cd /tmp/workspace/attachments/f3446268-5a8d-4d41-a4c6-68b2be34be86/ && python3 -c " import pdfplumber pdf_path = 'Midterm Mock 2 (1).pdf.pdf' with pdfplumber.open(pdf_path) as pdf: print(f'Total pages: {len(pdf.pages)}') for i, page in enumerate(pdf.pages): text = page.extract_text() print(f'--- PAGE {i+1} ---') print(text) print() "

cd /tmp/workspace/attachments/f3446268-5a8d-4d41-a4c6-68b2be34be86/ && python3 -c " from pypdf import PdfReader import base64, json reader = PdfReader('Midterm Mock 2 (1).pdf.pdf') print(f'Pages: {len(reader.pages)}') for i, page in enumerate(reader.pages): text = page.extract_text() print(f'Page {i+1} text length: {len(text) if text else 0}') # Check for images resources = page.get('/Resources', {}) xobject = resources.get('/XObject', {}) print(f'Page {i+1} XObjects: {list(xobject.keys()) if xobject else None}') "

cd /tmp/workspace/attachments/f3446268-5a8d-4d41-a4c6-68b2be34be86/ && python3 -c " from pypdf import PdfReader from PIL import Image import io, os reader = PdfReader('Midterm Mock 2 (1).pdf.pdf') page = reader.pages[0] resources = page['/Resources'] xobject = resources['/XObject'] for key in xobject: obj = xobject[key] print(f'Key: {key}, Type: {obj.get(\"/Subtype\")}') if obj.get('/Subtype') == '/Image': data = obj.get_data() print(f'Image data length: {len(data)}') filter_type = obj.get('/Filter') print(f'Filter: {filter_type}') width = obj.get('/Width') height = obj.get('/Height') print(f'Dimensions: {width}x{height}') # Try to save try: img = Image.open(io.BytesIO(data)) img.save('/tmp/workspace/page1_img.png') print('Saved as PNG') except Exception as e: print(f'PIL error: {e}') # Try saving raw with open('/tmp/workspace/page1_img.raw', 'wb') as f: f.write(data) print('Saved raw') "

cd /tmp/workspace/attachments/f3446268-5a8d-4d41-a4c6-68b2be34be86/ && python3 -c " import fitz # PyMuPDF doc = fitz.open('Midterm Mock 2 (1).pdf.pdf') print(f'Pages: {len(doc)}') for page_num in range(len(doc)): page = doc[page_num] # Extract images image_list = page.get_images(full=True) print(f'Page {page_num+1}: {len(image_list)} images') for img_index, img in enumerate(image_list): xref = img[0] base_image = doc.extract_image(xref) image_bytes = base_image['image'] image_ext = base_image['ext'] print(f' Image {img_index}: {image_ext}, size={len(image_bytes)}') with open(f'/tmp/workspace/page{page_num+1}_img{img_index}.{image_ext}', 'wb') as f: f.write(image_bytes) print(f' Saved to /tmp/workspace/page{page_num+1}_img{img_index}.{image_ext}') "

pip install --user pymupdf --quiet && python3 -c "import fitz; print('OK')"

python3 -c " import fitz doc = fitz.open('/tmp/workspace/attachments/f3446268-5a8d-4d41-a4c6-68b2be34be86/Midterm Mock 2 (1).pdf.pdf') print(f'Pages: {len(doc)}') for page_num in range(len(doc)): page = doc[page_num] # Try text extraction first text = page.get_text() print(f'Text length page {page_num+1}: {len(text)}') if text.strip(): print(text[:3000]) # Extract images image_list = page.get_images(full=True) print(f'Images on page {page_num+1}: {len(image_list)}') for img_index, img in enumerate(image_list): xref = img[0] base_image = doc.extract_image(xref) image_bytes = base_image['image'] image_ext = base_image['ext'] out_path = f'/tmp/workspace/page{page_num+1}_img{img_index}.{image_ext}' with open(out_path, 'wb') as f: f.write(image_bytes) print(f' Saved: {out_path} ({len(image_bytes)} bytes)') # Also render page as image mat = fitz.Matrix(2, 2) # 2x zoom for higher resolution clip = page.rect pix = page.get_pixmap(matrix=mat) out_path = f'/tmp/workspace/page{page_num+1}_render.png' pix.save(out_path) print(f' Rendered page to: {out_path}') "

python3 -c " from PIL import Image import os # Resize the image to something manageable img = Image.open('/tmp/workspace/page1_img0.png') print(f'Original size: {img.size}') # Resize to ~1200px wide while maintaining aspect ratio w, h = img.size new_w = 1200 new_h = int(h * new_w / w) img_resized = img.resize((new_w, new_h), Image.LANCZOS) out_path = '/tmp/workspace/page1_small.jpg' img_resized.save(out_path, 'JPEG', quality=85) print(f'Saved resized image: {out_path} ({os.path.getsize(out_path)} bytes)') print(f'New size: {img_resized.size}') "

python3 -c " from PIL import Image import os img = Image.open('/tmp/workspace/page1_small.jpg') w, h = img.size print(f'Image size: {w}x{h}') # Split into 3 sections sections = 4 section_h = h // sections for i in range(sections): top = i * section_h bottom = min((i+1) * section_h, h) cropped = img.crop((0, top, w, bottom)) out_path = f'/tmp/workspace/section_{i+1}.jpg' cropped.save(out_path, 'JPEG', quality=90) print(f'Section {i+1}: y={top}-{bottom}, saved to {out_path} ({os.path.getsize(out_path)} bytes)') "

cd /tmp/workspace && python3 -m http.server 8765 & sleep 1 echo "Server started" curl -s -o /dev/null -w "%{http_code}" http://localhost:8765/section_1.jpg

python3 -c "import pytesseract; print('OK')" 2>&1; python3 -c "import easyocr; print('OK')" 2>&1; python3 -c "import cv2; print('OK')" 2>&1

which tesseract 2>&1; tesseract --version 2>&1 | head -3

python3 -c " from markitdown import MarkItDown md = MarkItDown() result = md.convert('/tmp/workspace/attachments/f3446268-5a8d-4d41-a4c6-68b2be34be86/Midterm Mock 2 (1).pdf.pdf') print(result.text_content[:5000]) " 2>&1