Search Operations

Search Operations

The RTRAConnector class provides a suite of methods to perform web searches, aggregate metadata, and generate AI-driven responses based on live search data.

Initialization

Before performing search operations, you must initialize the connector with your Google Custom Search credentials.

from rtrapy.rtra_connector import RTRAConnector

# Initialize with Google API Key and Custom Search Engine ID
connector = RTRAConnector(api_key="YOUR_GOOGLE_API_KEY", engine_id="YOUR_ENGINE_ID")

Basic Search Retrieval

The search method performs a direct query against the Google Custom Search API and returns the raw result set.

search(query)

Executes a query and returns a list of result items.

• Parameters:
  • query (str): The search string to look up.
• Returns:
  • list[dict]: A list of dictionaries containing search result data (e.g., title, link, snippet). Returns an empty list if the request fails.

Example:

results = connector.search("Python asynchronous programming")
for item in results:
    print(item.get('title'))

Result Aggregation

The combine_search_results method is used to extract and concatenate text from multiple search results, providing a simplified context for further processing.

combine_search_results(query)

Retrieves the top 5 search results and merges their text snippets into a single string.

• Parameters:
  • query (str): The search string to aggregate.
• Returns:
  • str: A concatenated string of snippets, each separated by a newline.

Example:

context = connector.combine_search_results("Quantum computing breakthroughs 2024")
print(context)

AI-Generated Responses

The generate_detailed_response method implements a Retrieval-Augmented Generation (RAG) workflow. It fetches live search data and uses a hosted LLM (Mistral-7B) to synthesize a comprehensive answer.

generate_detailed_response(query)

Performs a search, aggregates the top results, and sends them to the Hugging Face Inference API to generate a natural language response.

• Parameters:
  • query (str): The question or topic to generate a report on.
• Returns:
  • str: The AI-generated text based on the search context. Returns an empty string if an error occurs during the API call.

Example:

response = connector.generate_detailed_response("What are the latest updates on space exploration?")
print(f"Detailed Analysis: {response}")

Note: This method utilizes the Hugging Face Inference API. Ensure you have a stable internet connection for the model to process the aggregated search context.