Re: Jupyter notebooks / Dynabook / JupyterTalk

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
47 messages Options
123
Reply | Threaded
Open this post in threaded view
|

Re: Jupyter notebooks / Dynabook / JupyterTalk

Hannes Hirzel
Hello

On 8/24/18, H. Hirzel <[hidden email]> wrote:

> On 8/23/18, Jesús Marí <[hidden email]> wrote:
>> I'll take a look to *nteract*, seems that it uses the same protocol to
>> communicate with kernels.
>
> Good, actually my goal is to have a simple installation in a new Linux
> installation,
> something like
>
> a) download nteract (https://nteract.io/desktop), a 50MB executable file
> b) set up Pharo 6.1 with JupyterTalk
> c) whatever else is necessary, hopefully not to much
>

I have now a working JupyterTalk installation, see the attached screen shot.
What needs to be done is to come up with a reviewed, reproducible,
easy to use installation process.

The steps I used are

a) a fresh new Ubuntu 18.04.1 64 bit bionic beaver installation
    then
    sudo apt-get update

b) python --version

    Command 'python' not found, but can be installed with:

    python3 --version
    Python 3.6.5


b) Instead of anaconda I tried then pip3 approach
    sudo apt install python3-pip

    python3 -m pip install jupyter

But
    jupyter notebook

did not work.


After

    sudo apt install jupyter-core
    sudo apt install jupyter-notebook

This only downloaded 50MB instead of the 600MB for anaconda

The command

   jupyter-notebook

worked. Note the dash, i.e. it is  a different command

Then

   sudo apt-get install curl

and the Pharo 6.1 64 bit installation


   curl get.pharo.org/64/61+vm | bash

Then the Metacello installation taken from https://github.com/jmari/JupyterTalk.

Creation of the kernel.json file as described in
https://github.com/jmari/JupyterTalk was not necessary

The content of
 ls /usr/share/jupyter/kernels/

is only

python3

Interestingly, no entry for a Pharo kernel.

The CLI command
   jupyter-notebook
brings up a working Jupyter dashboard with the Pharo Smalltalk menu
entry installed in the 'New' menu.

--Hannes

On 8/24/18, H. Hirzel <[hidden email]> wrote:

> On 8/23/18, Jesús Marí <[hidden email]> wrote:
>> I'll take a look to *nteract*, seems that it uses the same protocol to
>> communicate with kernels.
>
> Good, actually my goal is to have a simple installation in a new Linux
> installation,
> something like
>
> a) download nteract (https://nteract.io/desktop), a 50MB executable file
> b) set up Pharo 6.1 with JupyterTalk
> c) whatever else is necessary, hopefully not to much
>

Reply | Threaded
Open this post in threaded view
|

Re: Jupyter notebooks / Dynabook / JupyterTalk

Jesús Marí
Last update  works also on *nteract*

In the other hand, I could not understand how Jupyter knows about Pharo
kernel without the configuration file!!! Unbelievable!




--
Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html

Reply | Threaded
Open this post in threaded view
|

Re: Jupyter notebooks / Dynabook / JupyterTalk

Hannes Hirzel
Actually there is a probably a configuration file, but not in the
place where I checked, I assume.

I have Ubuntu or Ubuntu based  test installations in partitions
sda6,7,8,9,10. I can not recall in which partition in actually
happened.

However I found out about the command [1]

    jupyter --paths

And I did this in some of  the different test installations [2]

In sda8 I  have

user8@Latitude2:~$ ls /usr/share/jupyter
    kernels  nbextensions

user8@Latitude2:~$ ls /usr/share/jupyter/kernels
    python3

user8@Latitude2:~$ ls /usr/local/share/jupyter/kernels
    pharo


So the probable explanation is that there was actually a configuration
file but not  in

                                 /usr/local/share/jupyter/kernels/pharo

as you indicate in https://github.com/jmari/JupyterTalk/blob/master/README.md

but in

    /usr/share/jupyter/kernels


So my guess is that a download *nteract*  (https://www.nteract.io/)
alone and pharo6.1 with Juyptertalk (thus no python notebook) should
work in a new Linux installation if there is a

a kernel.json file with Pharo Smalltalk [3]

in either

     /usr/local/share/jupyter/kernels/pharo
OR
    /usr/share/jupyter/kernels/pharo

I will test that and report to the list.

--Hannes




[1] jupyter help
=================

user10@Latitude2:~$ jupyter --help
usage: jupyter [-h] [--version] [--config-dir] [--data-dir] [--runtime-dir]
               [--paths] [--json]
               [subcommand]

Jupyter: Interactive Computing

positional arguments:
  subcommand     the subcommand to launch

optional arguments:
  -h, --help     show this help message and exit
  --version      show the jupyter command's version and exit
  --config-dir   show Jupyter config dir
  --data-dir     show Jupyter data dir
  --runtime-dir  show Jupyter runtime dir
  --paths        show all Jupyter paths. Add --json for machine-readable
                 format.
  --json         output paths as machine-readable json

Available subcommands: bundlerextension console kernel kernelspec migrate
nbconvert nbextension notebook qtconsole run serverextension troubleshoot
trust



