garbled_text_output

import os
from pypdf import PdfReader

pdf_path = 'document.pdf'
reader = PdfReader(pdf_path)

# This may return garbled text for PDFs with custom fonts
text = reader.pages[0].extract_text()  # <-- Line that produces garbled output
print(text)
print(repr(text))  # Shows mojibake: '\x00H\x00e\x00l\x00l\x00o' or 'àáâã'

import os
from pypdf import PdfReader
import pdfplumber

pdf_path = 'document.pdf'

# Strategy 1: Try pdfplumber first (better encoding handling)
try:
    with pdfplumber.open(pdf_path) as pdf:
        text = pdf.pages[0].extract_text()
        # Validate extraction quality
        if '\ufffd' in text or len(text.strip()) < 5:
            raise ValueError('Low quality extraction — fallback to OCR')
        print('Extracted (pdfplumber):', text)
except Exception as e:
    print(f'pdfplumber failed: {e} — trying pypdf with layout mode')
    
    # Strategy 2: pypdf with layout extraction
    reader = PdfReader(pdf_path)
    text = reader.pages[0].extract_text(extraction_mode='layout')
    if '\ufffd' not in text and len(text.strip()) > 5:
        print('Extracted (pypdf layout):', text)
    else:
        # Strategy 3: Fall back to OCR for scanned/complex PDFs
        print('Resorting to OCR for this PDF...')
        try:
            import pytesseract
            from pdf2image import convert_from_path
            images = convert_from_path(pdf_path, first_page=1, last_page=1)
            text = pytesseract.image_to_string(images[0])
            print('Extracted (OCR):', text)
        except ImportError:
            print('Install pytesseract and pdf2image for OCR: pip install pytesseract pdf2image')