ingest.py

"""
Mixing Encyclopedia - Knowledge Ingestion Script
Loads mixing knowledge from JSON into ChromaDB
"""

import chromadb
import json
from pathlib import Path
import sys
import io

# Fix Windows console encoding for emojis
if sys.platform == "win32":
    sys.stdout = io.TextIOWrapper(sys.stdout.buffer, encoding='utf-8')

def ingest_knowledge():
    """Load knowledge from JSON file into ChromaDB"""
    
    # Initialize ChromaDB
    print("🔧 Initializing ChromaDB...")
    client = chromadb.PersistentClient(path="./chroma_db")
    
    # Delete existing collection if it exists
    try:
        client.delete_collection(name="mixing_knowledge")
        print("🗑️  Deleted existing collection")
    except:
        pass
    
    # Create new collection
    collection = client.create_collection(
        name="mixing_knowledge",
        metadata={"description": "Personal mixing knowledge base"}
    )
    print("✅ Created new collection")
    
    # Load knowledge from JSON
    knowledge_file = Path("data/knowledge.json")
    if not knowledge_file.exists():
        print(f"❌ Error: {knowledge_file} not found!")
        print("📝 Create data/knowledge.json with your mixing knowledge")
        return
    
    with open(knowledge_file, 'r', encoding='utf-8') as f:
        knowledge = json.load(f)
    
    print(f"📚 Loading {len(knowledge)} entries...")
    
    # Prepare data for ChromaDB
    documents = []
    metadatas = []
    ids = []
    
    for i, entry in enumerate(knowledge):
        documents.append(entry['content'])
        metadatas.append({
            'topic': entry['topic'],
            'category': entry['category'],
            'source': entry['source']
        })
        ids.append(f"entry_{i}")
    
    # Add to collection
    collection.add(
        documents=documents,
        metadatas=metadatas,
        ids=ids
    )
    
    print(f"✅ Successfully ingested {len(knowledge)} entries!")
    print(f"📊 Total entries in database: {collection.count()}")
    print("\n🎉 Ready to use! Run the Jupyter notebook to query your knowledge.")

if __name__ == "__main__":
    ingest_knowledge()