250 lines
9.5 KiB
Python
250 lines
9.5 KiB
Python
"""
|
|
Gradio interface for the intelligent research system.
|
|
This module provides a web interface for users to interact with the research system.
|
|
"""
|
|
|
|
import os
|
|
import json
|
|
import gradio as gr
|
|
import sys
|
|
import time
|
|
from pathlib import Path
|
|
|
|
# Add the parent directory to the path to allow importing from other modules
|
|
sys.path.append(str(Path(__file__).parent.parent))
|
|
|
|
from query.query_processor import QueryProcessor
|
|
from execution.search_executor import SearchExecutor
|
|
from execution.result_collector import ResultCollector
|
|
|
|
|
|
class GradioInterface:
|
|
"""Gradio interface for the intelligent research system."""
|
|
|
|
def __init__(self):
|
|
"""Initialize the Gradio interface."""
|
|
self.query_processor = QueryProcessor()
|
|
self.search_executor = SearchExecutor()
|
|
self.result_collector = ResultCollector()
|
|
self.results_dir = Path(__file__).parent.parent / "results"
|
|
self.results_dir.mkdir(exist_ok=True)
|
|
|
|
def process_query(self, query, num_results=10, use_reranker=True):
|
|
"""
|
|
Process a query and return the results.
|
|
|
|
Args:
|
|
query (str): The query to process
|
|
num_results (int): Number of results to return
|
|
use_reranker (bool): Whether to use the Jina Reranker for semantic ranking
|
|
|
|
Returns:
|
|
tuple: (markdown_results, json_results_path)
|
|
"""
|
|
try:
|
|
# Process the query
|
|
print(f"Processing query: {query}")
|
|
processed_query = self.query_processor.process_query(query)
|
|
print(f"Processed query: {processed_query}")
|
|
|
|
# Get available search engines and print their status
|
|
available_engines = self.search_executor.get_available_search_engines()
|
|
print(f"Available search engines: {available_engines}")
|
|
|
|
# Check which handlers are actually available
|
|
for engine_name, handler in self.search_executor.available_handlers.items():
|
|
print(f"Handler {engine_name} available: {handler.is_available()}")
|
|
if not handler.is_available():
|
|
print(f" - Reason: API key may be missing for {engine_name}")
|
|
|
|
# Add search engines if not specified
|
|
if 'search_engines' not in processed_query:
|
|
processed_query['search_engines'] = available_engines
|
|
print(f"Using search engines: {available_engines}")
|
|
|
|
# Execute the search - request more results from each engine
|
|
print(f"Executing search...")
|
|
search_results = self.search_executor.execute_search(
|
|
structured_query=processed_query,
|
|
num_results=num_results
|
|
)
|
|
|
|
# Print which engines returned results
|
|
for engine, results in search_results.items():
|
|
print(f"Engine {engine} returned {len(results)} results")
|
|
|
|
# Add the query to each result for reranking
|
|
for engine, results in search_results.items():
|
|
for result in results:
|
|
result["query"] = processed_query.get("enhanced_query", processed_query.get("original_query", query))
|
|
|
|
# Process the results - don't limit the number of results
|
|
print(f"Processing results...")
|
|
processed_results = self.result_collector.process_results(
|
|
search_results, dedup=True, max_results=None, use_reranker=use_reranker
|
|
)
|
|
print(f"Processed {len(processed_results)} results")
|
|
|
|
# Save results to file
|
|
timestamp = int(time.time())
|
|
results_file = self.results_dir / f"results_{timestamp}.json"
|
|
|
|
# Ensure the results are not empty before saving
|
|
if processed_results:
|
|
with open(results_file, "w") as f:
|
|
json.dump(processed_results, f, indent=2)
|
|
print(f"Results saved to {results_file}")
|
|
file_path = str(results_file)
|
|
else:
|
|
error_message = "No results found. Please try a different query or check API keys."
|
|
print(error_message)
|
|
file_path = None
|
|
return f"## No Results Found\n\n{error_message}", file_path
|
|
|
|
# Format results for display
|
|
markdown_results = self._format_results_as_markdown(processed_results)
|
|
|
|
return markdown_results, file_path
|
|
|
|
except Exception as e:
|
|
error_message = f"Error processing query: {str(e)}"
|
|
print(f"ERROR: {error_message}")
|
|
import traceback
|
|
traceback.print_exc()
|
|
return f"## Error\n\n{error_message}", None
|
|
|
|
def _format_results_as_markdown(self, results):
|
|
"""
|
|
Format results as markdown.