[2] jupyter paths in different test installations
=================================================


sda7
-----
 jupyter --paths
config:
    /home/user25/.jupyter
    /home/user25/anaconda3/etc/jupyter
    /usr/local/etc/jupyter
    /etc/jupyter
data:
    /home/user25/.local/share/jupyter
    /home/user25/anaconda3/share/jupyter
    /usr/local/share/jupyter
    /usr/share/jupyter
runtime:
    /run/user/1000/jupyter


sda8
-------

jupyter --paths
config:
    /home/user8/.jupyter
    /usr/etc/jupyter
    /usr/local/etc/jupyter
    /etc/jupyter
data:
    /home/user8/.local/share/jupyter
    /usr/local/share/jupyter
    /usr/share/jupyter
runtime:
    /run/user/1000/jupyter



user8@Latitude2:~$ ls /usr/share/jupyter
kernels  nbextensions
user8@Latitude2:~$ ls /usr/share/jupyter/kernels
python3
user8@Latitude2:~$ ls /usr/local/share/jupyter/kernels
pharo
user8@Latitude2:~$



sda10
-----



user10@Latitude2:~$ jupyter --paths
config:
    /home/user10/.jupyter
    /usr/etc/jupyter
    /usr/local/etc/jupyter
    /etc/jupyter
data:
    /home/user10/.local/share/jupyter
    /usr/local/share/jupyter
    /usr/share/jupyter
runtime:
    /run/user/1000/jupyter




[3] Example kernel.json file for Pharo
======================================

ls /usr/local/share/jupyter/kernels/pharo
kernel.json
user8@Latitude2:~$ cat /usr/local/share/jupyter/kernels/pharo/kernel.json
{
  "argv": [
    "/home/user8/pharo-vm/bin/pharo",
    "/home/user8/Pharo.image",
    "ipharo",
    "{connection_file}"
  ],
  "display_name": "Pharo Smalltalk",
  "language": "smalltalk"
}




On 8/24/18, Jesús Marí <[hidden email]> wrote:

> Last update  works also on *nteract*
>
> In the other hand, I could not understand how Jupyter knows about Pharo
> kernel without the configuration file!!! Unbelievable!
>
>
>
>
> --
> Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html
>
>

Reply | Threaded
Open this post in threaded view
|

Re: Jupyter notebooks / Dynabook / JupyterTalk

Hannes Hirzel
Hello Jesús Marí

The JavaScript kernel has custom output functions [1] such as

   $$.html(htmlString)
   $$.svg(svgString)
   $$.png(base64String)
   $$.jpeg(base64String)
   $$.mime(mimeBundle)

What is the equivalent for $$.svg(svgString) for example?

Regards
Hannes




[1] http://n-riesco.github.io/ijavascript/doc/custom.ipynb.html

On 8/27/18, H. Hirzel <[hidden email]> wrote:

