6.9rc3 : Ipython notebook seems seriously out of whack.

[Emmanuel Charpentier]

I just found another nice one with 6.9rc3.

Not only it has problems with $SAGE_ROOT/local/lib/libtinfo* (see here), but it also spews nonsense :

In [2]:

%display simple

solve(x^2+1==0,x,to_poly_solve="force")
Out[2]:

[]

Huh ?

Command-line sage :
sage: solve(x^2+1==0, x)
[x == -I, x == I]

Sage via emacs :
sage: solve(x^2+1==0,x)
[x == -I, x == I]

This is nicely typeset in the emacs window, BTW...).

I also checked that the sage notebook is indemn of this folly...

And, BTW, I also tried Trac#19374 with the same result (and no typeset output, by the way !).

I think this should be a blocker...

For the record : this is on a tree resulting from a fresh tarball, where I moved $SAGE_ROOT/local/lib/libtinfo* out of the way, which passes ptestlong, and where I did git trac chechout 19374.
 
Any idea/opinion on how to write the ticket ?

HTH,

--
Emmanuel Charpentier

[Jeroen Demeyer]

On 2015-10-10 21:24, Emmanuel Charpentier wrote:
> In [2]:
> %display simple
>
> solve(x^2+1==0,x,to_poly_solve="force")
> Out[2]:
>
> []
>
>
> Huh ?

Works for me...

In any case, it would be very unlikely that a Sage command actually
gives different output in the command line and in the Jupyter notebook.

[Thierry Dumont]

>jupyter notebook
sage 6.9 rc3
This works for me too.
t;d.

[Emmanuel Charpentier]

Works for me in a new sheet. Doesn't in a sheet initially created by 6.9beta1. As far as I can tell, nothing in the sheet explains this behaviour...

And #19374 still doesn't work for me neither in the "old" or the "new" sheet).

HTH,

--
Emmanuel Charpentier

[Eric Gourgoulhon]

Le dimanche 11 octobre 2015 09:53:49 UTC+2, Emmanuel Charpentier a écrit :

And #19374 still doesn't work for me neither in the "old" or the "new" sheet).

Works fine for me with the just released sage 6.9 (which includes #19374).

- System: Ubuntu 14.04 LTS 64bits with libssl-dev package installed
- Installation of sage 6.9 and launch of Jupyter notebook:
    git clone git://github.com/sagemath/sage.git
    cd sage
    make
    ./sage -n jupyter
 
Best wishes,

Eric.

[HG]

I did the same with sage 6.9 and sagemanifold, some command are working other not.
I got this message if I try a worksheet done :

Failed to retrieve MathJax from '../nbextensions/mathjax/MathJax.js'

Math/LaTeX rendering will be disabled.

If you have administrative access to the notebook server and a working internet connection, you can install a local copy of MathJax for offline use with the following command on the server at a Python or Jupyter prompt:

>>> from Jupyter.external import mathjax; mathjax.install_mathjax()

This will try to install MathJax into the Jupyter source directory.

If Jupyter is installed to a location that requires administrative privileges to write, you will need to make this call as an administrator, via 'sudo'.

When you start the notebook server, you can instruct it to disable MathJax support altogether:

$ ipython notebook --no-mathjax

which will prevent this dialog from appearing.

Used from git with a new worksheet some commands works

But not very stable at the moment

ubuntu 14.04 fresh install with git sage and sagemanifold

regards

Henri

[Jeroen Demeyer]

On 2015-10-11 18:30, HG wrote:
> I did the same with sage 6.9 and sagemanifold, some command are working
> other not.
> I got this message if I try a worksheet done :
>
>
> Failed to retrieve MathJax from '../nbextensions/mathjax/MathJax.js'

Probably there is something wrong with your installation. Did you run
"make"?

[HG]

I recompiled sage and sagemanifolds (git), if I start from a terminal ./sage -n=ipython it works fine. I I make an icon I still got jupiter not running. I guess I have to indicate the path somewhere ?
I notice from few months on, that ubunty python is always sending a message of security which seems to pertube fonctionning. Kind of "your path can be attacked by people bad intentioned ..." Last year I hadn't this.
Well I wait for sage-6.9 deb as I didn't solve this problem (which is only anoying).
Henri

[Emmanuel Charpentier]

I went back o the "develop" branch", fetched the last release (6.9) and re-made. $SAGE_ROOT/local/lib/libtinfo* is still out of the way.

The resulting system can open a sample sheet created with 6.9rc3, correctly solves x^2+1==0 and typesets the output if necessary.

Two bizarreries :

• Example of typeset output (cut from Jupyter to Google groups) :

In [5]:

solve(x^2+1==0, x)

Out[5]:

[x=(−i),x=i]

 (i. e. \newcommand{\Bold}[1]{\mathbf{#1}}\left[x = \left(-i\right), x = i\right]). The "simple" output doesn't have the superfetatory parentheses around the -I solution...

• The mathjax menu seems difficult to reach....

ISTR that something analogous has been discussed recently, but can't retrieve it at the moment...

HTH,

--
Emmanuel Charpentier

[Eric Gourgoulhon]

Le lundi 12 octobre 2015 09:33:21 UTC+2, Emmanuel Charpentier a écrit :

Two bizarreries :

The "simple" output doesn't have the superfetatory parentheses around the -I solution...

This has nothing to do with the LaTeX rendering in the jupyter notebook: it results from sage command latex and is already here in the console:

sage: latex([x == -I, x == I])

\left[x = \left(-i\right), x = i\right]

• The mathjax menu seems difficult to reach....

Have you tried to right-click on the displayed formula ? (it works for me)

Best wishes,

Eric.

[Emmanuel Charpentier]

It seems more complicated that that. See my followup in Trac#19371

HTH,

--
Emmanuel Charpentier

[HG]

The new version sage 6.9 linux binary for ubuntu 15.04 64 bits works like a charm. (Well I couldn't run properly my older files without the error message of mathjax).

[HG]

J'ai remarqué que si le fichier est dans home alors il trouve bien mathjax, donc c'est un problème de chemin vers un sous répertoire mais je ne sais pas comment le régler.

Le samedi 10 octobre 2015 21:24:51 UTC+2, Emmanuel Charpentier a écrit :

[Dominique Laurain]

HG's last post translated (as accurate as I can :-)) ...

I I found the following hint : when that file is located in $HOME directory, then mathjax can be launched.
It looks like the path to subdirectory is not well handled....and i don't know how to fix it.

Dominique, for english language readers