|
|
|
|
Args:
|
|
results (list): List of result dictionaries
|
|
|
|
Returns:
|
|
str: Markdown formatted results
|
|
"""
|
|
if not results:
|
|
return "## No Results Found\n\nNo results were found for your query."
|
|
|
|
# Count results by source
|
|
source_counts = {}
|
|
for result in results:
|
|
source = result.get("source", "unknown")
|
|
source_counts[source] = source_counts.get(source, 0) + 1
|
|
|
|
# Create source distribution string
|
|
source_distribution = ", ".join([f"{source}: {count}" for source, count in source_counts.items()])
|
|
|
|
markdown = f"## Search Results\n\n"
|
|
markdown += f"*Sources: {source_distribution}*\n\n"
|
|
|
|
for i, result in enumerate(results):
|
|
title = result.get("title", "Untitled")
|
|
url = result.get("url", "")
|
|
snippet = result.get("snippet", "No snippet available")
|
|
source = result.get("source", "unknown")
|
|
authors = result.get("authors", "Unknown")
|
|
year = result.get("year", "Unknown")
|
|
score = result.get("relevance_score", 0)
|
|
|
|
markdown += f"### {i+1}. {title}\n\n"
|
|
markdown += f"**Source**: {source}\n\n"
|
|
markdown += f"**URL**: [{url}]({url})\n\n"
|
|
markdown += f"**Snippet**: {snippet}\n\n"
|
|
markdown += f"**Authors**: {authors}\n\n"
|
|
markdown += f"**Year**: {year}\n\n"
|
|
markdown += f"**Score**: {score}\n\n"
|
|
markdown += "---\n\n"
|
|
|
|
return markdown
|
|
|
|
def create_interface(self):
|
|
"""
|
|
Create and return the Gradio interface.
|
|
|
|
Returns:
|
|
gr.Blocks: The Gradio interface
|
|
"""
|
|
with gr.Blocks(title="Intelligent Research System") as interface:
|
|
gr.Markdown("# Intelligent Research System")
|
|
gr.Markdown(
|
|
"""
|
|
This system helps you research topics by searching across multiple sources
|
|
including Google (via Serper), Google Scholar, and arXiv.
|
|
|
|
The system will return ALL results from each search engine, up to the maximum
|
|
number specified by the "Results Per Engine" slider. Results are ranked by
|
|
relevance across all sources.
|
|
"""
|
|
)
|
|
|
|
with gr.Row():
|
|
with gr.Column(scale=4):
|
|
query_input = gr.Textbox(
|
|
label="Research Query",
|
|
placeholder="Enter your research question here...",
|
|
lines=3
|
|
)
|
|
with gr.Column(scale=1):
|
|
num_results = gr.Slider(
|
|
minimum=5,
|
|
maximum=50,
|
|
value=20,
|
|
step=5,
|
|
label="Results Per Engine"
|
|
)
|
|
use_reranker = gr.Checkbox(
|
|
label="Use Semantic Reranker",
|
|
value=True,
|
|
info="Uses Jina AI's reranker for more relevant results"
|
|
)
|
|
search_button = gr.Button("Search", variant="primary")
|
|
|
|
gr.Examples(
|
|
examples=[
|
|
["What are the latest advancements in quantum computing?"],
|
|
["Compare transformer and RNN architectures for NLP tasks"],
|
|
["Explain the environmental impact of electric vehicles"]
|
|
],
|
|
inputs=query_input
|
|
)
|
|
|
|
with gr.Row():
|
|
with gr.Column():
|
|
results_output = gr.Markdown(label="Results")
|
|
|
|
with gr.Row():
|
|
with gr.Column():
|
|
file_output = gr.Textbox(
|
|
label="Results saved to file",
|
|
interactive=False
|
|
)
|
|
|
|
search_button.click(
|
|
fn=self.process_query,
|
|
inputs=[query_input, num_results, use_reranker],
|
|
outputs=[results_output, file_output]
|
|
)
|
|
|
|
return interface
|
|
|
|
def launch(self, **kwargs):
|
|
"""
|
|
Launch the Gradio interface.
|
|
|
|
Args:
|
|
**kwargs: Keyword arguments to pass to gr.Interface.launch()
|
|
"""
|
|
interface = self.create_interface()
|
|
interface.launch(**kwargs)
|
|
|
|
|
|
def main():
|
|
"""Main function to launch the Gradio interface."""
|
|
interface = GradioInterface()
|
|
interface.launch(share=True)
|
|
|
|
|
|
if __name__ == "__main__":
|
|
main()
|