> Actually there is a probably a configuration file, but not in the
> place where I checked, I assume.
>
> I have Ubuntu or Ubuntu based  test installations in partitions
> sda6,7,8,9,10. I can not recall in which partition in actually
> happened.
>
> However I found out about the command [1]
>
>     jupyter --paths
>
> And I did this in some of  the different test installations [2]
>
> In sda8 I  have
>
> user8@Latitude2:~$ ls /usr/share/jupyter
>     kernels  nbextensions
>
> user8@Latitude2:~$ ls /usr/share/jupyter/kernels
>     python3
>
> user8@Latitude2:~$ ls /usr/local/share/jupyter/kernels
>     pharo
>
>
> So the probable explanation is that there was actually a configuration
> file but not  in
>
>                                  /usr/local/share/jupyter/kernels/pharo
>
> as you indicate in
> https://github.com/jmari/JupyterTalk/blob/master/README.md
>
> but in
>
>     /usr/share/jupyter/kernels
>
>
> So my guess is that a download *nteract*  (https://www.nteract.io/)
> alone and pharo6.1 with Juyptertalk (thus no python notebook) should
> work in a new Linux installation if there is a
>
> a kernel.json file with Pharo Smalltalk [3]
>
> in either
>
>      /usr/local/share/jupyter/kernels/pharo
> OR
>     /usr/share/jupyter/kernels/pharo
>
> I will test that and report to the list.
>
> --Hannes
>
>
>
>
> [1] jupyter help
> =================
>
> user10@Latitude2:~$ jupyter --help
> usage: jupyter [-h] [--version] [--config-dir] [--data-dir] [--runtime-dir]
>                [--paths] [--json]
>                [subcommand]
>
> Jupyter: Interactive Computing
>
> positional arguments:
>   subcommand     the subcommand to launch
>
> optional arguments:
>   -h, --help     show this help message and exit
>   --version      show the jupyter command's version and exit
>   --config-dir   show Jupyter config dir
>   --data-dir     show Jupyter data dir
>   --runtime-dir  show Jupyter runtime dir
>   --paths        show all Jupyter paths. Add --json for machine-readable
>                  format.
>   --json         output paths as machine-readable json
>
> Available subcommands: bundlerextension console kernel kernelspec migrate
> nbconvert nbextension notebook qtconsole run serverextension troubleshoot
> trust
>
>
>
> [2] jupyter paths in different test installations
> =================================================
>
>
> sda7
> -----
>  jupyter --paths
> config:
>     /home/user25/.jupyter
>     /home/user25/anaconda3/etc/jupyter
>     /usr/local/etc/jupyter
>     /etc/jupyter
> data:
>     /home/user25/.local/share/jupyter
>     /home/user25/anaconda3/share/jupyter
>     /usr/local/share/jupyter
>     /usr/share/jupyter
> runtime:
>     /run/user/1000/jupyter
>
>
> sda8
> -------
>
> jupyter --paths
> config:
>     /home/user8/.jupyter
>     /usr/etc/jupyter
>     /usr/local/etc/jupyter
>     /etc/jupyter
> data:
>     /home/user8/.local/share/jupyter
>     /usr/local/share/jupyter
>     /usr/share/jupyter
> runtime:
>     /run/user/1000/jupyter
>
>
>
> user8@Latitude2:~$ ls /usr/share/jupyter
> kernels  nbextensions
> user8@Latitude2:~$ ls /usr/share/jupyter/kernels
> python3
> user8@Latitude2:~$ ls /usr/local/share/jupyter/kernels
> pharo
> user8@Latitude2:~$
>
>
>
> sda10
> -----
>
>
>
> user10@Latitude2:~$ jupyter --paths
> config:
>     /home/user10/.jupyter
>     /usr/etc/jupyter
>     /usr/local/etc/jupyter
>     /etc/jupyter
> data:
>     /home/user10/.local/share/jupyter
>     /usr/local/share/jupyter
>     /usr/share/jupyter
> runtime:
>     /run/user/1000/jupyter
>
>
>
>
> [3] Example kernel.json file for Pharo
> ======================================
>
> ls /usr/local/share/jupyter/kernels/pharo
> kernel.json
> user8@Latitude2:~$ cat /usr/local/share/jupyter/kernels/pharo/kernel.json
> {
>   "argv": [
>     "/home/user8/pharo-vm/bin/pharo",
>     "/home/user8/Pharo.image",
>     "ipharo",
>     "{connection_file}"
>   ],
>   "display_name": "Pharo Smalltalk",
>   "language": "smalltalk"
> }
>
>
>
>
> On 8/24/18, Jesús Marí <[hidden email]> wrote:
>> Last update  works also on *nteract*
>>
>> In the other hand, I could not understand how Jupyter knows about Pharo
>> kernel without the configuration file!!! Unbelievable!
>>
>>
>>
>>
>> --
>> Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html
>>
>>
>

Reply | Threaded
Open this post in threaded view
|

Re: Jupyter notebooks / Dynabook / JupyterTalk

Jesús Marí
You can display text or html using
self display openInJupyter:'*hello world in bolt*'

or anything in a morph (a jpg, png ..a Roassal drawing).
self display openInJupyter: aMorph

you can expecify the extent
self display openInJupyter: aMorph extent: aPoint.

if you have installed Roassal group (it's a hack of many Roassal Classes)
then you can visualize Interactive Roassal drawings doing:

"this will load a modified version of Roassal.js that opens a Roassal view
inside <div>"
self loadScript: IPRoassal js.

self display
        interactionOn;
        openInJupyter: aRoassalView.






--
Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html

Reply | Threaded
Open this post in threaded view
|

Re: Jupyter notebooks / Dynabook / JupyterTalk

Hannes Hirzel
On 8/29/18, Jesús Marí <[hidden email]> wrote:
> You can display text or html using

> self display openInJupyter:'*hello world in bolt*'
>
> or anything in a morph (a jpg, png ..a Roassal drawing).
> self display openInJupyter: aMorph
>
> you can expecify the extent
> self display openInJupyter: aMorph extent: aPoint.

Thanks! This is convenient!

> if you have installed Roassal group (it's a hack of many Roassal Classes)
> then you can visualize Interactive Roassal drawings doing:
>
> "this will load a modified version of Roassal.js that opens a Roassal view
> inside <div>"
> self loadScript: IPRoassal js.
>
> self display
>         interactionOn;
>         openInJupyter: aRoassalView.
>

Reply | Threaded
Open this post in threaded view
|

Re: About describing the JupzterTalk installation process: Which Linux distribution are you using?

Jesús Marí
In reply to this post by SergeStinckwich
Hi Mr.Hirzel
about opening SVG, files, in fact you can open SVG as a morph  using (ASVGMorph fromFile: 'file.svg') but first you need to install XML parser from the catalog. I think there is a missing dependency. so you can do self display openInJupyter:(ASVGMorph fromFile: 'file.svg'). About nteract and jupytertalk, I mean you need to install jupyter if you want to use other kernels but node kernel so your plan about installing only nteract+jupyterTalk wonn't be factible. I already have JupyterTalk working on Pharo 32 bits(windows) but it is harder to install than Linux or Mac because you need to install zeromq 32 bits...it is not so complicated but you need to install an additional dependency. In a few days it will be available in JupyterTalk repository.

El mié., 29 ago. 2018 a las 19:38, H. Hirzel (<[hidden email]>) escribió:
About Windows ... do not worry, time will tell...

