Some magics (e. g %%mathematica) have vanished

[EmmanuelCharpentier]

Is there an existing issue for this?

• I have searched the existing issues for a bug report that matches the one I want to file, without success.

Did you read the documentation and troubleshoot guide?

• I have read the documentation and troubleshoot guide

Environment

- **OS**: Debian testing
- **Sage Version**: 10.0.beta1 and 10.0.beta3

Steps To Reproduce

sage: %%mathematica
....: D[Sin[x], x]
....: 
....: 

Expected Behavior

Cos[x]

Actual Behavior

UsageError: Cell magic `%%mathematica` not found.

Additional Information

FWIW, in a console :

sage: %display ascii_art # see https://github.com/sagemath/sage/issues/30661
sage: %lsmagic
Available line magics:
%alias  %alias_magic  %attach  %autoawait  %autocall  %autoindent  %automagic  %bookmark  %cat  %cd  %clear  %colors  %conda  %config  %cp  %cpaste  %crun  %debug  %dhist  %dirs  %display  %doctest_mode  %ed  %edit  %env  %gap  %gp  %gui  %hist  %history  %iload  %killbgscripts  %ldir  %less  %lf  %lk  %ll  %load  %load_ext  %loadpy  %logoff  %logon  %logstart  %logstate  %logstop  %ls  %lsmagic  %lx  %macro  %magic  %magma  %man  %matplotlib  %maxima  %mkdir  %more  %mv  %notebook  %page  %paste  %pastebin  %pdb  %pdef  %pdoc  %pfile  %pinfo  %pinfo2  %pip  %polymake  %popd  %pprint  %precision  %prun  %psearch  %psource  %pushd  %pwd  %pycat  %pylab  %quickref  %recall  %rehashx  %reload_ext  %rep  %rerun  %reset  %reset_selective  %rm  %rmdir  %run  %runfile  %sage0  %save  %sc  %set_env  %singular  %store  %sx  %system  %tb  %time  %timeit  %unalias  %unload_ext  %who  %who_ls  %whos  %xdel  %xmode

Available cell magics:
%%!  %%HTML  %%SVG  %%bash  %%capture  %%cython  %%debug  %%file  %%fortran  %%gap  %%gp  %%html  %%javascript  %%js  %%latex  %%magma  %%markdown  %%maxima  %%perl  %%polymake  %%prun  %%pypy  %%python  %%python2  %%python3  %%ruby  %%sage0  %%script  %%sh  %%singular  %%svg  %%sx  %%system  %%time  %%timeit  %%writefile

Automagic is ON, % prefix IS NOT needed for line magics.

The %sage0 and %%sage0 entries look suspicious...

[EmmanuelCharpentier]

Not able to add/modify labels (I may not yet be in "triage"). Should be t:bug, c:core, p:minor

Furthermore @John H Palmieri, but don't know his Github handle...

[miguelmarco]

I also found that %%lisp has disappeared.

[miguelmarco]

My guess is that it has to do with moving their import to lazy_import in 5f08ad8

After that change, the creation of the magic commands ignores it:

sage/src/sage/repl/interface_magic.py

Lines 99 to 101 in 7404764

	for name, obj in sage.interfaces.all.__dict__.items():
	if isinstance(obj, sage.interfaces.interface.Interface):
	yield cls(name, obj)

[miguelmarco]

A workaround i found is to register the magic command manually:

from sage.interfaces.lisp import lisp
from sage.repl.interface_magic import InterfaceMagic
shell = get_ipython()
interface=InterfaceMagic("lisp", lisp)
shell.register_magic_function(interface.line_magic_factory(), magic_name="lisp", magic_kind='line')
shell.register_magic_function(interface.cell_magic_factory(), magic_name=interface._name, magic_kind='cell')

[jhpalmieri]

See also #34547 (comment).

[miguelmarco]

See also #34547 (comment).

Thanks. I guess we have the answer to your question about whether we care. Would it be possible to register the magic commands but still lazy_import them?

[jhpalmieri]

I don't know, partly because I don't know why lazily importing broke the commands.

[miguelmarco]

The code that registers the magic commands specifically checks if they are instances of Interface :

sage/src/sage/repl/interface_magic.py

Lines 99 to 101 in 7404764

	for name, obj in sage.interfaces.all.__dict__.items():
	if isinstance(obj, sage.interfaces.interface.Interface):
	yield cls(name, obj)

[jhpalmieri]

Should be fixed by #35680. Emmanuel, can you verify?

[EmmanuelCharpentier]

Seen a bit late... Sorry for the delay ! Le jeudi 25 mai 2023 à 09:58 -0700, John H. Palmieri a écrit :

Should be fixed by #35680. Emmanuel, can you verify?

Nice diagnosis ! Seems good...

…

-- Emmanuel Charpentier

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: ***@***.***>

[EmmanuelCharpentier]

Merging #35680 fixed some of these problems. But some other magics are still missing :

sage: %bg foo=(x/(x^5+a^5)).integrate(x, algorithm="fricas")
UsageError: Line magic function `%bg` not found.

The lack of %bg justifies (IMHO) upgrading the priority if this bug to major.

