python3 -c " import pdfplumber with pdfplumber.open('/home/daytona/workspace/attachments/5fafd891-08db-4070-8d5d-d7159091c4c3/pedia drugs.pdf') as pdf: print(f'Total pages: {len(pdf.pages)}') for i, page in enumerate(pdf.pages): text = page.extract_text() print(f'\\n--- PAGE {i+1} ---') print(text) "

python3 -c " import subprocess result = subprocess.run(['tesseract', '--version'], capture_output=True, text=True) print(result.stdout) print(result.stderr) "

cd /home/daytona/workspace && mkdir -p pedia-ocr && python3 -c " from pdf2image import convert_from_path import pytesseract import os pdf_path = '/home/daytona/workspace/attachments/5fafd891-08db-4070-8d5d-d7159091c4c3/pedia drugs.pdf' output_dir = '/home/daytona/workspace/pedia-ocr' # Convert first 5 pages to check content images = convert_from_path(pdf_path, dpi=200, first_page=1, last_page=5) for i, img in enumerate(images): text = pytesseract.image_to_string(img) print(f'\\n=== PAGE {i+1} ===') print(text[:2000]) "

cd /home/daytona/workspace/pedia-ocr && convert -density 150 '/home/daytona/workspace/attachments/5fafd891-08db-4070-8d5d-d7159091c4c3/pedia drugs.pdf[0]' page1.png 2>&1 && echo "Done page 1" && tesseract page1.png page1_text && cat page1_text.txt | head -100

mkdir -p /home/daytona/workspace/pedia-ocr && convert -density 150 '/home/daytona/workspace/attachments/5fafd891-08db-4070-8d5d-d7159091c4c3/pedia drugs.pdf[0]' /home/daytona/workspace/pedia-ocr/page1.png 2>&1 && echo "Done" && tesseract /home/daytona/workspace/pedia-ocr/page1.png /home/daytona/workspace/pedia-ocr/page1_text && cat /home/daytona/workspace/pedia-ocr/page1_text.txt

mkdir -p /home/daytona/workspace/pedia-ocr && python3 -c " from pdf2image import convert_from_path import pytesseract pdf_path = '/home/daytona/workspace/attachments/5fafd891-08db-4070-8d5d-d7159091c4c3/pedia drugs.pdf' # Convert just page 1 at lower DPI to avoid memory issues images = convert_from_path(pdf_path, dpi=120, first_page=1, last_page=1) img = images[0] img.save('/home/daytona/workspace/pedia-ocr/page1.png') text = pytesseract.image_to_string(img) print(text[:3000]) " 2>&1

python3 -c " from pdf2image import convert_from_path import base64, os pdf_path = '/home/daytona/workspace/attachments/5fafd891-08db-4070-8d5d-d7159091c4c3/pedia drugs.pdf' # Save pages 1-6 as images for analysis images = convert_from_path(pdf_path, dpi=150, first_page=1, last_page=6) for i, img in enumerate(images): out_path = f'/home/daytona/workspace/pedia-ocr/page{i+1}.png' img.save(out_path) print(f'Saved {out_path}') print('Done') " 2>&1

python3 -c " from pdf2image import convert_from_path pdf_path = '/home/daytona/workspace/attachments/5fafd891-08db-4070-8d5d-d7159091c4c3/pedia drugs.pdf' # Save just page 1 images = convert_from_path(pdf_path, dpi=150, first_page=1, last_page=1) images[0].save('/home/daytona/workspace/pedia-ocr/page1.png') print('Saved page 1') " 2>&1

python3 -c " from pdf2image import convert_from_path pdf_path = '/home/daytona/workspace/attachments/5fafd891-08db-4070-8d5d-d7159091c4c3/pedia drugs.pdf' # Save pages 1-3 at lower dpi for pg in range(1, 4): images = convert_from_path(pdf_path, dpi=100, first_page=pg, last_page=pg) out = f'/home/daytona/workspace/pedia-ocr/pg{pg}.jpg' images[0].save(out, 'JPEG', quality=70) import os print(f'Page {pg}: {os.path.getsize(out)} bytes -> {out}') " 2>&1

python3 -c " import pytesseract from PIL import Image # Try OCR on the already-saved jpg img = Image.open('/home/daytona/workspace/pedia-ocr/pg2.jpg') text = pytesseract.image_to_string(img, config='--psm 6') print(text[:3000]) " 2>&1