Closed dsche-cyber closed 3 years ago
Uploading the problematic .ipynb here, if that's possible, would help me reproduce the problem.
In this folder (from a github textbook) I tried to open calibrating-qubits-pulse.ipynb
:
Folder
Thanks for reporting the bug. fe63100b should help.
The first time I tried to open it, it worked, but now it does not. Maybe it is important to know, that this is an online book, so this has probably some html parts and also it has latex formulas in it. So together there are a lot of formats that need to be parsed correctly...
In M-x ein:log-pop-to-request-buffer
I get:
[debug] request--callback: executing success
[debug] request--callback: executing complete
[debug] /tmp/curl-trace unreadable
[debug] request--curl-callback: event finished
[debug] request--callback: UNPARSED
[debug] request--callback: PARSED
[debug] request--callback: executing success
[debug] request--callback: executing complete
[debug] /tmp/curl-trace unreadable
Edit: gc-cons-threshold, not gc-operation
grep -- gc-cons-threshold /home/dsche/.emacs.d/.local/straight/build/ein/*
should return nothing.
Here I open the notebook, run ein:stop
, then reopen the notebook.
Indeed you command returns nothing. For you the notebook opens automatically, for me not. Finally I got it: I needed to manually open it (C-x b, and choose the buffer, I think this is not intended behaviour, also the old buffer still persist with the plain text, for you its automatically closing).
Moreover I get a Warning after C-c C-o. Also Latex text is not displaying.
If it weren't already obvious, I think your EIN experience under your current doom configuration may not be worth it.
EIN cannot handle any undo-mode other than the default. It deftly sidesteps undo-tree-mode but I did not foresee undo-fu-mode, and noticed it's not easily sidestepped without messing with a user's global preferences. EIN can only warn "Disabling undo for undo-fu-only-undo" and consequently undo won't work at all.
There is also an outstanding memory leak for single-line ipynb files in #740 .
As for the central issue of "Notebook pending open! Retry?", I suspect an error is getting swallowed. You'd need to M-: (setq debug-on-error t)
to get emacs to stop and backtrace when it happens.
Latex rendering is a known deficiency https://github.com/millejoh/emacs-ipython-notebook/projects/3
Problem description
Issue same as #480 Open a .ipynb in doom emacs.
C-c C-o
. Choose parent dir of opened .ipynb.There are three different cases that appear: 1.)
Notebook .ipynb pending open" Retry? (y or n)
after y nothing happens (ksysguard shows a new process jupyter-notebook) 2.) Notebook does not open in any buffer, instead minibuffer prints Worksheet is ready. 3.) For some notebooks it works (the notebook opens in a new buffer)Steps to reproduce the problem
After ein:notebook-open, click on Open for some notebooks.
Info asked for in #480
After
M-x ein:log-pop-to-all-buffer
In
M-x ein:log-pop-to-request-buffer
I get:The debug report:
System info: