton commit 253169ff7b83df66d20846983ac9ce2982f3a18e Merge: 35f34ee0f6 3dc65b5797 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:16 2024 +0200 gh-37707: Implement an iterator for absolute number fields       We use the $\mathbb{Q}$ vector space structure. Although because of #37706, we go through `cartesian_product`. If there is an easy way to implement this in full generality (i.e., all number fields), please let me know. ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [x] I have created tests covering the changes. - [x] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37707 Reported by: Travis Scrimshaw Reviewer(s): Matthias Köppe commit 35f34ee0f66904d9f2766b646664ae67cedd3aae Merge: 0a16c78b4e e077f1edf7 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:11 2024 +0200 gh-37703: Implement an is_nilpotent() method for matrices       Checking if a matrix is nilpotent is a fundamental test that Sage is (surprisingly) missing. We check this by using the char poly. ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [x] I have created tests covering the changes. - [x] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37703 Reported by: Travis Scrimshaw Reviewer(s): Frédéric Chapoton, Martin Rubey commit 0a16c78b4e2b677ba30b198fdc781d38e595c9d8 Merge: 9d788cb197 7e8aa98fa6 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:04 2024 +0200 gh-37699: Add `approximate_closest_vector` to `IntegerLattice`, using Babai's nearest plane algorithm This PR implements Babai's nearest plane algorithm to find vectors close to a given vector in an integer lattice. I would appreciate if someone could verify the bound stated in the docstring, it is inferred from Babai's original paper but parameterized to use the `delta` parameter of the LLL algorithm. Ideas for improvements: - The doctest is currently the same as for `closest_vector`, this could be expanded - The `delta` parameter for LLL needs to be kept track of by the user. How "much" the basis has been reduced (i.e to how high a `delta` value) could be kept track of by the class and if a further reduction is needed for some given bound then it could be performed. ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [x] I have created tests covering the changes. - [x] I have updated the documentation accordingly. URL: https://github.com/sagemath/sage/pull/37699 Reported by: TheBlupper Reviewer(s): grhkm21, TheBlupper commit 9d788cb19732f9f750c6d5a25c28ccb958414112 Merge: 6925269840 d63ab2607e Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:02 2024 +0200 gh-37689: Fix changes.html in doc preview       by changing the base doc url. The base doc url is where the doc preview for the released version is uploaded. It changed recently: https://github.com/sagemath/sage/issues/34874#issuecomment-1870589626, but we didn't fix accordingly here: [.ci/create-changes- html.sh](https://github.com/sagemath/sage/blob/develop/.ci/create- changes-html.sh) ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37689 Reported by: Kwankyu Lee Reviewer(s): commit 6925269840b81d3a1085ae900037e71dfd440c87 Merge: 12a003eddd 3b137b50a0 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:59 2024 +0200 gh-37688: typo Small typo URL: https://github.com/sagemath/sage/pull/37688 Reported by: Dennis Yurichev Reviewer(s): Martin Rubey commit 12a003eddd08bcac14c88becd82bf52af599fe6c Merge: 3c9a67a2f0 6184336d73 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:56 2024 +0200 gh-37687: provide a construction functor using a single functor for families We enable sage to find common parents also when symmetric functions are involved. In particular, we provide a functorial construction that takes a commutative (coefficient) ring and produces a (commutative) ring of symmetric functions over this ring. For Macdonald polynomials and other symmetric functions that involve parameters we cheat a bit: the categories should actually be commutative rings with distinguished elements (namely, the parameters). However, having these is very likely overkill. This is an alternative to #37220, relying on individual construction functors instead of passing around strings, and #37686, which provides an individual functor for each family. URL: https://github.com/sagemath/sage/pull/37687 Reported by: Martin Rubey Reviewer(s): Travis Scrimshaw commit 3c9a67a2f0f8d42c1167f7fa192c70d29cdf6766 Merge: 1ab35e3a69 31645a4f35 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:53 2024 +0200 gh-37683: some shortcuts using bool fixes suggested by `ruff --select=SIM210` namely use `bool` for concision ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37683 Reported by: Frédéric Chapoton Reviewer(s): Matthias Köppe, Travis Scrimshaw commit 1ab35e3a696baf5a2fdd866a670c32611fa4a996 Merge: 654ed08b89 449864f0af Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:49 2024 +0200 gh-37680: a few more uses of "in Fields()" Just replacing some tests by `R in Fields()` in some pyx files outside the `rings` folder ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37680 Reported by: Frédéric Chapoton Reviewer(s): Matthias Köppe commit 654ed08b899e3a0bcd2d72a9ea66e5f1a96569a8 Merge: 042153b0d9 a8671aef33 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:44 2024 +0200 gh-37679: simplifications in symmetric_ideal.py some simplifications in `symmetric_ideal` ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37679 Reported by: Frédéric Chapoton Reviewer(s): Matthias Köppe commit 042153b0d976f8ec7b7489c404a17084321e3e06 Merge: 0de7848873 5d4a91fa85 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:41 2024 +0200 gh-37677: some simplifications in moment-angle complex a bunch of code details in the modified file fixes #36217 ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. URL: https://github.com/sagemath/sage/pull/37677 Reported by: Frédéric Chapoton Reviewer(s): Travis Scrimshaw commit 0de7848873bcda21b916e13ad71dd0dbe6aa9ff0 Merge: 014ccc31c4 e0ef88414d Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:38 2024 +0200 gh-37674: some pep8 and ruff cleanups in modules/ some ruff and pep8 cleanup in the `modules` folder ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37674 Reported by: Frédéric Chapoton Reviewer(s): Travis Scrimshaw commit 014ccc31c4118b929bb46c0811331f828041aee4 Merge: 93cc706453 42b4b671c3 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:34 2024 +0200 gh-37673: upgrade msolve to 0.6.5 routine upgrade, dropping already applied patch URL: https://github.com/sagemath/sage/pull/37673 Reported by: Dima Pasechnik Reviewer(s): Dima Pasechnik, Marc Mezzarobba, Matthias Köppe commit 93cc7064532b3456e5520833121e761078078576 Merge: cf431c8d40 f9ddc324b8 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:31 2024 +0200 gh-37672: src/sage/modular/quasimodform/ring.py: Fix pycodestyle warning       ### :memo: Checklist  - [ ] The title is concise and informative. - [ ] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37672 Reported by: Matthias Köppe Reviewer(s): Frédéric Chapoton commit cf431c8d40397205bb0f2e1f2843beec16788236 Merge: 2c71f81f0d 378ec74794 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:28 2024 +0200 gh-37671: pkgs/sagemath-standard: Support gpep517 From https://github.com/sagemath/sage/pull/37138#issuecomment-2018937633       ### :memo: Checklist  - [x] The title is concise and informative. - [ ] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37671 Reported by: Matthias Köppe Reviewer(s): François Bissey commit 2c71f81f0d804932deb11f3d28eb6e7962593076 Merge: 5801382f06 b8757d814e Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:24 2024 +0200 gh-37668: Wrong results in is_isotopic method of the Link class for certain chiral link       The following result is wrong, since `K10_67` is a chiral knot: ``` sage: L = KnotInfo.K10_67 ....: L1 = L.link() ....: L1r = L.link().reverse() ....: L1.is_isotopic(L1r) True ``` The problem is that the `get_knotinfo` method currently does not distinguish between all four symmetry mutants of a chiral knot or link. In the example, `get_knotinfo` detects both knots as *not mirrored* to the symmetry mutant of `K10_67` recorded in the KnotInfo database, as you can see from the verbose messages: ``` sage: set_verbose(1) sage: L1.is_isotopic(L1r) verbose 1 (4365: link.py, is_isotopic) KnotInfo other: KnotInfo.K10_67 mirrored False True ``` The wrong conclusion in `is_isotopic` is that they must be isotopic. The idea in this PR is to replace the boolean `mirrored` in the output of `get_knotinfo` with the value of an enum representing all symmetry mutants for an entry of the KnotInfo database. Of course, this requires major revisions in the `get_knotinfo` and `is_isotopic` methods. The key to this is a new local method `_knotinfo_matching_dict` that builds on `_knotinfo_matching_list` and returns the matching list for all symmetry mutants. On the occasion of that revision, I also fix other wrong results for example this one of a multi-component link which is not amphicheiral, but reversible: ``` sage: L = KnotInfo.L6a2_0 sage: L1 = L.link() sage: L2 = L.link(L.items.braid_notation) sage: L1.is_isotopic(L2) False ``` This bug is due to `get_knotinfo` returning the wrong mirror version for `L1` concerning `L6a2_1`: ``` ... sage: L1.get_knotinfo(unique=False) [(<KnotInfo.L6a2_0: 'L6a2{0}'>, False), (<KnotInfo.L6a2_1: 'L6a2{1}'>, False)] sage: L2.get_knotinfo(unique=False) [(<KnotInfo.L6a2_0: 'L6a2{0}'>, False), (<KnotInfo.L6a2_1: 'L6a2{1}'>, True)] ``` In addition to these fixes, improvements and corresponding adjustments, I correct some document formatting. ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [x] I have created tests covering the changes. - [x] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37668 Reported by: Sebastian Oehms Reviewer(s): Sebastian Oehms, Travis Scrimshaw commit 5801382f0638c7c8f50ae5a906b85306e5350fb5 Merge: cf365e5e51 bcc326d972 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:11 2024 +0200 gh-37667: Fix noexcept clauses (#37560) In #36507 I added a lot of `noexcept` clauses guided by a warning which is not quite right (see https://github.com/cython/cython/pull/5999#issuecomment-1986868208). A new warning in 3.0.9 shows the mistake (incorrectly added `noexcept` clauses). This broke some doctests (#37560) and a workaround was implemented for 10.3 (#37583) since we were too close to release. This PR now does the proper fix, removing all the incorrect `noexcept`, and also adding a few missing `noexcept`. Note: if one tries this PR with cython <= 3.0.8 it seems it's wrong in the sense that it will show 40k more warnings after the PR. These are *incorrect* warnings. If one uses cython 3.0.9 each of these 40k lines give a correct warning before this PR and an incorrect warnings after the PR, and there is no way to avoid 40k warnings. To confirm this PR is a good one, one needs to use cython 3.0.9 + https://github.com/cython/cython/pull/6087 or wait for cython 3.0.10. In this case, the warnings are correct and one will get 40k warnings before the PR and no warning after the PR. The last commit reverts the workaround from #37583 so we don't silence these warnings on `cython()` now that we have our own code right. @vbraun: this is "almost trivial" but touches too many files. Could we merge it in beta0 to avoid conflicts? I did this "kind of" automatically and I've been testing it for two weeks in almost all my builds. The changes made should not affect behaviour at all in the legacy mode we are using cython. ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. URL: https://github.com/sagemath/sage/pull/37667 Reported by: Gonzalo Tornaría Reviewer(s): Gonzalo Tornaría, Matthias Köppe commit cf365e5e51d30495f823ed928e0fe027166545d0 Merge: 59bf21efe7 a47b98b32e Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:08 2024 +0200 gh-37663: sage/rings/{complex,real}*: Untitlecase titles, add refs to libraries       This cleans up the table of contents https://deploy-preview-37663-- sagemath.netlify.app/html/en/reference/rings_numerical/ ### :memo: Checklist  - [x] The title is concise and informative. - [ ] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    - Depends on #37406 (merged here to resolve merge conflict) URL: https://github.com/sagemath/sage/pull/37663 Reported by: Matthias Köppe Reviewer(s): Marc Mezzarobba commit 59bf21efe755cac20496bb417285a7e3545bc790 Merge: 2fd04522f3 210c5d8389 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:06 2024 +0200 gh-37661: some cross-references btw doc of RR, RBF, ... (mainly intended to make RBF and CBF, which are the most powerful implementations for most purposes, more discoverable) URL: https://github.com/sagemath/sage/pull/37661 Reported by: Marc Mezzarobba Reviewer(s): commit 2fd04522f3caaca5fff0f78f1f1dccc6c0027475 Merge: 03f81897de 16a5261134 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:02 2024 +0200 gh-37660: Fix e302 libs pyx This is adding missing empty lines (pycodestyle E302) in pyx files in the `modules` and `libs` folders. Also some little tweak in code in the singular libs files. ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37660 Reported by: Frédéric Chapoton Reviewer(s): David Coudert commit 03f81897de2bf0aaf6b7fa82982b17251b0c9fd1 Merge: 81b4ddccd1 23c45b1d61 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:59 2024 +0200 gh-37659: fix ruff codes UP012 and UP023 scripted using `ruff` ; only two minor changes see https://docs.astral.sh/ruff/rules/#pyupgrade-up ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37659 Reported by: Frédéric Chapoton Reviewer(s): Matthias Köppe commit 81b4ddccd1de963f7731661106e1e425efd2e76b Merge: 74a82d7ce4 4fa6f09984 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:54 2024 +0200 gh-37655: Removed caching of determinant in LLL for the `NTL:LLL`-algorithm The method previously computed the square root of the squared determinant to cache, thus caching the absolute value of the determinant instead of the determinant. Therefore we remove this caching completely, which fixes #37236. URL: https://github.com/sagemath/sage/pull/37655 Reported by: Sebastian A. Spindler Reviewer(s): Vincent Delecroix commit 74a82d7ce445d8d9e66595887e9fcdf40f838702 Merge: 34539cdb4b 51d423d5a2 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:52 2024 +0200 gh-37652: fix typos and no "Algebra" in the doc This fixes a bunch of typos in the doc, and also rewrites one example to use `Parent` instead of the auld `Algebra` ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37652 Reported by: Frédéric Chapoton Reviewer(s): Matthias Köppe commit 34539cdb4b65fbb8f34f738c15f41a800de25141 Merge: 7a5ba77a93 5f32f5f882 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:49 2024 +0200 gh-37651: use parent in asymptotic ring trying to use the modern `Parent` and categories in the asymptotic ring, instead of the auld `Algebra` ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37651 Reported by: Frédéric Chapoton Reviewer(s): Marc Mezzarobba commit 7a5ba77a93cef0ee7a44f4228070e48b890d2b73 Merge: d4ab336474 4fa36454b5 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:45 2024 +0200 gh-37640: Modified `.is_reduced()` of `binary_qf.py` to avoid square root computation Adapted the checks in `.is_reduced()` to avoid the square root computation, which fixes #37635. Furthermore added a non-singularity check. This adaption is based on the observation that $$|\sqrt{D} - 2 \cdot |a|| < b < \sqrt{D}$$ if and only if $$b > 0, \ a \cdot c < 0 \text{ and } (a-c)^2 < D$$ are all satisfied simultaneously (where $D = b^2 - 4ac > 0$). The above can be proven in a straightforward manner by taking squares and square roots while making sure that the numbers being squared are always positive, so that inequalities are preserved. URL: https://github.com/sagemath/sage/pull/37640 Reported by: Sebastian A. Spindler Reviewer(s): Lorenz Panny commit d4ab336474b9f40b9e7f637c9b3ba6ac8bbc2135 Merge: 3c174ef6a6 bf0c3695cc Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:41 2024 +0200 gh-37639: add Legendre transform and suspension for lazy symmetric functions This adds two useful functions, particularly important when working with operads and Koszul duality. ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [x] I have created tests covering the changes. - [x] I have updated the documentation accordingly. URL: https://github.com/sagemath/sage/pull/37639 Reported by: Frédéric Chapoton Reviewer(s): Martin Rubey, Travis Scrimshaw commit 3c174ef6a69d80e733dc7461b44fbc8b646ac30d Merge: 6ab45f666b d7021096c4 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:38 2024 +0200 gh-37638: use strings as label in modular-decomposition trees Fixing #37631 and adding a note saying that labels of rooted trees must be comparable ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [x] I have created tests covering the changes. - [x] I have updated the documentation accordingly. URL: https://github.com/sagemath/sage/pull/37638 Reported by: Frédéric Chapoton Reviewer(s): cyrilbouvier commit 6ab45f666ba665ee9a0c517233830bafbea544b9 Merge: 8e26e27572 c4c26ee1bf Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:36 2024 +0200 gh-37637: Update jupyter-sphinx to version 0.5.3 and pin thebe to version 0.8.2       jupyter-sphinx that we are shipping was patched to use thebe@latest, which is the cause of the current breakdown of the sage live doc. The patch to jupyter-sphinx is partly based on https://github.com/jupyter/jupyter-sphinx/pull/231 ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37637 Reported by: Kwankyu Lee Reviewer(s): Matthias Köppe commit 8e26e27572073bc9f54839143f52d2d7e4c710e0 Merge: 13a555d970 af97e6da28 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:28 2024 +0200 gh-37634: The Glaisher-Franklin bijections on integer partitions The Glaisher-Franklin bijections, for a positive integer $s$, map the set of parts divisible by $s$ to the set of parts which occur at least $s$ times, see https://www.findstat.org/MapsDatabase/Mp00312 for the $s=2$ case. This generalizes and streamlines the code from findstat. URL: https://github.com/sagemath/sage/pull/37634 Reported by: Martin Rubey Reviewer(s): Martin Rubey, Travis Scrimshaw commit 13a555d970eefceddd20728c6a3275447713ea7e Merge: 58e8c578fd d45422e64e Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:23 2024 +0200 gh-37606: Border matrix       Plain TeX users may remember `\bordermatrix`. Here we implement this as options of the `Matrix` class's `str` method. ``` sage: M = matrix([[1,2,3], [4,5,6], [7,8,9]]) sage: M.subdivide(None, 2) sage: print(M.str(unicode=True, ....: top_border=['ab', 'cde', 'f'], ....: bottom_border=['*', '', ''], ....: left_border=[1, 10, 100], ....: right_border=['', ' <', ''])) ab cde f 1⎛ 1 2│ 3⎞ 10⎜ 4 5│ 6⎟ < 100⎝ 7 8│ 9⎠ * ``` Follow-up PR: As the guiding application for this feature, we equip finite-dimensional modules with basis with methods `_repr_matrix`, `_ascii_art_matrix`, `_unicode_art_matrix` that can be used as in this example: ``` sage: M = matrix(ZZ, [[1, 0, 0], [0, 1, 0]], ....: column_keys=['a', 'b', 'c'], ....: row_keys=['v', 'w']); M Generic morphism: From: Free module generated by {'a', 'b', 'c'} over Integer Ring To: Free module generated by {'v', 'w'} over Integer Ring sage: M._unicode_art_ = M._unicode_art_matrix sage: unicode_art(M) # indirect doctest a b c v⎛1 0 0⎞ w⎝0 1 0⎠ ``` ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37606 Reported by: Matthias Köppe Reviewer(s): David Coudert, Matthias Köppe commit 58e8c578fdc79d94dba5de30ae3a5ba6c08630ef Merge: 294eed4041 c1d38c77d2 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:20 2024 +0200 gh-37563: trying a change in unique_representation namely break the loop over `mro` once found what we want I am not sure at all that this keeps the expected properties, but this is more efficient.. ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37563 Reported by: Frédéric Chapoton Reviewer(s): Travis Scrimshaw commit 294eed4041447cb7420462f9548bc9195e376565 Merge: 14da417689 def8e56c30 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:17 2024 +0200 gh-37546: Sphinx ext links for Sage source files       In the documentation, we can now write `` :sage_root:`src/setup.py` `` or `` :sage_root:`src/doc/en/installation` ``, and it will be formatted uniformly, and a link to the file in the repository will be created. - Rebased version of #33756 Fixes #33756. ### :memo: Checklist  - [x] The title is concise and informative. - [ ] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37546 Reported by: Matthias Köppe Reviewer(s): David Coudert, Kwankyu Lee commit 14da417689b26c72b4369c8a352d3f8345a3c259 Merge: 59ced5e363 1e66857af1 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:13 2024 +0200 gh-37497: remove zombie code We remove zombie code for solving linear systems over a finite field, which was (likely accidentally) created in #23214, but leads to errors when solving sparse linear equations over finite fields. This does not quite fix the issue, because we now fall back to generic code. Fixes #28586 URL: https://github.com/sagemath/sage/pull/37497 Reported by: Martin Rubey Reviewer(s): Matthias Köppe commit 59ced5e363eef8f18c34b65fbde75f007015ee0f Merge: fbaec798ee 6704b34380 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:09 2024 +0200 gh-37425: Remove mention of patchbot, remove 'make buildbot-python3'      ### :memo: Checklist     - [x] The title is concise, informative, and self-explanatory. - [ ] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies  - Depends on https://github.com/sagemath/sage/pull/37421  URL: https://github.com/sagemath/sage/pull/37425 Reported by: Matthias Köppe Reviewer(s): Frédéric Chapoton commit fbaec798ee1a0df0e9ded0dd2dd0811387cbb797 Merge: 7855be0849 cb4e7184b3 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:06 2024 +0200 gh-37391: Make installation of "wheel" packages less noisy   Some care for `build/bin/sage-spkg`, reducing unhelpful verbosity. In particular, for "wheel" packages (https://deploy-livedoc-- sagemath.netlify.app/html/en/developer/packaging#package-source-types), there is no build step, so there is no point in talking about what C compiler is in use. Also saving a few lines of output (and making it clearer who does what) by prefixing the output from the various `spkg-...` scripts. Example: ``` $ make packaging-no-deps [packaging-23.2] Using cached file /Users/mkoeppe/s/sage/sage- rebasing/worktree-pristine/upstream/packaging-23.2-py3-none-any.whl [packaging-23.2] Setting up build directory /Users/mkoeppe/s/sage/sage- rebasing/worktree-pristine/local/var/lib/sage/venv- python3.11/var/tmp/sage/build/packaging-23.2 [packaging-23.2] [spkg-piprm] Found existing installation: packaging 23.2 [packaging-23.2] [spkg-piprm] Uninstalling packaging-23.2: [packaging-23.2] [spkg-piprm] Successfully uninstalled packaging-23.2 [packaging-23.2] Removing stamp file /Users/mkoeppe/s/sage/sage- rebasing/worktree-pristine/local/var/lib/sage/venv- python3.11/var/lib/sage/installed/packaging-23.2 [packaging-23.2] [spkg-install] Staged wheel file, staged /Users/mkoeppe/s/sage/sage-rebasing/worktree- pristine/local/var/lib/sage/venv- python3.11/var/lib/sage/scripts/packaging/spkg-requirements.txt [packaging-23.2] Moving package files from temporary location /Users/mkoeppe/s/sage/sage-rebasing/worktree- pristine/local/var/lib/sage/venv- python3.11/var/tmp/sage/build/packaging-23.2/inst to /Users/mkoeppe/s/sage/sage-rebasing/worktree- pristine/local/var/lib/sage/venv-python3.11 [packaging-23.2] [spkg-pipinst] Using pip 23.3.1 from /Users/mkoeppe/s/sage/sage-rebasing/worktree- pristine/local/var/lib/sage/venv-python3.11/lib/python3.11/site- packages/pip (python 3.11) [packaging-23.2] [spkg-pipinst] Looking in links: /Users/mkoeppe/s/sage/sage-rebasing/worktree- pristine/local/var/lib/sage/venv-python3.11/var/lib/sage/wheels [packaging-23.2] [spkg-pipinst] Processing /Users/mkoeppe/s/sage/sage- rebasing/worktree-pristine/local/var/lib/sage/venv- python3.11/var/lib/sage/wheels/packaging-23.2-py3-none-any.whl (from -r /Users/mkoeppe/s/sage/sage-rebasing/worktree- pristine/local/var/lib/sage/venv- python3.11/var/lib/sage/scripts/packaging/spkg-requirements.txt (line 1)) [packaging-23.2] [spkg-pipinst] Installing collected packages: packaging [packaging-23.2] [spkg-pipinst] ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts. [packaging-23.2] [spkg-pipinst] meson-python 0.15.0 requires meson>=0.63.3; python_version < "3.12", which is not installed. [packaging-23.2] [spkg-pipinst] tox 4.11.1 requires cachetools>=5.3.1, which is not installed. [packaging-23.2] [spkg-pipinst] tox 4.11.1 requires chardet>=5.2, which is not installed. [packaging-23.2] [spkg-pipinst] Successfully installed packaging-23.2 [packaging-23.2] Deleting build directory /Users/mkoeppe/s/sage/sage- rebasing/worktree-pristine/local/var/lib/sage/venv- python3.11/var/tmp/sage/build/packaging-23.2 [packaging-23.2] Finished installing packaging-23.2 ``` Also displaying build times for each package, even when using `make V=0`, but suppressing very short build times: ``` $ make V=0 ... [setuptools_scm-8.0.4] installing. Log file: /Users/mkoeppe/s/sage/sage- rebasing/worktree-pristine/logs/pkgs/setuptools_scm-8.0.4.log [hatchling-1.20.0] installing. Log file: /Users/mkoeppe/s/sage/sage- rebasing/worktree-pristine/logs/pkgs/hatchling-1.20.0.log [setuptools_scm-8.0.4] successfully installed. [hatchling-1.20.0] successfully installed. ... [sage_conf-10.3.beta8] successfully installed (real 11.36 user 8.32 sys 2.67). [pyproject_metadata-0.7.1] successfully installed (real 16.57 user 10.70 sys 2.97). [ipykernel-6.27.1] successfully installed (real 19.31 user 11.35 sys 3.71). ```    ### :memo: Checklist     - [x] The title is concise, informative, and self-explanatory. - [ ] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies   URL: https://github.com/sagemath/sage/pull/37391 Reported by: Matthias Köppe Reviewer(s): Kwankyu Lee commit 7855be084987c6e200d4c65fee0f9d819510a819 Merge: 031064b9ce 236361bce6 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:57:55 2024 +0200 gh-37361: Add iterators, Whitney numbers, characteristic polynomial, etc Changes (mostly in `matroid.pyx`): - Add `_iterator` functions for `bases`, `circuits`, etc, and use them internally (this avoids memory consumption/blowups) - Add Whitney numbers functions (for the first and second kind) - Correct `f_vector` definition (the definition up till now referred to the Whitney numbers of second kind) - Add characteristic polynomial - Correct `is_valid` function to be `False` on a negative rank value - Optimize `circuits` function (e.g. `sage: %time C = matroids.CompleteGraphic(8).circuits()` drops from `4m` to `1m`) - Optimize `broken_circuit_complex` function - Improve docstrings (1-line outputs and other changes) ### ⌛ Dependencies * Depends on #37667 URL: https://github.com/sagemath/sage/pull/37361 Reported by: gmou3 Reviewer(s): gmou3, Matthias Köppe, Travis Scrimshaw commit 031064b9cee8f7bec802ff0af0c9c28e3fb31b44 Merge: ad0a4b03b3 34564a3962 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:57:47 2024 +0200 gh-37340: Add class FlatsMatroid This matroid subclass allows the definition and internal representation of a matroid based on its flats (closed sets). This representation can be advantageous for some algorithms. This completes the list of the most standard matroid definitions (bases, circui…

