Shift+click stops working until page refreshed

[aiqc]

Description

These important shortcuts frequently stop responding entirely:

Shift + click (select multiple cells)
c (copy one or more cells)
v (paste one or more cells)
d, d (delete one or more cells)

So I got into a habit of frequently refreshing the page to get them working again. I recently forgot to save my work before refreshing, which resulted in data loss.

Right now I need to copy/paste a group of 5 cells for 20 different scenarios. Jupyter is unusable for this.

By the time I refresh, I forget where on the page I was and what I was trying to do

---

Reproduce

The only bug I have been able to figure out is shift+click

1. Refresh the jupyterlab page
2. shift+click multiple cells
3. c
4. At this point i can still shift+click other cells successfully
5. v
6. It should break here. Repeat steps 1-5 until shift+click breaks

Sometimes it works v.s. sometimes it breaks all of shift+click

It also breaks when I use the edit dropdown

---

Context

There is no error in the browser's console.

• jupyterlab = 3.4.2
• Chrome = Version 101.0.4951.64 (Official Build) (x86_64)
• macOS = 11.6.5

This has been an bug for years

[JasonWeill]

On the main branch (4.x), using Chrome 102, I was unable to get c and v shortcuts to work (they appear to do nothing). SHIFT+click continued to work, though.

[aiqc]

Sometimes tab stops working for me in the same manner. When cell controls break it's annoying, but if I can't reliably edit text anymore it becomes time to look for an alternative tool. Especially since other IDEs that embraced ipynb have been adding data science features.

[jasongrout]

As another data point: On Chrome 102 and Firefox 101 on MacOS 12.4, I repeatedly (about 20-30 times) did steps 2-5 of the reproduction steps in the OP, starting with a notebook with about 5 simple code cells, with a fresh JupyterLab 3.4.3 install in a clean environment. I saw no issues. Copy/paste and shift-click always seemed to work great. I repeated with JupyterLab 4.0.0a26 with the same results.

[aiqc]

@jasongrout perhaps it has to do with memory/ amount of content. How many cells did you have? How many lab tabs open?

[jasongrout]

@jasongrout perhaps it has to do with memory/ amount of content. How many cells did you have? How many lab tabs open?

in jupyterlab, less than 5 tabs. At the end of the testing, I had probably 100-200 small cells.