Update app.py

app.py

@@ -1,203 +1,203 @@
-import gradio as gr
-import google.generativeai as genai
-from duckduckgo_search import DDGS
-import requests
-from bs4 import BeautifulSoup
-import time
-from urllib.parse import urlparse
-import re
-import json
-from typing import List, Dict, Any
-
-# Search the web for relevant information using DuckDuckGo
-def web_search(query: str, max_results: int = 10) -> List[Dict[str, str]]:
-    """Search the web for relevant information using DuckDuckGo"""
-    try:
-        with DDGS() as ddgs:
-            results = list(ddgs.text(query, max_results=max_results))
-            return results
-    except Exception as e:
-        print(f"Search error: {e}")
-        return []
-
-# Fetch and extract content from a URL
-def fetch_url_content(url: str) -> str:
-    """Fetch content from a URL and extract meaningful text"""
-    try:
-        headers = {
-            'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36'
-        }
-        response = requests.get(url, headers=headers, timeout=10)
-        response.raise_for_status()
-        
-        soup = BeautifulSoup(response.content, 'html.parser')
-        
-        # Remove unwanted elements
-        for element in soup(['script', 'style', 'nav', 'footer', 'header', 'aside']):
-            element.decompose()
-        
-        # Get text content
-        text = soup.get_text()
-        
-        # Clean up text
-        lines = (line.strip() for line in text.splitlines())
-        chunks = (phrase.strip() for line in lines for phrase in line.split("  "))
-        text = ' '.join(chunk for chunk in chunks if chunk)
-        
-        return text[:5000]  # Limit content length to avoid token limits
-    except Exception as e:
-        print(f"Error fetching {url}: {e}")
-        return ""
-
-# Research function using web search and content extraction
-def perform_research(query: str, max_sources: int = 5) -> Dict[str, Any]:
-    """Perform research by searching and extracting content from multiple sources"""
-    print(f"Researching: {query}")
-    
-    # Search for relevant sources
-    search_results = web_search(query, max_results=max_sources*2)  # Get extra results to account for failed fetches
-    
-    sources = []
-    content_chunks = []
-    
-    for i, result in enumerate(search_results[:max_sources]):
-        print(f"Fetching content from {result['href']}")
-        content = fetch_url_content(result['href'])
-        
-        if content and len(content) > 200:  # Only include if we got meaningful content
-            sources.append({
-                'title': result.get('title', 'No title'),
-                'url': result.get('href', ''),
-                'content': content
-            })
-            content_chunks.append(f"SOURCE {i+1}:\nURL: {result.get('href', '')}\nCONTENT:\n{content}\n")
-        
-        # Be polite with delays between requests
-        time.sleep(1)
-    
-    research_context = "\n".join(content_chunks)
-    
-    return {
-        'sources': sources,
-        'research_context': research_context,
-        'query': query
-    }
-
-# Generate a research report using Gemini
-def generate_research_report(research_data: Dict[str, Any], gemini_api_key: str) -> str:
-    """Generate a comprehensive research report using Gemini"""
-    if not gemini_api_key:
-        return "Gemini API key is required to generate the report."
-    
-    try:
-        # Initialize Gemini
-        genai.configure(api_key=gemini_api_key)
-        model = genai.GenerativeModel('gemini-pro')
-        
-        prompt = f"""
-        RESEARCH TOPIC: {research_data['query']}
-        
-        RESEARCH CONTEXT FROM VARIOUS SOURCES:
-        {research_data['research_context']}
-        
-        Please analyze this research and create a comprehensive, well-structured report with:
-        1. Key findings and insights
-        2. Detailed explanations of complex concepts
-        3. Relevant examples and case studies
-        4. Real-world applications
-        5. Future predictions and trends
-        6. Citations for all sources with links
-        
-        Format your response using Markdown with appropriate headings, subheadings, bullet points, and bold text for emphasis.
