Error (ein): Undo failure diagnostic (with MRE)

[veprbl]

Problem description

Exception pops up when notebook is running, the undo is broken afterwards.

Steps to reproduce the problem

1. Start Emacs (ein from the current master branch which is at 34e9ef39d161cc60d7cd823d9b2610dc0ad10ed9):

   emacs -Q -f package-initialize --eval "(setq debug-on-error t)"

2. M-x, ein:notebooklist-login, connect to a jupyter-notebook
3. Open a new notebook
4. Enter following into the cell

   import time
   while True:
   for i in range(10):
       print(f"{i}\r", end="")
       time.sleep(0.1)

5. Execute: C-c C-c
6. Enter minibuffer and leave: M-x C-g
7. Observe popup:

   Error (ein): Undo failure diagnostic (nil (nil comint-must-overwrite nil 114 . 115) (6 . 114) (
   . 116) (5 . 114) (
   . 116) (nil comint-must-overwrite t 114 . 115) (nil comint-must-overwrite nil 114 . 115) (4 . 114) (
   . 116) (3 . 114) (
   . 116) (nil comint-must-overwrite t 114 . 115) (nil comint-must-overwrite nil 114 . 115) (2 . 114) (
   . 116) (1 . 114) (
   . 116) (nil comint-must-overwrite t 114 . 115) (nil comint-must-overwrite nil 114 . 115) (0 . 114) (
   . 116) (9 . 114) (
   . 116) (nil comint-must-overwrite t 114 . 115) (nil comint-must-overwrite nil 114 . 115) (8 . 114) (
   . 116) (7 . 114) (
   . 116) (nil comint-must-overwrite t 114 . 115) (nil comint-must-overwrite nil 114 . 115) (6 . 114) (
   . 116) (5 . 114) (
   . 116) (nil comint-must-overwrite t 114 . 115) nil (nil rear-nonsticky nil 112 . 113) (nil fontified nil 10 . 113) (10 . 113) (t . 0)) (1e7d8 1e7d8 1e7d8 1e7d8 1e7d8) | 37

8. Interrupt notebook
9. Observe that undo doesn't work anymore
10. Switch to the *Messages* buffer, observe

    ein: [info] WS action [(error ein:worksheet--jigger-undo-list: aborting)] on-message (ws://127.0.0.1:8888/api/kernels/09aedf37-9456-4fdf-aec0-d93fb59896b9/channels?session_id=9fed7b1a-7aaf-43d0-b2a6-50ba3d36e56c)

System info:

("EIN system info"
 :emacs-version
 "GNU Emacs 26.3 (build 1, x86_64-apple-darwin17.7.0, NS appkit-1561.60 Version 10.13.6 (Build 17G65))"
 :emacs-bzr-version nil
 :window-system ns
 :emacs-variant nil
 :os
 (:uname
  "This is a macOS system"
  :lsb-release nil)
 :notebook nil
 :ipython
 "/nix/store/rrqk34hkyi38kdls1gzizwbpvpncpq6d-emacs-26.3/bin/emacs: /usr/local/bin/ipython: No such file or directory
"
 :image-types
 (svg png gif tiff jpeg xpm xbm pbm)
 :image-types-available
 (svg png gif tiff jpeg xpm xbm pbm)
 :request
 (:backend curl)
 :ein
 (:version "20200219.1501"
       :source-dir
       "/nix/store/m4n0z4kx8rlk762s3j9dzafv3bz1a5zj-emacs-packages-deps/share/emacs/site-lisp/elpa/ein-20200219.1501/")
 :lib
 ((:name "websocket"
     :path
     "/nix/store/m4n0z4kx8rlk762s3j9dzafv3bz1a5zj-emacs-packages-deps/share/emacs/site-lisp/elpa/websocket-20200102.637/websocket.elc"
     :featurep t
     :version-var websocket-version
     :version "1.12")
  (:name "anaphora"
     :path
     "/nix/store/m4n0z4kx8rlk762s3j9dzafv3bz1a5zj-emacs-packages-deps/share/emacs/site-lisp/elpa/anaphora-20180618.2200/anaphora.elc"
     :featurep t
     :version-var nil
     :version nil)
  (:name "request"
     :path
     "/nix/store/m4n0z4kx8rlk762s3j9dzafv3bz1a5zj-emacs-packages-deps/share/emacs/site-lisp/elpa/request-20200219.2257/request.elc"
     :featurep t
     :version-var request-version
     :version "0.3.0")
  (:name "deferred"
     :path
     "/nix/store/m4n0z4kx8rlk762s3j9dzafv3bz1a5zj-emacs-packages-deps/share/emacs/site-lisp/elpa/deferred-20170901.1330/deferred.elc"
     :featurep t
     :version-var deferred:version
     :version "0.5.0")
  (:name "polymode"
     :path
     "/nix/store/m4n0z4kx8rlk762s3j9dzafv3bz1a5zj-emacs-packages-deps/share/emacs/site-lisp/elpa/polymode-20191208.1239/polymode.elc"
     :featurep t
     :version-var nil
     :version nil)
  (:name "dash"
     :path
     "/nix/store/m4n0z4kx8rlk762s3j9dzafv3bz1a5zj-emacs-packages-deps/share/emacs/site-lisp/elpa/dash-20200119.2310/dash.elc"
     :featurep t
     :version-var nil
     :version nil)))

[dickmao]

Thanks for finding this bug and the MRE. 4403b65 should fix. I also added some undo caveats to the README.

[veprbl]

@dickmao I've tested the fix and it appears to be working just fine. Thank you for fixing this!