Incorrect Batch Regex Replacement

[snyhai]

Environment data

• VS Code version: 1.95.3
• Jupyter Extension version (available under the Extensions sidebar): v2024.10.0
• Python Extension version (available under the Extensions sidebar): v2024.20.0
• OS (Windows | Mac | Linux distro) and version: Mac 15.2 Beta (24C5089c)
• Python and/or Anaconda version: Python 3.12.2 conda 24.9.2
• Type of virtual environment used (N/A | venv | virtualenv | conda | ...): conda
• Jupyter server running: Local

Expected behaviour

When performing a regex-based batch replacement for the pattern (\d)\+(\d) with replacement $1 + $2, all occurrences across cells should be updated consistently.

Input ipynb JSON:

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "plaintext"
    }
   },
   "outputs": [],
   "source": [
    "3+4\n",
    "2+3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "plaintext"
    }
   },
   "outputs": [],
   "source": [
    "2+3\n",
    "1+2\n",
    "0+1"
   ]
  }
 ],
 "metadata": {
  "language_info": {
   "name": "python"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}

Expected single and batch replacement result:

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "plaintext"
    }
   },
   "outputs": [],
   "source": [
    "3 + 4\n",
    "2 + 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "plaintext"
    }
   },
   "outputs": [],
   "source": [
    "2 + 3\n",
    "1 + 2\n",
    "0 + 1"
   ]
  }
 ],
 "metadata": {
  "language_info": {
   "name": "python"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}

Actual behaviour

When performing batch replacement, the second cell’s content is processed incorrectly, resulting in an incorrect update. The output looks as follows:

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "plaintext"
    }
   },
   "outputs": [],
   "source": [
    "3 + 4\n",
    "2 + 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "plaintext"
    }
   },
   "outputs": [],
   "source": [
    "3 + 4\n",
    "2 + 3\n",
    "2 + 3"
   ]
  }
 ],
 "metadata": {
  "language_info": {
   "name": "python"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}

Steps to reproduce:

1. Use the above input ipynb file.
2. Perform regex replacement with pattern (\d)\+(\d) and replacement $1 + $2 in the Jupyter notebook.
3. Compare results for single and batch replacement modes.

Logs

There is no relevant output in the Jupyter panel.

I don’t know whether this is an issue with VS Code or Jupyter, but this kind of error only seems to occur in .ipynb files.