Separate Collections per Client or Domain

Use one RAGWire instance per collection. This is useful when serving multiple clients or keeping domains isolated.

from ragwire import RAGWire
import yaml, tempfile, os

def make_rag(collection_name: str) -> RAGWire:
    with open("config.yaml") as f:
        config = yaml.safe_load(f)
    config["vectorstore"]["collection_name"] = collection_name
    with tempfile.NamedTemporaryFile(mode="w", suffix=".yaml", delete=False) as tmp:
        yaml.dump(config, tmp)
        tmp_path = tmp.name
    try:
        return RAGWire(tmp_path)
    finally:
        os.unlink(tmp_path)

rag_legal   = make_rag("legal_docs")
rag_finance = make_rag("financial_docs")
rag_hr      = make_rag("hr_docs")

rag_legal.ingest_directory("data/legal/")
rag_finance.ingest_directory("data/finance/")

When to use this pattern:

• Multi-tenant SaaS: one collection per customer so data is fully isolated
• Domain isolation: legal, financial, and HR docs are searched independently
• Different embedding models per domain: create each RAGWire with a different config.yaml

All instances share the same Qdrant server

Collections are logically separated inside Qdrant, so you don't need a separate database per client.