pdf-extraction-fallback-7db3aa

PDF Extraction Fallback Strategy

Purpose

When processing complex documents (tax forms, legal documents, scanned materials), PDF extraction often fails on the first attempt. This skill provides a systematic fallback approach that tries multiple extraction methods in sequence until one succeeds.

When to Use

• Initial PDF reading tools return errors or empty content
• Document appears to be scanned/image-based rather than text-based
• Previous extraction attempts produced incomplete or garbled output
• Working with forms, tables, or structured documents that need reliable extraction

Fallback Sequence

Tier 1: Shell-Based Extraction (pdftotext)

Start with command-line tools that often handle edge cases better:

# Extract text maintaining layout
pdftotext -layout input.pdf output.txt

# Extract raw text (faster, less formatting)
pdftotext input.pdf output.txt

# Extract specific page range
pdftotext -f 1 -l 3 input.pdf output.txt

Check if output contains meaningful content before proceeding.

Tier 2: Python-Based Parsing

If shell tools fail, use Python libraries with different extraction approaches:

# Using PyPDF2 for basic text extraction
import PyPDF2
with open('document.pdf', 'rb') as f:
    reader = PyPDF2.PdfReader(f)
    text = ''.join(page.extract_text() for page in reader.pages)

# Using pdfplumber for tables and structured content
import pdfplumber
with pdfplumber.open('document.pdf') as pdf:
    for page in pdf.pages:
        text = page.extract_text()
        tables = page.extract_tables()

# Using pypdf for newer PDF features
from pypdf import PdfReader
reader = PdfReader('document.pdf')
text = ''.join(page.extract_text() for page in reader.pages)

Tier 3: OCR Tools (for Scanned Documents)

If the PDF contains images or scanned content, use OCR:

# Using tesseract via command line
tesseract input.pdf output --psm 6

# Using Python with pytesseract
import pytesseract
from pdf2image import convert_from_path

images = convert_from_path('document.pdf')
text = ''.join(pytesseract.image_to_string(img) for img in images)