Spyder is opening black DOS windows now

[stonebig]

Description of your problem

What steps will reproduce the problem?

1. spyder dev of 2016-09-15
2. start spyder on Windows 10
3. spyder start , but a side-python-screen (process) appears

What is the expected output? What do you see instead?

the black side window/process is new, is it a consequence of new ipython connexion method ?

I would suppose maybe we don't want to see it

Please provide any additional information below

Versions and main components

• Spyder Version: github 2016-06-15
• Python Version: 3.4
• Operating system: Windows 10

Dependencies

jedi >=0.8.1 : 0.9.0 (OK)
matplotlib >=1.0 : 2.0.0b4 (OK)
nbconvert >=4.0 : 4.2.0 (OK)
numpy >=1.7 : 1.11.1 (OK)
pandas >=0.13.1 : 0.19.0rc1 (OK)
pep8 >=0.6 : 1.7.0 (OK)
pyflakes >=0.6.0 : 1.3.0 (OK)
pygments >=2.0 : 2.1.3 (OK)
pylint >=0.25 : 1.6.4 (OK)
qtconsole >=4.2.0: 4.2.1 (OK)
rope >=0.9.4 : 0.9.4-1 (OK)
sphinx >=0.6.6 : 1.4.6 (OK)
sympy >=0.7.3 : 1.0 (OK)

Please go to the menu entry Help > Optional Dependencies (or
Help > Dependencies), press the button Copy to clipboard
and paste the contents below:

[ccordoba12]

What happens if you open more IPython consoles?

[ccordoba12]

I know how to fix this one, it's just a matter of using pythonw to start IPython kernels.

[stonebig]

3 Ipython consoles ==> 3 black DOS windows

[ccordoba12]

Yep, just what I imagined. Thanks for checking it @stonebig :-)

[ilayn]

I can confirm this behavior on Windows 10, Python 3.6

jedi >=0.8.1 : 0.9.0 (OK)
matplotlib >=1.0 : 1.5.1 (OK)
nbconvert >=4.0 : 4.2.0 (OK)
numpy >=1.7 : 1.11.1 (OK)
pandas >=0.13.1 : 0.18.1 (OK)
pep8 >=0.6 : 1.7.0 (OK)
pyflakes >=0.6.0 : 1.3.0 (OK)
pygments >=2.0 : 2.1.3 (OK)
pylint >=0.25 : 1.6.4 (OK)
qtconsole >=4.2.0: 4.2.1 (OK)
rope >=0.9.4 : 0.9.4-1 (OK)
sphinx >=0.6.6 : 1.4.6 (OK)
sympy >=0.7.3 : None (NOK)

[ccordoba12]

@stonebig, could you show the full title of cmd window open by Spyder? Right now I can only see up to winpython-64bit-3.4.x.5\...

[stonebig]

[ccordoba12]

@stonebig, I think there's a bug in jupyter-client that prevents me to fix this.

Could you locate this line:

https://github.com/jupyter/jupyter_client/blob/master/jupyter_client/launcher.py#L123

and change it to

proc = Popen(cmd, shell=True, **kwargs)

to see if the cmd window goes away?

[stonebig]

testing now .... works as expected ... just still that kenel death even for a new Ipython console .

==> is it a traitlets timer that could be changed ?

[ccordoba12]

I have a different fix now. Please help me test PR #3442 :-)

[stonebig]

greuuuh... looking at the other fix now

[ccordoba12]

Sorry, sorry :-p

[stonebig]

it works too. still even for each new ipython console, having it (announced) dying once is quite a short timeout (3 seconds ?). in total it takes 4 seconds to open a new console on my pc

[stonebig]

by the way, on Python 3.4 / Windows, you are under PyQt5.6 or PyQt5.5 ? (left on PyQt5.5.1 binary package by riverbanks, a real pain)

[ccordoba12]

5.6, don't know what's happening with the new kernel dying but I don't think it's that serious ;-)

[stonebig]

it's just a papercut, but you see the "death" at every new kernel, and that gives a (false) feeling of instability. if it's a traitlets thing, maybe jupyter friends could change the default time out (from 3 to 5 seconds)

[ccordoba12]

I'll take a look at it for Spyder 3.1. Please open a new bug so we don't forget about it :-)