At the moment I have it on Linux and that is fine.

I am now more interested in actually USING it.

See my question about giving back SVG graphics results


On 8/29/18, Jesus Mari Aguirre <[hidden email]> wrote:
> I think Windows will not be possible by now, there's not Pharo 64 bits on
> windows...I don't know what version of zeroMQ uses nteract, I think there
> is a 32 bits version...
> I'll do my best but I can not guarantee to have JupyterTalk on Windows.
>
>
> El mié., 29 ago. 2018 a las 6:30, H. Hirzel (<[hidden email]>)
> escribió:
>
>> Thank you for keeping me informed.
>>
>> I now manage to get Linux installations working, but the process is
>> still not consistent, see the installation log for Ubuntu 18.04 below.
>>
>> It could be that it is much easier for Windows, at the moment I only
>> have a 32bit Windows installation, so I cannot check.
>>
>> I'd like to focus now on _using_ the Pharo Smalltalk kernel, see
>> upcoming mail on the list.
>>
>> --Hannes
>>
>> ========================================
>>
>>
>> New Ubuntu 18.04 installation
>>
>> sudo apt-get update
>>
>> sudo apt install python3-pip
>>
>>
>> pip3 install jupyter
>>
>> did not work the first time (network issues?)
>>
>> pip3 install jupyter
>>
>> sudo apt install jupyter-core
>>
>> sudo apt install jupyter-notebook
>>
>>
>>
>>
>>
>> ================================================================================
>> Installation log
>>
>> ================================================================================
>>
>> user9@Latitude2:~$
>>
>> python3 --version
>> Python 3.6.5
>>
>>
>> sudo apt-get update
>>
>> sudo apt install python3-pip
>>
>> user9@Latitude2:~$ sudo apt install python3-pip
>> [sudo] password for user9:
>> Reading package lists... Done
>> Building dependency tree
>> Reading state information... Done
>> The following additional packages will be installed:
>>   build-essential dh-python dpkg-dev fakeroot g++ g++-7 gcc gcc-7
>>   libalgorithm-diff-perl libalgorithm-diff-xs-perl
>> libalgorithm-merge-perl
>>   libasan4 libatomic1 libc-dev-bin libc6-dev libcilkrts5 libexpat1-dev
>>   libfakeroot libgcc-7-dev libitm1 liblsan0 libmpx2 libpython3-dev
>>   libpython3.6-dev libquadmath0 libstdc++-7-dev libtsan0 libubsan0
>>   linux-libc-dev make manpages-dev python-pip-whl python3-dev
>>   python3-distutils python3-lib2to3 python3-setuptools python3-wheel
>>   python3.6-dev
>> Suggested packages:
>>   debian-keyring g++-multilib g++-7-multilib gcc-7-doc libstdc++6-7-dbg
>>   gcc-multilib autoconf automake libtool flex bison gcc-doc
>> gcc-7-multilib
>>   gcc-7-locales libgcc1-dbg libgomp1-dbg libitm1-dbg libatomic1-dbg
>>   libasan4-dbg liblsan0-dbg libtsan0-dbg libubsan0-dbg libcilkrts5-dbg
>>   libmpx2-dbg libquadmath0-dbg glibc-doc libstdc++-7-doc make-doc
>>   python-setuptools-doc
>> The following NEW packages will be installed:
>>   build-essential dh-python dpkg-dev fakeroot g++ g++-7 gcc gcc-7
>>   libalgorithm-diff-perl libalgorithm-diff-xs-perl
>> libalgorithm-merge-perl
>>   libasan4 libatomic1 libc-dev-bin libc6-dev libcilkrts5 libexpat1-dev
>>   libfakeroot libgcc-7-dev libitm1 liblsan0 libmpx2 libpython3-dev
>>   libpython3.6-dev libquadmath0 libstdc++-7-dev libtsan0 libubsan0
>>   linux-libc-dev make manpages-dev python-pip-whl python3-dev
>>   python3-distutils python3-lib2to3 python3-pip python3-setuptools
>>   python3-wheel python3.6-dev
>> 0 upgraded, 39 newly installed, 0 to remove and 60 not upgraded.
>> Need to get 32.3 MB of archives.
>> After this operation, 141 MB of additional disk space will be used.
>> Do you want to continue? [Y/n] y
>>
>>
>>  pip3 install jupyter
>>
>>
>>
>> Error
>>
>>
>>  Downloading
>> https://files.pythonhosted.org/packages/94/dd/fe6c4d683b09eb05342bd2816b7779663f71762b4fa9c2d5203d35d17354/jupyter_client-5.2.3-py2.py3-none-any.whl
>> (89kB
>> <https://files.pythonhosted.org/packages/94/dd/fe6c4d683b09eb05342bd2816b7779663f71762b4fa9c2d5203d35d17354/jupyter_client-5.2.3-py2.py3-none-any.whl(89kB>
>> )
>>     22% |███████▍                        | 20kB 24kB/s eta
