Update app.py

app.py

@@ -1,8 +1,8 @@
 import streamlit as st
 import anthropic, openai, base64, cv2, glob, json, math, os, pytz, random, re, requests, time, zipfile
-from datetime import datetime 
+from datetime import datetime
 from audio_recorder_streamlit import audio_recorder
-from collections import defaultdict
+from collections import defaultdict, deque, Counter
 from dotenv import load_dotenv
 from gradio_client import Client
 from huggingface_hub import InferenceClient
@@ -12,14 +12,65 @@ from streamlit_marquee import streamlit_marquee
 import asyncio
 import edge_tts
 
-# App Config
-st.set_page_config(page_title="🚲TalkingAIResearcher🏆", page_icon="🚲🏆", layout="wide")
+# Core Configuration
+st.set_page_config(
+    page_title="🚲TalkingAIResearcher🏆",
+    page_icon="🚲🏆",
+    layout="wide",
+    initial_sidebar_state="auto",
+    menu_items={
+        'Get Help': 'https://huggingface.co/awacke1',
+        'Report a bug': 'https://huggingface.co/spaces/awacke1',
+        'About': "🚲TalkingAIResearcher🏆"
+    }
+)
+
+# Custom CSS
+st.markdown("""
+<style>
+    .main { background: linear-gradient(to right, #1a1a1a, #2d2d2d); color: #fff; }
+    .stMarkdown { font-family: 'Helvetica Neue', sans-serif; }
+    .stButton>button { margin-right: 0.5rem; }
+</style>
+""", unsafe_allow_html=True)
+
+# Constants
+EDGE_TTS_VOICES = [
+    "en-US-AriaNeural",  # Default voice
+    "en-US-GuyNeural", 
+    "en-US-JennyNeural",
+    "en-GB-SoniaNeural",
+    "en-GB-RyanNeural",
+    "en-AU-NatashaNeural",
+    "en-AU-WilliamNeural",
+    "en-CA-ClaraNeural",
+    "en-CA-LiamNeural"
+]
+
+FILE_EMOJIS = {
+    "md": "📝",
+    "mp3": "🎵", 
+    "wav": "🔊",
+    "txt": "📄",
+    "pdf": "📑",
+    "html": "🌐"
+}
+
+# Load environment variables
 load_dotenv()
 
-EDGE_TTS_VOICES = ["en-US-AriaNeural", "en-US-GuyNeural", "en-US-JennyNeural", "en-GB-SoniaNeural"]
-FILE_EMOJIS = {"md": "📝", "mp3": "🎵", "wav": "🔊", "txt": "📄", "pdf": "📑", "html": "🌐"}
+# API Setup
+openai_api_key = os.getenv('OPENAI_API_KEY', "")
+anthropic_key = os.getenv('ANTHROPIC_API_KEY', "")
+if 'OPENAI_API_KEY' in st.secrets:
+    openai_api_key = st.secrets['OPENAI_API_KEY']
+if 'ANTHROPIC_API_KEY' in st.secrets:
+    anthropic_key = st.secrets["ANTHROPIC_API_KEY"]
+
+openai_client = OpenAI(api_key=openai_api_key)
+claude_client = anthropic.Anthropic(api_key=anthropic_key)
 
-# Initialize session state
+# Initialize Session State
 state_vars = {
     'tts_voice': EDGE_TTS_VOICES[0],
     'audio_format': 'mp3',
@@ -39,10 +90,7 @@ for key, default in state_vars.items():
     if key not in st.session_state:
         st.session_state[key] = default
 
-# API clients setup
-openai_client = OpenAI(api_key=os.getenv('OPENAI_API_KEY'))
-claude_client = anthropic.Anthropic(api_key=os.getenv('ANTHROPIC_API_KEY'))
-
+# Core Functions
 @st.cache_resource
 def get_cached_audio_b64(file_path):
     with open(file_path, "rb") as f:
@@ -52,6 +100,54 @@ def beautify_filename(filename):
     name = os.path.splitext(filename)[0]
     return name.replace('_', ' ').replace('.', ' ')
 