commit a047ccfd4c4454411027afb9dd0981d0ae4a8293 Merge: cc1aa671e5 14a1e5c078 Author: Jackson Walters <jacksonwalters@gmail.com> Date: Tue Apr 9 14:41:10 2024 -0400 Merge branch 'pr/37748' of https://github.com/jacksonwalters/sage into pr/37748 commit cc1aa671e549388cc5d325cce8893062125c52dd Author: Jackson Walters <jacksonwalters@gmail.com> Date: Tue Apr 9 14:39:37 2024 -0400 Squashed commit of the following: commit 14a1e5c078d1684f751b9e7ef450f197e7bcf27f Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 23:01:35 2024 -0400 Update src/sage/combinat/symmetric_group_algebra.py Good idea. moved description from _dft_modular to dft docstring remove doctest with p=3, n=3 Update src/sage/combinat/symmetric_group_algebra.py Update src/sage/combinat/symmetric_group_algebra.py Co-Authored-By: Travis Scrimshaw <clfrngrown@aol.com> commit f3807cb141bae308906295f605939d4b2b205a5a Merge: 9ad2152942 8ea5214695 Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 20:12:53 2024 -0400 Merge branch 'develop' into pr/37748 commit 8ea5214695f742eccd34bd88600771b7121cdb9f Author: Release Manager <release@sagemath.org> Date: Tue Apr 9 00:29:19 2024 +0200 Updated SageMath version to 10.4.beta2 commit 9ad2152942819a55c20d80485ad8cd8925969c6c Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 13:38:43 2024 -0400 double backslash for math commit 619a48dea00a0a67b8d4e542948200dc23e9589a Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 13:33:00 2024 -0400 add examples commit 527ae2670dd3a37b25a192ab49dad7f4c16dd5aa Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 13:24:13 2024 -0400 form should default to None if form is None, check if characteristic of field divides order of group. if so, it's the modular case, and we return the projection onto idempotents change-of-basis, i.e. the MFT. if not, use semi-normal form. commit 488e931da94f4872047ff8e839edc4a0c99b163e Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 13:09:54 2024 -0400 missing parantheses commit 7e96c904b6bc47d2108930c039b95a14155d425b Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 13:00:57 2024 -0400 Update src/sage/combinat/symmetric_group_algebra.py Co-authored-by: Travis Scrimshaw <clfrngrown@aol.com> commit cf27c64dff0b77c6cca162849ec923ae14f4920f Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 13:00:14 2024 -0400 Update src/sage/combinat/symmetric_group_algebra.py I thought I could do this, don't know why I convinced myself I needed flatten. Co-authored-by: Travis Scrimshaw <clfrngrown@aol.com> commit 92699f42eba524047b09d527981dab2456802958 Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 12:59:16 2024 -0400 Update src/sage/combinat/symmetric_group_algebra.py Co-authored-by: Travis Scrimshaw <clfrngrown@aol.com> commit 68f2ef140f91356f55290d100ad8a5282f19c206 Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 12:58:30 2024 -0400 Update src/sage/combinat/symmetric_group_algebra.py Co-authored-by: Travis Scrimshaw <clfrngrown@aol.com> commit 618d5e1a8cb11006ef9fa6642e33f06d5728d6ae Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 12:57:18 2024 -0400 Update src/sage/combinat/symmetric_group_algebra.py Co-authored-by: Travis Scrimshaw <clfrngrown@aol.com> commit 04db911e3f6a08b3be1bf5c295e913a40ee8b28a Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 12:56:37 2024 -0400 Update src/sage/combinat/symmetric_group_algebra.py I see, so "None" if the user doesn't specify, and we determine which form based on whether p|n!. Sounds good. Co-authored-by: Travis Scrimshaw <clfrngrown@aol.com> commit e86f79410ebda01b8a85598a5754f9c383336ae1 Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 11:09:57 2024 -0400 add doctest to ensure issue 37751 is fixed commit f8c64d16b3d1c7b3d338d9d2339ae2e284df3c58 Merge: aedd2f3927 3079e2d897 Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 10:55:44 2024 -0400 Merge branch 'pr/37748' of https://github.com/jacksonwalters/sage into pr/37748 commit aedd2f3927aac05dceeb97bd484afa82557652a8 Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 10:55:43 2024 -0400 remove multiplication stars commit 3079e2d897396222b0f23784a4d07a40f25b3c1f Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 10:54:38 2024 -0400 Update src/sage/combinat/symmetric_group_algebra.py Co-authored-by: grhkm21 <83517584+grhkm21@users.noreply.github.com> commit b14565299a6f5a82e5fae5080de88b455ce696e5 Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 10:51:59 2024 -0400 docstring lines are wrapped at 80 characters ensure each line is less than 80 characters commit e3fcc1f59c4668f88c184649c18627480365e646 Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 10:24:51 2024 -0400 keep form argument keep form since there is a one year deprecation warning before removing arguments. show deprecation warning when default seminormal form is called, but suggest modular form which works in all cases commit 6d187b153b5b5c3e29fe07d9efe925e34bf5baa5 Merge: 551139c09f c8d260a6fe Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 23:34:58 2024 +0200 gh-37764: ECM-related tests fail after an incremental build ### TL;DR: This is because setting the default ECMBIN=ecm is only done when using the configure test, but not when skipping the configure test because the spkg is already installed. ### Steps to reproduce: * ecm is not installed in the base os * Sage make distclean && make succeeeds * Tests succeed * re-running ./bootstrap && ./configure && make works * Tests now fail with PermissionError: [Errno 13] Permission denied: '' ### Dependencies See also: * https://github.com/sagemath/sage/pull/37701 * https://github.com/sagemath/sage/pull/37011#issuecomment-2023089743 URL: https://github.com/sagemath/sage/pull/37764 Reported by: Volker Braun Reviewer(s): Matthias Köppe commit 71ad110e196d9f17934d030ce1c000e11b50fee8 Author: Jackson Walters <jacksonwalters@gmail.com> Date: Sun Apr 7 19:57:42 2024 -0400 trigger GitHub actions commit c8d260a6fe37cb0379690e57dab6fd2a3d5d8048 Author: Volker Braun <tarpit@vbraun.imap.cc> Date: Sun Apr 7 22:13:11 2024 +0200 Update build/pkgs/ecm/spkg-configure.m4 Co-authored-by: Matthias Köppe <mkoeppe@math.ucdavis.edu> commit fbf896681da3ad1fad6c164e4d5a7cf452ec2eec Author: Volker Braun <tarpit@vbraun.imap.cc> Date: Sun Apr 7 21:35:42 2024 +0200 Update build/pkgs/ecm/spkg-configure.m4 Co-authored-by: Matthias Köppe <mkoeppe@math.ucdavis.edu> commit 6531f7e415230d62fecd0a38fd804081341a9f32 Author: Volker Braun <tarpit@vbraun.imap.cc> Date: Sun Apr 7 21:35:38 2024 +0200 Update build/pkgs/ecm/spkg-configure.m4 Co-authored-by: Matthias Köppe <mkoeppe@math.ucdavis.edu> commit af0e24286a62a06d85e1baf49d0dc8f624965e5c Author: Volker Braun <vbraun.name@gmail.com> Date: Sun Apr 7 19:17:37 2024 +0200 ECM-related tests fail after an incremental build This is because setting the default ECMBIN=ecm is only done when using the configure test, but not when skipping the configure test because the spkg is already installed. Steps to reproduce: * ecm is not installed in the base os * Sage make distclean && make succeeeds * Tests succeed * re-running ./bootstrap && ./configure && make works * Tests now fail with PermissionError: [Errno 13] Permission denied: '' See also: * https://github.com/sagemath/sage/pull/37701 * https://github.com/sagemath/sage/pull/37011#issuecomment-2023089743 commit 551139c09f26a5da96b1187c3f0dd17b8d80ef84 Merge: 03be519274 547d502ed5 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:43 2024 +0200 gh-37763: Fix tolerance for scipy 1.13 After updating scipy to 1.13: ``` ********************************************************************** File "pkgs/sagemath-standard/build/lib.linux-x86_64-cpython- 312/sage/matrix/matrix_double_dense.pyx", line 3686, in sage.matrix.matrix_double_dense.Matrix_double_dense.exp Failed example: A.exp() # tol 1.1e-14 # needs sage.symbolic Expected: [-19.614602953804912 + 12.517743846762578*I 3.7949636449582176 + 28.88379930658099*I] [ -32.383580980922254 + 21.88423595789845*I 2.269633004093535 + 44.901324827684824*I] Got: [-19.61460295380496 + 12.517743846762578*I 3.7949636449581874 + 28.88379930658104*I] [-32.38358098092232 + 21.884235957898436*I 2.2696330040934853 + 44.901324827684896*I] Tolerance exceeded in 1 of 8: 2.269633004093535 vs 2.2696330040934853, tolerance 3e-14 > 1.1e-14 ********************************************************************** ``` This PR raises the tolerance to `3e-14` to fix this. ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37763 Reported by: Gonzalo Tornaría Reviewer(s): Antonio Rojas commit 03be519274f1e3aa424ea729f406180a10df1498 Merge: 4d88e898e2 cd41d1e414 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:37 2024 +0200 gh-37755: src/sage/graphs/graph_decompositions/tdlib.pyx: Use -std=c++11       This is needed with the current Xcode command line tools on macOS with boost from homebrew. Similar to https://github.com/sagemath/sage/pull/37646#issuecomment-2017108199 ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37755 Reported by: Matthias Köppe Reviewer(s): David Coudert commit 4d88e898e203d9b7297d224f40d659c213701ccb Merge: 9971cec642 6722976389 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:34 2024 +0200 gh-37753: Added rank check to doctest in `.gauss_on_polys` Fixes #37732. Also removed assignment of vectors `v` in the same doctest since they are not used. URL: https://github.com/sagemath/sage/pull/37753 Reported by: Sebastian A. Spindler Reviewer(s): Martin Rubey commit 9971cec64239679ee00ee7b22415b98ed53fbbf6 Merge: 3df3305a64 80239bc925 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:31 2024 +0200 gh-37736: Reimplementing the Witt (Sym Func) change of basis, caches, and omega involution       We reimplement the coercion from the Witt basis of symmetric functions to the $h$, $e$, and $p$ bases and their inverses by using the generating function identities and recursion for $w_n, h_n, e_n, p_n$ and the mulitplicative basis property for the general shape (in particular, we attempt to minimize the number of multiplications and maximize the use of the cache by recursively splitting the partition by its smallest part). We cache the result using `@cached_method` instead of custom classes. We also deprecate the `coerce_*` inputs. We also take advantage that $\omega w_n = w_n$ when $n$ is odd to compute the omega involution. This extends to a faster computation of the antipode. ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [x] I have created tests covering the changes. - [x] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37736 Reported by: Travis Scrimshaw Reviewer(s): Darij Grinberg, Martin Rubey commit 3df3305a6493f88ce86a1bc73e17c34ad01fbe33 Merge: 01333dfec1 ddcb3e45ca Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:27 2024 +0200 gh-37725: Fix typo/phrasing in README.md       - Removed the phrase "It assumes that you have already cloned the git repository" from the setup guide since the guide actually contains instructions on cloning the repository - Removed a random misplaced period ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37725 Reported by: Faisal Reviewer(s): gmou3 commit 01333dfec1ae7a518260c0178d4a0b254b886157 Merge: bcbb58f10a 8378fe6fdd Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:24 2024 +0200 gh-37721: Renamed "ring" argument of matrix contructor to "base_ring"       The matrix constructor used to accept ``ring`` as argument to specify a the base ring of a matrix. It now takes ``base_ring`` instead. ``ring`` is still accepted but deprecated (I added a deprecation in the code). This is for unification (``base_ring`` is what users expect from their experience with other classes). This solves @mkoeppe's issue #33380. ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [x] I have created tests covering the changes. - [x] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37721 Reported by: SandwichGouda Reviewer(s): Matthias Köppe commit bcbb58f10a840f43c1359934b85bfef9c143731c Merge: 08d5d97336 9b3ddf95fd Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:21 2024 +0200 gh-37720: pkgs/sagemath-repl/pyproject.toml.m4: Declare extra 'sphinx'       As discussed in https://github.com/sagemath/sage/pull/37589#issuecomment-2030456264, the Sage documentation system uses Sphinx to format the documentation in the terminal, for example in the help system. For the modularization project, a simple fallback for the case of Sphinx not being present was added. Here we declare the corresponding "extra" so that users can do `pip install sagemath-repl[sphinx]`. ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37720 Reported by: Matthias Köppe Reviewer(s): commit 08d5d97336d5cb87f4e9731b724b320e8cef771f Merge: 253169ff7b 5a61d3c103 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:18 2024 +0200 gh-37712: src/tox.ini (rst): Add missing Sphinx roles       Fixes #37711 ### :memo: Checklist  - [x] The title is concise and informative. - [ ] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37712 Reported by: Matthias Köppe Reviewer(s): Frédéric Chapoton commit 253169ff7b83df66d20846983ac9ce2982f3a18e Merge: 35f34ee0f6 3dc65b5797 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:16 2024 +0200 gh-37707: Implement an iterator for absolute number fields       We use the $\mathbb{Q}$ vector space structure. Although because of #37706, we go through `cartesian_product`. If there is an easy way to implement this in full generality (i.e., all number fields), please let me know. ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [x] I have created tests covering the changes. - [x] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37707 Reported by: Travis Scrimshaw Reviewer(s): Matthias Köppe commit 35f34ee0f66904d9f2766b646664ae67cedd3aae Merge: 0a16c78b4e e077f1edf7 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:11 2024 +0200 gh-37703: Implement an is_nilpotent() method for matrices       Checking if a matrix is nilpotent is a fundamental test that Sage is (surprisingly) missing. We check this by using the char poly. ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [x] I have created tests covering the changes. - [x] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37703 Reported by: Travis Scrimshaw Reviewer(s): Frédéric Chapoton, Martin Rubey commit 0a16c78b4e2b677ba30b198fdc781d38e595c9d8 Merge: 9d788cb197 7e8aa98fa6 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:04 2024 +0200 gh-37699: Add `approximate_closest_vector` to `IntegerLattice`, using Babai's nearest plane algorithm This PR implements Babai's nearest plane algorithm to find vectors close to a given vector in an integer lattice. I would appreciate if someone could verify the bound stated in the docstring, it is inferred from Babai's original paper but parameterized to use the `delta` parameter of the LLL algorithm. Ideas for improvements: - The doctest is currently the same as for `closest_vector`, this could be expanded - The `delta` parameter for LLL needs to be kept track of by the user. How "much" the basis has been reduced (i.e to how high a `delta` value) could be kept track of by the class and if a further reduction is needed for some given bound then it could be performed. ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [x] I have created tests covering the changes. - [x] I have updated the documentation accordingly. URL: https://github.com/sagemath/sage/pull/37699 Reported by: TheBlupper Reviewer(s): grhkm21, TheBlupper commit 9d788cb19732f9f750c6d5a25c28ccb958414112 Merge: 6925269840 d63ab2607e Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:02 2024 +0200 gh-37689: Fix changes.html in doc preview       by changing the base doc url. The base doc url is where the doc preview for the released version is uploaded. It changed recently: https://github.com/sagemath/sage/issues/34874#issuecomment-1870589626, but we didn't fix accordingly here: [.ci/create-changes- html.sh](https://github.com/sagemath/sage/blob/develop/.ci/create- changes-html.sh) ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37689 Reported by: Kwankyu Lee Reviewer(s): commit 6925269840b81d3a1085ae900037e71dfd440c87 Merge: 12a003eddd 3b137b50a0 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:59 2024 +0200 gh-37688: typo Small typo URL: https://github.com/sagemath/sage/pull/37688 Reported by: Dennis Yurichev Reviewer(s): Martin Rubey commit 12a003eddd08bcac14c88becd82bf52af599fe6c Merge: 3c9a67a2f0 6184336d73 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:56 2024 +0200 gh-37687: provide a construction functor using a single functor for families We enable sage to find common parents also when symmetric functions are involved. In particular, we provide a functorial construction that takes a commutative (coefficient) ring and produces a (commutative) ring of symmetric functions over this ring. For Macdonald polynomials and other symmetric functions that involve parameters we cheat a bit: the categories should actually be commutative rings with distinguished elements (namely, the parameters). However, having these is very likely overkill. This is an alternative to #37220, relying on individual construction functors instead of passing around strings, and #37686, which provides an individual functor for each family. URL: https://github.com/sagemath/sage/pull/37687 Reported by: Martin Rubey Reviewer(s): Travis Scrimshaw commit 3c9a67a2f0f8d42c1167f7fa192c70d29cdf6766 Merge: 1ab35e3a69 31645a4f35 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:53 2024 +0200 gh-37683: some shortcuts using bool fixes suggested by `ruff --select=SIM210` namely use `bool` for concision ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37683 Reported by: Frédéric Chapoton Reviewer(s): Matthias Köppe, Travis Scrimshaw commit 1ab35e3a696baf5a2fdd866a670c32611fa4a996 Merge: 654ed08b89 449864f0af Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:49 2024 +0200 gh-37680: a few more uses of "in Fields()" Just replacing some tests by `R in Fields()` in some pyx files outside the `rings` folder ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37680 Reported by: Frédéric Chapoton Reviewer(s): Matthias Köppe commit 654ed08b899e3a0bcd2d72a9ea66e5f1a96569a8 Merge: 042153b0d9 a8671aef33 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:44 2024 +0200 gh-37679: simplifications in symmetric_ideal.py some simplifications in `symmetric_ideal` ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37679 Reported by: Frédéric Chapoton Reviewer(s): Matthias Köppe commit 042153b0d976f8ec7b7489c404a17084321e3e06 Merge: 0de7848873 5d4a91fa85 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:41 2024 +0200 gh-37677: some simplifications in moment-angle complex a bunch of code details in the modified file fixes #36217 ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. URL: https://github.com/sagemath/sage/pull/37677 Reported by: Frédéric Chapoton Reviewer(s): Travis Scrimshaw commit 0de7848873bcda21b916e13ad71dd0dbe6aa9ff0 Merge: 014ccc31c4 e0ef88414d Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:38 2024 +0200 gh-37674: some pep8 and ruff cleanups in modules/ some ruff and pep8 cleanup in the `modules` folder ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37674 Reported by: Frédéric Chapoton Reviewer(s): Travis Scrimshaw commit 014ccc31c4118b929bb46c0811331f828041aee4 Merge: 93cc706453 42b4b671c3 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:34 2024 +0200 gh-37673: upgrade msolve to 0.6.5 routine upgrade, dropping already applied patch URL: https://github.com/sagemath/sage/pull/37673 Reported by: Dima Pasechnik Reviewer(s): Dima Pasechnik, Marc Mezzarobba, Matthias Köppe commit 93cc7064532b3456e5520833121e761078078576 Merge: cf431c8d40 f9ddc324b8 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:31 2024 +0200 gh-37672: src/sage/modular/quasimodform/ring.py: Fix pycodestyle warning       ### :memo: Checklist  - [ ] The title is concise and informative. - [ ] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37672 Reported by: Matthias Köppe Reviewer(s): Frédéric Chapoton commit cf431c8d40397205bb0f2e1f2843beec16788236 Merge: 2c71f81f0d 378ec74794 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:28 2024 +0200 gh-37671: pkgs/sagemath-standard: Support gpep517 From https://github.com/sagemath/sage/pull/37138#issuecomment-2018937633       ### :memo: Checklist  - [x] The title is concise and informative. - [ ] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37671 Reported by: Matthias Köppe Reviewer(s): François Bissey commit 2c71f81f0d804932deb11f3d28eb6e7962593076 Merge: 5801382f06 b8757d814e Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:24 2024 +0200 gh-37668: Wrong results in is_isotopic method of the Link class for certain chiral link       The following result is wrong, since `K10_67` is a chiral knot: ``` sage: L = KnotInfo.K10_67 ....: L1 = L.link() ....: L1r = L.link().reverse() ....: L1.is_isotopic(L1r) True ``` The problem is that the `get_knotinfo` method currently does not distinguish between all four symmetry mutants of a chiral knot or link. In the example, `get_knotinfo` detects both knots as *not mirrored* to the symmetry mutant of `K10_67` recorded in the KnotInfo database, as you can see from the verbose messages: ``` sage: set_verbose(1) sage: L1.is_isotopic(L1r) verbose 1 (4365: link.py, is_isotopic) KnotInfo other: KnotInfo.K10_67 mirrored False True ``` The wrong conclusion in `is_isotopic` is that they must be isotopic. The idea in this PR is to replace the boolean `mirrored` in the output of `get_knotinfo` with the value of an enum representing all symmetry mutants for an entry of the KnotInfo database. Of course, this requires major revisions in the `get_knotinfo` and `is_isotopic` methods. The key to this is a new local method `_knotinfo_matching_dict` that builds on `_knotinfo_matching_list` and returns the matching list for all symmetry mutants. On the occasion of that revision, I also fix other wrong results for example this one of a multi-component link which is not amphicheiral, but reversible: ``` sage: L = KnotInfo.L6a2_0 sage: L1 = L.link() sage: L2 = L.link(L.items.braid_notation) sage: L1.is_isotopic(L2) False ``` This bug is due to `get_knotinfo` returning the wrong mirror version for `L1` concerning `L6a2_1`: ``` ... sage: L1.get_knotinfo(unique=False) [(<KnotInfo.L6a2_0: 'L6a2{0}'>, False), (<KnotInfo.L6a2_1: 'L6a2{1}'>, False)] sage: L2.get_knotinfo(unique=False) [(<KnotInfo.L6a2_0: 'L6a2{0}'>, False), (<KnotInfo.L6a2_1: 'L6a2{1}'>, True)] ``` In addition to these fixes, improvements and corresponding adjustments, I correct some document formatting. ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [x] I have created tests covering the changes. - [x] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37668 Reported by: Sebastian Oehms Reviewer(s): Sebastian Oehms, Travis Scrimshaw commit 5801382f0638c7c8f50ae5a906b85306e5350fb5 Merge: cf365e5e51 bcc326d972 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:11 2024 +0200 gh-37667: Fix noexcept clauses (#37560) In #36507 I added a lot of `noexcept` clauses guided by a warning which is not quite right (see https://github.com/cython/cython/pull/5999#issuecomment-1986868208). A new warning in 3.0.9 shows the mistake (incorrectly added `noexcept` clauses). This broke some doctests (#37560) and a workaround was implemented for 10.3 (#37583) since we were too close to release. This PR now does the proper fix, removing all the incorrect `noexcept`, and also adding a few missing `noexcept`. Note: if one tries this PR with cython <= 3.0.8 it seems it's wrong in the sense that it will show 40k more warnings after the PR. These are *incorrect* warnings. If one uses cython 3.0.9 each of these 40k lines give a correct warning before this PR and an incorrect warnings after the PR, and there is no way to avoid 40k warnings. To confirm this PR is a good one, one needs to use cython 3.0.9 + https://github.com/cython/cython/pull/6087 or wait for cython 3.0.10. In this case, the warnings are correct and one will get 40k warnings before the PR and no warning after the PR. The last commit reverts the workaround from #37583 so we don't silence these warnings on `cython()` now that we have our own code right. @vbraun: this is "almost trivial" but touches too many files. Could we merge it in beta0 to avoid conflicts? I did this "kind of" automatically and I've been testing it for two weeks in almost all my builds. The changes made should not affect behaviour at all in the legacy mode we are using cython. ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. URL: https://github.com/sagemath/sage/pull/37667 Reported by: Gonzalo Tornaría Reviewer(s): Gonzalo Tornaría, Matthias Köppe commit cf365e5e51d30495f823ed928e0fe027166545d0 Merge: 59bf21efe7 a47b98b32e Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:08 2024 +0200 gh-37663: sage/rings/{complex,real}*: Untitlecase titles, add refs to libraries       This cleans up the table of contents https://deploy-preview-37663-- sagemath.netlify.app/html/en/reference/rings_numerical/ ### :memo: Checklist  - [x] The title is concise and informative. - [ ] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    - Depends on #37406 (merged here to resolve merge conflict) URL: https://github.com/sagemath/sage/pull/37663 Reported by: Matthias Köppe Reviewer(s): Marc Mezzarobba commit 59bf21efe755cac20496bb417285a7e3545bc790 Merge: 2fd04522f3 210c5d8389 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:06 2024 +0200 gh-37661: some cross-references btw doc of RR, RBF, ... (mainly intended to make RBF and CBF, which are the most powerful implementations for most purposes, more discoverable) URL: https://github.com/sagemath/sage/pull/37661 Reported by: Marc Mezzarobba Reviewer(s): commit 2fd04522f3caaca5fff0f78f1f1dccc6c0027475 Merge: 03f81897de 16a5261134 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:02 2024 +0200 gh-37660: Fix e302 libs pyx This is adding missing empty lines (pycodestyle E302) in pyx files in the `modules` and `libs` folders. Also some little tweak in code in the singular libs files. ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37660 Reported by: Frédéric Chapoton Reviewer(s): David Coudert commit 03f81897de2bf0aaf6b7fa82982b17251b0c9fd1 Merge: 81b4ddccd1 23c45b1d61 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:59 2024 +0200 gh-37659: fix ruff codes UP012 and UP023 scripted using `ruff` ; only two minor changes see https://docs.astral.sh/ruff/rules/#pyupgrade-up ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37659 Reported by: Frédéric Chapoton Reviewer(s): Matthias Köppe commit 81b4ddccd1de963f7731661106e1e425efd2e76b Merge: 74a82d7ce4 4fa6f09984 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:54 2024 +0200 gh-37655: Removed caching of determinant in LLL for the `NTL:LLL`-algorithm The method previously computed the square root of the squared determinant to cache, thus caching the absolute value of the determinant instead of the determinant. Therefore we remove this caching completely, which fixes #37236. URL: https://github.com/sagemath/sage/pull/37655 Reported by: Sebastian A. Spindler Reviewer(s): Vincent Delecroix commit 74a82d7ce445d8d9e66595887e9fcdf40f838702 Merge: 34539cdb4b 51d423d5a2 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:52 2024 +0200 gh-37652: fix typos and no "Algebra" in the doc This fixes a bunch of typos in the doc, and also rewrites one example to use `Parent` instead of the auld `Algebra` ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37652 Reported by: Frédéric Chapoton Reviewer(s): Matthias Köppe commit 34539cdb4b65fbb8f34f738c15f41a800de25141 Merge: 7a5ba77a93 5f32f5f882 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:49 2024 +0200 gh-37651: use parent in asymptotic ring trying to use the modern `Parent` and categories in the asymptotic ring, instead of the auld `Algebra` ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37651 Reported by: Frédéric Chapoton Reviewer(s): Marc Mezzarobba commit 7a5ba77a93cef0ee7a44f4228070e48b890d2b73 Merge: d4ab336474 4fa36454b5 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:45 2024 +0200 gh-37640: Modified `.is_reduced()` of `binary_qf.py` to avoid square root computation Adapted the checks in `.is_reduced()` to avoid the square root computation, which fixes #37635. Furthermore added a non-singularity check. This adaption is based on the observation that $$|\sqrt{D} - 2 \cdot |a|| < b < \sqrt{D}$$ if and only if $$b > 0, \ a \cdot c < 0 \text{ and } (a-c)^2 < D$$ are all satisfied simultaneously (where $D = b^2 - 4ac > 0$). The above can be proven in a straightforward manner by taking squares and square roots while making sure that the numbers being squared are always positive, so that inequalities are preserved. URL: https://github.com/sagemath/sage/pull/37640 Reported by: Sebastian A. Spindler Reviewer(s): Lorenz Panny commit d4ab336474b9f40b9e7f637c9b3ba6ac8bbc2135 Merge: 3c174ef6a6 bf0c3695cc Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:41 2024 +0200 gh-37639: add Legendre transform and suspension for lazy symmetric functions This adds two useful functions, particularly important when working with operads and Koszul duality. ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [x] I have created tests covering the changes. - [x] I have updated the documentation accordingly. URL: https://github.com/sagemath/sage/pull/37639 Reported by: Frédéric Chapoton Reviewer(s): Martin Rubey, Travis Scrimshaw commit 3c174ef6a69d80e733dc7461b44fbc8b646ac30d Merge: 6ab45f666b d7021096c4 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:38 2024 +0200 gh-37638: use strings as label in modular-decomposition trees Fixing #37631 and adding a note saying that labels of rooted trees must be comparable ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [x] I have created tests covering the changes. - [x] I have updated the documentation accordingly. URL: https://github.com/sagemath/sage/pull/37638 Reported by: Frédéric Chapoton Reviewer(s): cyrilbouvier commit 6ab45f666ba665ee9a0c517233830bafbea544b9 Merge: 8e26e27572 c4c26ee1bf Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:36 2024 +0200 gh-37637: Update jupyter-sphinx to version 0.5.3 and pin thebe to version 0.8.2       jupyter-sphinx that we are shipping was patched to use thebe@latest, which is the cause of the current breakdown of the sage live doc. The patch to jupyter-sphinx is partly based on https://github.com/jupyter/jupyter-sphinx/pull/231 ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37637 Reported by: Kwankyu Lee Reviewer(s): Matthias Köppe commit 8e26e27572073bc9f54839143f52d2d7e4c710e0 Merge: 13a555d970 af97e6da28 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:28 2024 +0200 gh-37634: The Glaisher-Franklin bijections on integer partitions The Glaisher-Franklin bijections, for a positive integer $s$, map the set of parts divisible by $s$ to the set of parts which occur at least $s$ times, see https://www.findstat.org/MapsDatabase/Mp00312 for the $s=2$ case. This generalizes and streamlines the code from findstat. URL: https://github.com/sagemath/sage/pull/37634 Reported by: Martin Rubey Reviewer(s): Martin Rubey, Travis Scrimshaw commit 13a555d970eefceddd20728c6a3275447713ea7e Merge: 58e8c578fd d45422e64e Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:23 2024 +0200 gh-37606: Border matrix       Plain TeX users may remember `\bordermatrix`. Here we implement this as options of the `Matrix` class's `str` method. ``` sage: M = matrix([[1,2,3], [4,5,6], [7,8,9]]) sage: M.subdivide(None, 2) sage: print(M.str(unicode=True, ....: top_border=['ab', 'cde', 'f'], ....: bottom_border=['*', '', ''], ....: left_border=[1, 10, 100], ....: right_border=['', ' <', ''])) ab cde f 1⎛ 1 2│ 3⎞ 10⎜ 4 5│ 6⎟ < 100⎝ 7 8│ 9⎠ * ``` Follow-up PR: As the guiding application for this feature, we equip finite-dimensional modules with basis with methods `_repr_matrix`, `_ascii_art_matrix`, `_unicode_art_matrix` that can be used as in this example: ``` sage: M = matrix(ZZ, [[1, 0, 0], [0, 1, 0]], ....: column_keys=['a', 'b', 'c'], ....: row_keys=['v', 'w']); M Generic morphism: From: Free module generated by {'a', 'b', 'c'} over Integer Ring To: Free module generated by {'v', 'w'} over Integer Ring sage: M._unicode_art_ = M._unicode_art_matrix sage: unicode_art(M) # indirect doctest a b c v⎛1 0 0⎞ w⎝0 1 0⎠ ``` ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37606 Reported by: Matthias Köppe Reviewer(s): David Coudert, Matthias Köppe commit 58e8c578fdc79d94dba5de30ae3a5ba6c08630ef Merge: 294eed4041 c1d38c77d2 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:20 2024 +0200 gh-37563: trying a change in unique_representation namely break the loop over `mro` once found what we want I am not sure at all that this keeps the expected properties, but this is more efficient.. ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37563 Reported by: Frédéric Chapoton Reviewer(s): Travis Scrimshaw commit 294eed4041447cb7420462f9548bc9195e376565 Merge: 14da417689 def8e56c30 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:17 2024 +0200 gh-37546: Sphinx ext links for Sage source files       In the documentation, we can now write `` :sage_root:`src/setup.py` `` or `` :sage_root:`src/doc/en/installation` ``, and it will be formatted uniformly, and a link to the file in the repository will be created. - Rebased version of #33756 Fixes #33756. ### :memo: Checklist  - [x] The title is concise and informative. - [ ] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37546 Reported by: Matthias Köppe Reviewer(s): David Coudert, Kwankyu Lee commit 14da417689b26c72b4369c8a352d3f8345a3c259 Merge: 59ced5e363 1e66857af1 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:13 2024 +0200 gh-37497: remove zombie code We remove zombie code for solving linear systems over a finite field, which was (likely accidentally) created in #23214, but leads to errors when solving sparse linear equations over finite fields. This does not quite fix the issue, because we now fall back to generic code. Fixes #28586 URL: https://github.com/sagemath/sage/pull/37497 Reported by: Martin Rubey Reviewer(s): Matthias Köppe commit 59ced5e363eef8f18c34b65fbde75f007015ee0f Merge: fbaec798ee 6704b34380 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:09 2024 +0200 gh-37425: Remove mention of patchbot, remove 'make buildbot-python3'      ### :memo: Checklist     - [x] The title is concise, informative, and self-explanatory. - [ ] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies  - Depends on https://github.com/sagemath/sage/pull/37421  URL: https://github.com/sagemath/sage/pull/37425 Reported by: Matthias Köppe Reviewer(s): Frédéric Chapoton commit fbaec798ee1a0df0e9ded0dd2dd0811387cbb797 Merge: 7855be0849 cb4e7184b3 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:06 2024 +0200 gh-37391: Make installation of "wheel" packages less noisy   Some care for `build/bin/sage-spkg`, reducing unhelpful verbosity. In particular, for "wheel" packages (https://deploy-livedoc-- sagemath.netlify.app/html/en/developer/packaging#package-source-types), there is no build step, so there is no point in talking about what C compiler is in use. Also saving a few lines of output (and making it clearer who does what) by prefixing the output from the various `spkg-...` scripts. Example: ``` $ make packaging-no-deps [packaging-23.2] Using cached file /Users/mkoeppe/s/sage/sage- rebasing/worktree-pristine/upstream/packaging-23.2-py3-none-any.whl [packaging-23.2] Setting up build directory /Users/mkoeppe/s/sage/sage- rebasing/worktree-pristine/local/var/lib/sage/venv- python3.11/var/tmp/sage/build/packaging-23.2 [packaging-23.2] [spkg-piprm] Found existing installation: packaging 23.2 [packaging-23.2] [spkg-piprm] Uninstalling packaging-23.2: [packaging-23.2] [spkg-piprm] Successfully uninstalled packaging-23.2 [packaging-23.2] Removing stamp file /Users/mkoeppe/s/sage/sage- rebasing/worktree-pristine/local/var/lib/sage/venv- python3.11/var/lib/sage/installed/packaging-23.2 [packaging-23.2] [spkg-install] Staged wheel file, staged /Users/mkoeppe/s/sage/sage-rebasing/worktree- pristine/local/var/lib/sage/venv- python3.11/var/lib/sage/scripts/packaging/spkg-requirements.txt [packaging-23.2] Moving package files from temporary location /Users/mkoeppe/s/sage/sage-rebasing/worktree- pristine/local/var/lib/sage/venv- python3.11/var/tmp/sage/build/packaging-23.2/inst to /Users/mkoeppe/s/sage/sage-rebasing/worktree- pristine/local/var/lib/sage/venv-python3.11 [packaging-23.2] [spkg-pipinst] Using pip 23.3.…

