139_spanish-voice-trainer/main.py

# main.py
import asyncio
from database.connection import init_db
from core.bulk_importer import BulkImporter
from core.clean_glossary import GlossaryCleaner

async def rebuild_pipeline():
    print("🚀 Initiating Clean Reconstruction Pipeline...")
    
    # 1. This will automatically recreate the blank .db file and all tables
    init_db()
    
    # 2. Run the layout-aware bulk import from the PDF layout text
    importer = BulkImporter()
    pdf_file = "aula_int_plus_1_glos_en_alfa.pdf"
    textbook = "Aula Internacional Plus 1"
    
    importer.import_pdf_glossary(pdf_file, textbook)
    
    # 3. Immediately run the metadata preservation and text cleaning pass
    cleaner = GlossaryCleaner()
    cleaner.process_database_clean()
    
    print("\n✨ Database completely rebuilt with pristine structured data!")

if __name__ == "__main__":
    asyncio.run(rebuild_pipeline())