Hi there,

I'm now trying to create a working notebook instance on iOS, but due to the 
restriction on system fork, I can only get a valid ipykernel running in the 
thread, so I came across a SO question talking about the threaded version 
IPYKernel, with that I can run a valid IPYKernel within a python thread. So now 
I'm trying to make the Jupyter partly working on the system.

After reading the doc(didn't find much useful though). I've starting to get 
into the source for the answer, but since it used quite a lot of Mixins, i 
didn't get it quite well the structure of those objects.

>From my current understanding, IPYkernel is a valid kernel listening on four 
>ports(hb/stdin/shell/iopub),
 In which i can use jupyter_client to directly talk with(kernel info/commands 
and all other actions etc).

 and what MultiKernelManager(KenrelManager) does is creating a kernel from 
subprocess(why not using ipykernel but using lauch_process here?) ,and fire up 
a Blocking client and starting a control_socket, so I guess the control_socket 
here works exactly the same with IPYkernel here? And what is this client does?

Basically what I want is if possible directly using what ipykernel generates 
and interface with what does notebook wants to talk like(from my own 
understanding, notebook is just a bridge between web socket and tcp 
communications between the four ports), is that correct?

I'm sorry if I didn't get it well, I've just touched the notebook code for two 
days,the docs and code itself is just too complicated to get a well 
understanding what it is and what is happening right here. Also on developing 
environment(iOS) I don't have a valid gdb for better debug environments, any 
helps or suggestions would be real appreciated :)

Thanks for your help

-- 
You received this message because you are subscribed to the Google Groups 
"Project Jupyter" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jupyter/5fca0887-a224-4a36-8f7e-e136442811d5%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to