examples

Examples

This directory contains example code demonstrating how to use FLAMEHAVEN FileSearch.

Available Examples

1. Basic Usage (basic_usage.py)

Demonstrates core library functionality:

• Initializing SovDefLite
• Creating stores
• Uploading files
• Searching for information
• Managing stores
• Getting metrics

Run:

export GEMINI_API_KEY="your-api-key"
python examples/basic_usage.py

What it covers:

• Store creation and management
• Single file upload
• Search with citations
• Metrics collection

2. API Client (api_example.py)

Shows how to interact with the SovDef API server:

• Using the API client class
• Uploading files via HTTP
• Searching via REST API
• Managing stores via API
• Error handling

Prerequisites:

# Terminal 1: Start API server
export GEMINI_API_KEY="your-api-key"
uvicorn flamehaven_filesearch.api:app --reload

# Terminal 2: Run example
python examples/api_example.py

What it covers:

• API client implementation
• File upload (single and multiple)
• Search (POST and GET methods)
• Store operations
• Health checks and metrics

Quick Start Examples

Minimal Example (3 lines)

from flamehaven_filesearch import SovDefLite

searcher = SovDefLite(api_key="your-key")
searcher.upload_file("document.pdf")
result = searcher.search("What are the key points?")
print(result['answer'])

API Example (curl)

# Upload
curl -X POST "http://localhost:8000/upload" \
  -F "file=@document.pdf"

# Search
curl "http://localhost:8000/search?q=key+points"

Creating Your Own Examples

Template Structure

from flamehaven_filesearch import SovDefLite
import os

# 1. Setup
api_key = os.getenv("GEMINI_API_KEY")
searcher = SovDefLite(api_key=api_key)

# 2. Upload
result = searcher.upload_file("your_file.pdf")
print(f"Upload: {result['status']}")

# 3. Search
answer = searcher.search("your question")
print(f"Answer: {answer['answer']}")
print(f"Sources: {len(answer['sources'])}")

Common Use Cases

1. Document Q&A

# Upload technical documentation
searcher.upload_file("api_docs.pdf", store_name="docs")

# Ask questions
answer = searcher.search("How do I authenticate?", store_name="docs")

2. Research Paper Analysis

# Upload multiple papers
papers = ["paper1.pdf", "paper2.pdf", "paper3.pdf"]
searcher.upload_files(papers, store_name="research")

# Find related work
answer = searcher.search(
    "What methodologies do these papers use?",
    store_name="research"
)

3. Legal Document Search

# Upload contracts
searcher.upload_file("contract.pdf", store_name="legal")

# Search for clauses
answer = searcher.search(
    "What are the termination clauses?",
    store_name="legal",
    temperature=0.1  # Lower temperature for factual queries
)

4. Multi-Store Organization

# Create separate stores
searcher.create_store("technical")
searcher.create_store("business")
searcher.create_store("legal")

# Upload to appropriate stores
searcher.upload_file("architecture.pdf", store_name="technical")
searcher.upload_file("business_plan.pdf", store_name="business")
searcher.upload_file("terms.pdf", store_name="legal")

# Search in specific context
tech_answer = searcher.search("system architecture", store_name="technical")
biz_answer = searcher.search("revenue model", store_name="business")

Testing Examples

Before running examples:

1. Set API key:

   export GEMINI_API_KEY="your-api-key"

2. Prepare test files:

   • Create a sample PDF, DOCX, or TXT file
   • Update file paths in examples

3. Run examples:

   python examples/basic_usage.py

Need Help?

• Check the main README
• Review API documentation
• Open an issue

Contributing Examples

Have a useful example? Submit a PR!

1. Create a descriptive .py file
2. Add comments explaining each step
3. Update this README
4. Test your example
5. Submit PR