+def display_marquee_controls():
+    st.sidebar.markdown("### 🎯 Marquee Settings")
+    cols = st.sidebar.columns(2)
+    with cols[0]:
+        bg_color = st.color_picker("🎨 Background", "#1E1E1E")
+        text_color = st.color_picker("✍️ Text", "#FFFFFF")
+    with cols[1]:
+        font_size = st.slider("📏 Size", 10, 24, 14)
+        duration = st.slider("⏱️ Speed", 1, 20, 10)
+    
+    return {
+        "background": bg_color,
+        "color": text_color,
+        "font-size": f"{font_size}px",
+        "animationDuration": f"{duration}s",
+        "width": "100%",
+        "lineHeight": "35px"
+    }
+
+def get_high_info_terms(text: str, top_n=10) -> list:
+    stop_words = set(['the', 'a', 'an', 'and', 'or', 'but', 'in', 'on', 'at', 'to', 'for'])
+    words = re.findall(r'\b\w+(?:-\w+)*\b', text.lower())
+    bi_grams = [' '.join(pair) for pair in zip(words, words[1:])]
+    combined = words + bi_grams
+    filtered = [term for term in combined if term not in stop_words and len(term) > 3]
+    return Counter(filtered).most_common(top_n)
+
+def generate_filename(prompt, response, file_type="md"):
+    prefix = datetime.now().strftime("%y%m_%H%M") + "_"
+    combined = (prompt + " " + response).strip()
+    name_parts = [term for term, _ in get_high_info_terms(combined, top_n=5)]
+    filename = prefix + "_".join(name_parts)[:150] + "." + file_type
+    return filename
+
+def create_file(prompt, response, file_type="md"):
+    filename = generate_filename(prompt.strip(), response.strip(), file_type)
+    with open(filename, 'w', encoding='utf-8') as f:
+        f.write(f"{prompt}\n\n{response}")
+    return filename
+
+def get_download_link(file_path, file_type="zip"):
+    with open(file_path, "rb") as f:
+        b64 = base64.b64encode(f.read()).decode()
+    ext_map = {'zip': '📦', 'mp3': '🎵', 'wav': '🔊', 'md': '📝'}
+    emoji = ext_map.get(file_type, '')
+    return f'<a href="data:application/{file_type};base64,{b64}" download="{os.path.basename(file_path)}">{emoji} Download {os.path.basename(file_path)}</a>'
+
+# Audio Processing
 def clean_speech_text(text):
     text = re.sub(r'\s+', ' ', text.strip())
     text = text.replace("</s>", "").replace("#", "")
@@ -79,39 +175,8 @@ def process_audio_file(audio_path):
         st.session_state.messages.append({"role": "user", "content": text})
         return text
 
-def process_with_gpt(text):
-    if not text: return
-    st.session_state.messages.append({"role": "user", "content": text})
-    with st.chat_message("user"): st.markdown(text)
-    with st.chat_message("assistant"):
-        response = openai_client.chat.completions.create(
-            model="gpt-4-turbo-preview",
-            messages=st.session_state.messages,
-            stream=False
-        )
-        answer = response.choices[0].message.content
-        st.write(f"GPT-4: {answer}")
-        create_file(text, answer, "md")
-        st.session_state.messages.append({"role": "assistant", "content": answer})
-        return answer
-
-def process_with_claude(text):
-    if not text: return
-    with st.chat_message("user"): st.markdown(text)
-    with st.chat_message("assistant"):
-        response = claude_client.messages.create(
-            model="claude-3-sonnet-20240229",
-            max_tokens=4000,
-            messages=[{"role": "user", "content": text}]
-        )
-        answer = response.content[0].text
-        st.write(f"Claude-3: {answer}")
-        create_file(text, answer, "md")
-        st.session_state.chat_history.append({"user": text, "claude": answer})
-        return answer
-
+# File Management
 def load_files_for_sidebar():
-    """Load and filter files by timestamp prefix"""
     files = []
     for f in glob.glob("*.*"):
         basename = os.path.basename(f)
@@ -189,6 +254,7 @@ def display_file_manager_sidebar(groups_sorted):
                             height=0
                         )
 
+# ArXiv Integration
 def perform_arxiv_search(query):
     client = Client("awacke1/Arxiv-Paper-Search-And-QA-RAG-Pattern")
     papers = client.predict(
@@ -260,12 +326,57 @@ def create_paper_audio(papers, query):
             st.write("### 📢 Summary")
             st.audio(summary_audio)
 