The related #30661 is still open :

sage: %display simple
sage: %lsmagic
<IPython.core.magics.basic.MagicsDisplay object at 0x7f25ff917650>
sage: %display ascii_art
sage: %lsmagic
Available line magics:
%Axiom  %ECM  %FriCAS  %Gap3  %Genus2reduction  %Gfan  %Giac  %Kash  %LiE  %Lisp  %Macaulay2  %Maple  %Mathematica  %Mathics  %Matlab  %Mupad  %Mwrank  %Octave  %PSage  %R  %alias  %alias_magic  %attach  %autoawait  %autocall  %autoindent  %automagic  %axiom  %bookmark  %cat  %cd  %clear  %colors  %conda  %config  %cp  %cpaste  %crun  %debug  %dhist  %dirs  %display  %doctest_mode  %ecm  %ed  %edit  %env  %four_ti_2  %fricas  %frobby  %gap  %gap3  %gap3_version  %genus2reduction  %gfan  %giac  %gnuplot  %gp  %gui  %hist  %history  %iload  %kash  %kash_version  %killbgscripts  %ldir  %less  %lf  %lie  %lisp  %lk  %ll  %load  %load_ext  %loadpy  %logoff  %logon  %logstart  %logstate  %logstop  %ls  %lsmagic  %lx  %macaulay2  %macro  %magic  %magma  %magma_free  %man  %maple  %mathematica  %mathics  %matlab  %matlab_version  %matplotlib  %maxima  %mkdir  %more  %mupad  %mv  %mwrank  %notebook  %octave  %page  %paste  %pastebin  %pdb  %pdef  %pdoc  %pfile  %pinfo  %pinfo2  %pip  %polymake  %popd  %povray  %pprint  %precision  %prun  %psearch  %psource  %pushd  %pwd  %pycat  %pylab  %qepcad  %qepcad_formula  %qepcad_version  %quickref  %r  %r_version  %read_data  %recall  %rehashx  %reload_ext  %rep  %rerun  %reset  %reset_selective  %rm  %rmdir  %run  %runfile  %sage0  %save  %sc  %scilab  %set_env  %singular  %store  %sx  %system  %tachyon_rt  %tb  %time  %timeit  %unalias  %unload_ext  %who  %who_ls  %whos  %xdel  %xmode

Available cell magics:
%%!  %%Axiom  %%ECM  %%FriCAS  %%Gap3  %%Genus2reduction  %%Gfan  %%Giac  %%HTML  %%Kash  %%LiE  %%Lisp  %%Macaulay2  %%Maple  %%Mathematica  %%Mathics  %%Matlab  %%Mupad  %%Mwrank  %%Octave  %%PSage  %%R  %%SVG  %%axiom  %%bash  %%capture  %%cython  %%debug  %%ecm  %%file  %%fortran  %%four_ti_2  %%fricas  %%frobby  %%gap  %%gap3  %%gap3_version  %%genus2reduction  %%gfan  %%giac  %%gnuplot  %%gp  %%html  %%javascript  %%js  %%kash  %%kash_version  %%latex  %%lie  %%lisp  %%macaulay2  %%magma  %%magma_free  %%maple  %%markdown  %%mathematica  %%mathics  %%matlab  %%matlab_version  %%maxima  %%mupad  %%mwrank  %%octave  %%perl  %%polymake  %%povray  %%prun  %%pypy  %%python  %%python2  %%python3  %%qepcad  %%qepcad_formula  %%qepcad_version  %%r  %%r_version  %%read_data  %%ruby  %%sage0  %%scilab  %%script  %%sh  %%singular  %%svg  %%sx  %%system  %%tachyon_rt  %%time  %%timeit  %%writefile

Automagic is ON, % prefix IS NOT needed for line magics.
sage: %display typeset
sage: %lsmagic
<IPython.core.magics.basic.MagicsDisplay object at 0x7f260048c990>

[jhpalmieri]

Do you have any idea where/how %bg was defined before? I don't think it should have been affected by #34547 — that should only have affected magic commands related to interfaces. When did %bg last work?

[jhpalmieri]

Could it have been in a previous version of IPython? I see https://pypi.org/project/ipython-bg/, but it hasn't been modified for 3 years. See also https://stackoverflow.com/questions/10819479/missing-background-magic-command-bg-in-ipython and ipython/ipython#844.

[EmmanuelCharpentier]

O dunno ; I just discovered its existence by reading thé tutorial, of all things (and for other reasons... :-)).One Nevers ceases to learn... 8=O) !Le 25 juin 2023 22:20, "John H. Palmieri" ***@***.***> a écrit : Do you have any idea where/how %bg was defined before? I don't think it should have been affected by #34547 — that should only have affected magic commands related to interfaces. When did %bg last work? —Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you authored the thread.Message ID: ***@***.***>

[jhpalmieri]

We should probably remove that from the tutorial. See #22934 (comment), or we can open a new issue.

[jhpalmieri]

I created #35836.

[mkoeppe]

Can this be closed as completed?

[jhpalmieri]

Okay with me. Emmanuel?

[fchapoton]

@EmmanuelCharpentier please confirm that this issue can be closed