-        """
-        
-        response = model.generate_content(prompt)
-        return response.text
-    except Exception as e:
-        return f"Error generating report: {str(e)}"
-
-# Main research function
-def run_research(topic: str, gemini_api_key: str):
-    """Run the complete research process"""
-    if not gemini_api_key:
-        return "Please enter your Gemini API key.", None, gr.update(visible=False)
-    
-    if not topic:
-        return "Please enter a research topic.", None, gr.update(visible=False)
-    
-    try:
-        # Perform research
-        research_data = perform_research(topic)
-        
-        if not research_data['sources']:
-            return "No relevant sources found. Please try a different search term.", None, gr.update(visible=False)
-        
-        # Generate report
-        report = generate_research_report(research_data, gemini_api_key)
-        
-        # Create a downloadable file
-        filename = f"{topic.replace(' ', '_')}_report.md"
-        
-        return report, filename, gr.update(visible=True)
-        
-    except Exception as e:
-        error_msg = f"An error occurred: {str(e)}"
-        return error_msg, None, gr.update(visible=False)
-
-# Gradio interface
-def create_interface():
-    with gr.Blocks(title="Gemini Deep Research Agent", theme=gr.themes.Soft()) as demo:
-        gr.Markdown("# 📘 Gemini Deep Research Agent")
-        gr.Markdown("This agent performs deep research on any topic using Google's Gemini and DuckDuckGo search")
-        
-        with gr.Row():
-            with gr.Column(scale=1):
-                gr.Markdown("## API Configuration")
-                gemini_key = gr.Textbox(
-                    label="Gemini API Key",
-                    type="password",
-                    placeholder="Enter your Gemini API key (get it from https://aistudio.google.com/)"
-                )
-                
-            with gr.Column(scale=2):
-                research_topic = gr.Textbox(
-                    label="Research Topic",
-                    placeholder="e.g., Latest developments in AI",
-                    lines=2
-                )
-                
-                research_btn = gr.Button("Start Research", variant="primary")
-                
-                output = gr.Markdown(
-                    label="Research Report",
-                    value="Your research report will appear here..."
-                )
-                
-                download_btn = gr.DownloadButton(
-                    "Download Report",
-                    visible=False
-                )
-        
-        # Set up the button action
-        research_btn.click(
-            fn=run_research,
-            inputs=[research_topic, gemini_key],
-            outputs=[output, download_btn, download_btn]
-        )
-        
-        # Set up download functionality
-        def create_file(content):
-            return content
-        
-        download_btn.click(
-            fn=create_file,
-            inputs=[output],
-            outputs=[download_btn]
-        )
-    
-    return demo
-
-# Main execution
-if __name__ == "__main__":
-    demo = create_interface()
+import gradio as gr
+import google.generativeai as genai
+from duckduckgo_search import DDGS
+import requests
+from bs4 import BeautifulSoup
+import time
+from urllib.parse import urlparse
+import re
+import json
+from typing import List, Dict, Any
+
+# Search the web for relevant information using DuckDuckGo
+def web_search(query: str, max_results: int = 10) -> List[Dict[str, str]]:
+    """Search the web for relevant information using DuckDuckGo"""
+    try:
+        with DDGS() as ddgs:
+            results = list(ddgs.text(query, max_results=max_results))
+            return results
+    except Exception as e:
+        print(f"Search error: {e}")
+        return []
+
+# Fetch and extract content from a URL
+def fetch_url_content(url: str) -> str:
+    """Fetch content from a URL and extract meaningful text"""
+    try:
+        headers = {
+            'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36'
+        }
+        response = requests.get(url, headers=headers, timeout=10)
+        response.raise_for_status()
+        
+        soup = BeautifulSoup(response.content, 'html.parser')
+        
+        # Remove unwanted elements
+        for element in soup(['script', 'style', 'nav', 'footer', 'header', 'aside']):
+            element.decompose()
+        
+        # Get text content
+        text = soup.get_text()
+        
+        # Clean up text
+        lines = (line.strip() for line in text.splitlines())
+        chunks = (phrase.strip() for line in lines for phrase in line.split("  "))
+        text = ' '.join(chunk for chunk in chunks if chunk)
+        
+        return text[:5000]  # Limit content length to avoid token limits
+    except Exception as e:
+        print(f"Error fetching {url}: {e}")
+        return ""
+
+# Research function using web search and content extraction
+def perform_research(query: str, max_sources: int = 5) -> Dict[str, Any]:
+    """Perform research by searching and extracting content from multiple sources"""
+    print(f"Researching: {query}")
+    
+    # Search for relevant sources
+    search_results = web_search(query, max_results=max_sources*2)  # Get extra results to account for failed fetches
+    
+    sources = []
+    content_chunks = []
+    
+    for i, result in enumerate(search_results[:max_sources]):
+        print(f"Fetching content from {result['href']}")
+        content = fetch_url_content(result['href'])
+        
+        if content and len(content) > 200:  # Only include if we got meaningful content
+            sources.append({
+                'title': result.get('title', 'No title'),
+                'url': result.get('href', ''),
+                'content': content
+            })
+            content_chunks.append(f"SOURCE {i+1}:\nURL: {result.get('href', '')}\nCONTENT:\n{content}\n")
+        
+        # Be polite with delays between requests
+        time.sleep(1)
+    
+    research_context = "\n".join(content_chunks)
+    
+    return {
+        'sources': sources,
+        'research_context': research_context,
+        'query': query
+    }
+
+# Generate a research report using Gemini
+def generate_research_report(research_data: Dict[str, Any], gemini_api_key: str) -> str:
+    """Generate a comprehensive research report using Gemini"""
+    if not gemini_api_key:
+        return "Gemini API key is required to generate the report."