>> 0:00:03Exception:
>> Traceback (most recent call last):
>>   File
>> "/usr/share/python-wheels/urllib3-1.22-py2.py3-none-any.whl/urllib3/response.py",
>> line 302, in _error_catcher
>>     yield
>>   File
>> "/usr/share/python-wheels/urllib3-1.22-py2.py3-none-any.whl/urllib3/response.py",
>> line 384, in read
>>
>>
>> REDO
>>
>>
>>  pip3 install jupyter
>>
>>
>> But then
>>
>> jupyter
>>
>> Command 'jupyter' not found, but can be installed with:
>>
>> sudo apt install jupyter-core
>>
>> user9@Latitude2:~$ sudo apt install jupyter-core
>> [sudo] password for user9:
>> Reading package lists... Done
>> Building dependency tree
>> Reading state information... Done
>> The following additional packages will be installed:
>>   python3-decorator python3-ipython-genutils python3-jupyter-core
>>   python3-traitlets
>> The following NEW packages will be installed:
>>   jupyter-core python3-decorator python3-ipython-genutils
>> python3-jupyter-core
>>   python3-traitlets
>> 0 upgraded, 5 newly installed, 0 to remove and 60 not upgraded.
>> Need to get 132 kB of archives.
>>
>>
>>
>>
>>
>>  jupyter --version
>> 4.4.0
>>
>>
>>  jupyter notebook
>> Error executing Jupyter command 'notebook': [Errno 2] No such file or
>> directory
>> user9@Latitude2:~$ Error executing Jupyter command 'notebook': [Errno
>> 2] No such file or directory
>> Error: command not found
>> user9@Latitude2:~$
>>
>>
>>
>>
>> sudo apt install jupyter-notebook
>> Reading package lists... Done
>> Building dependency tree
>> Reading state information... Done
>> The following additional packages will be installed:
>>   fonts-font-awesome fonts-mathjax javascript-common
>>   jupyter-nbextension-jupyter-js-widgets libjs-backbone libjs-bootstrap
>>   libjs-bootstrap-tour libjs-codemirror libjs-es6-promise libjs-jed
>> libjs-jquery
>>   libjs-jquery-typeahead libjs-jquery-ui libjs-marked libjs-mathjax
>> libjs-moment
>>   libjs-requirejs libjs-requirejs-text libjs-text-encoding
>> libjs-underscore libjs-xterm
>>   liblua5.1-0 libluajit-5.1-2 libluajit-5.1-common pandoc pandoc-data
>> python3-bleach
>>   python3-dateutil python3-entrypoints python3-html5lib python3-ipykernel
>>   python3-ipython python3-ipywidgets python3-jinja2 python3-jsonschema
>>   python3-jupyter-client python3-mistune python3-nbconvert
>> python3-nbformat
>>   python3-notebook python3-pandocfilters python3-pickleshare
>> python3-prompt-toolkit
>>   python3-pygments python3-simplegeneric python3-terminado
>> python3-testpath
>>   python3-tornado python3-wcwidth python3-webencodings python3-zmq
>> Suggested packages:
>>   apache2 | lighttpd | httpd python-ipywidgets libjs-jquery-lazyload
>> libjs-json
>>   libjs-jquery-ui-docs fonts-mathjax-extras fonts-stix libjs-mathjax-doc
>>   texlive-latex-recommended texlive-xetex texlive-luatex pandoc-citeproc
>>   texlive-latex-extra context wkhtmltopdf python-bleach-doc
>> python3-genshi python3-lxml
>>   python-ipywidgets-doc python-jinja2-doc python-notebook-doc
>> ttf-bitstream-vera
>>   python3-pycurl python-tornado-doc python3-twisted
>> The following NEW packages will be installed:
>>   fonts-font-awesome fonts-mathjax javascript-common
>>   jupyter-nbextension-jupyter-js-widgets jupyter-notebook libjs-backbone
>>   libjs-bootstrap libjs-bootstrap-tour libjs-codemirror
>> libjs-es6-promise libjs-jed
>>   libjs-jquery libjs-jquery-typeahead libjs-jquery-ui libjs-marked
>> libjs-mathjax
>>   libjs-moment libjs-requirejs libjs-requirejs-text libjs-text-encoding
>>   libjs-underscore libjs-xterm liblua5.1-0 libluajit-5.1-2
>> libluajit-5.1-common pandoc
>>   pandoc-data python3-bleach python3-dateutil python3-entrypoints
>> python3-html5lib
>>   python3-ipykernel python3-ipython python3-ipywidgets python3-jinja2
>>   python3-jsonschema python3-jupyter-client python3-mistune
>> python3-nbconvert
>>   python3-nbformat python3-notebook python3-pandocfilters
>> python3-pickleshare
>>   python3-prompt-toolkit python3-pygments python3-simplegeneric
>> python3-terminado
>>   python3-testpath python3-tornado python3-wcwidth
>> python3-webencodings python3-zmq
