Batch file rename (based on file content, possibly AI related)

hsehestedt said:

If you are going to do this via program, then you need to be able to tell the program something precise to look for in the contents of the file otherwise how will it know what criteria you want to use to perform a certain type of rename?

Click to expand...

hsehestedt said:

But beyond that, you would simply have to parse the contents of the file and look for something that will tell it what that file is.

Click to expand...

jro11 said:

AI generated over 90% of each file; when I asked ChatGPT, it looked at the document from which I addressed it (you can open CoPilot within Word) and provided me with a number of possible alternative names; If I bring up CoPilot I see a list of names it's already given to our 'conversations'--so in essence it's already done the hard work.

Figuring out a name is the least I can expect of AI. Where it seems to have a problem is in a) opening a document, and, b) renaming it.
(Instead Copilot (aka Microsoft) picked the most inane way of naming these documents...

For 200+ documents?

Click to expand...

hsehestedt said:

Sure, a program could parse thousands or tens of thousands of files rapidly.

Click to expand...

jro11 said:

And the name of that program is???

Click to expand...

hsehestedt said:

Seriously, it would be super easy to write in a few minutes once you know what you are specifically looking for.

Click to expand...

Batch file rename (based on file content, possibly AI related)​

jro11 said:

I have a slew of files CoPilot-generated files relating to questions I posed this AI engine. They are all named:
[]CopilotAnswers-unique number,docx[/]I brought up one of these files in Word and asked Copilot if it could rename the file based on its content. It replied it could do so (even provided example names...not bad, not bad at all). However, I want to do this as a batch process and CoPilot can't open or rename files. I understand there are Python routines and downloadable apps that can do this, as a Batch process.
I'm hoping the file can be renamed based on the content of the file, not just some numbering/date/... scheme independent of the file's content--a process AI seems particularly adept at.
Have you had any experience doing this? What has your experience been? What recommendations do you have?

Click to expand...

jro11 said:

Thanks

Click to expand...

jro11 said:

Seriously,

Batch file rename (based on file content, possibly AI related)​

Click to expand...

# Specify the folder containing the .docx files
$folderPath = "C:\Path\To\Your\Folder"

# Check if the folder exists
if (!(Test-Path $folderPath)) {
    Write-Host "Folder not found: $folderPath"
    exit
}

# Get all .docx files in the folder
$docxFiles = Get-ChildItem -Path $folderPath -Filter *.docx

# Loop through each .docx file
foreach ($file in $docxFiles) {
    try {
        # Open the document as a ComObject
        $word = New-Object -ComObject Word.Application
        $doc = $word.Documents.Open($file.FullName)

        # Extract the Subject property
        $subject = $doc.BuiltInDocumentProperties.Item("Subject").Value

        # Close the document and quit Word
        $doc.Close()
        $word.Quit()

        # Check if the Subject is not empty or null
        if (![string]::IsNullOrWhiteSpace($subject)) {
            # Define the new file name using the Subject property
            $newFileName = "$subject.docx"

            # Rename the file
            Rename-Item -Path $file.FullName -NewName $newFileName -ErrorAction Stop
            Write-Host "Renamed '$($file.Name)' to '$newFileName'"
        } else {
            Write-Host "Subject property is empty for '$($file.Name)'"
        }
    } catch {
        Write-Host "Error processing '$($file.Name)': $_"
    }
}

• Modify the `$folderPath` variable to the path where your `.docx` files are located.
• Copy and paste the script** into a `.ps1` file (e.g., `RenameDocxFiles.ps1`). Save it in the folder where the documents are.
• Run the script in PowerShell:
  • Open PowerShell as Administrator.
  • Navigate to the folder where the `.ps1` file is saved.
  • Execute the script by typing: `.\RenameDocxFiles.ps1`.

• The script uses the Microsoft Word COM object to extract the metadata (`Subject`) from each `.docx` file. As long as you have Word installed on your system. (You do)
• If a file’s "Subject" property is empty, the script will skip renaming that file.

import os
import re
from collections import Counter
from docx import Document

def infer_subject(text):
    """
    Analyze text content to determine the main subject.
    """
    # Basic word frequency analysis
    words = re.findall(r'\b\w+\b', text.lower())
    common_words = ["the", "and", "to", "of", "in", "a", "is", "for", "on", "that", "with", "as", "by", "at", "an"]
    keywords = [word for word in words if word not in common_words]
    most_common = Counter(keywords).most_common(3)  # Top 3 keywords
    return "_".join([item[0] for item in most_common])  # Combine keywords as the "subject"

def rename_files_in_folder(folder_path):
    """
    Rename all .docx files in a folder based on their inferred subject.
    """
    if not os.path.exists(folder_path):
        print(f"Folder not found: {folder_path}")
        return

    for file_name in os.listdir(folder_path):
        if file_name.endswith(".docx"):
            file_path = os.path.join(folder_path, file_name)
            try:
                # Extract text from the .docx file
                doc = Document(file_path)
                text = " ".join([para.text for para in doc.paragraphs])

                # Infer the subject of the document
                subject = infer_subject(text)
                if subject:
                    # Generate new file name based on the subject
                    new_file_name = f"{subject}.docx"
                    new_file_path = os.path.join(folder_path, new_file_name)

                    # Rename the file
                    os.rename(file_path, new_file_path)
                    print(f"Renamed '{file_name}' to '{new_file_name}'")
                else:
                    print(f"Could not infer subject for '{file_name}'")
            except Exception as e:
                print(f"Error processing '{file_name}': {e}")

# Specify the folder containing the .docx files
folder_path = r"C:\Path\To\Your\Folder"
rename_files_in_folder(folder_path)

• Reads `.docx` Content: Extracts text from each `.docx` file in the folder.
• Analyzes Text: Uses basic keyword frequency analysis to infer the subject matter (removing common filler words).
• Renames Files: Generates a new name based on the top 3 keywords (e.g., `project_plan_review.docx`).

1. File `Document1.docx` contains the text: *"This document discusses project plans and reviews regarding marketing."*
2. The script renames the file to `project_plans_reviews.docx`.