pdf-processing

name	pdf-processing
description	Skill for processing and extracting content from PDF files

PDF Processing Skill

This skill provides guidance and tools for processing PDF files, extracting text, and analyzing document content.

Overview

This skill includes Python scripts for common PDF processing tasks:

• Extracting text from PDF files
• Analyzing PDF structure
• Extracting tables from PDFs
• Converting PDF pages to images

Available Scripts

1. Extract Text from PDF

Use execute_script to run the text extraction script:

python /nfs/FM/gongoubo/new_project/Agent-Handbook/mini-agents/Mini_Agents/skills/document-skills/pdf/scripts/extract_text.py /path/to/your/file.pdf

2. Analyze PDF Structure

Use execute_script to run the structure analysis script:

python /nfs/FM/gongoubo/new_project/Agent-Handbook/mini-agents/Mini_Agents/skills/document-skills/pdf/scripts/analyze_structure.py /path/to/your/file.pdf

3. Extract Tables from PDF

Use execute_script to run the table extraction script:

python /nfs/FM/gongoubo/new_project/Agent-Handbook/mini-agents/Mini_Agents/skills/document-skills/pdf/scripts/extract_tables.py /path/to/your/file.pdf

4. Quick Code Execution

For quick PDF processing, use execute_code with inline Python code:

import PyPDF2

# Open PDF file
pdf_path = '/path/to/your/file.pdf'
with open(pdf_path, 'rb') as file:
    reader = PyPDF2.PdfReader(file)
    
    # Get basic info
    num_pages = len(reader.pages)
    print(f"Number of pages: {num_pages}")
    
    # Extract text from first page
    first_page = reader.pages[0]
    text = first_page.extract_text()
    print(f"First page text preview: {text[:200]}...")

Usage Examples

Example 1: Extract All Text from PDF

Use execute_code tool:

import PyPDF2

pdf_path = '/path/to/your/file.pdf'
with open(pdf_path, 'rb') as file:
    reader = PyPDF2.PdfReader(file)
    
    all_text = ""
    for page_num in range(len(reader.pages)):
        page = reader.pages[page_num]
        all_text += page.extract_text()
    
    print(all_text)

Example 2: Extract Text from Specific Page

Use execute_code tool:

import PyPDF2

pdf_path = '/path/to/your/file.pdf'
page_number = 2  # Extract from page 3 (0-indexed)

with open(pdf_path, 'rb') as file:
    reader = PyPDF2.PdfReader(file)
    
    if page_number < len(reader.pages):
        page = reader.pages[page_number]
        text = page.extract_text()
        print(f"Text from page {page_number + 1}:")
        print(text)
    else:
        print(f"Page {page_number + 1} does not exist. Total pages: {len(reader.pages)}")

Example 3: Get PDF Metadata

Use execute_code tool:

import PyPDF2

pdf_path = '/path/to/your/file.pdf'
with open(pdf_path, 'rb') as file:
    reader = PyPDF2.PdfReader(file)
    
    metadata = reader.metadata
    print("PDF Metadata:")
    if metadata:
        for key, value in metadata.items():
            print(f"  {key}: {value}")
    else:
        print("  No metadata available")

Best Practices

1. Always check if the PDF file exists before processing
2. Handle encoding issues gracefully (some PDFs may have special characters)
3. Use execute_code for simple text extraction tasks
4. Use execute_script for complex PDF processing workflows
5. Be aware that some PDFs may be password-protected or scanned images
6. For scanned PDFs, you may need OCR (Optical Character Recognition)

Required Libraries

• PyPDF2
• pdfplumber (for better text extraction and table extraction)
• tabula-py (for advanced table extraction)

Install with: pip install PyPDF2 pdfplumber tabula-py

Troubleshooting

Issue: Cannot extract text from PDF

• The PDF might be a scanned image. Use OCR tools like Tesseract.
• The PDF might be password-protected. Provide the password when opening.

Issue: Text extraction is garbled

• Try different PDF libraries (PyPDF2, pdfplumber, pdftotext)
• Some PDFs have complex layouts that are hard to parse automatically

Issue: Large PDF processing is slow

• Process pages in batches
• Consider using multiprocessing for parallel processing