>> 0 upgraded, 52 newly installed, 0 to remove and 60 not upgraded.
>> Need to get 20.7 MB of archives.
>> After this operation, 142 MB of additional disk space will be used.
>> Do you want to continue? [Y/n]
>>
>>
>>
>>
>> Setting up libjs-bootstrap-tour (0.12.0+dfsg-1) ...
>> Setting up python3-notebook (5.2.2-1) ...
>> Setting up jupyter-nbextension-jupyter-js-widgets (6.0.0-2) ...
>> /usr/lib/python3.6/runpy.py:125: RuntimeWarning:
>> 'notebook.nbextensions' found in sys.modules after import of package
>> 'notebook', but prior to execution of 'notebook.nbextensions'; this
>> may result in unpredictable behaviour
>>   warn(RuntimeWarning(msg))
>> Enabling notebook extension jupyter-js-widgets/extension...
>>       - Validating: OK
>> Setting up jupyter-notebook (5.2.2-1) ...
>> Setting up python3-ipywidgets (6.0.0-2) ...
>> Processing triggers for libc-bin (2.27-3ubuntu1) ...
>>
>> user9@Latitude2:~$
>>
>>
>>
>> jupyter notebook
>> [I 05:23:54.607 NotebookApp] Writing notebook server cookie secret to
>> /run/user/1000/jupyter/notebook_cookie_secret
>> [I 05:23:55.398 NotebookApp] Serving notebooks from local directory:
>> /home/user9
>> [I 05:23:55.398 NotebookApp] 0 active kernels
>> [I 05:23:55.398 NotebookApp] The Jupyter Notebook is running at:
>> [I 05:23:55.399 NotebookApp]
>>
>> http://localhost:8888/?token=f3e847ae72b205b350efa0739bfc0c575a44a52c4e02aa60
>> [I 05:23:55.399 NotebookApp] Use Control-C to stop this server and
>> shut down all kernels (twice to skip confirmation).
>> [C 05:23:55.406 NotebookApp]
>>
>>     Copy/paste this URL into your browser when you connect for the first
>> time,
>>     to login with a token:
>>
>> http://localhost:8888/?token=f3e847ae72b205b350efa0739bfc0c575a44a52c4e02aa60
>> [I 05:23:58.884 NotebookApp] Accepting one-time-token-authenticated
>> connection from 127.0.0.1
>> [Parent 5668, Gecko_IOThread] WARNING: pipe error (59): Connection
>> reset by peer: file
>>
>> /build/firefox-oscv9o/firefox-61.0.1+build1/ipc/chromium/src/chrome/common/ipc_channel_posix.cc,
>> line 353
>>
>>
>>
>>
>>
>>
>>
>>
>> On 8/27/18, Jesus Mari Aguirre <[hidden email]> wrote:
>> > *a) https://www.nteract.io/ <https://www.nteract.io/>b) Pharo 6.1c)
>> > JuypterTalkd) kernel.json in /usr/local/share/jupyter/kernels/pharo  *
>> >
>> > OK, I'll keep in mind in my instalation script. I'll keep you informed.
>> >
>> > Regarding on Windows instalation, probably it won't be so hard, It was
>> very
>> > easy from Mac os X to Linux, I'm sure Windows will not be different.
>> >
>> >
>> >
>> > El lun., 27 ago. 2018 a las 13:46, H. Hirzel
>> > (<[hidden email]>)
>> > escribió:
>> >
>> >> Thank you for the installation sequence. Good that you will add an
>> >> installation script to your repository. If you label it as 'example
>> >> installation script it gives people a clue what is involved and saves
>> >> time. And they will know that it might not work in all cases or
>> >> different solutions are possible.
>> >>
>> >> Good to know how to get a python3.6 environment in Ubuntu 16.04.
>> >>
>> >> My current Ubuntu 16.04 based distribution has
>> >>
>> >> python3 --version
>> >> Python 3.5.2
>> >>
>> >>
>> >> Just for trying I only did
>> >>
>> >>     sudo add-apt-repository ppa:deadsnakes/ppa
>> >>     pip3 install jupyter
>> >>
>> >> and then
>> >>
>> >>      jupyter notebook
>> >>
>> >> works.
>> >>
>> >> But as you see from the mail to the list I am primarily interested in
>> >> getting an easy to use installation sequence and it seems that the
>> >> following
>> >>
>> >> a) https://www.nteract.io/
>> >> b) Pharo 6.1
>> >> c) JuypterTalk
>> >> d) kernel.json in /usr/local/share/jupyter/kernels/pharo
>> >>
>> >> should be sufficient.
>> >>
>> >> However at the moment I now have some working jupyter notebook
>> >> installations and I will start to work with them. In particular I am
>> >> interested in how to output  html and svg contents from the Pharo
>> >> kernel.
>> >>
>> >> --Hannes
>> >>
>> >>
>> >>
>> >>
>> >> On 8/25/18, Jesus Mari Aguirre <[hidden email]> wrote:
>> >> > sorry... I press ctrl+tab in google mail... as I was saying
>> >> > I did:
>> >> >
>> >> >   sudo add-apt-repository ppa:deadsnakes/ppa
>> >> >   sudo apt-get update
>> >> >   sudo apt-get install python3.6
>> >> >   pip3 install jupyter
>> >> >
>> >> > I think this was the sequence of commands
>> >> >
>> >> > If you are going to use ipywidgets (still not supported by
>> JupyterTalk)
>> >> >   pip3 install ipywidgets
>> >> >   jupyter nbextension enable --py widgetsnbextension
>> >> >
>> >> > I'm preparing a bash script to make it easy, but jupyter is
>> >> > sometimes
>> a
>> >> > pain in the neck.
>> >> >
>> >> >
>> >> >
>> >> >
>> >> >
>> >> >
>> >> > El sáb., 25 ago. 2018 a las 21:07, Jesus Mari Aguirre (<
>> >> > [hidden email]>) escribió:
>> >> >
>> >> >> I did:
>> >> >>   sudo add-apt-repository ppa:deadsnakes/ppa
>> >> >>   sudo apt-get update
>> >> >>   sudo apt-get install python3.6
>> >> >> pip3 install jupyter
>> >> >>
>> >> >> El sáb., 25 ago. 2018 a las 16:39, H. Hirzel (<
>> [hidden email]
>> >> >)
>> >> >> escribió:
>> >> >>
>> >> >>> Hello Jesus Mari
>> >> >>>
>> >> >>> On 8/25/18, Jesus Mari Aguirre <[hidden email]> wrote:
>> >> >>> > I don't care about python version but Jupyter does. I think last
>> >> >>> version of
>> >> >>> > Jupyter needs Python 3 and Pip3.
>> >> >>>
>> >> >>> Going for Python 3 is what I do in Ubuntu 16.04 and Ubuntu 18.04
>> >> >>>
>> >> >>> The second approach on
>> >> >>> http://jupyter.org/install
>> >> >>> for "experienced" python users (I am not is)
>> >> >>>
>> >> >>> python3 -m pip install --upgrade pip
>> >> >>>
>> >> >>> This leaves me with an unusable pip3
>> >> >>>
>> >> >>> pip3 --version
>> >> >>> Traceback (most recent call last):
>> >> >>>   File "/usr/bin/pip3", line 9, in <module>
>> >> >>>     from pip import main
>> >> >>> ImportError: cannot import name 'main'
>> >> >>>
>> >> >>> So I did not upgrade pip and directly did
>> >> >>>
>> >> >>>     python3 -m pip install jupyter
>> >> >>>
>> >> >>> In one installation this morning this worked, in another one in
>> >> >>> the
>> >> >>> evening (network quality was not so good) it did not work even
>> >> >>> after
>> >> >>> several tries. (details below)
>> >> >>>
>> >> >>> > I installed everything on Ubuntu 14.
>> >> >>> Good to know that you can even have it in Ubuntu 14.
>> >> >>> I use Ubuntu 16.04
>> >> >>>
>> >> >>> How did you install Juypter in Ubuntu 14.04?
>> >> >>>
>> >> >>> Could you please send me the result of
>> >> >>>
>> >> >>> history | grep 'install jupyter'
>> >> >>>
>> >> >>>
>> >> >>> - one installation was fine
>> >> >>>
>> >> >>> - the other gave an error message.[1] Jupyter did not install.
>> >> >>>  It bothers me that I cannot repeat the process easily. In the end
>> >> >>> something like a snap package which contains everything the
>> necessary
>> >> >>> jupyter files and Pharo6.1 with JupyterTalk installed might be the
>> >> >>> way
>> >> >>> to go for installing it on different systems.
>> >> >>>
>> >> >>>
>> >> >>> > I'm very surprised you don't need to create the kernel.json
>> >> >>> > file,
>> I
>> >> >>> > can
>> >> >>> not
>> >> >>> > explain how does Jupyter starts Pharo without the path and arts
>> >> >>> > to
>> >> the
>> >> >>> > Pharo VM...
>> >> >>>
>> >> >>> That was in a third Ubuntu setup. It is probably caused by the
>> >> >>> installation sequence. I'll have to check again.
>> >> >>>
>> >> >>> That is in Ubuntu 18.04 and I used
>> >> >>>
>> >> >>> sudo apt-get install jupyter-core
>> >> >>> sudo apt-get install jupyter-notebook
>> >> >>>
>> >> >>> Seems to be in the repository, but it was not sufficient.
>> >> >>>
>> >> >>> This is why I like the *nteract* approach. One download which
>> >> >>> includes
>> >> >>> everything.
>> >> >>>
>> >> >>> Webbrowser with notebook, node kernel.
>> >> >>>
>> >> >>> --Hannes
>> >> >>>
>> >> >>> > El sáb., 25 ago. 2018 6:53, H. Hirzel <[hidden email]>
>> >> >>> escribió:
>> >> >>> >
>> >> >>> >> Hello Jesus Mari
>> >> >>> >>
>> >> >>> >> For my JupyterTalk installation tests I have been using Ubuntu
>> >> >>> >> 18.04.1
>> >> >>> so
>> >> >>> >> far.