commit 1d08fd591fb1694aff33cf5047f890cb8bf3b0e4 Merge: 14a1e5c078 a03a19d0be Author: Jackson Walters <jacksonwalters@gmail.com> Date: Tue Apr 9 15:10:30 2024 -0400 Merge branch 'pr/37748' of https://github.com/jacksonwalters/sage into pr/37748 commit a03a19d0be7b13f6ed9d082ecdef692dfcfb100f Author: Jackson Walters <jacksonwalters@gmail.com> Date: Tue Apr 9 14:58:25 2024 -0400 Reapply "Squashed commit of the following:" This reverts commit 794503ca9f61504c80874e26d7cc71a37de4debc. commit 794503ca9f61504c80874e26d7cc71a37de4debc Author: Jackson Walters <jacksonwalters@gmail.com> Date: Tue Apr 9 14:56:20 2024 -0400 Revert "Squashed commit of the following:" This reverts commit da28bff0c7bb201a2795bb8cc974df69239954ef. commit da28bff0c7bb201a2795bb8cc974df69239954ef Author: Jackson Walters <jacksonwalters@gmail.com> Date: Tue Apr 9 14:46:40 2024 -0400 Squashed commit of the following: commit a047ccfd4c4454411027afb9dd0981d0ae4a8293 Merge: cc1aa671e5 14a1e5c078 Author: Jackson Walters <jacksonwalters@gmail.com> Date: Tue Apr 9 14:41:10 2024 -0400 Merge branch 'pr/37748' of https://github.com/jacksonwalters/sage into pr/37748 commit cc1aa671e549388cc5d325cce8893062125c52dd Author: Jackson Walters <jacksonwalters@gmail.com> Date: Tue Apr 9 14:39:37 2024 -0400 Squashed commit of the following: commit 14a1e5c078d1684f751b9e7ef450f197e7bcf27f Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 23:01:35 2024 -0400 Update src/sage/combinat/symmetric_group_algebra.py Good idea. moved description from _dft_modular to dft docstring remove doctest with p=3, n=3 Update src/sage/combinat/symmetric_group_algebra.py Update src/sage/combinat/symmetric_group_algebra.py Co-Authored-By: Travis Scrimshaw <clfrngrown@aol.com> commit f3807cb141bae308906295f605939d4b2b205a5a Merge: 9ad2152942 8ea5214695 Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 20:12:53 2024 -0400 Merge branch 'develop' into pr/37748 commit 8ea5214695f742eccd34bd88600771b7121cdb9f Author: Release Manager <release@sagemath.org> Date: Tue Apr 9 00:29:19 2024 +0200 Updated SageMath version to 10.4.beta2 commit 9ad2152942819a55c20d80485ad8cd8925969c6c Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 13:38:43 2024 -0400 double backslash for math commit 619a48dea00a0a67b8d4e542948200dc23e9589a Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 13:33:00 2024 -0400 add examples commit 527ae2670dd3a37b25a192ab49dad7f4c16dd5aa Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 13:24:13 2024 -0400 form should default to None if form is None, check if characteristic of field divides order of group. if so, it's the modular case, and we return the projection onto idempotents change-of-basis, i.e. the MFT. if not, use semi-normal form. commit 488e931da94f4872047ff8e839edc4a0c99b163e Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 13:09:54 2024 -0400 missing parantheses commit 7e96c904b6bc47d2108930c039b95a14155d425b Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 13:00:57 2024 -0400 Update src/sage/combinat/symmetric_group_algebra.py Co-authored-by: Travis Scrimshaw <clfrngrown@aol.com> commit cf27c64dff0b77c6cca162849ec923ae14f4920f Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 13:00:14 2024 -0400 Update src/sage/combinat/symmetric_group_algebra.py I thought I could do this, don't know why I convinced myself I needed flatten. Co-authored-by: Travis Scrimshaw <clfrngrown@aol.com> commit 92699f42eba524047b09d527981dab2456802958 Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 12:59:16 2024 -0400 Update src/sage/combinat/symmetric_group_algebra.py Co-authored-by: Travis Scrimshaw <clfrngrown@aol.com> commit 68f2ef140f91356f55290d100ad8a5282f19c206 Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 12:58:30 2024 -0400 Update src/sage/combinat/symmetric_group_algebra.py Co-authored-by: Travis Scrimshaw <clfrngrown@aol.com> commit 618d5e1a8cb11006ef9fa6642e33f06d5728d6ae Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 12:57:18 2024 -0400 Update src/sage/combinat/symmetric_group_algebra.py Co-authored-by: Travis Scrimshaw <clfrngrown@aol.com> commit 04db911e3f6a08b3be1bf5c295e913a40ee8b28a Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 12:56:37 2024 -0400 Update src/sage/combinat/symmetric_group_algebra.py I see, so "None" if the user doesn't specify, and we determine which form based on whether p|n!. Sounds good. Co-authored-by: Travis Scrimshaw <clfrngrown@aol.com> commit e86f79410ebda01b8a85598a5754f9c383336ae1 Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 11:09:57 2024 -0400 add doctest to ensure issue 37751 is fixed commit f8c64d16b3d1c7b3d338d9d2339ae2e284df3c58 Merge: aedd2f3927 3079e2d897 Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 10:55:44 2024 -0400 Merge branch 'pr/37748' of https://github.com/jacksonwalters/sage into pr/37748 commit aedd2f3927aac05dceeb97bd484afa82557652a8 Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 10:55:43 2024 -0400 remove multiplication stars commit 3079e2d897396222b0f23784a4d07a40f25b3c1f Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 10:54:38 2024 -0400 Update src/sage/combinat/symmetric_group_algebra.py Co-authored-by: grhkm21 <83517584+grhkm21@users.noreply.github.com> commit b14565299a6f5a82e5fae5080de88b455ce696e5 Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 10:51:59 2024 -0400 docstring lines are wrapped at 80 characters ensure each line is less than 80 characters commit e3fcc1f59c4668f88c184649c18627480365e646 Author: Jackson Walters <jacksonwalters@gmail.com> Date: Mon Apr 8 10:24:51 2024 -0400 keep form argument keep form since there is a one year deprecation warning before removing arguments. show deprecation warning when default seminormal form is called, but suggest modular form which works in all cases commit 6d187b153b5b5c3e29fe07d9efe925e34bf5baa5 Merge: 551139c09f c8d260a6fe Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 23:34:58 2024 +0200 gh-37764: ECM-related tests fail after an incremental build ### TL;DR: This is because setting the default ECMBIN=ecm is only done when using the configure test, but not when skipping the configure test because the spkg is already installed. ### Steps to reproduce: * ecm is not installed in the base os * Sage make distclean && make succeeeds * Tests succeed * re-running ./bootstrap && ./configure && make works * Tests now fail with PermissionError: [Errno 13] Permission denied: '' ### Dependencies See also: * https://github.com/sagemath/sage/pull/37701 * https://github.com/sagemath/sage/pull/37011#issuecomment-2023089743 URL: https://github.com/sagemath/sage/pull/37764 Reported by: Volker Braun Reviewer(s): Matthias Köppe commit 71ad110e196d9f17934d030ce1c000e11b50fee8 Author: Jackson Walters <jacksonwalters@gmail.com> Date: Sun Apr 7 19:57:42 2024 -0400 trigger GitHub actions commit c8d260a6fe37cb0379690e57dab6fd2a3d5d8048 Author: Volker Braun <tarpit@vbraun.imap.cc> Date: Sun Apr 7 22:13:11 2024 +0200 Update build/pkgs/ecm/spkg-configure.m4 Co-authored-by: Matthias Köppe <mkoeppe@math.ucdavis.edu> commit fbf896681da3ad1fad6c164e4d5a7cf452ec2eec Author: Volker Braun <tarpit@vbraun.imap.cc> Date: Sun Apr 7 21:35:42 2024 +0200 Update build/pkgs/ecm/spkg-configure.m4 Co-authored-by: Matthias Köppe <mkoeppe@math.ucdavis.edu> commit 6531f7e415230d62fecd0a38fd804081341a9f32 Author: Volker Braun <tarpit@vbraun.imap.cc> Date: Sun Apr 7 21:35:38 2024 +0200 Update build/pkgs/ecm/spkg-configure.m4 Co-authored-by: Matthias Köppe <mkoeppe@math.ucdavis.edu> commit af0e24286a62a06d85e1baf49d0dc8f624965e5c Author: Volker Braun <vbraun.name@gmail.com> Date: Sun Apr 7 19:17:37 2024 +0200 ECM-related tests fail after an incremental build This is because setting the default ECMBIN=ecm is only done when using the configure test, but not when skipping the configure test because the spkg is already installed. Steps to reproduce: * ecm is not installed in the base os * Sage make distclean && make succeeeds * Tests succeed * re-running ./bootstrap && ./configure && make works * Tests now fail with PermissionError: [Errno 13] Permission denied: '' See also: * https://github.com/sagemath/sage/pull/37701 * https://github.com/sagemath/sage/pull/37011#issuecomment-2023089743 commit 551139c09f26a5da96b1187c3f0dd17b8d80ef84 Merge: 03be519274 547d502ed5 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:43 2024 +0200 gh-37763: Fix tolerance for scipy 1.13 After updating scipy to 1.13: ``` ********************************************************************** File "pkgs/sagemath-standard/build/lib.linux-x86_64-cpython- 312/sage/matrix/matrix_double_dense.pyx", line 3686, in sage.matrix.matrix_double_dense.Matrix_double_dense.exp Failed example: A.exp() # tol 1.1e-14 # needs sage.symbolic Expected: [-19.614602953804912 + 12.517743846762578*I 3.7949636449582176 + 28.88379930658099*I] [ -32.383580980922254 + 21.88423595789845*I 2.269633004093535 + 44.901324827684824*I] Got: [-19.61460295380496 + 12.517743846762578*I 3.7949636449581874 + 28.88379930658104*I] [-32.38358098092232 + 21.884235957898436*I 2.2696330040934853 + 44.901324827684896*I] Tolerance exceeded in 1 of 8: 2.269633004093535 vs 2.2696330040934853, tolerance 3e-14 > 1.1e-14 ********************************************************************** ``` This PR raises the tolerance to `3e-14` to fix this. ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37763 Reported by: Gonzalo Tornaría Reviewer(s): Antonio Rojas commit 03be519274f1e3aa424ea729f406180a10df1498 Merge: 4d88e898e2 cd41d1e414 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:37 2024 +0200 gh-37755: src/sage/graphs/graph_decompositions/tdlib.pyx: Use -std=c++11       This is needed with the current Xcode command line tools on macOS with boost from homebrew. Similar to https://github.com/sagemath/sage/pull/37646#issuecomment-2017108199 ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37755 Reported by: Matthias Köppe Reviewer(s): David Coudert commit 4d88e898e203d9b7297d224f40d659c213701ccb Merge: 9971cec642 6722976389 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:34 2024 +0200 gh-37753: Added rank check to doctest in `.gauss_on_polys` Fixes #37732. Also removed assignment of vectors `v` in the same doctest since they are not used. URL: https://github.com/sagemath/sage/pull/37753 Reported by: Sebastian A. Spindler Reviewer(s): Martin Rubey commit 9971cec64239679ee00ee7b22415b98ed53fbbf6 Merge: 3df3305a64 80239bc925 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:31 2024 +0200 gh-37736: Reimplementing the Witt (Sym Func) change of basis, caches, and omega involution       We reimplement the coercion from the Witt basis of symmetric functions to the $h$, $e$, and $p$ bases and their inverses by using the generating function identities and recursion for $w_n, h_n, e_n, p_n$ and the mulitplicative basis property for the general shape (in particular, we attempt to minimize the number of multiplications and maximize the use of the cache by recursively splitting the partition by its smallest part). We cache the result using `@cached_method` instead of custom classes. We also deprecate the `coerce_*` inputs. We also take advantage that $\omega w_n = w_n$ when $n$ is odd to compute the omega involution. This extends to a faster computation of the antipode. ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [x] I have created tests covering the changes. - [x] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37736 Reported by: Travis Scrimshaw Reviewer(s): Darij Grinberg, Martin Rubey commit 3df3305a6493f88ce86a1bc73e17c34ad01fbe33 Merge: 01333dfec1 ddcb3e45ca Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:27 2024 +0200 gh-37725: Fix typo/phrasing in README.md       - Removed the phrase "It assumes that you have already cloned the git repository" from the setup guide since the guide actually contains instructions on cloning the repository - Removed a random misplaced period ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37725 Reported by: Faisal Reviewer(s): gmou3 commit 01333dfec1ae7a518260c0178d4a0b254b886157 Merge: bcbb58f10a 8378fe6fdd Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:24 2024 +0200 gh-37721: Renamed "ring" argument of matrix contructor to "base_ring"       The matrix constructor used to accept ``ring`` as argument to specify a the base ring of a matrix. It now takes ``base_ring`` instead. ``ring`` is still accepted but deprecated (I added a deprecation in the code). This is for unification (``base_ring`` is what users expect from their experience with other classes). This solves @mkoeppe's issue #33380. ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [x] I have created tests covering the changes. - [x] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37721 Reported by: SandwichGouda Reviewer(s): Matthias Köppe commit bcbb58f10a840f43c1359934b85bfef9c143731c Merge: 08d5d97336 9b3ddf95fd Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:21 2024 +0200 gh-37720: pkgs/sagemath-repl/pyproject.toml.m4: Declare extra 'sphinx'       As discussed in https://github.com/sagemath/sage/pull/37589#issuecomment-2030456264, the Sage documentation system uses Sphinx to format the documentation in the terminal, for example in the help system. For the modularization project, a simple fallback for the case of Sphinx not being present was added. Here we declare the corresponding "extra" so that users can do `pip install sagemath-repl[sphinx]`. ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37720 Reported by: Matthias Köppe Reviewer(s): commit 08d5d97336d5cb87f4e9731b724b320e8cef771f Merge: 253169ff7b 5a61d3c103 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:18 2024 +0200 gh-37712: src/tox.ini (rst): Add missing Sphinx roles       Fixes #37711 ### :memo: Checklist  - [x] The title is concise and informative. - [ ] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37712 Reported by: Matthias Köppe Reviewer(s): Frédéric Chapoton commit 253169ff7b83df66d20846983ac9ce2982f3a18e Merge: 35f34ee0f6 3dc65b5797 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:16 2024 +0200 gh-37707: Implement an iterator for absolute number fields       We use the $\mathbb{Q}$ vector space structure. Although because of #37706, we go through `cartesian_product`. If there is an easy way to implement this in full generality (i.e., all number fields), please let me know. ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [x] I have created tests covering the changes. - [x] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37707 Reported by: Travis Scrimshaw Reviewer(s): Matthias Köppe commit 35f34ee0f66904d9f2766b646664ae67cedd3aae Merge: 0a16c78b4e e077f1edf7 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:11 2024 +0200 gh-37703: Implement an is_nilpotent() method for matrices       Checking if a matrix is nilpotent is a fundamental test that Sage is (surprisingly) missing. We check this by using the char poly. ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [x] I have created tests covering the changes. - [x] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37703 Reported by: Travis Scrimshaw Reviewer(s): Frédéric Chapoton, Martin Rubey commit 0a16c78b4e2b677ba30b198fdc781d38e595c9d8 Merge: 9d788cb197 7e8aa98fa6 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:04 2024 +0200 gh-37699: Add `approximate_closest_vector` to `IntegerLattice`, using Babai's nearest plane algorithm This PR implements Babai's nearest plane algorithm to find vectors close to a given vector in an integer lattice. I would appreciate if someone could verify the bound stated in the docstring, it is inferred from Babai's original paper but parameterized to use the `delta` parameter of the LLL algorithm. Ideas for improvements: - The doctest is currently the same as for `closest_vector`, this could be expanded - The `delta` parameter for LLL needs to be kept track of by the user. How "much" the basis has been reduced (i.e to how high a `delta` value) could be kept track of by the class and if a further reduction is needed for some given bound then it could be performed. ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [x] I have created tests covering the changes. - [x] I have updated the documentation accordingly. URL: https://github.com/sagemath/sage/pull/37699 Reported by: TheBlupper Reviewer(s): grhkm21, TheBlupper commit 9d788cb19732f9f750c6d5a25c28ccb958414112 Merge: 6925269840 d63ab2607e Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 14:00:02 2024 +0200 gh-37689: Fix changes.html in doc preview       by changing the base doc url. The base doc url is where the doc preview for the released version is uploaded. It changed recently: https://github.com/sagemath/sage/issues/34874#issuecomment-1870589626, but we didn't fix accordingly here: [.ci/create-changes- html.sh](https://github.com/sagemath/sage/blob/develop/.ci/create- changes-html.sh) ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37689 Reported by: Kwankyu Lee Reviewer(s): commit 6925269840b81d3a1085ae900037e71dfd440c87 Merge: 12a003eddd 3b137b50a0 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:59 2024 +0200 gh-37688: typo Small typo URL: https://github.com/sagemath/sage/pull/37688 Reported by: Dennis Yurichev Reviewer(s): Martin Rubey commit 12a003eddd08bcac14c88becd82bf52af599fe6c Merge: 3c9a67a2f0 6184336d73 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:56 2024 +0200 gh-37687: provide a construction functor using a single functor for families We enable sage to find common parents also when symmetric functions are involved. In particular, we provide a functorial construction that takes a commutative (coefficient) ring and produces a (commutative) ring of symmetric functions over this ring. For Macdonald polynomials and other symmetric functions that involve parameters we cheat a bit: the categories should actually be commutative rings with distinguished elements (namely, the parameters). However, having these is very likely overkill. This is an alternative to #37220, relying on individual construction functors instead of passing around strings, and #37686, which provides an individual functor for each family. URL: https://github.com/sagemath/sage/pull/37687 Reported by: Martin Rubey Reviewer(s): Travis Scrimshaw commit 3c9a67a2f0f8d42c1167f7fa192c70d29cdf6766 Merge: 1ab35e3a69 31645a4f35 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:53 2024 +0200 gh-37683: some shortcuts using bool fixes suggested by `ruff --select=SIM210` namely use `bool` for concision ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37683 Reported by: Frédéric Chapoton Reviewer(s): Matthias Köppe, Travis Scrimshaw commit 1ab35e3a696baf5a2fdd866a670c32611fa4a996 Merge: 654ed08b89 449864f0af Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:49 2024 +0200 gh-37680: a few more uses of "in Fields()" Just replacing some tests by `R in Fields()` in some pyx files outside the `rings` folder ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37680 Reported by: Frédéric Chapoton Reviewer(s): Matthias Köppe commit 654ed08b899e3a0bcd2d72a9ea66e5f1a96569a8 Merge: 042153b0d9 a8671aef33 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:44 2024 +0200 gh-37679: simplifications in symmetric_ideal.py some simplifications in `symmetric_ideal` ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37679 Reported by: Frédéric Chapoton Reviewer(s): Matthias Köppe commit 042153b0d976f8ec7b7489c404a17084321e3e06 Merge: 0de7848873 5d4a91fa85 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:41 2024 +0200 gh-37677: some simplifications in moment-angle complex a bunch of code details in the modified file fixes #36217 ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. URL: https://github.com/sagemath/sage/pull/37677 Reported by: Frédéric Chapoton Reviewer(s): Travis Scrimshaw commit 0de7848873bcda21b916e13ad71dd0dbe6aa9ff0 Merge: 014ccc31c4 e0ef88414d Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:38 2024 +0200 gh-37674: some pep8 and ruff cleanups in modules/ some ruff and pep8 cleanup in the `modules` folder ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37674 Reported by: Frédéric Chapoton Reviewer(s): Travis Scrimshaw commit 014ccc31c4118b929bb46c0811331f828041aee4 Merge: 93cc706453 42b4b671c3 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:34 2024 +0200 gh-37673: upgrade msolve to 0.6.5 routine upgrade, dropping already applied patch URL: https://github.com/sagemath/sage/pull/37673 Reported by: Dima Pasechnik Reviewer(s): Dima Pasechnik, Marc Mezzarobba, Matthias Köppe commit 93cc7064532b3456e5520833121e761078078576 Merge: cf431c8d40 f9ddc324b8 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:31 2024 +0200 gh-37672: src/sage/modular/quasimodform/ring.py: Fix pycodestyle warning       ### :memo: Checklist  - [ ] The title is concise and informative. - [ ] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37672 Reported by: Matthias Köppe Reviewer(s): Frédéric Chapoton commit cf431c8d40397205bb0f2e1f2843beec16788236 Merge: 2c71f81f0d 378ec74794 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:28 2024 +0200 gh-37671: pkgs/sagemath-standard: Support gpep517 From https://github.com/sagemath/sage/pull/37138#issuecomment-2018937633       ### :memo: Checklist  - [x] The title is concise and informative. - [ ] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37671 Reported by: Matthias Köppe Reviewer(s): François Bissey commit 2c71f81f0d804932deb11f3d28eb6e7962593076 Merge: 5801382f06 b8757d814e Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:24 2024 +0200 gh-37668: Wrong results in is_isotopic method of the Link class for certain chiral link       The following result is wrong, since `K10_67` is a chiral knot: ``` sage: L = KnotInfo.K10_67 ....: L1 = L.link() ....: L1r = L.link().reverse() ....: L1.is_isotopic(L1r) True ``` The problem is that the `get_knotinfo` method currently does not distinguish between all four symmetry mutants of a chiral knot or link. In the example, `get_knotinfo` detects both knots as *not mirrored* to the symmetry mutant of `K10_67` recorded in the KnotInfo database, as you can see from the verbose messages: ``` sage: set_verbose(1) sage: L1.is_isotopic(L1r) verbose 1 (4365: link.py, is_isotopic) KnotInfo other: KnotInfo.K10_67 mirrored False True ``` The wrong conclusion in `is_isotopic` is that they must be isotopic. The idea in this PR is to replace the boolean `mirrored` in the output of `get_knotinfo` with the value of an enum representing all symmetry mutants for an entry of the KnotInfo database. Of course, this requires major revisions in the `get_knotinfo` and `is_isotopic` methods. The key to this is a new local method `_knotinfo_matching_dict` that builds on `_knotinfo_matching_list` and returns the matching list for all symmetry mutants. On the occasion of that revision, I also fix other wrong results for example this one of a multi-component link which is not amphicheiral, but reversible: ``` sage: L = KnotInfo.L6a2_0 sage: L1 = L.link() sage: L2 = L.link(L.items.braid_notation) sage: L1.is_isotopic(L2) False ``` This bug is due to `get_knotinfo` returning the wrong mirror version for `L1` concerning `L6a2_1`: ``` ... sage: L1.get_knotinfo(unique=False) [(<KnotInfo.L6a2_0: 'L6a2{0}'>, False), (<KnotInfo.L6a2_1: 'L6a2{1}'>, False)] sage: L2.get_knotinfo(unique=False) [(<KnotInfo.L6a2_0: 'L6a2{0}'>, False), (<KnotInfo.L6a2_1: 'L6a2{1}'>, True)] ``` In addition to these fixes, improvements and corresponding adjustments, I correct some document formatting. ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [x] I have created tests covering the changes. - [x] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37668 Reported by: Sebastian Oehms Reviewer(s): Sebastian Oehms, Travis Scrimshaw commit 5801382f0638c7c8f50ae5a906b85306e5350fb5 Merge: cf365e5e51 bcc326d972 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:11 2024 +0200 gh-37667: Fix noexcept clauses (#37560) In #36507 I added a lot of `noexcept` clauses guided by a warning which is not quite right (see https://github.com/cython/cython/pull/5999#issuecomment-1986868208). A new warning in 3.0.9 shows the mistake (incorrectly added `noexcept` clauses). This broke some doctests (#37560) and a workaround was implemented for 10.3 (#37583) since we were too close to release. This PR now does the proper fix, removing all the incorrect `noexcept`, and also adding a few missing `noexcept`. Note: if one tries this PR with cython <= 3.0.8 it seems it's wrong in the sense that it will show 40k more warnings after the PR. These are *incorrect* warnings. If one uses cython 3.0.9 each of these 40k lines give a correct warning before this PR and an incorrect warnings after the PR, and there is no way to avoid 40k warnings. To confirm this PR is a good one, one needs to use cython 3.0.9 + https://github.com/cython/cython/pull/6087 or wait for cython 3.0.10. In this case, the warnings are correct and one will get 40k warnings before the PR and no warning after the PR. The last commit reverts the workaround from #37583 so we don't silence these warnings on `cython()` now that we have our own code right. @vbraun: this is "almost trivial" but touches too many files. Could we merge it in beta0 to avoid conflicts? I did this "kind of" automatically and I've been testing it for two weeks in almost all my builds. The changes made should not affect behaviour at all in the legacy mode we are using cython. ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. URL: https://github.com/sagemath/sage/pull/37667 Reported by: Gonzalo Tornaría Reviewer(s): Gonzalo Tornaría, Matthias Köppe commit cf365e5e51d30495f823ed928e0fe027166545d0 Merge: 59bf21efe7 a47b98b32e Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:08 2024 +0200 gh-37663: sage/rings/{complex,real}*: Untitlecase titles, add refs to libraries       This cleans up the table of contents https://deploy-preview-37663-- sagemath.netlify.app/html/en/reference/rings_numerical/ ### :memo: Checklist  - [x] The title is concise and informative. - [ ] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    - Depends on #37406 (merged here to resolve merge conflict) URL: https://github.com/sagemath/sage/pull/37663 Reported by: Matthias Köppe Reviewer(s): Marc Mezzarobba commit 59bf21efe755cac20496bb417285a7e3545bc790 Merge: 2fd04522f3 210c5d8389 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:06 2024 +0200 gh-37661: some cross-references btw doc of RR, RBF, ... (mainly intended to make RBF and CBF, which are the most powerful implementations for most purposes, more discoverable) URL: https://github.com/sagemath/sage/pull/37661 Reported by: Marc Mezzarobba Reviewer(s): commit 2fd04522f3caaca5fff0f78f1f1dccc6c0027475 Merge: 03f81897de 16a5261134 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:59:02 2024 +0200 gh-37660: Fix e302 libs pyx This is adding missing empty lines (pycodestyle E302) in pyx files in the `modules` and `libs` folders. Also some little tweak in code in the singular libs files. ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37660 Reported by: Frédéric Chapoton Reviewer(s): David Coudert commit 03f81897de2bf0aaf6b7fa82982b17251b0c9fd1 Merge: 81b4ddccd1 23c45b1d61 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:59 2024 +0200 gh-37659: fix ruff codes UP012 and UP023 scripted using `ruff` ; only two minor changes see https://docs.astral.sh/ruff/rules/#pyupgrade-up ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37659 Reported by: Frédéric Chapoton Reviewer(s): Matthias Köppe commit 81b4ddccd1de963f7731661106e1e425efd2e76b Merge: 74a82d7ce4 4fa6f09984 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:54 2024 +0200 gh-37655: Removed caching of determinant in LLL for the `NTL:LLL`-algorithm The method previously computed the square root of the squared determinant to cache, thus caching the absolute value of the determinant instead of the determinant. Therefore we remove this caching completely, which fixes #37236. URL: https://github.com/sagemath/sage/pull/37655 Reported by: Sebastian A. Spindler Reviewer(s): Vincent Delecroix commit 74a82d7ce445d8d9e66595887e9fcdf40f838702 Merge: 34539cdb4b 51d423d5a2 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:52 2024 +0200 gh-37652: fix typos and no "Algebra" in the doc This fixes a bunch of typos in the doc, and also rewrites one example to use `Parent` instead of the auld `Algebra` ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37652 Reported by: Frédéric Chapoton Reviewer(s): Matthias Köppe commit 34539cdb4b65fbb8f34f738c15f41a800de25141 Merge: 7a5ba77a93 5f32f5f882 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:49 2024 +0200 gh-37651: use parent in asymptotic ring trying to use the modern `Parent` and categories in the asymptotic ring, instead of the auld `Algebra` ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37651 Reported by: Frédéric Chapoton Reviewer(s): Marc Mezzarobba commit 7a5ba77a93cef0ee7a44f4228070e48b890d2b73 Merge: d4ab336474 4fa36454b5 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:45 2024 +0200 gh-37640: Modified `.is_reduced()` of `binary_qf.py` to avoid square root computation Adapted the checks in `.is_reduced()` to avoid the square root computation, which fixes #37635. Furthermore added a non-singularity check. This adaption is based on the observation that $$|\sqrt{D} - 2 \cdot |a|| < b < \sqrt{D}$$ if and only if $$b > 0, \ a \cdot c < 0 \text{ and } (a-c)^2 < D$$ are all satisfied simultaneously (where $D = b^2 - 4ac > 0$). The above can be proven in a straightforward manner by taking squares and square roots while making sure that the numbers being squared are always positive, so that inequalities are preserved. URL: https://github.com/sagemath/sage/pull/37640 Reported by: Sebastian A. Spindler Reviewer(s): Lorenz Panny commit d4ab336474b9f40b9e7f637c9b3ba6ac8bbc2135 Merge: 3c174ef6a6 bf0c3695cc Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:41 2024 +0200 gh-37639: add Legendre transform and suspension for lazy symmetric functions This adds two useful functions, particularly important when working with operads and Koszul duality. ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [x] I have created tests covering the changes. - [x] I have updated the documentation accordingly. URL: https://github.com/sagemath/sage/pull/37639 Reported by: Frédéric Chapoton Reviewer(s): Martin Rubey, Travis Scrimshaw commit 3c174ef6a69d80e733dc7461b44fbc8b646ac30d Merge: 6ab45f666b d7021096c4 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:38 2024 +0200 gh-37638: use strings as label in modular-decomposition trees Fixing #37631 and adding a note saying that labels of rooted trees must be comparable ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [x] I have created tests covering the changes. - [x] I have updated the documentation accordingly. URL: https://github.com/sagemath/sage/pull/37638 Reported by: Frédéric Chapoton Reviewer(s): cyrilbouvier commit 6ab45f666ba665ee9a0c517233830bafbea544b9 Merge: 8e26e27572 c4c26ee1bf Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:36 2024 +0200 gh-37637: Update jupyter-sphinx to version 0.5.3 and pin thebe to version 0.8.2       jupyter-sphinx that we are shipping was patched to use thebe@latest, which is the cause of the current breakdown of the sage live doc. The patch to jupyter-sphinx is partly based on https://github.com/jupyter/jupyter-sphinx/pull/231 ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37637 Reported by: Kwankyu Lee Reviewer(s): Matthias Köppe commit 8e26e27572073bc9f54839143f52d2d7e4c710e0 Merge: 13a555d970 af97e6da28 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:28 2024 +0200 gh-37634: The Glaisher-Franklin bijections on integer partitions The Glaisher-Franklin bijections, for a positive integer $s$, map the set of parts divisible by $s$ to the set of parts which occur at least $s$ times, see https://www.findstat.org/MapsDatabase/Mp00312 for the $s=2$ case. This generalizes and streamlines the code from findstat. URL: https://github.com/sagemath/sage/pull/37634 Reported by: Martin Rubey Reviewer(s): Martin Rubey, Travis Scrimshaw commit 13a555d970eefceddd20728c6a3275447713ea7e Merge: 58e8c578fd d45422e64e Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:23 2024 +0200 gh-37606: Border matrix       Plain TeX users may remember `\bordermatrix`. Here we implement this as options of the `Matrix` class's `str` method. ``` sage: M = matrix([[1,2,3], [4,5,6], [7,8,9]]) sage: M.subdivide(None, 2) sage: print(M.str(unicode=True, ....: top_border=['ab', 'cde', 'f'], ....: bottom_border=['*', '', ''], ....: left_border=[1, 10, 100], ....: right_border=['', ' <', ''])) ab cde f 1⎛ 1 2│ 3⎞ 10⎜ 4 5│ 6⎟ < 100⎝ 7 8│ 9⎠ * ``` Follow-up PR: As the guiding application for this feature, we equip finite-dimensional modules with basis with methods `_repr_matrix`, `_ascii_art_matrix`, `_unicode_art_matrix` that can be used as in this example: ``` sage: M = matrix(ZZ, [[1, 0, 0], [0, 1, 0]], ....: column_keys=['a', 'b', 'c'], ....: row_keys=['v', 'w']); M Generic morphism: From: Free module generated by {'a', 'b', 'c'} over Integer Ring To: Free module generated by {'v', 'w'} over Integer Ring sage: M._unicode_art_ = M._unicode_art_matrix sage: unicode_art(M) # indirect doctest a b c v⎛1 0 0⎞ w⎝0 1 0⎠ ``` ### :memo: Checklist  - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. - [ ] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies    URL: https://github.com/sagemath/sage/pull/37606 Reported by: Matthias Köppe Reviewer(s): David Coudert, Matthias Köppe commit 58e8c578fdc79d94dba5de30ae3a5ba6c08630ef Merge: 294eed4041 c1d38c77d2 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:20 2024 +0200 gh-37563: trying a change in unique_representation namely break the loop over `mro` once found what we want I am not sure at all that this keeps the expected properties, but this is more efficient.. ### :memo: Checklist - [x] The title is concise and informative. - [x] The description explains in detail what this PR is about. URL: https://github.com/sagemath/sage/pull/37563 Reported by: Frédéric Chapoton Reviewer(s): Travis Scrimshaw commit 294eed4041447cb7420462f9548bc9195e376565 Merge: 14da417689 def8e56c30 Author: Release Manager <release@sagemath.org> Date: Sun Apr 7 13:58:17 2024 +0200 gh-37546: Sphinx ext links for Sage source files       In the documentation, we can now write `` :sage_root:`src/setup.py` `` or `` :sage_root:`src/doc/en/installation` ``, and it will be formatted uniformly, and a link to the file in the repository will be created. - Rebased version of #33756 Fixes #33756. ### :memo: Checklist  - [x] The title is concise and informative. - [ ] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### :hourglass: Dependencies  …

