cd /home/daytona/workspace/attachments/6f5d5f1f-abe1-4dab-a24b-2f67d4c73272/ && python3 -c " import pdfplumber with pdfplumber.open('__Pharma_Pre_proff_2021_solved_copy.pdf') as pdf: print(f'Total pages: {len(pdf.pages)}') for i, page in enumerate(pdf.pages): text = page.extract_text() if text: print(f'--- PAGE {i+1} ---') print(text) print() "

python3 -c " from pypdf import PdfReader reader = PdfReader('/home/daytona/workspace/attachments/6f5d5f1f-abe1-4dab-a24b-2f67d4c73272/__Pharma_Pre_proff_2021_solved_copy.pdf') print(f'Pages: {len(reader.pages)}') for i, page in enumerate(reader.pages[:5]): text = page.extract_text() print(f'--- PAGE {i+1} ---') print(repr(text[:500]) if text else 'EMPTY') "

python3 -c " import subprocess, os # Convert first few pages to images then OCR result = subprocess.run( ['pdftoppm', '-r', '200', '-l', '3', '/home/daytona/workspace/attachments/6f5d5f1f-abe1-4dab-a24b-2f67d4c73272/__Pharma_Pre_proff_2021_solved_copy.pdf', '/tmp/pharma_page'], capture_output=True, text=True ) print('STDOUT:', result.stdout) print('STDERR:', result.stderr) print('RC:', result.returncode) import os files = [f for f in os.listdir('/tmp') if f.startswith('pharma_page')] print('Files created:', files) "

python3 -c " import pytesseract from PIL import Image for i in range(1, 4): img = Image.open(f'/tmp/pharma_page-0{i}.ppm') text = pytesseract.image_to_string(img) print(f'=== PAGE {i} ===') print(text[:2000]) print() "

python3 -c " import subprocess, os, pytesseract from PIL import Image # Convert ALL pages result = subprocess.run( ['pdftoppm', '-r', '250', '/home/daytona/workspace/attachments/6f5d5f1f-abe1-4dab-a24b-2f67d4c73272/__Pharma_Pre_proff_2021_solved_copy.pdf', '/tmp/pharma_all'], capture_output=True, text=True ) files = sorted([f for f in os.listdir('/tmp') if f.startswith('pharma_all')]) print(f'Total image files: {len(files)}') all_text = '' for f in files: img = Image.open(f'/tmp/{f}') text = pytesseract.image_to_string(img) page_num = f.split('-')[-1].replace('.ppm','') all_text += f'\n\n=== PAGE {page_num} ===\n{text}' with open('/tmp/pharma_ocr.txt', 'w') as fout: fout.write(all_text) print('OCR complete. Total chars:', len(all_text)) print(all_text[:3000]) "

# Try fewer pages at a time - OCR pages 1-10 only python3 -c " import subprocess, os, pytesseract from PIL import Image # Convert pages 1-10 only (faster) result = subprocess.run( ['pdftoppm', '-r', '200', '-f', '1', '-l', '10', '/home/daytona/workspace/attachments/6f5d5f1f-abe1-4dab-a24b-2f67d4c73272/__Pharma_Pre_proff_2021_solved_copy.pdf', '/tmp/pharma_p'], capture_output=True, text=True, timeout=60 ) files = sorted([f for f in os.listdir('/tmp') if f.startswith('pharma_p-')]) print(f'Files: {files}') for f in files[:10]: img = Image.open(f'/tmp/{f}') text = pytesseract.image_to_string(img) print(f'=== {f} ===') print(text[:1500]) print() " 2>&1 | head -200

# Check if the previously OCR'd page 2 text is useful and get more pages one by one quickly # First check what was already extracted from /tmp/pharma_page images python3 -c " import pytesseract from PIL import Image # Just page 2 which had readable text img = Image.open('/tmp/pharma_page-02.ppm') text = pytesseract.image_to_string(img) print(text) " 2>&1

# Convert just page 2 to PNG and display - faster than OCR all pages # Also try to get page structure info python3 -c " import subprocess # Convert just 5 key pages to PNG at lower resolution for faster OCR result = subprocess.run( ['pdftoppm', '-r', '150', '-f', '2', '-l', '2', '-png', '/home/daytona/workspace/attachments/6f5d5f1f-abe1-4dab-a24b-2f67d4c73272/__Pharma_Pre_proff_2021_solved_copy.pdf', '/tmp/pg2'], capture_output=True, text=True, timeout=15 ) print('RC:', result.returncode) import os print(os.listdir('/tmp')) "

python3 << 'EOF' import pytesseract from PIL import Image import os results = {} files = sorted([f for f in os.listdir('/tmp') if f.startswith('pharma_all-')]) print(f"Processing {len(files)} pages...") for f in files: page_num = int(f.replace('pharma_all-','').replace('.ppm','')) img = Image.open(f'/tmp/{f}') text = pytesseract.image_to_string(img) results[page_num] = text # Save with open('/tmp/pharma_ocr_all.txt', 'w') as fout: for pn in sorted(results.keys()): fout.write(f'\n\n=== PAGE {pn} ===\n{results[pn]}') print("Done! Pages with substantial text:") for pn in sorted(results.keys()): txt = results[pn].strip() if len(txt) > 50: print(f" Page {pn}: {len(txt)} chars") EOF