>> >> >>> >> Python 3.
>> >> >>> >>
>> >> >>> >> My production environment is 16.04. So I still need to do the
>> >> >>> >> installation there.
>> >> >>> >>
>> >> >>> >> Is your python environment version 2 or 3?
>> >> >>> >>
>> >> >>> >> Regards
>> >> >>> >> Hannes
>> >> >>> >>
>> >> >>> >
>> >> >>> ---------------------------
>> >> >>>
>> >> >>> [1]
>> >> >>> user26@Latitude-E64102:~$ pip3 install jupyter
>> >> >>> Collecting jupyter
>> >> >>>   Using cached
>> >> >>>
>> >>
>> https://files.pythonhosted.org/packages/83/df/0f5dd132200728a86190397e1ea87cd76244e42d39ec5e88efd25b2abd7e/jupyter-1.0.0-py2.py3-none-any.whl
>> >> >>> Collecting nbconvert (from jupyter)
>> >> >>>   Using cached
>> >> >>>
>> >>
>> https://files.pythonhosted.org/packages/39/ea/280d6c0d92f8e3ca15fd798bbcc2ea141489f9539de7133d8fe10ea4b049/nbconvert-5.3.1-py2.py3-none-any.whl
>> >> >>> Collecting ipykernel (from jupyter)
>> >> >>>   Using cached
>> >> >>>
>> >>
>> https://files.pythonhosted.org/packages/ab/3f/cd624c835aa3336a9110d0a99e15070f343b881b7d651ab1375ef226a3ac/ipykernel-4.8.2-py3-none-any.whl
>> >> >>> Collecting ipywidgets (from jupyter)
>> >> >>>   Using cached
>> >> >>>
>> >>
>> https://files.pythonhosted.org/packages/34/3a/5b258ea6d584f5a8527c2295d0ebf7ffb1654e3de38d37697f88bbef6621/ipywidgets-7.4.0-py2.py3-none-any.whl
>> >> >>> Collecting notebook (from jupyter)
>> >> >>>   Downloading
>> >> >>>
>> >>
>> https://files.pythonhosted.org/packages/5e/7c/7fd8e9584779d65dfcad9fa2e09c76131a41f999f853a9c7026ed8585586/notebook-5.6.0-py2.py3-none-any.whl
>> >> >>> (8.9MB)
>> >> >>>     52% |█████████████████               | 4.7MB 50kB/s eta
>> >> >>> 0:01:22Exception:
>> >> >>> Traceback (most recent call last):
>> >> >>>   File
>> >> >>>
>> >>
>> "/usr/share/python-wheels/urllib3-1.13.1-py2.py3-none-any.whl/urllib3/response.py",
>> >> >>> line 226, in _error_catcher
>> >> >>>     yield
>> >> >>>   File
>> >> >>>
>> >>
>> "/usr/share/python-wheels/urllib3-1.13.1-py2.py3-none-any.whl/urllib3/response.py",
>> >> >>> line 301, in read
>> >> >>>     data = self._fp.read(amt)
>> >> >>>   File
>> >> >>>
>> >>
>> "/usr/share/python-wheels/CacheControl-0.11.5-py2.py3-none-any.whl/cachecontrol/filewrapper.py",
>> >> >>> line 49, in read
>> >> >>>     data = self.__fp.read(amt)
>> >> >>>   File "/usr/lib/python3.5/http/client.py", line 448, in read
>> >> >>>     n = self.readinto(b)
>> >> >>>   File "/usr/lib/python3.5/http/client.py", line 488, in readinto
>> >> >>>     n = self.fp.readinto(b)
>> >> >>>   File "/usr/lib/python3.5/socket.py", line 575, in readinto
>> >> >>>     return self._sock.recv_into(b)
>> >> >>>   File "/usr/lib/python3.5/ssl.py", line 929, in recv_into
>> >> >>>     return self.read(nbytes, buffer)
>> >> >>>   File "/usr/lib/python3.5/ssl.py", line 791, in read
>> >> >>>     return self._sslobj.read(len, buffer)
>> >> >>>   File "/usr/lib/python3.5/ssl.py", line 575, in read
>> >> >>>     v = self._sslobj.read(len, buffer)
>> >> >>> socket.timeout: The read operation timed out
>> >> >>>
>> >> >>> During handling of the above exception, another exception
>> >> >>> occurred:
>> >> >>>
>> >> >>> Traceback (most recent call last):
>> >> >>>   File "/usr/lib/python3/dist-packages/pip/basecommand.py", line
>> 209,
>> >> in
>> >> >>> main
>> >> >>>     status = self.run(options, args)
>> >> >>>   File "/usr/lib/python3/dist-packages/pip/commands/install.py",
>> line
>> >> >>> 328, in run
>> >> >>>     wb.build(autobuilding=True)
>> >> >>>   File "/usr/lib/python3/dist-packages/pip/wheel.py", line 748, in
>> >> build
>> >> >>> ....
>> >> >>> 231, in _error_catcher
>> >> >>>     raise ReadTimeoutError(self._pool, None, 'Read timed out.')
>> >> >>> requests.packages.urllib3.exceptions.ReadTimeoutError:
>> >> >>> HTTPSConnectionPool(host='files.pythonhosted.org', port=443): Read
>> >> >>> timed out.
>> >> >>> You are using pip version 8.1.1, however version 18.0 is
>> >> >>> available.
>> >> >>> You should consider upgrading via the 'pip install --upgrade pip'
>> >> >>> command.
>> >> >>>
>> >> >>
>> >> >
>> >>
>> >
>>
>
123