tobiasdiez · 2024-04-11T03:49:56Z

Setting to positive review based on the counting in #37138 (comment)

github-actions · 2024-04-11T04:28:21Z

Documentation preview for this PR (built with commit 5bcb9d4; changes) is ready! 🎉
This preview will update shortly after each push to this PR.

vbraun · 2024-04-13T06:11:35Z

merge conflict

… to all files, remove remaining `# coding: utf-8`" This reverts commit 6ecb1d8, reversing changes made to 8ea5214.

As noted in sagemath#37024 and in sagemath#36489, `sage-conf` is actually not needed on a few systems now. For this reason, we make the installation of it optional (as there are no optional install requires as far as I know, this means we remove it from `pyproject.toml` and `setup.cfg`). We also remove it from "install all dependencies via conda" as its not needed there.

tobiasdiez · 2024-04-14T15:13:09Z

merge conflict

Cleanly rebased on top of #37796.

vbraun · 2024-04-16T22:20:22Z

merge conflict

tobiasdiez commented Jan 22, 2024

View reviewed changes

tornaria mentioned this pull request Jan 23, 2024

Support python 3.12 on sagemath-standard #36407

Merged

tobiasdiez added the s: needs review label Feb 11, 2024

dimpase force-pushed the conf-optional branch from 7442c71 to 617d26a Compare March 30, 2024 11:42

kwankyu reviewed Apr 8, 2024

View reviewed changes

tobiasdiez added the s: positive review label Apr 11, 2024

vbraun added the s: needs work label Apr 13, 2024

roed314 and others added 3 commits April 13, 2024 02:36

Revert "sagemathgh-36964: Add # sage_setup: distribution directives…

f16eb4b

… to all files, remove remaining `# coding: utf-8`" This reverts commit 6ecb1d8, reversing changes made to 8ea5214.

Remove outdated comment

5bcb9d4

tobiasdiez force-pushed the conf-optional branch from 67c1e61 to 5bcb9d4 Compare April 14, 2024 15:11

tobiasdiez removed the s: needs work label Apr 14, 2024

vbraun added the s: needs work label Apr 16, 2024

tornaria mentioned this pull request Apr 23, 2024

sagemath-standard: include sage_setup in sdist #37287

Merged

2 tasks

vbraun merged commit 5bcb9d4 into sagemath:develop Apr 27, 2024
14 of 33 checks passed

github-actions bot removed s: positive review s: needs work labels Apr 27, 2024

mkoeppe added this to the sage-10.4 milestone Apr 27, 2024

tobiasdiez deleted the conf-optional branch May 7, 2024 07:00

Make sage-conf optional #37138

Make sage-conf optional #37138

Conversation

tobiasdiez commented Jan 22, 2024 • edited Loading

📝 Checklist

⌛ Dependencies

tobiasdiez left a comment

Choose a reason for hiding this comment

mkoeppe commented Jan 22, 2024

tornaria commented Jan 22, 2024

mkoeppe commented Jan 22, 2024

tornaria commented Jan 22, 2024

mkoeppe commented Jan 22, 2024

tornaria commented Jan 22, 2024

mkoeppe commented Jan 22, 2024

tornaria commented Jan 22, 2024

tornaria commented Jan 22, 2024

mkoeppe commented Jan 22, 2024

tornaria commented Jan 22, 2024

mkoeppe commented Jan 22, 2024

mkoeppe commented Jan 22, 2024

mkoeppe commented Jan 22, 2024

tornaria commented Jan 22, 2024

mkoeppe commented Jan 22, 2024

tornaria commented Jan 22, 2024

mkoeppe commented Jan 22, 2024

orlitzky commented Jan 23, 2024

mkoeppe commented Jan 23, 2024

orlitzky commented Jan 23, 2024

tornaria commented Jan 23, 2024

orlitzky commented Jan 23, 2024

mkoeppe commented Jan 23, 2024

orlitzky commented Jan 23, 2024

mkoeppe commented Jan 23, 2024

tobiasdiez commented Jan 24, 2024

kwankyu commented Mar 29, 2024

kwankyu commented Mar 29, 2024 • edited Loading

dimpase commented Mar 29, 2024 • edited Loading

standard package classification (sans sagelib+sage-conf+bootstrap)

tornaria commented Apr 8, 2024

kwankyu Apr 8, 2024

Choose a reason for hiding this comment

tobiasdiez Apr 11, 2024

Choose a reason for hiding this comment

tobiasdiez commented Apr 11, 2024

github-actions bot commented Apr 11, 2024 • edited Loading

vbraun commented Apr 13, 2024

tobiasdiez commented Apr 14, 2024

vbraun commented Apr 16, 2024

tobiasdiez commented Jan 22, 2024 •

edited

Loading

kwankyu commented Mar 29, 2024 •

edited

Loading

dimpase commented Mar 29, 2024 •

edited

Loading

github-actions bot commented Apr 11, 2024 •

edited

Loading