app.py

import gradio as gr
from pyserini.search.lucene import LuceneSearcher
import os

# Download the index if it doesn't exist
if not os.path.exists('msmarco-passage'):
    os.system('python -c "from pyserini.search import LuceneSearcher; LuceneSearcher.from_prebuilt_index(\'msmarco-passage\')"')

def search_pyserini(query):
    try:
        searcher = LuceneSearcher('msmarco-passage')
        searcher.set_bm25(k1=0.9, b=0.4)
        hits = searcher.search(query, k=10)
        
        results = []
        for i, hit in enumerate(hits):
            doc = searcher.doc(hit.docid)
            content = doc.raw()[:200] + "..." if len(doc.raw()) > 200 else doc.raw()
            results.append(f"Rank: {i+1}\nDoc ID: {hit.docid}\nScore: {hit.score:.4f}\nContent: {content}\n")
        
        return "\n".join(results)
    except Exception as e:
        return f"An error occurred: {str(e)}"

css = """
.gradio-container {
    font-family: 'Arial', sans-serif;
}
.output-text {
    white-space: pre-wrap;
    font-family: 'Courier New', monospace;
    font-size: 14px;
    line-height: 1.5;
    padding: 10px;
    border: 1px solid #ccc;
    border-radius: 5px;
    background-color: #f9f9f9;
}
"""

with gr.Blocks(css=css) as iface:
    gr.Markdown("# Pyserini Search Interface")
    gr.Markdown("Enter a query to search using Pyserini with BM25 scoring (k1=0.9, b=0.4).")
    
    with gr.Row():
        query_input = gr.Textbox(
            lines=1, 
            placeholder="Enter your search query here...",
            label="Search Query"
        )
    
    with gr.Row():
        search_button = gr.Button("Search", variant="primary")
    
    with gr.Row():
        output = gr.Textbox(
            lines=20, 
            label="Search Results", 
            elem_classes=["output-text"]
        )
    
    search_button.click(
        fn=search_pyserini,
        inputs=query_input,
        outputs=output
    )

iface.launch()