import os
from pypdf import PdfReader

SRC = '/home/maxi/Projekt/wtc.waiswuaschd.de/data/import_2026'
IMG_OUT = '/home/maxi/Projekt/wtc.waiswuaschd.de/app/static/uploads/recipes'

for f in sorted(os.listdir(SRC)):
    if not f.lower().endswith('.pdf'):
        continue
    path = os.path.join(SRC, f)
    base = os.path.splitext(f)[0]
    safe_name = "".join(c for c in base if c.isalnum() or c in (' ', '_', '-')).strip().replace(' ', '_').lower()
    
    try:
        reader = PdfReader(path)
        img_idx = 0
        best_img = None
        best_size = 0
        best_ext = '.jpg'
        for page in reader.pages:
            for img in page.images:
                data = img.data
                if len(data) > 20000 and len(data) > best_size: # > 20KB
                    best_size = len(data)
                    best_img = data
                    best_ext = os.path.splitext(img.name)[1].lower() or '.jpg'
        
        if best_img:
            target_file = f"{safe_name}{best_ext}"
            target_path = os.path.join(IMG_OUT, target_file)
            with open(target_path, 'wb') as out_f:
                out_f.write(best_img)
            print(f"Extracted PDF image for {f}: {target_file} ({best_size} bytes)")
    except Exception as e:
        print(f"Error reading PDF {f}: {e}")