+def display_papers(papers):
+    st.write("## Research Papers")
+    for i, paper in enumerate(papers[:20], 1):
+        with st.expander(f"{i}. 📄 {paper['title']}", expanded=True):
+            st.markdown(f"**{paper['date']} | {paper['title']} | ⬇️**")
+            st.markdown(f"*{paper['authors']}*")
+            st.markdown(paper[previous code] ... st.markdown(paper['summary'])
+            if paper.get('audio'):
+                st.write("📚 Paper Audio")
+                st.audio(paper['audio'])
+
+def process_with_gpt(text):
+    if not text: return
+    st.session_state.messages.append({"role": "user", "content": text})
+    with st.chat_message("user"): 
+        st.markdown(text)
+    with st.chat_message("assistant"):
+        response = openai_client.chat.completions.create(
+            model="gpt-4-turbo-preview",
+            messages=st.session_state.messages,
+            stream=False
+        )
+        answer = response.choices[0].message.content
+        st.write(f"GPT-4: {answer}")
+        create_file(text, answer, "md")
+        st.session_state.messages.append({"role": "assistant", "content": answer})
+        return answer
+
+def process_with_claude(text):
+    if not text: return
+    with st.chat_message("user"):
+        st.markdown(text)
+    with st.chat_message("assistant"):
+        response = claude_client.messages.create(
+            model="claude-3-sonnet-20240229",
+            max_tokens=4000,
+            messages=[{"role": "user", "content": text}]
+        )
+        answer = response.content[0].text
+        st.write(f"Claude-3: {answer}")
+        create_file(text, answer, "md")
+        st.session_state.chat_history.append({"user": text, "claude": answer})
+        return answer
+
+# Main App
 def main():
-    st.sidebar.markdown("### 🚲BikeAI🏆 Research Assistant")
+    st.sidebar.title("🚲 Research Assistant")
     
     # Voice settings
-    st.sidebar.markdown("### 🎤 Voice Config")
-    voice = st.sidebar.selectbox("Voice:", EDGE_TTS_VOICES, 
+    st.sidebar.markdown("### 🎤 Voice Config") 
+    voice = st.sidebar.selectbox("Voice:", EDGE_TTS_VOICES,
                                index=EDGE_TTS_VOICES.index(st.session_state['tts_voice']))
     fmt = st.sidebar.radio("Format:", ["MP3", "WAV"], index=0)
     
@@ -287,7 +398,7 @@ def main():
         text = st.text_area("Message:", height=100).strip()
         if st.button("Send"):
             process_with_gpt(text)
-        
+            
         st.subheader("History")
         tab1, tab2 = st.tabs(["Claude", "GPT-4"])
         with tab1:
@@ -309,6 +420,7 @@ def main():
                 st.success("Saved!")
                 st.session_state.should_rerun = True
 
+    # File management
     groups = load_files_for_sidebar()
     display_file_manager_sidebar(groups)
     
@@ -316,19 +428,19 @@ def main():
         st.session_state.should_rerun = False
         st.rerun()
 
-sidebar_md = """# 📚 Research Papers
+sidebar_md = """# 📚 Research
 ## AGI Levels
 L0 ❌ No AI
 L1 🌱 ChatGPT [2303.08774](https://arxiv.org/abs/2303.08774) | [PDF](https://arxiv.org/pdf/2303.08774.pdf)
 L2 💪 Watson [2201.11903](https://arxiv.org/abs/2201.11903) | [PDF](https://arxiv.org/pdf/2201.11903.pdf)
 L3 🎯 DALL·E [2204.06125](https://arxiv.org/abs/2204.06125) | [PDF](https://arxiv.org/pdf/2204.06125.pdf)
 L4 🏆 AlphaGo [1712.01815](https://arxiv.org/abs/1712.01815) | [PDF](https://arxiv.org/pdf/1712.01815.pdf)
-L5 🚀 AlphaFold [L5 🚀 AlphaFold [2203.15556](https://arxiv.org/abs/2203.15556) | [PDF](https://arxiv.org/pdf/2203.15556.pdf)
+L5 🚀 AlphaFold [2203.15556](https://arxiv.org/abs/2203.15556) | [PDF](https://arxiv.org/pdf/2203.15556.pdf)
 
-## 🧬 AlphaFold2 
+## 🧬 AlphaFold2
 [2203.15556](https://arxiv.org/abs/2203.15556) | [PDF](https://arxiv.org/pdf/2203.15556.pdf)
-1. 🧬 Input → 2. 🔍 Search → 3. 🧩 MSA
-4. 📑 Templates → 5. 🔄 Evoformer → 6. 🧱 Structure 
+1. 🧬 Input → 2. 🔍 Search → 3. 🧩 MSA  
+4. 📑 Templates → 5. 🔄 Evoformer → 6. 🧱 Structure
 7. 🎯 3D Predict → 8. ♻️ Recycle"""
 
 st.sidebar.markdown(sidebar_md)