+    
+    try:
+        # Initialize Gemini
+        genai.configure(api_key=gemini_api_key)
+        model = genai.GenerativeModel('gemini-2.0-flash')
+        
+        prompt = f"""
+        RESEARCH TOPIC: {research_data['query']}
+        
+        RESEARCH CONTEXT FROM VARIOUS SOURCES:
+        {research_data['research_context']}
+        
+        Please analyze this research and create a comprehensive, well-structured report with:
+        1. Key findings and insights
+        2. Detailed explanations of complex concepts
+        3. Relevant examples and case studies
+        4. Real-world applications
+        5. Future predictions and trends
+        6. Citations for all sources with links
+        
+        Format your response using Markdown with appropriate headings, subheadings, bullet points, and bold text for emphasis.
+        """
+        
+        response = model.generate_content(prompt)
+        return response.text
+    except Exception as e:
+        return f"Error generating report: {str(e)}"
+
+# Main research function
+def run_research(topic: str, gemini_api_key: str):
+    """Run the complete research process"""
+    if not gemini_api_key:
+        return "Please enter your Gemini API key.", None, gr.update(visible=False)
+    
+    if not topic:
+        return "Please enter a research topic.", None, gr.update(visible=False)
+    
+    try:
+        # Perform research
+        research_data = perform_research(topic)
+        
+        if not research_data['sources']:
+            return "No relevant sources found. Please try a different search term.", None, gr.update(visible=False)
+        
+        # Generate report
+        report = generate_research_report(research_data, gemini_api_key)
+        
+        # Create a downloadable file
+        filename = f"{topic.replace(' ', '_')}_report.md"
+        
+        return report, filename, gr.update(visible=True)
+        
+    except Exception as e:
+        error_msg = f"An error occurred: {str(e)}"
+        return error_msg, None, gr.update(visible=False)
+
+# Gradio interface
+def create_interface():
+    with gr.Blocks(title="Gemini Deep Research Agent", theme=gr.themes.Soft()) as demo:
+        gr.Markdown("# 📘 Gemini Deep Research Agent")
+        gr.Markdown("This agent performs deep research on any topic using Google's Gemini and DuckDuckGo search")
+        
+        with gr.Row():
+            with gr.Column(scale=1):
+                gr.Markdown("## API Configuration")
+                gemini_key = gr.Textbox(
+                    label="Gemini API Key",
+                    type="password",
+                    placeholder="Enter your Gemini API key (get it from https://aistudio.google.com/)"
+                )
+                
+            with gr.Column(scale=2):
+                research_topic = gr.Textbox(
+                    label="Research Topic",
+                    placeholder="e.g., Latest developments in AI",
+                    lines=2
+                )
+                
+                research_btn = gr.Button("Start Research", variant="primary")
+                
+                output = gr.Markdown(
+                    label="Research Report",
+                    value="Your research report will appear here..."
+                )
+                
+                download_btn = gr.DownloadButton(
+                    "Download Report",
+                    visible=False
+                )
+        
+        # Set up the button action
+        research_btn.click(
+            fn=run_research,
+            inputs=[research_topic, gemini_key],
+            outputs=[output, download_btn, download_btn]
+        )
+        
+        # Set up download functionality
+        def create_file(content):
+            return content
+        
+        download_btn.click(
+            fn=create_file,
+            inputs=[output],
+            outputs=[download_btn]
+        )
+    
+    return demo
+
+# Main execution
+if __name__ == "__main__":
+    demo = create_interface()
     demo.launch()